Thiazolidinone derivatives play crucial roles in anticancer (breast cancer, lung cancer and leukaemia) drug discovery process.¹ Especially, 5-arylidene-2-aminothiazolidinones are found to show antimitotic activities against MCF7 breast cancer cells.² However, bioactive molecules are known to undergo polymorphic modifications under certain conditions and polymorphs often exhibit distinct physicochemical and biopharmaceutical properties.³ Indeed, polymorphism has been the critical issue in drug development process.⁴ Therefore, systematic characterization of polymorphism in bioactive molecules is highly essential. Here, we report the discovery of polymorphism on 5-arylidene-2-aminothiazolidinones derivatives and their systematic characterizations via single-crystal X-ray diffraction, thermal analyses and spectroscopic analysis. The estimation of energies in terms of interaction energies, lattice energies and energy frameworks bring out the energetic stabilities of the polymorphs. Solubility, dissolution rate and phase stability experiments confirm that the thermodynamically most stable form exist with least solubility and dissolution rate. Further, we investigated the extent of inhibition imposed by our library of polymorphs on the proliferation of MCF7 breast cancer cell lines and also the extent of their binding to the target enzyme (g-enolase). Our preliminary cellular assay suggest that in general specific polymorphic forms are indeed more potent than the corresponding bulk form. The g-enolase binding assay also demonstrated a trend similar to that of the phenotypic screening against MCF7. Furthermore, the binding affinity of the polymorphs with g-enolase as estimated via molecular dynamic simulations is in well agreement with the binding assay results. The aforementioned experiments in general emphasized the importance of polymorphism in improving the biological potency of the molecules. We believe that the studies of this kind would help screening potent drug molecules in pharmaceutical industries.

Last decades there is an ongoing interest in materials with reversible temperature-triggered single-crystal-to-single-crystal phase transitions (SCSC-PT) because of the huge fundamental attraction to this phenomenon as well as the promising practical applications of SCSC-PT as different kinds of switches (e.g. to switch dielectric properties) and memory-devices [1-2].

On the other hand, hydrogen bonds are considered as the powerful and flexible synthon in the field of crystal engineering particularly to obtain materials possessing phase transitions. Hence, keeping it in mind we have obtained a series of complexes with the general formula [CaZn₂(piv)₆(L)₂], where piv – pivalic acid anion, L – MeOH, EtOH, THF.

Structures of all coordination complexes were unambiguously determined using single-crystal X-Ray Diffraction (XRD) analysis. All compounds form supramolecular 1D-chains either via hydrogen bonds (in the case of L = MeOH and EtOH) or via CH…O-contacts (in the case of L = THF). The XRD analysis revealed that complex [CaZn₂(piv)₆(EtOH)₂] has interesting temperature-induced isosymmetric reversible phase transition (about 250K on cooling and about 280K on heating) with large hysteresis loop where ethanol groups’ rotation being as a main driving force of the process. As a result of rearrangement of hydrogen bonds network, molecules of the low temperature phase become considerably bent with the angle Zn…Ca…Zn 166.45° compared to completely linear molecules at room temperature.

On the contrary, complex [CaZn₂(piv)₆(MeOH)₂] does not show phase transitions in the available temperature range 100K – 293K. The geometry of its molecules is similar to the low temperature phase of [CaZn₂(piv)₆(EtOH)₂], but the angle Zn…Ca…Zn is 167.30° at 150K. Noteworthy, even though the XRD analysis did reveal rotation of one of the MeOH groups, the cell parameters are roughly the same between structures obtained at RT and 150K. Apparently, because of much less steric hindrance and spatial degrees of freedom of MeOH groups, the temperature of SCSC-PT lies much higher beyond 293K to be revealed.

[1] Ge, J.-Zh, Fu, X.-Q., Hang, T., Ye, Q., & Xiong R.-G., (2010). Cryst. Growth Des., 10, 8, 3632.

[2] Kendin, M. & Tsymbarenko, D. (2020). Cryst. Growth Des., 20, 5, 3316.

Ferroelectric effects are observed in wide variety of materials such as composite ceramics, solids, polymers, crystals and liquid. Various inorganic compounds, such as, barium titanate (BaTiO3), potassium dihydrogen phosphate (KDP), Lithium niobate (LiNbO3), lead zirconate titanate (PZT) etc are well known for decades. On the other hand, organic ferroelectrics in recent times have attracted considerable attention in the material science due to their potential applications as ferroelectric random access memories (FeRAM), non volatile random access memories (NVRAMs) and dynamic random access memories (DRAMs) etc. However, single component organic compounds that show multifunctional properties have hardly investigated and very few reports are available in the literature.

In the present investigation, we report the crystal structures of two N-Benzylideneaniline analogues (BOA and HBOA) and analyse their hydrogen bonding interactions, second harmonic generation efficiency, ferroelectric and dielectric properties. Both the molecules are crystallized in a non-centrosymmetric with monoclinic space group P2₁ and stabilized by strong intermolecular interactions through O-H…O and N-H…O hydrogen bonds. SHG activity of BOA and HBOA also confirms their non-centrosymmetric nature which is prerequisite characteristics for ferroelectric materials. SHG efficiency was found to be 68mv and 140mv for BOA and HBOA respectively with respect to KDP (75mv, standard). Further large thermal hysteresis was observed from the DSC scan of BOA and HBOA at the range of 100 ºC and 87 ºC respectively. Furthermore both the compounds have shown significant reversible mechanofluorochromic (MFC) behaviour upon grinding and fuming. The reversible MFC behaviour was confirmed by cognate techniques like powder X-ray diffraction (PXRD), thermal analysis and fluorescence studies. Both the systems were further investigated to demonstrate the structure-property relationship for realizing their ferroelectric behaviour by PUND (positive up and negative down). Interestingly, both the molecules are fascinating ferroelectric behaviour at high electric field and tolerate upto 181-182 kV/cm without any breakdown (upto instrument limitation). The origin of ferroelectricity has been observed due to the proton transfer in both the systems (Shown in Fig. 1). The dielectric loss-factor was measured in a wide range of frequency at room temperature. High dielectric constant and low loss of energy indicate that the materials can be good candidate for energy storage application. The combination of findings provides a potential perspective for designing new organic multifunctional materials for the applications in light displaying and memory devices.

Lapachol (systematic name: 2-hydroxi-3-(3-methyl-but-2-enyl)-[1,4]naphtoquinone) is a 1,4-naphtoquinone, representative of a large class of natural compounds that exhibit a wide range of biological effects, acting as antibacterial, antifungal, antiparasitic, antiviral, antileishmanial and anticancer, among others [1]. The crystal structures of two polymorphs of lapachol LAPA I (triclinic P ̅1, a= 5.960(1), b= 9.569(2), c= 10.679(2) Å, α= 96.82(2), β= 98.32(2) and γ= 90.32(2) °) and LAPA II (monoclinic P2₁/c, a= 6.035(1), b= 9.427(2), c= 20.918(5) Å and β=98.27(2) °) were determined by Larsen et al. at 105 K [2]. In the course of our research on novel synthetic approaches and crystal structure determination of known and new derivatives of lapachol [3], crystals of a new lapachol polymorph were obtained. Lapachol (Fig. 1a) was obtained from an extract of Handroanthus heptaphyllus (Vell.) Mattos (pink trumpet tree or lapacho negro) and purified by recrystallization by slow evaporation from an EtOH:Et₂O mixture to obtain yellow crystalline plates that correspond, mostly, to LAPA II (monoclinic P2₁/c, a= 6.0550(2), b= 9.5769(2), c= 21.2391(5) Å and β= 98.2910(10) °, T=293 K). However, when lapachol was recrystallized from ethyl acetate, large yellow plates also containing lapachol molecules but arranged in a new crystalline form, were obtained. This new polymorph, LAPA III, also belongs to the monoclinic system, with space group P2₁/c and cell parameters a= 9.6134(5), b= 6.0119(3), c= 21.5464(11) Å and β= 96.760(2) ° at 293 K.

Lapachol molecules show the same conformation in the three structures and crystallize forming H-bonded dimers between nearby OH and ketone groups from two molecules related by an inversion center. In the three structures the dimers stack together in a staggered manner to form double layers of molecules with the butenyl moiety pointing outward. These layers seem to be very stable, since the C face of the unit cell is identical in the three compounds with a≈ 6.0(1), b≈ 9.6(1) Å and γ≈ 90.0(4) °. The difference between them lies in the way successive layers stack to form the crystal. In the triclinic LAPA I all the layers stack in identical orientation with alternating butenyl residues from the bottom and top layers (Fig. 1b). The stacking in LAPA II, however, occurs with the next layer rotated 180° and shifted b/2 respect to an axis parallel to b between the layers (z= ½ of the triclinic cell or ¼ of the monoclinic ones). This rotation axis transforms b in the monoclinic axis of the cell and also introduces a c-glide plane normal to b. This makes the two parallel layers different along c, so LAPA II exhibits a doubled c-axis. Additionally to a, b and γ, β (≈ 98.3°) is also conserved between both structures because the addition of symmetry requires no change in the relative position of the layers along b. LAPA III shows exactly the same relation with LAPA I except that the addition of symmetry occurs in the triclinic a-axis, so the conventional monoclinic unit cell given above shows a and b axes exchanged respect to LAPA I. Again, the addition of a 2₁ axis and a c-glide makes a-axis the unique one, and produces a unit cell that conserves a, b, γ (as LAPA II) and the triclinic α angle (≈ 96.7°) transformed to b, and a doubled c-axis (Fig. 1c). Considering this, it is only the stacking of lapachol double layers what determines the polymorphic form. Hirschfeld surface analysis and intermolecular interaction energy calculations were performed to identify the main interlayer interactions to better explain the differences between the crystal structures of the three lapachol polymorphs.

En este trabajo sintetizamos, caracterizamos y estudiamos por difracción de rayos X monocristal y polvo, dispersión Raman y cálculos de mecánica cuántica, la estructura de una serie de bifenilos sustituidos en posiciones 3,3´, 4,4´- con una variedad de grupos R conectados al núcleo de bifenilo a través de átomos de oxígeno (R: metilo, acetilo, hexilo). [1,2] La serie de seis miembros se dividió en dos grupos con diferencias notables en la conformación molecular, así como en los puntos de fusión (a saber ., en el estado sólido tres miembros son estrictamente planos y presentan un pf significativamente más bajo, mientras que los tres restantes están muy retorcidos, con un pf mayor). Así, el objetivo del trabajo es conocer si alguno de los fragmentos moleculares intervinientes ejerce alguna influencia decisiva sobre la planaridad molecular así como sobre la estabilidad térmica de los compuestos.

La conformación molecular, en particular el ángulo de torsión entre anillos aromáticos, se ha estudiado extensamente tanto en estado sólido como líquido. Los resultados muestran que los tres compuestos que aparecen como rigurosamente planos en el sólido (según lo evaluado por Difracción de rayos X de monocristal) presentan en lugar de una conformación retorcida en la masa fundida (como se describe mediante experimentos Raman y / o cálculos de mecánica cuántica). El tema sólido vs fundido sugiere fuertemente que las razones se encuentran en las restricciones del empaque, aunque no es fácil encontrar una explicación sencilla: en algunos casos (como aquellos con las cadenas de sustitución más cortas) las interacciones combinadas no unidas pueden ser señalados específicamente como responsables de los efectos, mientras que en algunos otros (como en los más largos), pueden ser más sutiles y difusos, no atribuible directamente a interacciones específicas. [2] Finalmente, se sugiere una “regla empírica” para el diseño de bifenilos con diferente conformación molecular, basada en la selección del OR utilizado. [2,3]

[1] Zelcer, A., Cecchi, F., Alborés, P., Guillon, D., Heinrich, B. y Cukiernik, FD (2013). Liq.Cryst. , 40 , 1121. [2] Vadra, N., Suarez, SA, Slep, LD, Manzano, VE, Halac, EB, Baggio, RF & Cukiernik, FD (2020). Acta Cryst. B, cumbre. [3] Grineva, OV (2009). J. Struct. Chem. 50 , 727.

Agradecemos el apoyo económico de la Universidad de Buenos Aires (becas UBACyT 20020130100776BA y20020170100512BA) y CONICET (número de beca PIP20110101035 y becas a NV (PhD) y VEM (postdoc)).

Meloxicam (MLX), a widely used anti-inflammatory drug, can crystallize as one of four neat polymorphic forms and one hydrated form, as reported in the original MLX patent [1]. However, only its commercially available form I and hydrated form IV have their crystal structure determined and were obtained in their pure forms by other researchers, with forms II, III and V remaining elusive. Recently, having found an admixture of a small amount of form III of MLX in one of the commercially available drug formulations, Freitas et al. attempted to obtain pure form III by repeating crystallization procedures described in the patent, but all trials remained unsuccessful [2].

In this contribution we describe our efforts to obtain and characterize three elusive polymorphs of MLX using Crystal Structure Prediction (CSP) calculations and various crystallization approaches. Each of the three solid forms required a different approach, among which were meticulous and numerous repetitions of examples described in the patent, application of various crystallization additives and gel phase crystallization. Crystal Structure Prediction calculations were exploited in two different ways. First, they were used to indicate the prevalent hydrogen bonding patterns found in energetically favourable crystal structures of MLX in order to rationalize possible crystallization routes. It was found, for example, that form II, identified on the calculated CSP landscape by comparison of the experimental and theoretical powder X-Ray diffractograms, display different hydrogen bonding (HB) pattern than commercially available form I (Figure 1). This led us to perform crystallization experiments using nitrogen-containing heterocycles as additives [3] and gel phase crystallization with a hope to re-direct the crystallization route from form I to the three elusive polymorphs.

The second way of exploiting CSP calculations was to characterize the obtained polymorphs of MLX, in combination with NMR crystallography approach. All of the elusive polymorphs of MLX crystallized as microcrystalline powders, which excluded a possibility of characterizing them with single-crystal X-Ray diffraction. Instead, solid-state NMR under very fast magic angle spinning conditions and PXRD diffraction were used to guide the calculations, identify crystal structures of the elusive polymorphs on the CSP landscape, as well as to validate the calculated structures. The presented results are a step towards understanding a fine interplay between the crystal structure and the method of crystallization of elusive polymorphs.

[1] Coppi, L., Sanmarti, M. B. & Clavo, M. C. (2003). US patent 2003/0109701 A1. [2] Freitas, J. T. J., Santos Viana, O. M. M., Bonfilio, R. Doriguetto, A. C. & de Araujo, M. B. (2017). Eur. J. Pharm. Sci. 109, 347. [3] Thomas, L. H., Wales, C. & Wilson, C. C. (2016). Chem. Commun. 52, 7372.

This work was financially supported by Polish National Science Center (UMO-2018/31/D/ST4/01995). PL-GRID is is gratefully acknowledged for providing computational resources.

Active research of some organic compounds, which are actual or promising drugs, has led in the last decade to the production of a large number of their new polymorphic modifications [1]. To date, examples of compounds with up to 12 structurally studied modifications are known. From the point of view of crystal chemistry, such highly polymorphic systems are of great interest for studying the relationship between structure and properties, since their composition is fixed. At the same time, in the presence of suitable methods of analysis, one can try to establish the influence of the slightest changes in the geometry of molecules (conformation) or their packing on the properties of the final substances. Such changes inevitably affect interatomic interactions in the crystal structures of these compounds.

To study both chemical bonds and van der Waals interactions, classical crystal chemistry uses a comparison of interatomic distances with different tabulated systems of radii. However, practice shows that this approach is insufficient for such complex objects as highly polymorphic compounds. Another method for studying noncovalent interactions, the Hirshfeld surface method, implies a visual one-by-one comparison of compounds and takes into account only about 95% of the crystal volume, which also leads to many limitations.

In the course of this long-term project, we consider highly polymorphic systems within the framework of the stereoatomic model of crystal structures [2]. The advantage of this model is that it takes into account all possible interactions (chemical bonds and intra- and intermolecular noncovalent contacts) from uniform positions in all 100% of the crystal volume. In addition, this method allows computer programming and, as a consequence, automatic analysis in large data samples, which is especially important given the volume of already accumulated and progressively increasing information about crystal structures.

Many known highly polymorphic systems have already been successfully analysed using the stereoatomic model of crystal structures. This list includes such champions in terms of the number of structurally studied modifications as flufenamic acid (8 modifications) [3], aripiprazole (9 modifications) [4], galunisertib (10 modifications) [5], etc. For all compounds, various parameters of all types of interatomic interactions in crystal structures were calculated, which can be used for subsequent searches for all kinds of relationships. One of the main results, which was first shown using the example of these compounds, not qualitatively, but quantitatively, is that each polymorphic modification corresponds to a unique set of noncovalent contacts, what, in our opinion, is one of the reasons for the existence of polymorphism.

In the course of this project, additional tools for crystal chemical analysis were developed. For example, one of them is the k-Φ criterion, which makes it possible to unambiguously and objectively identify conformational polymorphs [6]. The (RF, d) distributions were introduced in assistance to the k-Φ criterion to get an idea of the relationship between the ranks of faces and the corresponding interatomic distances [5]. Also, a method was developed to visualize changes in noncovalent interactions when changing the geometry of molecules [7].

As a result, all the work done brings us closer to understanding how the differences in the energetics of polymorphic modifications can be explained from the standpoint of individual noncovalent interactions - this is the question posed by Professor J. Bernstein in his book in 2002 [1].

The study was funded by a grant of the Russian Science Foundation (project number 20-73-10250).

[1] Bernstein, J. (2002). Polymorphism in Molecular Crystals. Oxford University Press: New York. [2] Serezhkin, V. N. (2007). Some Features of Stereochemistry of U(VI). In Structural Chemistry of Inorganic Actinide Compounds, edited by S. Krivovichev, P. Burns & I. Tananaev, pp. 31–65. Elsevier Science. [3] Serezhkin, V. N. & Savchenkov, A. V. (2015). Cryst. Growth Des. 15, 2878. [4] Serezhkin, V. N. & Savchenkov, A. V. (2020). Cryst. Growth Des. 20, 1997. [5] Serezhkin, V. N. & Savchenkov, A. V. (2021). CrystEngComm. 23, 562. [6] Serezhkin, V. N. & Serezhkina, L. B. (2012). Crystallogr. Rep. 57, 33. [7] Savchenkov, A. V. & Serezhkin, V. N. (2018). Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 74, 137.

Since the early 2000’s several free-distribution and commercial computer programs have advanced crystal structure determination from powders (SDPD). The free-distribution software WinPSSP [1] consists of a graphical user interface for an essentially unmodified version of PSSP [2], which uses a reconstructed X-ray powder diffraction pattern (a pseudo-pattern) made of correlated peak intensities calculated from the results of a Le Bail fit [3] (FWHM, peak positions, multiplicity factors and squared structure-factor amplitudes). The simulated annealing algorithm [4] is used to locate the asymmetric unit (provided in Cartesian coordinates) in the unit cell, using a space group symmetry candidate, a cost function (quantitatively evaluating the agreement between the diffracted intensity in the pseudo-pattern and that calculated from a large number of trial models), once a set of structural parameters (fragment positions, Eulerian angles and torsional angles) have been defined by the user. The crystal and molecular structures of more than fifty small-molecule organics have been solved with PSSP and WinPSSP. After Rietveld refinement (typically done with GSAS [5]), those have been published in peer-reviewed journals, or the atomic coordinates have been submitted together with reference powder diffraction patterns to the Powder Diffraction File database [6]. In most cases, synchrotron X-ray powder diffraction from capillary transmission geometry has been used.

This work analyses the distribution of various indicators among the structures solved, such as internal, external and total degrees of freedom, space group symmetries, Z, Z’, unit cell volumes, number of atoms in the asymmetric unit (with and without hydrogens) and number of fragments to independently locate. The uses of solid-state characterization techniques providing information that enabled or confirmed SDPD (thermogravimetry, differential scanning calorimetry, NMR, optical microscopy and DFT calculations) is also highlighted. SDPD common impediments (e.g., preferred orientation from laboratory powder diffraction data) will be indicated.

Even though simple structures can be solved in the order of minutes using WinPSSP [2], due to the approximately exponential increase of the number of trial models required while the number of structural parameters increases [2], a challenge for the current search algorithm is to simultaneously and correctly locate many crystallographically independent fragments (typically three or more) using a reasonable number of trial models, comparable with that used in SDPD with small to moderate size asymmetric units. WinPSSP [2] facilitates the relocation of fragments incorrectly positioned in initial runs (as deemed by the user, considering data from varied solid-state characterization techniques and likely chemical bonding), and examples of this capability toward solving large structures will be discussed.

Another work goal is to attract undergraduate students to SDPD, whenever possible reducing the black box use of crystallography software. Guided SDPD examples have been presented in two workshops and will be used in a yearly SDPD workshop at Old Dominion University.

[1] Pagola, S., Polymeros, A. & Kourkoumelis, N. (2017). J. Appl. Cryst. 50, 293-303. [2] Pagola, S. & Stephens, P. W. (2010). J. Appl. Cryst. 43, 370-376. [3] Le Bail, A. (2005). Powder Diffraction 20, 316-326. [4] Kirkpatrick, S., Gellat, C. D. & Vecchi, M. P. (1983). Science 220, 671-680.

[5] Larson, A. C. & Von Dreele, R. B. (2004). General Structure Analysis System (GSAS). Los Alamos National Laboratory Report 86-748.

[6] Gates-Rector, S. & Blanton, T. (2019). Powder Diffraction 34, 352-360.

Keywords: powder diffraction; direct-space methods; simulated annealing; SDPD; crystal structure solution from powders

Use of the National Synchrotron Light Source, Brookhaven National Laboratory, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-98CH10886. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. SP gratefully acknowledges partial funding from the International Centre for Diffraction Data (ICDD) through GIA 08-04 and workshop sponsorship, and partial funding and other support from the Chemistry & Biochemistry Department at Old Dominion University.

Coordination compounds are obtained by the reaction between Lewis acids (metal) and Lewis bases (ligand). According to Pearson’s acid-base theory, also known as HSAB theory, the stability of the compound depends on the hardness of the acid and the base and their affinity. Hard bases tend to react with hard acids and soft bases prefer to react with soft acids. By just varying the hardness of the metal, the ligand or the solvent it is possible to substantially change the structure of the complex, giving a plethora of possibilities for the synthesis of different coordination compounds. This phenomenon was evidenced in coordination compounds of Cu (II) and Zn (II), where changing the ligand from 3,5-dinitrobenzoate to pyrazole derivatives affects the number of metallic centers present after crystallization [1]. Interested in these results we decided to study the effect of the recrystallization solvent on the coordination sphere of the complex [Co(3,5-dinitrobenzoate-O,O’)₂]. Depending on the donor capabilities of the solvent, the complex undergoes a change on its coordination sphere, changing from a trinuclear Co (II) complex when the solvent is a soft base to mononuclear Co (II) complex when the solvent is a hard base. These structural changes are of great interest because materials and molecules that include cobalt in their structure have several spin states, which gives it interesting magnetic properties [2].

Due to its anti-adherent and lubricating properties, magnesium stearate is the most used additive in pharmaceutical products [1]. Most products contain a few percent of magnesium stearate. The compound exists in several hydrated states, and the state of hydration has important consequences for the lubricating functionality [2]. Yet, none of the crystalline phases has been structurally determined despite the extensive use of this compound in pharmaceutical and other industries for over several decades [3]. The reason for that might be that commercially available samples are usually not pure; they contain a significant amount of magnesium palmitate – a stearate homologue differing by two CH₂ groups. Furthermore, it seems to be problematic to obtain large enough single crystals suitable for a conventional X-ray diffraction experiment due to extremely low solubility of this material. We were able to synthesize highly pure magnesium stearate and obtain micrometre size single crystals suitable for a microdiffraction experiment at an X-ray synchrotron facility. The structure of magnesium stearate trihydrate could be determined. This is the first structurally characterized magnesium stearate hydrate phase. Our work could facilitate structure determination of other magnesium stearate phases as well those of magnesium palmitate from PXRD data. The structural data would be immensely useful to understand lubricating and other structure-determined properties of this extensively used material.

[1] Lakio, S., Vajna, B., Farkas, I., Salokangas, H., Marosi, G., Ylirussi, J. (2013). AAPS PharmSciTech. 14, 435-444.

[2] Ertel, K. D., Carstensen, J. T. (1988). J. Pharm. Sci. 77, 625-629.

[3] Bracconi, P., Andrès, C., Ndiaye, A. (2003). Int. J. Pharm. 262, 109-124.