doi.org/10.26434/chemrxiv.6489425.v1

Solid State Analysis and Theoretical Explorations on Polymorphic andHydrate Forms of p-Hydroxybenzaldehyde Isonicotinichydrazone: Effectof Additives on Polymorphic CrystallizationLincy Tom, Victoria A. Smolenski, Jerry P. Jasinski, M.R. Prathapachandra Kurup

Submitted date: 12/06/2018 • Posted date: 12/06/2018Licence: CC BY-NC-ND 4.0Citation information: Tom, Lincy; Smolenski, Victoria A.; Jasinski, Jerry P.; Kurup, M.R. Prathapachandra(2018): Solid State Analysis and Theoretical Explorations on Polymorphic and Hydrate Forms ofp-Hydroxybenzaldehyde Isonicotinichydrazone: Effect of Additives on Polymorphic Crystallization. ChemRxiv.Preprint.

The reaction of p-hydroxybenzaldehyde with an equimolar amount of isonicotinic hydrazide afforded twopolymorphic and hydrate forms of p-hydroxybenzaldehyde isonicotinichydrazone (HBIH) by varying theexperimental reaction conditions. The compounds are fully characterized by means of single crystal andpowder diffraction methods, vibrational spectroscopy (FT-IR and Raman), thermal and elemental analysis.The compound crystallizes in three different forms in two different space groups, P21/c (form PA and PB) andPbca (PC). The Hirshfeld surface analysis shows the differences in the relative contributions of intermolecularinteractions to the total Hirshfeld surface area for the HBIH molecules. The calculated pairwise interactionenergies (104-116 kJ/mol) can be related to the stability of the crystals. Energy framework analysis identifiesthe interaction hierarchy and their topology. The geometry and conformation of the three forms are essentiallysimilar which differ only by packing arrangement.

File list (1)

download fileview on ChemRxivmrp 201.pdf (1.93 MiB)

Solid state analysis and theoretical explorations on polymorphic and hydrate

forms of p-hydroxybenzaldehyde isonicotinichydrazone: Effect of additives on

polymorphic crystallization

Lincy Toma, Victoria A. Smolenskib, Jerry P. Jasinskib, M.R. Prathapachandra Kurupa,c,

aDepartment of Applied Chemistry, Cochin University of Science and Technology, Kochi 682 022, Kerala, India, bDepartment of Chemistry, Keene State College, 229 Main Street, Keene, NH03435-2001, USA, cDepartment of

Chemistry, School of Physical Sciences, Central University of Kerala, Nileshwar-671314

Abstract

The reaction of p-hydroxybenzaldehyde with an equimolar amount of isonicotinic hydrazide afforded two

polymorphic and hydrate forms of p-hydroxybenzaldehyde isonicotinichydrazone (HBIH) by varying the

experimental reaction conditions. The compounds are fully characterized by means of single crystal and

powder diffraction methods, vibrational spectroscopy (FT-IR and Raman), thermal and elemental

analysis. The compound crystallizes in three different forms in two different space groups, P21/c (form PA

and PB) and Pbca (PC). The Hirshfeld surface analysis shows the differences in the relative contributions

of intermolecular interactions to the total Hirshfeld surface area for the HBIH molecules. The calculated

pairwise interaction energies (104-116 kJ/mol) can be related to the stability of the crystals. Energy

framework analysis identifies the interaction hierarchy and their topology. The geometry and

conformation of the three forms are essentially similar which differ only by packing arrangement.

Keywords: polymorphism, Raman spectroscopy, X-ray crystal structure, packing polymorphism,

Hirshfeld surface analysis, energy frameworks

1. Introduction

Polymorphism is the existence of a solid material in more than one form of crystal structure or phases that

have different arrangements or conformations of the molecules in the crystal lattice. It offers a useful tool

for examining the structure–property relationship due to the different intermolecular interactions and has

received considerable attention from pharmaceutical industry and the academic community [1,2]. The

difference in the internal structure of polymorphs alters their physicochemical properties, including

Corresponding author E mail address: mrpcusat@gmail.com, mrp@cukerala.ac.in (M.R. Prathapachandra Kurup) Phone: +91-9895226115 Fax: +91-484-2575804

solubility, bioavailability, functionality and so forth [3,4]. Enhancements in physicochemical properties

can also be achieved by altering the physical forms such as hydrates, solvates (pseudopolymorphism)

amorphous salts, co-crystals, etc.[5]. The molecular arrangement in crystals most frequently decides the

functionality and properties of these materials. The existence of polymorphs delivers an opening to

investigate and appreciate the nature of intermolecular interactions during the formation of molecular

packing. The presence of functional hydrogen bonding groups and other secondary interactions helps in

molecular recognition, self-assembly and different packing of molecular components.

The crystallization of polymorphs and solvated crystals is affected by several factors like concentration,

solvents, additives, seed crystals and crystallization methods, etc. [4,6–9]. However, the mechanism of

these effects is not known and the quantitative relationship between the operational factors and the

crystallization characteristics of polymorphs is not clearly understood.

Polymorphism is only exhibited in the solid state. On melting, vaporization, or dissolution, polymorphic

structures are lost. Accordingly, for the comprehensive and systematic study of the chemistry of the title

compound, we have collected all the structural and physicochemical data of polymorphic and hydrate

forms of this compound by solid state studies. In this way, we got a better understanding of crystal

structure assembly and assessment of its crystal structure. Further, various intermolecular interactions are

well supported by Hirshfeld surface and fingerprint plot analysis. The interaction energy and energy

framework analysis were also studied to obtain adequate information about its electronic characteristics.

2. Experimental

2.1. Materials

The reactants for syntheses were procured from Aldrich Sigma Ltd. For all solution crystallizations

analytical grade solvents were used

2.2. Methods

Spectroscopy: Raman spectra were collected in the shift region of 0−3700 cm-1 using a WITec

Alpha300RA Raman instrument, equipped with a Andor Back illuminated CCD camera and a 532 nm

DPSS laser. The IR absorption spectra were performed with a JASCO FT-IR-5300 Spectrometer in the

4000-400 cm-1 region by making their KBr discs.

X-ray diffraction: PXRD patterns were determined on a Bruker D8 Advance diffractometer with

graphite-monochromated Cu-Kα radiation. The Mercury software version 3.10.1, with the conditions of

start angle, 5° stop angle, 50° step size, 0.02, was used to calculate the simulated powder X-ray pattern for

a Cu Kα radiation with wavelength 1.54056 Å. Crystallographic data were collected with a Bruker

SMART APEX diffractometer with graphite monochromated Mo Kα (λ = 0.71073 Å) X-ray source (PC),

Bruker D8 QUEST with Photon detector (PB) and Rigaku Gemini Eos Diffractometer (PA). The

structures for all three were solved by direct methods and refined by full-matrix least-squares calculations

with the SHELXL-2015 software package [10]. All non-hydrogen atoms were refined with anisotropic

displacement parameters and positions of hydrogen atoms were located in the difference Fourier maps

and were placed in calculated positions and refined as riding atoms. Bond lengths and angles were

restrained in all three structure refinements to ensure proper geometry using DFIX and DANG

instructions. All the drawings were made using ORTEP 3, Diamond 3.2k and Mercury 3.10.1 programs

[11–13].

Thermal and Elemental Analysis: Thermal behavior, mass loss, and melting points of all samples were

determined using simultaneous thermal analysis (STA), comprising differential scanning calorimetry

(DSC) and thermal gravimetric analysis (TGA). A PerkinElmer STA6000/8000 system with Pyris

Software was used for this study. Samples were placed in 10 μL aluminum pans, and a heating ramp was

set in the range of 25−400 ° at a rate of 10 °C min-1. The elemental analysis of carbon, hydrogen and

nitrogen were carried out using an Elementar model Vario EL III elemental analyzer.

2.3. Computational details

Hirshfeld surface analysis: The package CrystalExplorer 17.5 [14] was used to generate and visualize the

three-dimensional Hirshfeld surfaces [15] and their respective two-dimensional fingerprint plots [16],

which accepts a structure input file in the CIF format. As crystal packing is dominated by van der Waals

interactions, these plots are useful to enumerate the intermolecular interactions and packing in the crystal

lattice. The analyses were carried out on crystal geometries of the respective forms. The 3D dnorm surfaces

were mapped over a fixed color scale of -0.65 au (red) 1.27 au (blue) and curvedness -4.000 to +0.400 Å.

The 2D fingerprint plots were displayed in the 0.6 - 2.4 Å range including reciprocal contacts.

Interaction energy calculation: CrystalExplorer 17.5 was used to explore the intermolecular interaction

energies [16] of polymorphs and hydrated forms. The energy calculation was performed for a cluster of

radius 3.8 Å around the selected atom in each form. The energy components were expressed in terms of

electrostatic, polarization, dispersion and exchange-repulsion and finally the total interaction energy. The

prevalence of these interactions are illustrated with the aid of B3LYP/6-31G energy model using

symmetry independent molecules of all three forms.

Energy frameworks analysis: The three dimensional topology of interactions for all three forms were

constructed based on the obtained values of interaction energies and the frameworks [17] were visualized

using CrystalExplorer 17.5. The tube size (scale factor) used in all the energy frameworks was 80 and the

energy threshold (cut-off) value was set to zero.

2.4. General description for synthesis and single crystal growth

The two polymorphs (PA and PB) and one hydrated form (PC) were prepared by the condensation reaction

between the p-hydroxybenzaldehyde and isonicotinic hydrazide. The controlling factors for polymorphic

crystallization lie on several factors and herein, we describe the effect of solvents and additives on

polymorphic crystallization.

3. Result and discussion

3.1. Comments on synthesis

Crystallization of HBIH molecules led to the formation of three different solid forms, which were

governed by the employed conditions. Single crystals of PA were obtained by slow evaporation of the

solution of p-hydroxybenzaldehyde isonicotinic acid hydrazone in methanol at 293 K. Forms PB and PC

were reprepared by adopting different synthetic routes than their previously reported structures and

therefore their crystal structures were redetermined by single-crystal X-ray diffraction to compare with

the originally published structures [18,19]. All the crystallographic parameters are similar for the

structures PB and PC. The solid state morphology of three forms are different and can be visibly identified

by their appearances (Fig. S1).

Table 1. Details of crystallization experiment

Solvents/Additives Reaction condition Crystal quality Form

Methanol RT, Reflux Block PA

Ethanol N2 atm, Reflux Plate PB [18]

Ethanol RT, Reflux Block PC [19]

DMF, Cd2+ RT, Reflux Block PB

DMF/MeOH, Cd2+ Stirring Block PB

Methanol, Zn2+ RT, Reflux Block PB

Methanol, Mn2+ Hydrothermal Crystalline powder -

Methanol, Succinic acid RT, Reflux Block PC

Methanol, Oxalic acid RT, Reflux Block PC

DMF, Succinic acid RT, Reflux Block PC

DMF/CH3CN (3:1), Cd2+ RT, Reflux Block PB

Methanol, Cd2+ Hydrothermal Crystalline powder -

Methanol/H2O (3:1) RT, Reflux Crystalline powder -

Polymorph PA was difficult to crystallize as diffraction quality and only obtained after a series of crystal

growth experiments from methanol. The additive effects on different polymorphs were studied by a series

of crystal growth experiments. The additives influence each process of nucleation, growth and

transformation [20] and the influence of solvent, metal ion and dicarboxylic acids on the crystallization of

p-hydroxybenzaldehydeisonicotinic hydrazone are discussed below. Details of the crystallization

experiments are given in Table 1.

Methanol solution of pyridine-4-carboxylic acid hydrazide (0.137 g, 1 mmol) was added to a methanol

solution of p-hydroxybenzaldehyde (0.122 g, 1 mmol), and the mixture was stirred under reflux for 1

hour. Slow evaporation of the solution yielded X-ray quality crystals of PA. Polymorph PB and PC were

obtained using metal ions and dicarboxylic acids as additives. Equimolar amount of the additive was

added to the reactant mixtures to yield different forms. Much effort had been made to screen the potential

polymorphs from a range of single solvents or combinations thereof, but only two polymorphs PB and PC

have been obtained so far.

3.2. Crystal structures

In order to cognize the similarities and the dissimilarities of the three forms, the crystal structures are

scrutinized via connectivity and then packing patterns and the motifs were compared. An ORTEP plot of

the HBIH (PA) molecule is shown in Figure 1. The crystallographic data for all three forms are

summarized in Table S1. The molecular conformation and the geometry of hydrogen bonds in the three

forms are presented in Table 2. There are only trivial differences in bond lengths, angles and torsion

angles between the three forms (Table S2, S3 and S4). Crystal structures of PA and PB were solved in the

monoclinic system with the P21/c space group containing four independent molecules per unit cell. Form

PC, crystallizes in the orthorhombic Pbca space group.

Figure 1. ORTEP diagram of HBIH (PA) drawn with 30% displacement ellipsoids

3.2.1. Conformational analysis

X-ray crystal structures were analyzed to understand the interplay of intramolecular (conformer) and

intermolecular interactions related to the stability of the polymorphs and the hydrated forms. The overlay

of the all symmetry independent HBIH crystal structures (PA, PB and PC) is shown in Figure 2. It is clear

that they exhibit nearly identical molecular conformations. The pyridine nitrogen (N3) and carbonyl

oxygen (O2) point in the same direction in all the three forms. The OH group of the hydrated form is

twisted differently with respect to C1–O1 with a torsion angle (C6–C1–O1–H1) of 176.63°, which is

equal to 4.44o and 0.706° respectively for PA and PB.

Figure 2. Superimposition of the HBIH geometries extracted from the crystal structures of PA, PB and PC

3.2.2. Molecular packing and Intermolecular Interactions

Representation of molecular packing in the HBIH polymorphs and in their solvated form, illustrated in

Figure 3, indicates that their crystalline structures have notable differences. According to the literature,

the polymorphism shown by the PA and PB forms can be assigned as a crystal packing polymorphism

[21,22]. The molecular packing and hydrogen bonding patterns are unique in these three forms. It

involves different combinations of hydrogen donor and acceptor sites in each form. The geometry and

conformation of both polymorphic molecules (PA and PB) are very similar. The crystal packing is strongly

determined by the hydrogen bonds between the HBIH molecules.

On probing into the crystal structure, it was inferred that both forms PA and PB are associated with a

different supramolecular network. This dissimilarity results from the effects of intermolecular interactions

in the lattice. The supramolecular structure fragments of polymorphs PA and PC are shown in Figure 4.

The skeleton molecule possesses six hydrogen-bond donors (three of which are very good N–H/O–H

donors and three that are potential C–H donors) and five very good (N, O) acceptors. Therefore, the self-

organization events among the molecules can take place with different interactions and geometric

configurations which in turn might lead to a different solid-state association and the formation of variable

hydrogen-bonded synthons. The possible supramolecular synthons and hydrogen-bonded motifs which

can be formed between the HBIH molecules are presented in Figure 5.

Figure 3. Packing diagrams of polymorphs and hydrate viewed along the ‘b’ axis.

Table 2. Hydrogen bonding interactions in PA, PB and PC

D–H···A D–H (Å) H···A (Å) D···A (Å) ∠ D–H···A

Polymorph PA

N(2)–H(2’)···O(2)a 0.871 2.321(11) 3.1495(15) 158.8(14)

O(1)–H(1)···N(3)b 0.82 1.94 2.7131(16) 158

C(7)–H(7)···O(2)a 0.93 2.43 3.2507(18) 147

Polymorph PB

N(2)–H(2’)···N(3)p 0.881(12) 2.396(14) 3.236(2) 159.5(15)

O(1)–H(1)···O(2)q 0.848(14) 2.043(17) 2.8384(17) 156(2)

O(1)–H(1)···N(1)q 0.848(14) 2.505(16) 3.1185(18) 130.1(18)

C(7)–H(7)···N(3)p 0.93 2.62 3.460(2)) 150

C(5)–H(5)···O(1)r 0.93 2.60 3.487(2) 160

Solvated form PC

N(2)–H(2’)···O(1S)x 0.884(13) 2.037(15) 2.886(3) 161(2)

O(1S)–H(1A)···O(2) 0.86(3) 2.00(3) 2.837(4) 163(3)

O(1S)–H(1B)···O(2)y 0.86(3) 2.23(3) 2.909(4) 136(3)

O(1S)–H(1B)···N(1)y 0.86(3) 2.43(3) 3.200(4) 150(3)

O(1)–H(1)···N(3)z 0.82 1.97 2.772(3) 167

C(10)–H(10)···O(1S)x 0.93 2.50 3.398(4) 161

C(7)–H(7)···O(1S)x 0.93 2.36 3.184(4) 147

PA PB

Pc

a= 1+x,y,z;b= x,1+y,z

p= -x,1/2+y,1/2-z, q= 1-x,1/2+y,3/2-z, r= 1-x,-1/2+y,3/2-z

x= 1-x,-1/2+y,1/2-z; y= -1/2+x,y,1/2-z; z= x,1/2-y,1/2+z

Form PA is characterized by promising a ‘unidirectional columnar packing’ arrangement in the crystal

state. The stacking is unidirectional along the crystallographic ‘a’ and ‘b’ axis (Fig. 4). Two independent

hydrogen bonds are influential in forming the supramolecular motifs. This is identified by the (6) ring

interaction which is favored by N(2)–H(2’)···O(2) and C(7)–H(7)···O(2) intermolecular hydrogen bonds

establishing a one-dimensional assembly parallel to ‘a’ axis. The connectivity pattern of the ribbons is

depicted in synthon 5. The shortest intermolecular distance between the nitrogen atom and the oxygen

atom from the NH-C=O group (the N(2)–H(2’)···O(2) close contact) belonging to different neighbouring

molecules is 3.1495 Å. This ID motif interacts further through a C(14) chain motif (O(1)–H(1)···N(3)),

occurs at a distance of 2.7131 Å, and tightens the intracolumnar packing which, therefore, expands the

supramolecular aggregation into a two-dimensional framework.

Figure 4. Crystal packing showing an array of molecules which forms a sheet (PA and PC)/chain (PB)

containing intermolecular hydrogen bonding interactions

The form for PB, unlike the above mentioned, contains a different hydrogen bonding pattern. Most

notably, the packing is characterized by an elusive columnar/herringbone dual crystal packing [23] motif.

The molecules are arranged in columns along the ‘a’ axis, however, related to one another by inversion

and hence running antiparallel, while the HBIH moieties favor a herringbone fashion parallel to the ‘c’

axis. The hydrogen atoms, N(2)–H(2’) and O(1)–H(1), play an important role in the intermolecular

bonding as in the case for PA. The crystal lattice consists of HBIH molecules arranged in layers parallel

to the ‘ab’ plane. Within each layer, molecules are related by three cyclic hydrogen bonded ring motifs,

PA

PC

PB

N(2)–H(2’)···N(3), O(1)–H(1)···O(2), O(1)–H(1)···N(1) and non-covalent C(7)–H(7)···N(3), C(5)–

H(5)···O(1) interactions and can be represented by , and graph-set motif notations.

These five hydrogen bonds together form a 2D supramolecular motif.

Figure 5. Hydrogen bonding patterns in HBIH forms A, B and C

The comparison of the hydrogen-bonding network for PA and PB reveals that they are only concomitant

polymorphs. The supramolecular organization is comprised of two dimensional layers in which molecules

are stacked in columns. The subtle difference in supramolecular architecture arises due to the differences

in the number of intermolecular interactions.

3.3. Simulated powder X-ray diffraction analysis

The most reliable way to distinguish these polymorphs is by powder X-ray diffraction analysis. The

experimental and simulated diffraction patterns (Figure 6) for each form of HBIH (PA, PB and PC) are

compared to highlight the subtle differences of the phases in each of these three forms. As shown, the

Synthon 1 (in form PC) Synthon 2 (in form PC) Synthon 3 (in form PC)

Synthon 4 (in form PB) Synthon 5 (in form PB)

Synthon 6 (in form PB) Synthon 7 (in form PA and PC)

simulated PXRD patterns are almost similar, establishing a close relationship between the structures, and

justifies the phase purity of the compounds. Variations in relative intensities and positions are most likely

being accounted for by some preferred orientation of the samples in the powder X-ray experiment.

Figure 6. Experimental and simulated powder patterns of PA, PB and PC.

3.4. Vibrational spectra and band assignments

The Raman and IR spectra of the three forms were recorded from solid state samples at room temperature.

The spectra exhibit nearly identical features for all three polymorphs. They are presented in Figures 7-9.

The vibrational frequencies and the tentative assignments of the observed bands are gathered in Table 3.

Forms PA and PB produce subtle differences in the peak positions across the entire spectrum which differ

only by <8 cm-1 and provides additional evidence for the structural similarity of the two forms. However,

form PC showed significant differences in position and intensity of the spectral bands due to lattice water.

The OH stretching vibrations centered around 3430 cm-1 are very intense in the IR spectrum, as the two

atoms involved in the vibrational mode have a large electronegativity difference, which is much less in

the Raman spectrum because the bond electronic density that can be deformed under the action of the

electric field is small. In PA and PB the band around 3020 cm-1 is attributed to aromatic CH stretching. An

additional band at 3287 cm-1 is found in the IR spectrum of polymorph PA, which is weak in PB and PC

corresponding to NH stretching vibrations. The bands at 1644, 1642 and 1658 cm-1 in the IR spectra are

assigned purely to ν(C=N) which are inactive in Raman [24, 25]. The spectrum of PC shows some

additional peaks at 3221 and 1549 cm-1 making them particularly diagnostic. The broad band at 3221 cm-1

corresponds to the O–H stretching of lattice water. The region 1500- 1700 cm-1 is conquered by the

bands associated with C=O, C=N and benzene ring stretching vibrations, which are partially coupled to

in-plane CH and NH deformations [23]. The C=N stretch appears only in the IR spectrum, at 1644 in PA

PA PB PC

Simulated

Experimental

and at 1642 cm-1 in PB. A medium Intensity Raman band at 1553 cm-1 is assigned to the C=O stretching

vibration. The aromatic breathing mode observed around 1600 cm-1 is very intense in the Raman spectra,

but with much reduced intensity in the IR spectra of PA and PB.

The bands around 1445 (Raman) and 1510 cm-1 (Raman and IR) in the spectra of PA, PB and PC are

attributed to modes involving benzene ring in-plane CH and NH deformations. The CO and CC stretching

vibrations in the Raman spectra are associated with the bands at 1287 and 1221 cm-1 respectively for PA

and PB. The contribution of the CO stretching is mainly evident in bands observed in the region nearby to

1287 cm-1 in both Raman and IR spectra. The region 700-1000 cm-1 is predominantly assigned to

aromatic out-of-plane CH deformation and CN ring deformations which give rise to weak bands in the IR

spectrum. These spectral alterations between the polymorphs reflects changes in the nature of hydrogen-

bond geometries.

Figure 7. Comparative FT-IR spectra of HBIH Figure 8. Comparative FT-Raman spectra of

HBIH

Figure 9. FT-IR and Raman spectra of PA

Table 3. Experimental Raman and IR spectral parameters for PA, PB and PC

PA PB PC Assignment

Raman IR Raman IR Raman IR

3428 m 3427 m 3435 m ν(OH)

3287 vw ν(NH)

3022 w 3020 w 3030 w ν(CH)

3221 m ν(OH)

1644 s 1642 s 1658 s ν(C=N)

1598 s 1607 w 1600 s 1609 w 1601 s 1592 s ν(C=C)

1553 m 1583 s 1553 m 1580 s 1553 m 1558 s ν(C=O)

1509 w 1517 s 1510 w 15015 s 1512 m 1518 s δip(NH)

1445 m 1443 m 1445 m δip(CH)

1287 m 1286 s 1287 m 1286 s 1287 s 1286 s ν(C–O)

1221 m 1221 m 1228 s 1223 w ν(C–N)

1169 m 1162 m 1161 m 1162 m 1154 s 1162 m ν(C–C)

1058 m 1063 1051 w 1054 1058 w 1063 δip(CH)

991 w 980 vw 991 w 980 vw 984 m 980 vw δop(CH)

831 w 831 w 839 s δop(CH)

748 w 748 w 748 m δop(CN)

3.5. Thermal Analysis

A comparison was performed between the TG, DTA and DSC traces of all three polymorphs (Figure 10),

and the profiles of the curves reveal that all of the polymorphs have identical decomposition patterns. The

DTA traces are slightly different and have only one peak corresponding to the melting of the compound.

Form PA has a slightly higher decomposition temperature (316 °C) than that of PB (302 °C). Analyzing

the thermogravimetric curve of PC, it is possible to observe a first weight loss in the temperature range of

82-131 °C that corresponds to the loss of water of crystallization (obsd. 3.63 %, cald. 3.47 %) [26]. This

assignment was confirmed by differential scanning calorimetry and DTA observations.

Figure 10. TG (a), DTA (b) and DSC (c) profiles of form PA, PB and PC

c a b

The melting points of PA (302.76 °C) and PB (302.31 °C) were identified from DSC traces. PC (292.07 °C)

have much lower melting points compared to the unhydrated forms. The polymorphs are stable upto 255

°C and at this phase the compounds start to disintegrate with significant weight loss. Maximum rate of

decomposition is shown at this stage. It was not possible to calculate the final mass residue due to

complete decomposition of compounds.

The elemental analysis of HBIH in all three forms are in good agreement with observed and calculated

values (Table S5).

3.6. Computational details

3.6.1. Hirshfeld surface analysis

In order to visualize and quantify the contributions of intermolecular interactions to the supramolecular

assembly across the three forms, Hirshfeld surface analysis was carried out. By using the

crystallographic data, the contact points were visualized through dark red spots on the Hirshfeld surface

(HS). The analysis were based on crystal geometries and performed on the asymmetric unit of each form.

The shape of the generated HS’s for polymorphs PA and PB show marked differences (Figure 11), which

are related to dissimilarities in packing and intermolecular interactions in the crystal lattices of both

polymorphs. The curvature mapped on the HS demonstrate the apparent surface dissemblance, which is

presented in Figure S2. The volume and total area of the HS are slightly larger for form PB than for the PA

form (278.93 Å3 vs. 281.23 Å2 and 275.32 Å3 vs. 279.77 Å2, respectively). However, the HS of form PC

shows close resemblance with the polymorphic PB form. The similarities result from nearly identical

distribution of the contact interactions.

Figure 11. dnorm mapped on Hirshfeld surfaces of the asymmetric unit of the three forms PA, PB and PC

respectively.

PA PB Pc

Figure 12. Percentage contributions to the Hirshfeld surface area for the various close intermolecular

contacts in the three forms

Figure 13. Intermolecular interactions of PA, PB and PC mapped on dnorm surfaces.

The molecule HBIH in all three forms exhibit different intermolecular contacts. However, the

discrepancies in interactions can be gauged from the changes in area of the HS surfaces and

corresponding fingerprint plots. The comparability and dissimilarities of intermolecular contacts in

different molecular environments of the three forms are quantified and the resulting histogram is depicted

in Figure 12. Hirshfeld surfaces and the related fingerprint plots together help quantifying intermolecular

contacts conveniently [27].

PA PA

PB

PC

Figure 14. 2D fingerprint plots for A, B and C. The spikes labelled with pink (H···H), blue (C···H),

green (O···H) and red (N···H) triangles depict the characteristic features of the fingerprint plots.

The predominant interactions between HBIH molecules in the three forms can be seen in the Hirshfeld

surfaces as deep red spots obtained as a result of hydrogen bond acceptors of types N−H∙∙∙O, C−H∙∙∙O and

O−H∙∙∙O on the HS of PA and due to hydrogen bond acceptors of types N−H∙∙∙N, C−H∙∙∙O, O−H∙∙∙O,

O−H∙∙∙N and C−H∙∙∙N on the HS of PB. O−H∙∙∙N and N−H∙∙∙O, C−H∙∙∙O, O−H∙∙∙O interactions associated

with lattice H2O significantly contribute to the HS of PC (Figure 13).

H···H, C···H, N···H and O···H interactions play a major role covering 95%, 82% and 88% of the total

HS areas in the case of PA, PB and PC respectively. The H∙∙∙H interactions, which are mirrored at the

center of the scattered points cover most area in the 2-D fingerprint plots of PB and PC, and therefore are

more responsible for stabilization of the crystal packing. (Figure 14). The distributions of de and di

distances are very similar for the N∙∙∙H and O∙∙∙H interactions which appear as a pair of very sharp spikes

of almost equal length in three forms. The most significant differences between the two polymorphs are

found for the C∙∙∙H (ca. 21% more in PA than PB) and for H∙∙∙H interactions (ca. 13.5% more in PB than

PA). Additionally, the C∙∙∙H and H∙∙∙C wing patterns illustrated in the FP plots are less symmetric for PB

than for form PA (Figure S3). The contribution of C∙∙∙C interactions to the total Hirshfeld surface of PB

(ca. 8.1%) is apparently larger and shore up the presence of a face-to-face stacking interaction between

the molecules. Evidently, distributions of intermolecular contacts visualized in the form of the fingerprint

plots, symmetry of such a distribution, and density of the points, seem to be perfectly suited for

characterization of interactions in the crystal phase [9].

A B C

3.6.2. Interaction energy calculation

The interaction energy values (Table 4) for all the three forms were explored via a cluster of nearest

neighboring molecules of radius 3.8 Å around the molecule in the asymmetric unit. The pair-wise

individual energy profiles (Eele, Epol, Edis, Erep) and total energy profile, Etot, for each form are given in

Figures 15, S4 and S5. In each cluster, the molecular pairs are uniquely color coded (Figure 16, S6 and

S7). The average interaction energy values calculated for forms PA, PB and PC are very close (-113.8, -

116.8 and -104.95, respectively, for forms PA, PB and PC). Not surprisingly, the values for the

polymorphic forms PA and PB are fairly comparable and form PB appears to be the energetically most

stable form.

Table 4. Interaction Energies calculated from the CrystalExplorer (kJ/mol)

Forms Eelec Epol Edis Erepl Etot Average Etot

PA -154 -39.4 -206.4 233.1 -227.6 -113.8

PB -119.3 -36.3 -215.8 174.1 -233.6 -116.8

PC -116 -35.6 -179.4 153.8 209.9 -104.95

Figure 15. Molecular pairs and the interaction energies (kJ/mol) obtained from energy framework

calculation for PA

Figure 16. Color coding for the neighboring molecules around the asymmetric unit in PB. The asymmetric

unit is shown with atom type color.

The quantitative differences in interaction energy values between the three forms arises due the

differences in crystal packing. Different symmetry independent molecules are shown by different color

codes. Furthermore, we discerned that from the contact interactions in terms of energetics. In form PA the

distance (5.36 Å) between the molecular centroids associated with N(2)–H(2’)···O(2), C(7)–H(7)···O(2)

interaction is slightly lower than the distance (9.08 and 8.83 Å) between molecular centroids associated

with O(1)–H(1)···O(2), O(1)–H(1)···N(1) and C(5)–H(5)···O(1) interactions in PB.

3.6.3. Energy frameworks analysis

The three dimensional topology of interactions for all three forms were constructed based on the obtained

values of interaction energies and the frameworks visualized using CrystalExplorer 17.5. The tube size

(scale factor) used in all the energy frameworks was 80 and the energy threshold (cut-off) value was set to

zero.

It is observed that hydrogen bonding interaction dominates the crystal packing of all three of these

compounds. In order to visualize and analyze these supramolecular features qualitatively, ‘energy

frameworks’ were constructed using CrystalExplorer. The obtained values of interaction energies were

used for the construction of energy frameworks (Figure 17). The pairwise interaction energies in the

crystal structures are represented as cylinders joining the molecules. The radii of these cylinders are set

relative to the strength of the intermolecular interaction [28]. Energy frameworks show that the dispersion

term is the major contribution to the total intermolecular energy in all the forms. Significant variations in

the dimensions of the pillars, crossbars and columns are observed between forms PA, PB and PC. Different

donor-acceptor pairs with different energies bring about the incomparable energy distribution patterns for

these compounds.

Electrostatic term Dispersion term Total interaction energy

PC

c-a

xis

PB

c-ax

is

PA

b-a

xis

Figure 17. Energy frameworks corresponding to the different energy components and the total interaction

energy for forms PA, PB and PC.

Conclusion

To summarize, the preparation and rational analysis of two polymorphic forms and one hydrated form of

p-hydroxybenzaldehydeisonicotinicacid hydrazone have been reported and characterized with different

crystallographic, spectroscopic and thermal methods. The intensive effect of additives for the

crystallization of polymorphs were studied. The solvent effect was also observed in polymorphism. The

comparative crystal structure analysis revealed that the polymorphs have different packing patterns. All

three forms of the compound are slightly different in terms of molecular conformation and consequently

in terms of supramolecular synthons. Hydrogen bonding is the dominating interaction in the construction

of supramolecular architecture. The X-ray structure analysis was supported by a detailed Hirshfeld

surface analysis. Symmetry and the shapes of the distributions of intermolecular contacts have been

analyzed and visualized in terms of the fingerprint plots.

Appendix A. Supplementary data

CCDC 1843609, 1846303-304 contains the supplementary crystallographic data for compound

PA, PB and PC respectively. Copies of this information may be obtained free of charge via

www.ccdc.cam.ac.uk/conts/retrieving.html or from the Director, CCDC, 12 Union Road, Cambridge,

CB2, IEZ, UK (fax: +44-1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).

Acknowledgements

Lincy Tom gratefully acknowledges CSIR, New Delhi, India for financial support in the form of a Senior

Research Fellowship. The authors are thankful to the Sophisticated Analytical Instrumentation Facility

(SAIF), Kochi India, SPAP, M.G. University Kottayam, for single crystal X-ray diffraction

measurements. We thank the School of Physical Sciences, Central University of Kerala for thermal and

PXRD analysis. We also thank SAIF, M.G. University, Kottayam, for Raman studies. JPJ acknowledges

the NSF--MRI program (Grant No. CHE-1039027) for funds to purchase the Rigaku X-ray

diffractometer.

References

[1] G. R. Desiraju, Angew. Chemie - Int. Ed., 46 (2007) 8342–8356.

[2] G. P. Stahly, Cryst. Growth Des., 7 (2007) 1007–1026.

[3] R. Censi and P. Di Martino, Molecules., 20 (2015) 18759–18776.

[4] S. S. Sreekumar, N. Mohan and M. R. P. Kurup, Polyhedron (2017) 6493–6502.

[5] A. M. Healy, Z. A. Worku, D. Kumar and A. M. Madi, Adv. Drug Deliv. Rev., 117 (2017) 25–46.

[6] D. Palanisamy and S. Karuppannan, Procedia Eng., 141 (2016) 70–77.

[7] A. Kons, A. Be and A. Actin, Cryst. Growth Des. 17 (2017) 1146-1158.

[8] A. L. Belladona, A. R. Meyer, N. Zanatta, G. Bonacorso and C. P. Frizzo, CrystEngComm., 18 (2016)

3866–3876.

[9] A. A. Hoser and D. M. Kamin, (2013) 1978–1988.

[10] G. M. Sheldrick, Acta Cryst. C71 (2015) 3-8.

[11] L.J. Farrugia, J. Appl. Cryst. 30 (1997) 565.

[12] C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock, G. P. Shields, R. Taylor, M. Towler and J. Van De

Streek, J. Appl. Crystallogr., 39 (2006) 453–457.

[13] S. Weber, J. Appl. Crystallogr., 32 (1999) 1028–1028.

[14] M. J. Turner, J. J. McKinnon, D. Jayatilaka and M. A. Spackman, CrystEngComm., 13 (2011) 1804–1813.

[15] M. A. Spackman and D. Jayatilaka, CrystEngComm., 11 (2009) 19–32.

[16] M. A. Spackman and J. J. McKinnon, CrystEngComm., 4 (2002) 378–392.

[17] M. J. Turner, S. P. Thomas, M. W. Shi, D. Jayatilaka and M. A. Spackman, Chem. Commun., 51 (2015)

3735–3738.

[18] Q.-L. Deng, M. Yu, X. Chen, C.-H. Diao, Z.-L. Jing and Z. Fan, Acta Crystallogr. Sect. E Struct. Reports

Online., 61 (2005) o2545–o2546.

[19] X. S. Tai, F. Y. Kong and J. Yin, Acta Crystallogr. Sect. E Struct. Reports Online., 63 , (2007) 79–80.

[20] M. Kitamura, CrystEngComm., 11 (2009) 949.

[21] T. Gelbrich, D. S. Hughes, M. B. Hursthouse and T. L. Threlfall, CrystEngComm., 10 (2008) 1328.

[22] N. C. Kasuga, Y. Saito, H. Sato and K. Yamaguchi, Acta Cryst. E71 (2015) 483–486.

[23] B. B. Shrestha, S. Higashibayashi and H. Sakurai, BeNstein J.Org. Chem.10 (2014) 841–847.

[24] M.M. Fousiamol, M. Sithambaresan, V.A. Smolenski, J.P. Jasinski, M. R.P. Kurup, Polyhedron 141 (2018)

60-68.

[25] N. Aiswarya, M. Sithambaresan, S. S. Sreejith, S. Weng and M. R. P. Kurup, Inorganica Chim. Acta., 443

(2016) 251–266.

[26] D. Kuriakose, A. A. Aravindakshan and M. R. P. Kurup, Polyhedron, 127 (2017) 84–96.

[27] K. K. Jha, S. Dutta, V. Kumar and P. Munshi, CrystEngComm., 18 (2016) 8497–8505.

[28] D. Dey, S. Bhandary, S. P. Thomas, M. A. Spackman and D. Chopra, Phys. Chem. Chem. Phys., 18 (2016)

31811–31820.

download fileview on ChemRxivmrp 201.pdf (1.93 MiB)