Conformational and crystal energetics of a polymorphic cyclized product of Napafenac: The Z′ and crystal stability correlation

Article abstract of DOI:10.1016/j.molstruc.2013.10.061

We report the single crystal diffraction study of a dimorphic 7-benzoyl-1,3-dihydroindol-2-one system and evaluate its stability relationships through thermal, grinding and slurry methods. Computational methods are invoked to understand the stability relationships. The form I crystallizes in the triclinic space group P1ˉ with two molecules in the asymmetric unit (Z′ = 2), whereas, form II crystallizes in the monoclinic space group P2₁/c with a single molecule in the asymmetric unit (Z′ = 1). The molecules exhibit subtle conformational variations along the bond connecting the phenyl and indole rings and adopt different crystal packing. While both polymorphs show similar amide dimer of NH?O interactions, they do differ significantly in the way the amide dimers are packed. The Differential Scanning Calorimetry (DSC), phase transformation studies and lattice energy calculations clearly established the greater stability of high Z′ form over low Z′ form, which contradicts the widely accepted notion that Z′ > 1 structures are usually kinetic (or metastable). The choice of two different orientations of phenyl rings in form I enforces shorter π?π interactions and additional CH?π contacts to result in the extra stabilization energy over form II crystal packing with a single conformer. The Z′ and stability correlation is established in 83 reported polymorphic systems which indicate that the number of stable crystal structures decrease as the Z′ value increases. However, the polymorphic systems with Z′ = 1 and 2 combinations are more frequently observed than any other combinations.

Full text of DOI:10.1016/j.molstruc.2013.10.061

Journal of Molecular Structure xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Conformational and crystal energetics of a polymorphic cyclized product
of Napafenac: The Z0 and crystal stability correlation
a,
a
b
Jagadeesh Babu Nanubolu a, , Krishnan Ravikumar , Balasubramanian Sridhar , Bojja Sreedhar
⇑
⇑
a Centre for X-ray Crystallography, CSIR-Indian Institute of Chemical Technology, Hyderabad 500607, India
b Inorganic and Physical Chemistry Division, CSIR-Indian Institute of Chemical Technology, Hyderabad 500607, India
g r a p h i c a l a b s t r a c t
h i g h l i g h t s
A cluster of four molecules showing five strongest intermolecular potentials (kcal molꢂ1) in two poly-
morphs is depicted. Form I with two different conformations exhibits stronger intermolecular potentials
than form II with a single conformation.
ꢀ
Single crystal diffraction analysis and
thermal study of a new polymorphic
system.
ꢀ
ꢀ
The high Z0 form is interestingly more
stable than its low Z0 form.
Conformer energies and
intermolecular potentials in two
polymorphs are discussed.
ꢀ
ꢀ
Collective weak interactions are
stronger than the conventional N–
Hꢁ ꢁ ꢁO bonds.
The Z0 and crystal stability correlation
is established in 83 polymorph sets.
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online xxxx
We report the single crystal diffraction study of a dimorphic 7-benzoyl-1,3-dihydroindol-2-one system
and evaluate its stability relationships through thermal, grinding and slurry methods. Computational
methods are invoked to understand the stability relationships. The form I crystallizes in the triclinic space
0
ꢀ
Keywords:
group P1 with two molecules in the asymmetric unit (Z = 2), whereas, form II crystallizes in the mono-
clinic space group P21/c with a single molecule in the asymmetric unit (Z0 = 1). The molecules exhibit sub-
tle conformational variations along the bond connecting the phenyl and indole rings and adopt different
crystal packing. While both polymorphs show similar amide dimer of NAHꢁ ꢁ ꢁO interactions, they do dif-
fer significantly in the way the amide dimers are packed. The Differential Scanning Calorimetry (DSC),
phase transformation studies and lattice energy calculations clearly established the greater stability of
high Z0 form over low Z0 form, which contradicts the widely accepted notion that Z0 > 1 structures are usu-
ally kinetic (or metastable). The choice of two different orientations of phenyl rings in form I enforces
Polymorph Z0
Multiple molecules
Asymmetric unit
Stable
Metastable
Intermolecular potential
shorter
p
ꢁ ꢁ ꢁ
p
interactions and additional CAHꢁ ꢁ ꢁ
p contacts to result in the extra stabilization energy over
form II crystal packing with a single conformer. The Z0 and stability correlation is established in 83
reported polymorphic systems which indicate that the number of stable crystal structures decrease as
the Z0 value increases. However, the polymorphic systems with Z0 = 1 and 2 combinations are more fre-
quently observed than any other combinations.
Ó 2013 Elsevier B.V. All rights reserved.
⇑
Corresponding authors. Tel.: +91 (0) 4027193118 (J.B. Nanubolu).
E-mail addresses: jagadeesh81@gmail.com (J.B. Nanubolu), sshiya@yahoo.com
(K. Ravikumar).
0022-2860/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2013.10.061
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

2
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
literature, for which the high Z0 form is genuinely more stable than
1. Introduction
low Z0 form [5b,13a–d] or the compounds crystallize exclusively
with Z0 > 1 structures [4c,13e,f].
The Z0 parameter is conventionally used in the crystallography
to denote the number of molecules in the asymmetric unit [1].
Strictly it is defined as the number of formula units in the unit cell
(Z) divided by the number of independent general positions of the
crystal space group [2]. In general, 90% of the molecules crystallize
with a single molecule or less (Z0 6 1), while a small but significant
fraction of the compounds (8.3%) contain more than one molecule
in the asymmetric unit (Z0 > 1) [3]. The incidence of structures with
Z0 > 1 has interestingly remained almost invariant over the two
decades (8.3% in 1990, 8.8% in 2006, 11.5% in 2009) [3,4]. With
the rising interest in organic crystal structure prediction and appli-
cations of organic materials in the pharmaceutical industry, it is
necessary to understand this phenomenon to ascertain if there is
any direct relationship between its occurrence and attributes such
as the molecular structure, chemical functionality, crystal system
and etc. [5]. Few authors have made an attempt in the past and
concluded that Z0 > 1 situation is often observed in low symmetry
triclinic and monoclinic space groups P1, P21/c and Pc [3,6a]. The
chiral molecules without internal symmetry have noticeably high-
er percentage of Z0 > 1 structures [4b]. Further, the precedence of
Z0 > 1 is more for certain class of compounds, such as nucleosides
and nucleotides (20.8%) and steroids (18.8%). This indicates a gen-
eral packing problem of the awkwardly shaped molecules which
often cannot crystallize out in simple packing modes with Z0 = 1
[4b]. Gavezzotti and Fillippini [6b] noted Z0 > 1 cases are more pro-
nounced in the case of hydrogen bonded crystals. Brock and Dun-
can [6a] provided logical reasons to the anomalous space group
frequencies for mono alcohols (ꢃ51%). Nangia [7] suggested that
there may be several low-lying, interconverting conformers in
solution and more than one may crystallize simultaneously be-
cause of the kinetic factors. Craven [8a], Desiraju [8b], Zorky [8c],
Britton [8d], Kuleshova [8e], Gavezzotti [8f], Dunitz and Bernstein
[8g] have emphasized the presence of pseudosymmetry in Z0 > 1
structures.
In an attempt to understand the dependence of polymorph sta-
bility on the number of molecules in the asymmetric unit, we have
analyzed a dimorphic system of 7-benzoyl-1,3-dihydroindol-2-one
in the present article. Form I crystallizes in the triclinic space
0
ꢀ
group, P1 with Z = 2 and form II crystallizes in the monoclinic
space group, P21/c, with Z0 = 1. Polymorphs are characterized by
single crystal X-ray diffraction (SC-XRD), powder X-ray diffraction
(PXRD) and Infrared spectroscopy (IR). The stability relationship is
established by Differential Scanning Calorimetry (DSC) and phase
transformation studies by grinding, slurry and thermal methods.
The conformer and lattice energy calculations are performed for
better understanding of the stability of crystal structures. Finally
a statistical analysis on polymorphic pairs with variable Z0 is car-
ried out to indicate the trends.
2. Experimental section
2.1. Materials and methods
Prior to setting up various crystallizations, napafenac (In-house
sources from IICT) was recrystallized and confirmed to be pure by
1H NMR. Further, the materials were identified as single phase
crystalline form by PXRD.
2.2. Synthesis and crystal growth
100 mg of napafenac was dissolved in 20 mL ethanol solution
under hot conditions. A clear yellow solution was obtained to
which four drops of conc. HCl were added. The yellowish ethanolic
solution turned to colorless within 2–3 min. Napafenac was cy-
clized to 7-benzoyl-1,3-dihydroindol-2-one. Crystals of the cy-
clized product (form I and II) were concomitantly observed upon
slow evaporation of the ethanol solvent.
Steed [2a] in his seminal work has critically reviewed the rea-
sons for Z0 > 1 structures. Their structural origins are linked to
the early events of crystallization such as molecular aggregation
and crystal nucleation. Accordingly, these have been referred as
‘‘fossil relic’’ of the fastest growing crystal nucleus or ‘‘early snap-
shot or kinetic form’’ [2a]. Desiraju [4a] suggested that the secrets
about the high Z0 structures may be best revealed in a polymorphic
system with variable Z0, as the high Z0 form can be directly com-
pared with its low Z0 (typically 1) polymorph. The term ‘‘poly-
morph’’ indicates a solid crystalline phase of a given compound
resulting from the possibility of at least two crystalline arrange-
ments of molecules of that compound, a phenomenon which has
high relevance to the pharmaceutical industry [9]. The polymorph
with higher Z0 structure is suggested to be ‘‘crystal on the way’’
meaning that it represents to a less stable high energy polymorph
(or in other words a kinetic form) in the crystallization pathway to-
wards the final thermodynamic crystal form with the lower Z0 [10].
The kinetic nature of high Z0 structures is seemed to be evident
from the higher proportion of Z0 > 1 structures grown from melt
or sublimation (fast crystallization conditions) as noted by Nangia
[11]. Clegg [12], on similar lines showed that the early stages of
crystal nuclei (poor quality crystals which are formed too rapidly
or incompletely crystallized) determined by synchrotron radiation
have greater proportion of Z0 > 1 (21%) followed by conventional X-
ray determined structures (12%). The slowly formed large crystals
grown for neutron diffraction have the lowest incidence of multi-
ple molecules (5%). While the above studies clearly suggest the
high Z0 structures to be metastable fossil relics or forms of arrested
crystallization, there are however few counter examples in the
2.3. Spectroscopic (NMR, IR and Mass) data
1H NMR of 7-benzoyl-1,3-dihydroindol-2-one (DMSO-d6):
d3.61(2H, s), 6.92 (1H, d), 7.06(1H, t), 7.31(1H, t), 7.46(1H, d),
7.55(2H, t), 7.70(2H, m), 10.35(1H, s). The m/z: 237; IR (KBr):
3268.1, 3047.6, 2910.7, 1733.4, 1711.8, 1647.0, 1595.7, 1475.5,
1447.3, 1383.0, 1333.1, 1304.5, 1252.9, 1206.2, 1171.1, 1070.9,
1036.2, 1013.6 cmꢂ1 1H NMR of napafenac (DMSO-d6): d3.42
.
(2H, s), 6.53(1H, t), 7.07(3H, d), 7.20(1H, dd), 7.25(1H, dd),
7.55(6H, m). IR: 3444.8, 3323.3, 3193.1, 1676.7, 1616.4, 1555.1,
1437.5, 1437.5, 1400.9, 1333.9, 1282.7, 1243.9, 1194.3,
1171.1 cmꢂ1
. The NMR spectrum is recorded on BRUKER
AVANCE-300 NMR spectrometer, FT-IR spectra on Thermo Nicolet
spectrometer, Model Nexus 670, USA.
2.4. Single crystal X-ray diffraction (SC-XRD)
Crystal data were collected at RT using a Bruker Smart Apex
CCD diffractometer with graphite monochromated Mo K
a radia-
tion (k = 0.71073 Å) with -scan method. Preliminary lattice
x
parameters and orientation matrices were obtained from four sets
of frames. Integration and scaling of intensity data was accom-
plished using SAINT program [14a]. The structures were solved
by Direct Methods using SHELXS97 and refinement was carried
out by full-matrix least-squares technique using SHELXL97 [14b].
Anisotropic displacement parameters were included for all non-
hydrogen atoms. All H atoms were positioned geometrically and
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
3
treated as riding on their parent C atoms, with CAH distances of
0.93–0.97 Å, and with Uiso(H) = 1.2Ueq (C) or 1.5Ueq for methyl
atoms. The NAH atoms were located from the difference Fourier
map.
the journals pertaining to crystal engineering and pharmaceutical
research community. For some of the well known polymorphic
systems, the stability data was inferred from the articles other than
CSD links using other search engines. Compounds which have sim-
ilar Z0 values in all of its polymorphs were ignored. A total of 83
polymorphic systems could be identified which were segregated
into two sets based on the thermodynamic stability at RT and their
preference for high or low Z0 value (Table 4).
2.5. Powder X-ray diffraction (PXRD)
The PXRD patterns were recorded at RT using a Bruker diffrac-
tometer with Cu Ka radiation (k = 1.5418 Å), running at 40Kv and
30 mA. The 2-theta range was covered from 2.00 to 50.00 degrees
with a step size of 0.02 degrees and a count time of 1.00 s per each step.
2.9. Phase transformation studies
Polymorph mixtures were prepared by rotary evaporation
method. They were subjected to slurry method by suspending
the material (approximately 100 mg of material) in 5–8 mL of ace-
tone in a 25 mL round bottomed flask. The flasks were sealed and
the material was constantly stirred at RT at 800 rpm. The sus-
pended material was filtered after 1 day and a PXRD was recorded
to monitor changes. The phase transformations were performed by
grinding and heating the polymorph mixtures at 120 °C in two sep-
arate experiments.
2.6. Computations
Conformer energy calculations were performed using the crys-
tallographic coordinates of triclinic and monoclinic conformations
of 7-benzoyl-1,3-dihydroindol-2-one. Since the observed con-
former in the crystal structure is normally different from the gas
phase minimised conformer, a constrained optimization of the
crystal conformer [7a] was carried out by fixing the torsion angles
during the optimization. The density functional theory (DFT) meth-
od was applied at the B3LYP hybrid exchange correlation function
level using the 6-31G(d,p) basis set as implemented in GAUSS-
IAN03 [15a]. X-seed [15b] was used for preparing crystal packing
diagrams. Intermolecular interactions and the lattice energies were
investigated using Unified pair-potentials (UNI) as implemented in
the CCDC Mercury CSD 3.1 [16a,b,7]. The hydrogen atom positions
were normalized to the standard values of neutron diffraction
bond distances. The quality of the lattice energy calculations ob-
tained using UNI force filed parameters is often similar to that of
quantum chemical calculations at a fraction of the computational
cost [16c,d]. This method allowed the calculation of lattice energies
in good agreement with crystal sublimation enthalpies for a wide
selection of organic compounds, and also performs well in energy
ranking for polymorphs of organic crystal structures [16e].
3. Results and discussion
3.1. Synthesis and characterization
Napafenac (2-amino-3-benzoylbenzeneacetamide) is a non-ste-
roidal anti inflammatory pro-drug [18a]. It is sold commercially as
an ophthalmic formulation by Alcon Pharmaceuticals (NEVNAC™)
and is used in the cataract surgery and post cataract recovery stage
for the treatment of pain and inflammation in the eye [18b,c]. The
onset of action is initiated by pro-drug’s penetration into the eye-
cornea followed by the hydrolysis to its chemically active amfenac
(3-amino-3-benzoyl phenyl acetate) which is thought to inhibit
the COX-1 and COX-2 receptor sites to bring about the resultant ac-
tion [18c]. When this pro-drug is attempted to crystallize from eth-
anol by adding few drops of concentrated hydrochloride with the
aim of obtaining the protonation of the amide group in the solid
form of napafenac [18d], it undergoes acid hydrolysis and cyclizes
to produce 7-benzoyl-1,3-dihydroindol-2-one. The chemical reac-
tion is very rapid, indicated by a distinct color change in solution
from yellow to colourless within first two minutes of adding acid
drops. The product is extracted as crystalline material after slow
evaporation of the ethanol. Incidentally the reaction product is also
one of the metabolites observed in patients after the treatment
with amfenac sodium [18e,f]. The chemical reaction is represented
in Scheme 1 and its suggested mechanism is provided in the sup-
porting information (Fig. S1 of ESI). The chemical reaction is
thought to proceed via the protonation of primary amide group
of napafenac and goes one step beyond by the nucleophillic attack
of adjacent phenyl amine group on the primary amide carbonyl
and finally rearranges to a 5-membered cyclic amide by knocking
out the ammonia. The chemical reaction is also observed under
very weak acetic acid conditions.
2.7. Thermal analysis
Differential Scanning Calorimetry (DSC) measurements were
made on a Mettler Toledo DSC 821 instrument using aluminum
sealed pans in a nitrogen atmosphere. The scans were recorded be-
tween 25 and 180 °C at a constant heating rate of 5 °C/min. The en-
thalpy of fusion values (DHfus) was estimated by drawing a straight
line connecting the baseline before and after the melting peak, and
integrating the area under the peak. Heat-cool-heat experiments
were carried out on triclinic form I material.
2.8. CSD analysis
A search of the Cambridge Structural Database (Version 5.33,
November 2012 release with May 2013 update) [17] with search
key words ‘‘polymorph’’ or ‘‘form’’ was undertaken. All structures
with 3D atomic co-ordinates determined, no errors, no polymeric
and only organics were the search filter criteria, resulting 9305
hits. A further search criteria were applied by restricting the struc-
tures to only Z0 > 1, which resulted in 1708 structures. All transition
metals and metals were removed leaving 1662 structures for final
analysis. Since the stability data of polymorphic systems was not
available in the CSD, the original articles had to be referred manu-
ally. The six letter code of each compound, REFCODE, was given as
a separate search query in the database and all of its structures
including low Z0 were extracted, followed by linking the refcodes
to the original articles through the URL links given under author/
journal information. The articles were searched for correlation be-
tween polymorph stability at RT and the number of molecules in
the asymmetric unit. The stability data were reported mostly in
Interpreting the IR spectra of napafenac indicates two distinct
NAH stretching bands at 3445 and 3323 cm–1 which correspond
Scheme 1. Chemical reaction of napafenac to the cyclized product under acidic
conditions.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

4
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
to stretching vibrational modes of aniline (Ph-NH2) and amide
(CONH2) groups, respectively. However, these bands appear at
3268 and 3200 cm–1 in the cyclized product. The C@O stretching
in napafenac at 1676.7 cmꢂ1 shifts to 1733.7 in cmꢂ1 in the cy-
clized product. This blue shift is attributed to the result of an inter-
nal strain in the formation five membered lactam ring. In the NMR
spectrum, the appearance of a broad peak at d10.35 ppm for cyclic
amide NH peak is the most important observation to distinguish
the product from the napafenac. Other spectral changes are sum-
marized in the experimental section. The mass spectrum provides
the intense m/z peak at 237 which nicely matches with the molec-
ular weight of the proposed structure. The spectra are provided in
the ESI (Figs. S2–S3 of ESI).
3.2. Polymorphs – crystal structure analysis
A single crystal X-ray diffraction is performed on the light pink
colored product crystals obtained from ethanol solvent. The mole-
cule crystallizes in the triclinic space group P1. An ORTEP picture is
Fig. 2. Simulated XRD from triclinic form is compared with the PXRD of the bulk
material. Majority of the peaks in the bulk material nicely match those of triclinic.
The unaccounted peaks are highlighted with asterisks which nicely match with the
simulated XRD pattern from monoclinic form II. The 2h positions nicely match for
bulk and simulated patterns, however the subtle intensity differences arise from
preferred orientation effects, especially the 2h peak at ꢃ31°.
ꢀ
provided in the Fig. 1a depicting two molecules in the asymmetric
unit. The Z0 = 2 for the title molecule prompted us to look for an-
other polymorphic form, considering that the high Z0 structures
are usually kinetic and are referred as ‘‘crystal on the way’’ to
low Z0 structure [10]. The bulk material of the cyclized product is
subjected for a routine powder X-ray diffraction analysis to estab-
lish the identity with the single crystal. A comparison of the simu-
lated powder pattern from triclinic structure with the bulk PXRD is
provided in Fig. 2. While majority of the peaks of the bulk material
nicely match with the simulated lines of the triclinic structure,
several minor peaks do not (marked as asterisks in Fig. 2). The
presence of these additional lines suggested that there could be
certainly another form in the bulk material.
In pursuit of unraveling the identity of the second form, another
crystallization batch was set up by employing similar crystalliza-
tion conditions as before. The grown crystals after three days were
carefully subjected to unit cell checks on single crystal diffractom-
eter. While most of the larger needle crystals gave unit cell dimen-
sions similar to triclinic, a few fine slender needles which normally
appear as bunch and form nearer to the corner of the beaker how-
ever gave a unit cell in the monoclinic crystal system. The crystal
structure refinement on the monoclinic unit cell confirmed that
this was in fact a second polymorph of the title molecule. Interest-
ingly, this polymorph contained a single molecule in the asymmet-
ric unit (Z0 = 1). Crystal data are provided in Table 1 and the ORTEP
picture is provided in Fig. 1b. The unaccounted reflections in the
initial bulk powder pattern are clearly matched by the simulated
lines from monoclinic crystal structure as shown in Fig. 2. It can
be seen that the both forms are obtained concomitantly with form
I as the major and form II as the minor quantity in the bulk mate-
rial. The triclinic polymorph which appeared first is designated as
form I and the monoclinic polymorph is designated as form II.
From the structural point of view, the title molecule consists of
two ring fragments (indole and phenyl ring) which are bridged by a
carbonyl group. It exhibits torsional freedom along the bond con-
necting the two rings defined by s1 (C9AC7AC8AC1) and s2
(C8AC7AC1AC6). An overlay of conformations from form I [labeled
as T1 and T2] and form II (labeled as M) is shown in Fig. 3. For three
conformations, the s1 angles are 72.8(2)°, 49.2(2)° and 59.2(2)° and
the s2 angles are 0.8(2)°, 15.6(2)° and 15.0(2)°, respectively. The
preference of an intramolecular NAHꢁ ꢁ ꢁO hydrogen bond possibly
restricts the s2 torsion angle window to be in the narrow range
but s1 can vary significantly. The Density Function Theory (DFT)
calculations performed in Gaussian03 estimated the relative con-
former energies to be 0 kcal mol–1 for T2, 1.631 kcal molꢂ1 for T1
and 0.798 kcal mol–1 for M. The least s1 angle conformer is shown
to be the lowest in energy (stable) and the highest s1 angle con-
former is higher in energy (metastable).
The crystal structures analysis reveals that both polymorphs
contain a similar amide dimer synthon, sustained by two symme-
try independent molecules in form I and inversion related pair in
form II (Figs. 4a and b, Tables 2 and 3). This dimer involves two
Fig. 1. ORTEP diagrams with atom-labeling. (a) Form I. (b) Form II. The thermal
ellipsoids are drawn at 30% probability. Hydrogen atoms are shown as spheres of
arbitrary radii.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
5
Table 1
Crystal data of two polymorphs.
Compound reference
Form I
Form II
Chemical formula
Formula mass
C
15H11NO2
C15H11NO2
237.25
237.25
Crystal system
Triclinic
8.3262(12)
9.7741(15)
14.849(2)
96.341(3)
97.050(4)
105.904(3)
1140.2(3)
294(2)
Monoclinic
8.5639(5)
5.2265(3)
25.7471(14)
90.00
94.1400(10)
90.00
1149.41(11)
294(2)
a/Å
b/Å
c/Å
a
/°
b/°
/°
c
Unit cell volume/Å3
Temperature/K
ꢀ
Space group
P21/c
P1
No. of formula units per unit cell, Z
No. of formula units per asymmetric
unit, Z0
4
2
4
1
Density (Mg/cm3)
1.382
1.371
Radiation type
Mo K
a
Mo Ka
No. of reflections measured
No. of independent reflections
Rint
11044
3999
0.0216
10370
2020
0.0189
Final R1 values (I > 2
r
(I))
r
0.0406
0.1055
0.0489
0.1115
0.0344
0.0908
0.0367
0.0928
Final wR(F2) values (I > 2
Final R1 values (all data)
(I))
Final wR(F2) values (all data)
Goodness of fit on F2
1.051
1.077
Largest diff. peak and hole/e Å3
0.197 and
ꢂ0.147
2.19 and 25.00
964907
0.163 and
ꢂ0.129
2.38 and 25.00
964908
h range
CCDC number
Fig. 3. Overlay of three conformations from two crystal structures. Form
I
conformations are labeled as T1 and T2 and form II conformation is labeled as M.
A schematic diagram of the title molecule is shown on the left. Overlay is made with
the phenyl ring of the Indole portion. The RMS deviation is 0.0137 Å for T1 and T2
and 0.00938 Å for T1 and M.
NAH donors and two C@O acceptors forming an eight membered
ring via two NAHꢁ ꢁ ꢁO hydrogen bonds [N1AH1 Nꢁ ꢁ ꢁO4 (2.09 Å,
2.939(2) Å, 160°) and N2AH2 Nꢁ ꢁ ꢁO2 (2.18 Å, 3.028(2) Å, 158°) in
form I; N1AH1 Nꢁ ꢁ ꢁO2 (2.06 Å, 2.910(2) Å, 159°) in form II]. The di-
mers are extended into identical one dimensional ribbons via
CAHꢁ ꢁ ꢁO interactions in both the forms [C11AH11ꢁ ꢁ ꢁO3 (2.75 Å,
3.663(2) Å, 167°) and C26AH26ꢁ ꢁ ꢁO1 (2.55 Å, 3.419(2) Å, 155°) in
form I; C11AH11ꢁ ꢁ ꢁO1, 2.68 Å, 3.605(2) Å, 173° in form II]. Addi-
Fig. 4. Two-dimensional intra-layer interactions are shown for polymorphs (a)
Form I. (b) Form II. The amide dimers are formed by two symmetry independent
molecules in form I (shown as ball and stick and capped stick models) and a single
molecule in form II. A one dimensional ribbon is formed through the interplay of
tionally, a weak C25AH25ꢁ ꢁ ꢁ
p interaction (3.05 Å, 3.804(2) Å,
139°) is observed in the 1D ribbon of form I (Fig. 4a). The close
packing of the phenyl rings from adjacent 1D ribbons leads to a
planar 2D layered structure in form I and a zig-zag (wave-like)
structure in form II (Fig. 4c and d). Such layers are stacked along
the third dimension. Form I shows three types of interactions in
between the layers [C6AH6ꢁ ꢁ ꢁO3, 2.67 Å, 3.498(2)Å, 149ꢁ;
CAHꢁ ꢁ ꢁO and CAHꢁ ꢁ ꢁp in form I and only CAHꢁ ꢁ ꢁO in form II. The two dimensional
sheet is observed from the close packing of the hydrophobic phenyl rings. The three
dimensional inter-layer interactions are shown in polymorphs (c) Form I. (d) Form
II. Form I is sustained by a combination of CAHꢁ ꢁ ꢁO and CAHꢁ ꢁ ꢁ
p interaction,
whereas a CAHꢁ ꢁ ꢁO connects these layers in form II. The 2D sheet is planar in form I,
whereas it is a zig-zag or wave like in form II. Please refer to Tables 2 and 3 for
symmetry codes of the hydrogen bonds.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

6
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
Table 2
List of intra- and inter-molecular interactions (NAHꢁ ꢁ ꢁO, CAHꢁ ꢁ ꢁO, CAHꢁ ꢁ ꢁ
p, pꢁ ꢁ ꢁp and dipolar contacts) in form I.
\DAH ꢁ ꢁ ꢁ A=ꢄ
S. No
Interactions
DAH /Å
Hꢁ ꢁ ꢁA /Å
Dꢁ ꢁ ꢁA/Å
Dimension
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
N1AH1Nꢁ ꢁ ꢁO1i
N2AH2Nꢁ ꢁ ꢁO3i
N1AH1N...O4i
N2AH2Nꢁ ꢁ ꢁO2i
C11AH11ꢁ ꢁ ꢁO3ii
C26AH26ꢁ ꢁ ꢁO1iii
C25AH25ꢁ ꢁ ꢁCg2iii
C6AH6ꢁ ꢁ ꢁO3iv
C2AH2ꢁ ꢁ ꢁCg6v
C30AH30Bꢁ ꢁ ꢁO2v
C15AH15Aꢁ ꢁ ꢁO4iv
Cg1ꢁ ꢁ ꢁCg7v
0.89
0.89
0.89
0.90
0.93
0.93
0.93
0.93
0.97
0.97
0.97
–
–
–
–
–
–
–
–
0.97
–
2.33
2.34
2.09
2.18
2.75
2.55
3.05
2.67
2.83
2.79
2.81
–
–
–
–
–
–
–
–
2.89
–
2.812(2)
2.813(2)
2.939(2)
3.028(2)
3.663(2)
3.419(2)
3.804(2)
3.498(2)
3.595(2)
3.464(2)
3.474(2)
3.741(2)
3.623(2)
3.933(2)
3.546(2)
3.799(2)
3.417(2)
3.517(2)
3.390(2)
3.702(2)
3.448(2)
3.362(2)
3.552(2)
3.587(2)
3.407(2)
3.516(2)
114
112
160
158
167
155
139
149
140
127
127
–
–
–
–
–
–
–
–
142
–
Intramolecular
Intramolecular
Amide dimer
Amide dimer
1D ribbon
1D ribbon
1D ribbon
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T1–T1)
3D, Type II (T2–T2)
3D, Type II (T2–T2)
3D, Type II (T2–T2)
3D, Type II (T2–T2)
3D, Type II (T2–T2)
3D, Type II (T2–T2)
C14AO2ꢁ ꢁ ꢁCg5v
Cg1ꢁ ꢁ ꢁCg1iv
(d+)C7ꢁ ꢁ ꢁO2(dꢂ)iv
(d+)C14ꢁ ꢁ ꢁO1(d–)iv
C14ꢁ ꢁ ꢁN1iv
C13ꢁ ꢁ ꢁO2iv
C8ꢁ ꢁ ꢁO2iv
C15AH15Aꢁ ꢁ ꢁO1iv
(d+)C22ꢁ ꢁ ꢁO4(dꢂ)v
(d+)C29ꢁ ꢁ ꢁO3(dꢂ)v
C23ꢁ ꢁ ꢁO4v
–
–
–
–
–
–
–
–
–
–
–
–
C28ꢁ ꢁ ꢁO4v
C29ꢁ ꢁ ꢁN2v
C30AH30Bꢁ ꢁ ꢁO3v
0.97
2.84
128
ii
iv
v
ix, y, z; ꢂ1ꢂx, ꢂy, ꢂz; iii1ꢂx, 1ꢂy, ꢂz;
ꢂx, ꢂy, ꢂz; ꢂx, 1ꢂy, ꢂz; Cg = centroid; Cg1 = N1AC13AC12AC14AC15; Cg2 = C1AC2AC3AC4AC5AC6;
Cg5 = N2AC28AC27AC30AC29; Cg6 = C16AC17AC18AC19AC20AC21; Cg7 = C23AC24AC25AC26AC27AC28.
Table 3
List of intra- and inter-molecular interactions (NAH. . .O, CAH. . .O,
pꢁ ꢁ ꢁp and dipolar contacts) in form II.
\DAH ꢁ ꢁ ꢁ A=ꢄ
S. No
Interactions
DAH/Å
Hꢁ ꢁ ꢁA /Å
Dꢁ ꢁ ꢁA/Å
Dimension
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
N1AH1 Nꢁ ꢁ ꢁO1i
N1AH1 Nꢁ ꢁ ꢁO2ii
C11AH11ꢁ ꢁ ꢁO1iii
Cg1ꢁ ꢁ ꢁCg3v
0.89
0.89
0.93
–
–
–
–
–
–
–
–
0.93
–
–
–
2.37
2.06
2.68
–
–
–
–
–
–
–
–
2.59
–
–
–
2.849(2)
2.910(2)
3.605(2)
3.882(2)
3.490(2)
3.493(2)
3.720(2)
3.443(1)
3.631(2)
3.600(2)
3.529(2)
3.447(2)
3.486(2)
3.526(2)
3.130(2)
114
159
173
–
–
–
–
–
–
–
–
153
–
–
–
Intramolecular
Amide dimer
1D ribbon
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type I stack
3D, Type II (M–M)
3D, Type II (M–M)
3D, Type II (M–M)
3D, Type II (M–M)
C14AO2ꢁ ꢁ ꢁCg3vi
O2AC14ꢁ ꢁ ꢁCg3vi
C14AO2ꢁ ꢁ ꢁCg3vi
C13ꢁ ꢁ ꢁO2vi
vi
C8ꢁ ꢁ ꢁO2
C12ꢁ ꢁ ꢁO2vi
vi
N1ꢁ ꢁ ꢁC9
C2AH2ꢁ ꢁ ꢁO2iv
C7AO1ꢁ ꢁ ꢁCg1iv
(d+)C7ꢁ ꢁ ꢁO2(dꢂ)iv
(d+)C14ꢁ ꢁ ꢁO1(dꢂ)iv
iii
ix, y, z; ii1ꢂx, ꢂ1ꢂy, 1ꢂz;
ꢂ1 + x, y, z; iv1ꢂx, ꢂy, 1ꢂz; vx, ꢂ1/2ꢂy, 1/2 + z; vix, ꢂ1 + y, z; Cg = centroid; Cg1@N1AC13AC12AC15AC14; Cg2@C1AC2AC3AC4AC5AC6;
Cg3@C8AC9AC10AC11AC12AC13.
C2AH2ꢁ ꢁ ꢁ
and 3], whereas, form II shows only two interactions
stacking, 3.882(2) Å,
Tables 2 and 3]. The stacking interactions are slightly shorter in
form I. Polymorphism arises from the differences in conformations
[7] and crystal packing [9b]. The role of non-covalent interactions
and their contribution to crystal stability will be discussed later.
p
, 2.83Å, 3.595(2)Å, 140ꢁ;
p
ꢁ ꢁ ꢁ
p
stacking, 3.7412Å; Tables
were recorded from 25 to 200 °C at a constant heating rate of
5 °C/min. The form I showed a single melting endotherm at
156.4 °C. The melting began at 155.2 °C and was completed by
159.3 °C (Fig. 5a). The DSC scan on form II crystals showed a
non-linear background as the temperature ramped up, presumably
due to the non-uniform contact of very limited sample of form II
crystals (ꢃ1 mg). Nevertheless it indicated very interesting results
(Fig. S4 of ESI). It contained two endotherm events, the first one ap-
peared at 118.6 °C followed by a second one at 156.2 °C. It ap-
peared that the first endotherm corresponded to the solid state
phase transformation of form II to I and the second endotherm to
the melting of the transformed form I. Attempts to obtain the
metastable form II in its pure state were unsuccessful and resulted
always as polymorph mixture (Please see ESI for the detailed dis-
cussion). A DSC scan on the readily obtained polymorphic mixture
contained two well defined endotherms (Fig. 5b), whereas, the
pure form I material contained only a single endotherm as
2
[C2AH2ꢁ ꢁ ꢁO2, 2.59 Å, 3.447(2) Å, 153ꢁ;
p p
ꢁ ꢁ ꢁ
3.3. Polymorph stability
To understand the stability of two crystal structures, we con-
ducted Differential Scanning Calorimetry (DSC) studies. Crystal
morphological differences allowed the concomitant mixture of
the two polymorphs to be carefully separated under the optical
microscope into pure Form I and II crystals. To ensure that the cor-
rect form was used in the DSC experiment, unit cells were checked
on few representative crystals from each group and DSC scans
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
7
Table 4
The high Z0 and stability correlation in 83 polymorphic systems. The six letter REFXCODE is given for each polymorphic system.
S.
REFCODE
The order of stability at RT
No
The low Z0 form is more stable than the high Z0 form in 53 systems (64%)
1
2
3
4
5
6
6
7
8
HXACAN
UCAYED
UCAYIH
WOBQEK
IJIBEJ
ETEDOX
FACCOB
FEQDIQ
FIYBEV
INDMET
MOVTIB
ODOBIT
PUQYOQ
TOKSAO
VAMBOA
WETFAD
GOGQOJ
AXOGAW
BICCIZ
EHOWIH
NICOAM
FURACL
KAXXAI
PICAMD
REHGUI
UHAVAB
VUSDIX
DONTIJ
Form I (Z0 = 1) > Form II (Z0 = 1) > Form III (Z0 = 3)
Form 1L (Z0 = 1) > Form 1H (Z0 = 3)
Form 2L (Z’ = 2) > Form 2H (Z0 = 8)
Form I (Z0 = 1, ratio 1:0.5) > Form II (Z0 = 2, ratio 2:1)
Form I (Z0 = 1, ratio 1:1) > Form 2 (Z0 = 2, ratio 2:2)
Form I (Z0 = 1) > Form II (Z0 = 1) > Form III (Z0 = 2)
Form 2a (Z0 = 2) > Form 2b (Z0 = 3)
Form I (Z0 = 1) > Form II (Z0 = 2)
Form II (Z0 = 1) > Form I (Z0 = 2).
9
Gamma (Z0 = 1) > Alpha (Z0 = 3)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Form I (Z0 = 2, ration 2:2) > Form II (Z0 = 8, ratio 8:8)
Form II (Z0 = 1, ratio 1:0.5) > Form I (Z0 = 2, ratio 2:1)
Form IV (Z0 = 1) > Form II (Z0 = 1) > Form III (Z0 = 2) > Form I (Z0 = 1)
Form
a
(Z0 = 1) > Form b (Z0 = 2), Form
(Z0 = 2).
c
(Z0=4) and Form d (Z0 = 2).
Form b (Z0 = 1) > Form
a
Form 1S (Z0 = 1) > Form 1 M (Z0 = 2).
Form II (Z0 = 1) > Form I (Z0 = 4).
Form B (Z0 = 1) > Form A (Z0 = 2)
Form
a
(Z0 = 1) > Form b (Z0 = 2)
Form Iso2 (Z0 = 1) > Form Iso3 (Z0 = 1) > Form Iso1 (Z0 = 2). Metastable Form IV (Z0 = 3), Form V (Z0 = 1) are discovered later.
Form Nic1 (Z0 = 1) > Form Nic2 (Z0 = 4)
Form II (Z0 = 1) > Form I (Z0 = 4).
Forms I (Z0 = 1) > Form II (Z0 = 1) > Forms III (Z0 = 2), Form IV (Z0 = 3), Form V (Z0 = 1)
Form I (Z0 = 1) > Form II (Z0 = 2)
Form II (Z0 = 1) > Form I (Z0 = 2)
Form orthorhombic (Z0 = 1) > Form triclinic (Z0 = 4).
Form XXV (Z0 = 1) > Form IV (Z0 = 2)
Form I (Z0 = 1) > Form II (Z0 = 4).
MELFIT
BEWKUJ
BIPDEJ
CBMZPN
FAFWIS
FPAMCA
Form Xꢁ(Z0 = 1) > Form II (Z0 = 1) > Form III (Z0 = 2) > Form IV (Z0 = 2) > Form I (Z0 = 1). Metastable Form VI (Z0 = 2) was discovered later.
Form III (Z0 = 1) > Form II (Z0 = 1), Form IV (Z0 = 1), Form V (Z0 = 2), Form VI (Z0 = 2).
Monoclinic beta (Z0 = 1) > Triclinic alpha (Z0 = 2)
Form III (Z0 = 1) > Form I (Z0 = 4) > Form IV (Z0 = 1) > Form II (Z0 = 1). Metastable Form V (Z0 = 1) was discovered later.
Form I (Z0 = 1) > Form III (Z0 = 2)
Form III (Z0 = 1) > Form II (Z0 = 1) > Form I (Z0 = 1) > Form VII (Z0 = 2), Form IV (Z0 = 3), Form V (Z0 = 4), Form VI (Z0 = 6), Form VIII
(Z0 = 9.5)
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
FUQMOU
QNGHSU
SLFNMA
VETVOG
FUFTEG
EJEQAL
CERYAA
CAQYAW, CAQYIE
KONTIQ
MENMIB
NAZLAC
OHIBOW
TOKSES
VAWKUA
WUCGUW
DETBAA
BARBAC, IYAQOP
AMBNAC
BAQHOS
Form II (Z0 = 1) > Form I (Z0 = 2)
Form orthorhombic (Z0 = 1) > Form triclinic (Z0 = 4)
Form II (Z0 = 1) > Form I (Z0 = 2).
Form B (Z0 = 1) > Form A (Z0 = 2)
Form III (Z0 = 1) > Form IV (Z0 = 1) > Form I (Z0 = 1) > Form II (Z0 = 2).
Form I (Z0 = 1) > Form II (Z0 = 1) > Form III (Z0 = 2).
Form I (Z0 = 1) > Form II (Z0 = 3), Form IV (Z0 = 1)
Form D1a (Z0 = 1)>Form D1b (Z0 = 1), Form D2a (Z0 = 2), Form D2b (Z0 = 1)
MH-I (Z0 = 1, ratio 1:1) > MH-II (Z0 = 1, ratio 1:1) > MH-III (Z0 = 4, ratio 4:4) > MH-IV (Z0 = 1, ratio 1:1)
Form I beta (Z0 = 1) > Form II alpha (Z0 = 2), Form III (Z0 = 3)
Forms II (Z0 = 1), Form III (Z0 = 1) > Form I (Z0 = 4)
Form B (Z0 = 1, triclinic) > Form A (Z0 = 2, orthorhombic)
Form III (Z0 = 1) > Form II (Z0 = 2) > Form I (Z0 = 1)
Form orthorhombic (Z0 = 1, ratio 1:1) > Form triclinic (Z0 = 2, ratio 2:2)
Form b (Z0 = 1) > Form
a
(Z0 = 2)
Form I (Z0 = 1) > Form II (Z0 = 0.5) > Form IV (Z0 = 4)
Form IV (Z0 = 1) > Form II (Z0 = 2) > Form I (Z0 = 1)
Beta (Z0 = 1) > Alpha (Z0 = 2)
Form II (Z0 = 1) > Form I (Z0 = 2) > Form III (Z0 = 0.5)
Form II (Z0 = 2) > Form I (Z0 = 4) > Form III (Z0 = 2)
WERNOW
The high Z0 form is more stable than the low Z0 form in 30 systems (36%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
NIZVAU
HACTPH
CUNKAY
PHBARB
WOBMUV
GIZYIY
DIPGIS
FURSEM
VIFQIL
WINWUL
WURMOM
VOLZEC
CEBGOF
ZZZPUS
KARCOW
ZZZAUS
Form II (Z0 = 2) > Form I (Z0 = 1) > Form III (Z0 = 2)
Form II (Z0 = 2) > Form I (Z0 = 1)
Form I (Z0 = 2) > Forms II and III (Z0 = 1)
Forms I (Z0 = 3) > Form II (Z0 = 3) > Form III (Z0 = 1)
Form I (Z0 = 2) > Form I (Z0 = 1), Form II (Z0 = 1)
Form I (Z0 = 3) > Form II (Z0 = 1)
Form I (Z0 = 2) > Form II (Z0 = 1)
Form I (Z0 = 2) > Form I (Z0 = 1) > Form III (Z0 = 1)
Form II (Z0 = 2) > Form I (Z0 = 1)
Form I (Z0 = 2) > Form II (Z0 = 1)
Form I (Z0 = 2) > Form II (Z0 = 1)
Form II (Z0 = 4) > Form I (Z0 = 1).
Form I-I (Z0 = 3) > Form I-III (Z0 = 1), Form I-107 (Z0 = 1), I-112 (Z0 = 1)
Form II (Z0 = 4) > Form I (Z0 = 1) > Form III (Z0 = 1) > Form IV (Z0 = 1). Metastable form V (Z0 = 1) was discovered later.
Beta (Z0 = 2) > Alpha (Z0 = 1)
Form 0 (Z0 = 3) > Form X (Z0 = 2) > Form VI (Z0 = 2)
(continued on next page)
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

8
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
Table 4 (continued)
S.
REFCODE
The order of stability at RT
No
17
18
19
ZZZPNG
MAGTUK
THBARB, PABNIR,
PABNAJ
JIWQAI
SUTHAZ
BENZIE
QULLUF
WUYPOW
BIJDON
EYOCUQ
GLUCIT
FEFREP
AMYTAL
BIRMEU
Form 1 (Z0 = 2) > Form 2 (Z0 = 2), Form 3 (Z0 = 1)
Form (Z0 = 4) > Form I (Z0 = 1)
Form IV (Z0 = 2) > Form VI (Z0 = 2), Form III (Z0 = 1) and Form V (Z0 = 4).
20
21
22
23
24
25
26
27
28
29
30
Form II (Z0 = 2) > Form I (Z0 = 1).
Form III (Z0 = 2, P21/c) > Form I (Z0 = 2, P21/c), Form II (Z0 = 1, P1121/b), Form IV (Z0 = 1, P1121/n), Form V (Z0 = 5, P21/c).
Form II (Z0 = 3) > Form III (Z0 = 1.5) > Form I (Z0 = 4.5), Form IV (Z0 = 4.5).
Form I (Z0 = 2, ratio 2:2) > Form II and III (Z0 = 1, ratio 1:1)
Form 2 (Z0 = 3) > Form I (Z0 = 0.5)
Form triclinic (Z0 = 2, ratio 2:2) > Form monoclinic (Z0 = 1, ratio 1:1).
Form triclinic (Z0 = 2) > Form monoclinic (Z0 = 1)
Form
c a e
(Z0 = 3) > Form (Z0 = 1), Form (Z0 = 2)
Form IV (Z0 = 2) > Form III (Z0 = 3) > Form II (Z0 = 1) > Form I (Z0 = 1)
Form II (Z0 = 2) > Form I (Z0 = 1)
Form I (Z0 = 2, ratio 2:2) > Form II (Z0 = 1, ratio 1:1)
Fig. 5. Differential Scanning Calorimetry scans of (a) form I (b) mixture of form I and II. The form I shows a single melting endotherm at 156.4 °C, whereas the polymorph
mixture shows two endotherms, a weak endothermic event at 118.9 °C followed by a strong endothermic event at 156.4 °C. The weak endotherm corresponds to the solid
state phase transformation of form II fraction of material to form I. Please see Fig. S4 of ESI for DSC scan of pure form II on a very limited sample.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
9
Fig. 6. PXRD patterns of the materials prepared by different crystallization
conditions and the solid state phase transformations. Rota-evaporation predomi-
nantly yields mixture of form I and II. These polymorph mixtures are stable at RT for
more than a month. However, when they were subjected to grinding, slurry or
thermal treatment, the monoclinic fraction in the polymorph mixture transformed
to the more stable form I. Simulated patterns from crystal structures are provided
for comparison.
discussed earlier. Thus the first small endotherm arose from
the polymorph transformation of the form II fraction in the
mixture to form I. This was later confirmed by heating the
polymorphic mixture at 120 °C for about half an hour and
recording a PXRD on the heated material. All peaks corresponding
to the form II fraction of the material transformed to the stable
form I (Fig. 6).
The greater stability of form I at room temperature was further
confirmed from slurry experiment [19]. In this typical experiment,
approximately 150 mg of polymorph mixture (readily obtained
using rotary evaporation method) was slurried in 8–10 mL of ace-
tone for 1 day. The undissolved solid was filtered, dried, and con-
firmed to be pure form
I by PXRD (Fig. 6). Grinding the
polymorphic mixture in ball mill also showed this transformation
within 15 min (Fig. 6). Grinding for another 15 min showed no
phase change indicating that triclinic form I is the stable most
form. The calculated density and packing fraction were completely
in agreement with the stability order inferred from the experimen-
tal results. The triclinic form was denser (1.382 g cmꢂ3) compared
to the monoclinic (1.371 g cmꢂ3) and obeyed the Kitaigorodskii’s
‘‘density rule’’ according to which the most stable form corre-
sponded to the structure with the highest density [20]. The greater
density was also reflected in the slightly higher packing index
(70.8% for form I vs 70.2% for form II).
Fig. 7. Crystal graphs of (a) form I and (b) form II displaying the interaction strength
(kcal mol–1) along the 2D sheet, indicated by dashed lines. Please see Fig. 4a and b
and Tables 2 and 3 for detailed list of interactions resulting in these intermolecular
potentials. The stronger interactions are shown by red labels, followed by moderate
interactions by green and weaker interactions by blue labels. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web
version of this article.)
3.4. Why is form I more stable?
Although the crystal structure analysis provided a qualitative
picture on the differences between two polymorphs, a computa-
tional analysis is sought after to establish a quantitative relation
between the intermolecular forces and their relation to crystal sta-
bility. For a given conformational system, the energy contribution
from the conformer (Econf) as well as packing (Ulatt) must be
calculated to arrive at the total energy of the polymorph
(Etotal = Ulatt + Econf) [8]. The conformer energies are calculated using
the density functional theory (DFT) methods at B3LYP 6-31G(d,p)
basis set as implemented in Gaussian03 [15a]. The form I contains
the most stable as well as the most metastable conformations (0
and 1.631 kcal mol–1 for conformers T2 and T1, respectively)
and form II contains an intermediate energy conformation
(0.798 kcal mol–1 for M) to form I conformers. The average energy
contribution per a single molecule in both polymorphs is very
close, 0.815 kcal mol–1 in form I and 0.789 kcal mol–1 in form II.
As a result, the Econf contribution to the Etotal is comparable in both
forms. In other words, the polymorph energy difference is expected
to arise mainly from the packing energies of respective crystals
(Ulatt). The Unified pair-potentials (UNI) force-field parameters
are used to calculate the packing energies in crystals [16a]. Form
I is estimated to be lower in energy, ꢂ37.362 kcal mol–1 compared
to form II with ꢂ35.645 kcal mol–1, implying former to be more
stable than the latter.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

10
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
Fig. 8. Crystal graphs of forms I and II displaying the interaction strength (kcal mol–1) along the 3D inter-layers and van der Waals representation of molecules. (a) Type-I
stacking in form I. (b) Type-I stacking in form II. (c) Type-II stacking in form I. (d) Type-II stacking in form II. Please see Tables 2 and 3 for significant interactions that result in
these stacked potential energies.
I arises from the combination of C11AH11ꢁ ꢁ ꢁO1 and C25AH25ꢁ ꢁ ꢁ
p.
The close packing of the phenyl rings and van der Waals interac-
tions contribute to ꢂ3.6 and ꢂ1.6 kcal mol–1 energies in form I
and ꢂ3.8 and ꢂ1.8 kcal mol–1 in form II, and extend the packing
along the 2D sheet. A cumulative intermolecular potentials of six
molecules that form 1D ribbon are ꢂ40.9 kcal mol–1 (form I) and
ꢂ42.8 kcal mol–1 (form II), and of nine molecules that make up
2D sheet are ꢂ60.2 kcal mol–1 (form I) and ꢂ63.0 kcal mol–1 (form
II).
While it is interesting to note that the less stable form II has
better or stronger intermolecular bonding along the 1D ribbon by
ꢂ2.0 kcal mol–1 and 2D sheet by ꢂ2.8 kcal mol–1, a comprehensive
stability picture emerges only when forces along the 3D interlayers
are analyzed (Fig. 8a–d). Stacking of translation related molecular
pair is labeled as type-I and inversion related pair as type-II. Form
I exhibits two types of type-1 potentials, ꢂ12.4 kcal mol–1 for T1–
T2 pair and ꢂ10.4 kcal mol–1 for T2–T1 pair (Fig. 8a), whereas, form
II exhibits a single potential of ꢂ10.7 kcal mol–1 for M–M pair
(Fig. 8b). The type-II stacked potential is observed with -10.5 kcal -
mol–1 for T1–T1 pair, ꢂ8.8 kcal mol–1 for T2–T2 pair in form I
(Fig. 8c), and ꢂ8.0 kcal mol–1 for M–M pair in form II (Fig. 8d).
Overall, form I shows more anisotropic distribution of intermolec-
ular potentials due to two different conformations and subtle dif-
ferences in their intermolecular contacts. The stacked pair
potentials are substantially more than what is normally expected
Fig. 9. The number of stable crystal structures and Z0 correlation in 83 polymorphic
systems.
With the knowledge that form I is energetically more favored
over II in line with the experimental stability order, we turn our
attention to quantify various intermolecular contacts that contrib-
ute to its increased stabilization energy. Along the 1D ribbon, the
strongest intermolecular potential is formed between the amide
dimer of NAHꢁ ꢁ ꢁO bonds in both the polymorphs (Fig. 7a and b).
This value is ꢂ8.7 kcal mol–1 in form I and ꢂ9.4 kcal mol–1 in form
II pointing to slightly better amide interaction in form II in accor-
dance with the shorter bond distances (Tables 2 and 3). The next
strongest intermolecular potentials are observed when the parallel
dimers are connected [three potentials in form I, ꢂ3.5, ꢂ1.9 and
ꢂ2.3 kcal mol–1 (Fig. 7a) and two potentials in form II, ꢂ2.6 and
ꢂ2.1 kcal mol–1 (Fig. 7b)]. The slightly stronger potential in form
for
a typical stacking between phenyl rings, for example,
ꢂ8.5 kJ mol–1 for the benzene dimer at a distance of 3.6 Å,
ꢂ18 kJ mol–1 for chlorobenzene [16c]. This is because, besides ring
stacking interactions [21a], the pairs have additional CAHꢁ ꢁ ꢁO con-
tacts [21b], CAHꢁ ꢁ ꢁ
p contact [21c] (observed only in form I), dipo-
lar interactions between carbonyls [21d], interactions between
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
11
Fig. 10. A cluster of four molecules showing five strongest intermolecular potentials is shown for form I and II. It can be seen that, at the expense of losing energy along the
amide dimer, form I allows stronger intermolecular potentials over form II owing to shorter
pꢁ ꢁ ꢁp phenyl ring interactions and additional CAHꢁ ꢁ ꢁp interactions.
amide N lone pair and aromatic ring [21e], interactions of C@O and
aromatic ring [21e], and other van der Waals interactions [21f]
which altogether contributed to the rather large intermolecular
potentials. Please refer to Tables 2 and 3 for detailed list of inter-
molecular contacts in the stacked pairs discussed above. Several
weak interactions resulting in strong intermolecular potentials
was reported earlier [16d,22]. Notably, the stacked pair potentials
are also stronger than the conventional NAHꢁ ꢁ ꢁO hydrogen bonded
dimer. The cumulative interaction energy of two types of stacked
pair potentials is estimated as ꢂ42.1 kcal molꢂ1 for form I and
ꢂ37.4 kcal mol–1 for form II, with form I lower in energy than II
by ꢂ4.7 kcal mol–1. Thus, what form I lacks along the 1D ribbon
Z0 = 1 and 2 combinations are more commonly observed (in 57 sys-
tems) any other Z0 combinations, such as, Z0 = 1 and 4 combination
(in 12 systems), Z0 = 1 and 3 (in 9 systems), Z0 = 2 and 3 (in 2 sys-
tems) and Z0 = 2 and 8 combination (in 2 systems).
The inclusion of a second symmetry independent molecule ap-
pears to be largely guided by the crystal structure optimization and
energy stabilization in some of the reported systems. In the case of
40-Hydroxyacetophenone [13a], the larger stability of form II
(Z0 = 2) over form I (Z0 = 1), at 298 K, is related to the differences
in CAHꢁ ꢁ ꢁO interactions, which are stronger (shorter) and more
numerous in form II. A second example is Leflunomide [23], in
which the cohesion in the more stable form I (Z0 = 2) is provided
and 2D sheet, it mostly makes up with its 3D interlayer interac-
by both H bonding as well as
pꢁ ꢁ ꢁp stacking while in its metastable
tions with an additional stabilization energy of ꢂ1.9 kcal mol–1
.
form I (Z0 = 1) it is only given by the H bonding. A third example is
2-(mesitylamino) nicotinic acid [13d], in which the occurrence of
the more stable form I is linked to better packing with two iso-
energetic conformations (Z0 = 2) than with a single conformer in
forms II and III (Z0 = 1). In line with the above said examples, the
present case also illustrates the requirement of two symmetry
independent molecules for tighter molecular packing and crystal
lattice stabilization. A cluster of four molecules which show five
strongest intermolecular potentials are shown for both poly-
morphs in Fig. 10. It can be seen that form I has stronger intermo-
lecular potentials compared to form II. Here, the amide dimer units
may be viewed as structure building growth units. These are rela-
tively stronger for metastable form II and may be expected to form
more easily in solution with a single conformation. Gradually these
may be transformed into more stable form I with two symmetry
independent conformations. The selection of two different confor-
mations has two benefits in form I, firstly it does not impose any
extra conformer energy burden and secondly it offers better crystal
packing. Thus the crystallization with two symmetry independent
molecules is genuinely justified for stability reasons.
Please note that an arbitrary cut-off of ꢂ1.5 kcal mol–1 was applied
for the above discussed intermolecular potentials and that many
interactions exist below this level of energy and contribute to the
overall lattice energy difference of ꢂ1.62 kcal mol–1. An in-depth
examination and appreciation of all intermolecular interactions,
not just those of classical hydrogen bonds is shown to be necessary
in the present case, to build an overall understanding on the crystal
stability.
3.5. High Z0 and stability correlation
To understand the statistical trends of high Z0 and their relation
to crystal stability, the latest version of the Cambridge Structural
Database (CSD) [17a] was searched for extracting all polymorphic
systems which contain variable Z0 instances. Only polymorphic
systems for which the stability data determined by experimental
methods like Differential Scanning Calorimetry, solvent mediated
phase transformations, competitive slurry or grinding methods
and solubility studies were included for more reliable statistical
trends. A total of 83 polymorphic systems were analyzed (Table 4).
Please see experimental section for detailed search criteria. As a
first observation, the low Z0 polymorph is thermodynamically more
stable than its high Z0 polymorph in 53 polymorphic systems (64%)
and the high Z0 form is thermodynamically more stable than its low
Z0 form in 30 systems (36%). This observation concurs with earlier
suggestions in the literature that high Z0 structures are usually ki-
netic. As a second observation, the stable most polymorph is ob-
served with structures of Z0 = 1 or 2 which represent 87% of the
total stable structures. The Z0 = 1 is the stable polymorph in 50 sys-
tems (about 60%), Z0 = 2 is the stable form in 23 systems (27%),
Z0 = 3 in 8 systems (10%) and Z0 = 4 in 2 systems (3%), and the stable
form with Z0 > 4 is not observed (Fig. 9). As a third observation, the
3.6. Summary
A complete crystal structure analysis of two polymorphs of 7-
benzoyl-1,3-dihydroindol-2-one is presented and intermolecular
interactions in the crystal structures are quantified to demonstrate
the requirement of a second molecule in the asymmetric unit of
form I for enhanced stability over the crystal structure with only
a single molecule in the asymmetric unit of form II. Both poly-
morphs are almost similar with respect to the propagation as
amide dimer by NAHꢁ ꢁ ꢁO bonds and 1D ribbon chain by CAHꢁ ꢁ ꢁO
hydrogen bonding, however, they considerably differ along the
extension into 2D sheet (which is planar in form I and wave-like
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

12
J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
(b) A. Gavezzotti, G. Fillippini, Geometry of the intermolecular XAHꢁ ꢁ ꢁY(X,
Y = N, O) hydrogen bond and calibration of the empirical hydrogen bond
potentials, J. Phys. Chem. 98 (1994) 4831–4837.
in form II) and the 3D interlayers. The stacking of the two different
conformations brings the close approach of phenyl rings to be in-
volved in shorter
tions, and other aromatic interactions with amide
p
ꢁ ꢁ ꢁ
p
interactions, additional CAHꢁ ꢁ ꢁ
p
interac-
and
[7] (a) A. Nangia, Conformational polymorphism in organic crystals, Acc. Chem.
Res. 41 (2008) 595–604;
N
(b) S. Roy, R. Banerjee, A. Nangia, G.J. Kruger, Conformational, concomitant
polymorphs of 4,4-diphenyl-2,5-cyclohexadienone. Conformation and lattice
energy compensation in kinetic and thermodynamic forms, Chem. -Eur. J. 12
(2006) 3777–3788;
carbonyl. Several of these weak interactions put together results
in stronger intermolecular potentials and provide greater cohesion
energy in form I. The differential scanning calorimetry confirmed
that form I is the stable polymorph with the melting point at
156.4 °C. The metastable form II undergoes a solid state phase
transformation to the stable form I at 118.6 °C. The grinding and
slurry method further established the greater stability of form I
at room temperature. Analysis of 83 polymorph systems in the lit-
erature with variable Z0 indicated that the low Z0 form is more sta-
ble than high Z0 form in 53 polymorphic systems and the high Z0
form is more stable than low Z0 form in the remaining 30 cases.
The number of stable crystal structures markedly decrease as the
Z0 value increases. Nevertheless the Z0 = 1 and 2 combinations are
more frequently observed than any other Z0 combinations.
(c) A. Nangia, Conformational polymorphs, multiple Z crystal structures and
phase transformations, J. Ind. Inst. Sci. 87 (2007) 133–147.
[8] (a) B.M. Craven, Pseudosymmetry in cholesterol monohydrate, Acta
Crystallogr. Sect. B 35 (1979) 1123–1128;
(b) G.R. Desiraju, J.C. Calabrese, R.L. Harlow, Pseudoinversion centres in the
ꢀ
space group P1 and
a
redetermination of the crystal structure of 3,4-
study of non-crystallographic symmetry, Acta
dimethoxycinnamic acid.
A
Crystallogr. Sect. B 47 (1991) 77–86;
(c) P.M. Zorky, Symmetry, pseudosymmetry and hypersymmetry in organic
crystals, J. Mol. Struct. 374 (1996) 9–28;
(d) D. Britton, Two-dimensional packing and pseudosymmetry, Acta
Crystallogr. Sect. B 56 (2000) 828–832;
(e) L.N. Kuleshova, M.Y. Antipin, I.V. Komkov, The role of molecular association
in the formation of crystals with Z > 1 of some hydroxyl containing
compounds, J. Mol. Struct. 647 (2003) 41–51;
(f) A. Gavezzotti, Structure and energy in organic crystals with two molecules
in the asymmetric unit: causality or chance?, CrystEngComm 10 (2008) 389–
398;
Acknowledgements
(g) J Bernstein, J.D. Dunitz, A. Gavezzotti, Polymorphic perversity: crystal
structures with many symmetry independent molecules in the unit cell, Cryst.
Growth. Des. 8 (2008) 2011–2018.
We acknowledge the CSIR, New Delhi, for financial support as a
part of XII five-year plan program under the title AARF (CSC0406).
The director Dr. M. Laksmikantham is thanked for kind encourage-
ment. Mr. Sivanarayan from our centre for X-ray crystallography is
profusely thanked for his support on PXRD data. We also express
our gratitude to Dr. Ravi Prasad and Ms. Rajani from the Analytical
Division, Dr. T. Venugopala Raju from NMR division and Mr. G. Sai
Krishna of Mass spectroscopy division for their kind support with
the relevant spectroscopic data.
[9] (a) W.C. McCrone, in: D. Fox, M.M. Labes, A. Weisemberg (Eds.), Polymorphism
in Physics and Chemistry of the Organic Solid-State, vol. 2, Interscience, New
York, 1965, pp. 725–767;
(b) T.M.R. Maria, M.E.S. Eusebio, J.A.E. Silva, A.J.F.N. Sobral, C. Cardoso, J.A.
Paixao, M.R. Silva, 2-Quinolinecarboxaldehyde: polymorphic behaviour of a
small rigid molecule, J. Mol. Struct. 1030 (2012) 67–74;
(c) J. Bernstein, Polymorphism in Molecular Crystals, Clarendon, Oxford, 2002;
(d) A. Borba, M. Albrecht, A. Gomez-Zavaglia, M.A. Suhm, R. Fausto, Low
temperature Infrared spectroscopy study of pyrazinamide: from the isolated
monomer to the stable low temperature crystalline phase, J. Phys. Chem. A 114
(2010) 151–161;
(e) R. Hilfiker, Polymorphism in the Pharmaceutical Industry, Wiley-VCH,
Weinheim, 2006;
Appendix A. Supplementary material
(f) H.G. Brittan, Polymorphism in Pharmaceutical Solids, Marcel Dekker, New
York, 1999.
[10] D. Das, R. Banerjee, R. Mondal, J.A.K. Howard, R. Boese, G.R. Desiraju, Synthon
evolution and unit cell evolution during crystallization. A study of symmetry
independent molecules (Z0 > 1) in crystals of some hydroxyl compounds,
Chem. Commun. (2006) 555–557.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.molstruc.
2013.10.061. These data include MOL files and InChiKeys of the
most important compounds described in this article.
[11] B.
Sarma,
S.
Roy,
A.
Nangia,
Polymorphs
of
1,1-bis(4-
hydroxyphenyl)cyclohexane and multiple Z0 crystal structures by melt and
sublimation crystallization, Chem. Commun. (2006) 4918–4920.
[12] G.S. Nichol, W. Clegg, Further thoughts on crystal structures with Z0 > 1:
analysis of single crystal structures determined using X-ray synchrotron and
neutron radiation in the Cambridge Structural Database, CrystEngComm 9
(2007) 959–960.
[13] (a) C.E.S. Bernardes, M.F.M. Piedade, M.E.M. da Piedade, Polymorphism in 4-
hydroxyacetophenone: structure and energetics, Cryst. Growth Des. 8 (2008)
2419–2430;
References
[1] C.P. Brock, Investigations of the systematics of crystal packing using the
Cambridge Structural Database, J. Res. Natl. Inst. Stand. Technol. 101 (1996)
321–325.
[2] (a) J.W. Steed, Should solid-state molecular packing have to obey the rules of
crystallographic symmetry?, CrystEngComm 5 (2003) 169–179;
(b) <http://wwwdur.ac.uk/zprime/introduction.html>.
[3] N. Padmaja, S. Ramakumar, M.A. Viswamitra, Space-group frequencies of
proteins and of organic compounds with more than one formula unit in the
asymmetric unit, Acta Crystallogr. Sect. A 46 (1990) 725–730.
[4] (a) G.R. Desiraju, On the presence of multiple molecules in the asymmetric unit
(Z > 1), CrystEngComm 9 (2007) 91–92;
(b) N.J. Babu, S. Cherukuvada, R. Thankuria, A. Nangia, Conformational and
synthon polymorphism in furosemide (Lasix), Cryst. Growth. Des. 10 (2010)
1979–1989;
(c) B. Sarma, P. Sanphui, A. Nangia, Polymorphism in isomeric
dihydroxybenzoic acids, Cryst. Growth. Des. 10 (2010) 2388–2399;
(d) S. Long, M.A. Siegler, A. Mattei, T. Li, Phase transition from two Z = 1 forms
to a Z = 2 form of a concomitant conformational polymorphic system, Cryst.
Growth Des. 11 (2011) 414–421;
(e) M. Rafilovich, J. Bernstein, Serendipity and four polymorphic structures of
benzidine, C12H12N2, J. Am. Chem. Soc. 128 (2006) 12185–12191;
(f) M. Pauchet, T. Morelli, S. Coste, J. Malaindain, G. Coquerel, Crystallization of
( ) modafinil in gel: access to form I, form II and twins, Cryst. Growth. Des. 6
(2006) 1881–1889.
(b) T. Steiner, Frequency of Z values in organic and organometallic crystal
structures, Acta Crystallogr. Sect. B 56 (2000) 673–676;
(c) K.M. Anderson, J.W. Steed, Comment ‘‘On the presence of multiple
molecules in the asymmetric unit (Z > 1)’’ by Gautam. R. Desiraju,
CrystEngComm 9 (2007) 328–330;
(d) K.M. Anderson, A.E. Goeta, K.S.B. Hancock, J.W. Steed, Unusual variations in
the incidence of Z0 > 1 in oxo-anion structures, Chem. Commun. (2006) 2138–
2140;
[14] (a) Bruker, SAINT and SMART, Bruker AXS Inc., Madison, Wisconsin, USA,
2001.;
(e) K.M. Anderson, M.R. Probert, C.N. Whiteley, A.M. Rowland, A.E. Goeta, J.W.
Steed, Designing cocrystals of pharmaceutically relevant compounds that
crystallize with Z > 1, Cryst. Growth. Des. 9 (2009) 1082–1087.
[5] (a) S.L. Price, From crystal structure prediction to polymorph prediction:
interpreting the crystal energy landscape, Phys. Chem. Chem. Phys. 10 (2008)
1996–2009;
(b) G.M. Sheldrick,
A short history of Acta Crystallogr Sect. A: Found.
Crystallogr. 64 (2008) 112–122.
[15] (a) 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, 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.
(b) S. Mohamed, S.A. Barnett, D.A. Tocher, S.L. Price, K. Shankland, C.K. Leech,
Discovery of three polymorphs of 7-fluoroisatin reveals challenges in using
computational crystal structure prediction as a complement to experimental
screening, CrystEngComm 10 (2008) 399–404;
(c) B.P. van Eijck, J. Kroon, Structure predictions allowing more than one
molecule in the asymmetric unit, Acta. Crystallogr. Sect. B 56 (2000) 535–542.
[6] (a) C.P. Brock, L.L. Duncan, Anamolous Space group frequencies for
monoalcohols CnHmOH, Chem. Mater. 6 (1994) 1307–1312;
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061

J.B. Nanubolu et al. / Journal of Molecular Structure xxx (2013) xxx–xxx
13
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, J.A. Pople, GAUSSIAN 03W (Version 6.1), Gaussian Inc.,
Wallingford, CT, 2004.;
(e) W.J. Welstead, H.W. Moran, H.F. Stauffer, L.B. Turnbull, L.F. Sancilio, Anti-
inflammatory agents. 1. Synthesis and anti-inflammatory activity of 2-amino-
3-benzoylphenyl acetic acid, J. Med. Chem. 22 (1979) 1074–1079;
(f) Y. Kato, M. Shimokawa, T. Yokoyama, K. Mohri, Simultaneous
determination of amfenac sodium and its metabolite (7-benzoyl-2-oxindole)
in human plasma by high-performance liquid chromatography, J. Chromatogr.
616 (1993) 67–71.
(b) L.J. Barbour, X-Seed, Graphical Interface to SHELX-97 and POVRay,
University of Missouri-Columbia, USA, 1999.
[16] (a) C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock,
L. Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, Mercury CSD 2.0 –
new features for the visualization and investigation of crystal structures, J.
Appl. Crystallogr. 41 (2008) 466–470;
[19] (a) M.D. Eddleston, S. Sivachelvam, W. Jones, Screening for polymorphs of
cocrystals: a case study, CrystEngComm 15 (2013) 175–181;
(b) P. Bergström, A. Fischer, L. Kloo, T. Sebhatu, Crystal structures and physical
properties of ropavacine HCl, J. Pharm. Sci. 95 (2006) 680–688.
[20] (a) A.I. Kitaigorodskii, Molecular crystals and molecules, Academic Press, New
York, 1973;
(b) A. Gavezzotti, Are crystal structures predictable?, Acc Chem. Res. 27 (1994)
309–314;
(c) A. Gavezzotti, Quantitative ranking of crystal packing modes by systematic
calculations on potential energies and vibrational amplitudes of molecular
dimers, J. Chem. Theor. Comput. 1 (2005) 834–840;
(d) G.K. Balendrian, N. Rath, A. Kotheimer, C. Miller, M. Zeller, N.P. Rath,
Biomolecular Chemistry of Isopropyl fibrates, J. Pharm. Sci. 101 (2012) 1555–
1569;
(e) A. Gavezzotti, Towards a realistic model for the quantitative evaluation of
intermolecular potentials and for the rationalization of organic crystal
structures. Part II. Crystal energy landscapes, CrystEngComm 5 (2003) 439–
446.
(b) A.I. Kitaigorodskii, Organic chemical crystallography, Consultants Bureau,
1961.
[21] (a) C.A. Hunter, K.R. Lawson, J. Perkins, C.J. Urch, Aromatic interactions, J.
Chem. Soc. Perkin Trans. 2. (2001) 651–659;
(b) G.R. Desiraju, The CAHꢁ ꢁ ꢁO hydrogen bond: structural implications and
supramolecular design, Acc. Chem. Res. 29 (1996) 441–449;
(c) R.T. Tiekink, J. Zukerman-schpector, The importance of Pi-interactions in
crystal engineering: frontiers in crystal engineering, John Wiley & Sons Ltd.,
2012;
(d) F.H. Allen, C.A. Baalham, J.P.M. Lommerse, P.R. Raithbay, Carbonyl–carbonyl
interactions can be competitive with hydrogen bonds, Acta. Crystallogr. Sect. B
54 (1998) 320–329;
[17] Cambridge Structural Database, ver. 5.32. ConQuest 1.15, November 2012
release, May 2013 update, CCDC, <www.ccdc.cam.ac.uk>.
[18] (a) M. Ahuja, A.S. Dhake, S.K. Sarma, D.K. Majumdar, Topical ocular delivery of
NSAIDs, AAPS J. 10 (2008) 229–241;
(e) M. Egli, S. Sarkel, Lone Pair – aromatic interactions: to stabilize or not to
stabilize, Acc. Chem. Res. 40 (2007) 197–205;
(b) Product Monograph on NEVNAC™ (nepafenac ophthalmic suspension) 0.1
w/v% by Alcon Research Inc., 2008, 1–25. <http://www.alcon.ca/pdf/
Product_pharma/Product_pharma_nevanac-eng.pdf>.;
(c) B.I. Gaynes, A. Onyekwuluje, Topical opthalmic NSAIDs: a discussion with
focus on napafenac ophthalmic suspension, Clin. Opthalmol. 2 (2008) 355–
368;
(f) V.A Parsegian, van der Waals forces: a handbook for biologists, chemists,
engineers, and physicists, Cambridge University Press, 2006.
[22] N. Schulthesis, M. Roe, S.X.M. Boerrigter, Cocrystals of neutraceutical p-
coumaric acid with caffeine and theophylline: polymorphism and solid-state
stability explored in detail using their crystal graphs, CrystEngComm 13
(2011) 611–619.
(d) J.B. Nanubolu, B. Sridhar, K. Ravikumar, Resonance-assisted amide
protonation in dutasteride hydrochloride salt, CrystEngComm 14 (2012)
2571–2578;
[23] D. Vega, A. Petragalli, D. Fernandez, J.A. Ellena, Polymorphism on leflunomide:
stability and crystal structures, J. Pharm. Sci. 95 (2006) 1075–1083.
Please cite this article in press as: J.B. Nanubolu et al., J. Mol. Struct. (2013), http://dx.doi.org/10.1016/j.molstruc.2013.10.061