Solvatomorphism of 2-(4-Fluorophenylamino)-5-(2,4

Jun 24, 2010 - E-mail: [email protected]. ... Crystal Growth & Design 2018 Article ASAP ... Crystal Growth & Design 2014 14 (5), 2654-2664...
0 downloads 0 Views 5MB Size
Download PDF
DOI: 10.1021/cg1003319

Solvatomorphism of 2-(4-Fluorophenylamino)-5(2,4-dihydroxybenzeno)-1,3,4-thiadiazole Chloride

2010, Vol. 10 3480–3488

Daniel M. Kami nski,*,† Anna A. Hoser,‡ Mariusz Gagos,§ Arkadiusz Matwijczuk,§ Marta Arczewska,§ Andrzej Niewiadomy,† and Krzysztof Wozniak‡ †

Department of Chemistry, University of Life Sciences in Lublin, Akademicka 15, 20-950 Lublin, Poland, ‡Department of Chemistry, Warsaw University, 02-093 Warszawa, Pasteura 1, Poland, and § Department of Biophysics, University of Life Sciences in Lublin, Akademicka 13, 20-950 Lublin, Poland Received March 12, 2010; Revised Manuscript Received May 29, 2010

ABSTRACT: 2-(4-Fluorophenylamino)-5-(2,4-dihydroxybenzeno)-1,3,4-thiadiazole (FABT) is a biologically active compound. It forms planar molecules and cations. Single crystals of the FABTHþ chloride grown from water solutions of different alcohols, such as methanol, propan-2-ol, and butanol, show structural changes induced mostly by hydrogen bond interactions with chloride anions and solvent molecules. For structures with the alcohol molecules built in the crystal lattice, the FABTHþ moiety takes the conformation with the o-OH (ortho position) group from the resorcin ring on the same side of the molecule as the sulfur atom in the thiadiazole ring, whereas, in the alcohol free crystals growth from a butanol-water mixture, this group is situated on the other side of the thiadiazole ring. The incorporation of the alcohol molecules into the crystal structures formed by FABTHþ cations strongly depends on their size, and it influences the properties of crystal lattices. In the case of the FABTHþClcrystallized from butanol, the crystal structure consists of columns of FABTHþ cations forming intermolecular channels containing two water molecules and two chloride anions related by centers of symmetry. The crystal structure of FABTHþClcrystallized from methanol is built of two separate layers consisting of FABTHþ cations and methanol and chloride anions repeating periodically. FABTHþCl- crystallized from propan-2-ol forms a 3D structure with separate water and propan-2-ol molecules glued by chloride anions and layers of the FABTHþ cations. The Hirshfeld surface analysis is a very useful tool in identifying subtle differences between the solvates. The DFT computations allow us to estimate the energy difference between the two conformers to be 3.2 kcal/mol and the rotational barrier to be 12.6 kcal/mol.

Introduction Polymorphism, i.e. the existence of a substance in structures characterized by different unit cells, where each of the forms consists of exactly the same elemental composition, and solvatomorphism, when crystal structures of a substance are defined by different unit cells differing in their elemental composition through the inclusion of one or more molecules of a solvent, are two phenomena of paramount importance, particularly for the pharmaceutical industry.1,2 Solvatomorphism is an especially important phenomenon because the active ingredients are delivered mostly as a solid phase. The properties of new solvates, new crystal phases, can vary markedly from those of the primary one.1 Understanding and controlling the solid state properties of pharmaceuticals helps to improve their bioavailability, purification process, stability, and other parameters of drugs. In this study, we present structures of different solvates of biologically active 2-(4-fluorophenylamino)-5-(2,4-dihydroxybenzeno)-1,3,4-thiadiazole chloride (hereafter abbreviated FABTHþCl-) obtained by single crystal X-ray diffraction (see Figure 1). The FABT molecule belongs to a large class of biologically active components, exhibiting antiproliferative and anticancer activities.3,4 Different structural and, as a consequence, electronic effects in the FABT structures can also influence the equilibrium of the keto/enol tautomerism5 possible in this molecule. The tautomerism is important in the fluidity of biological membranes.6 *To whom correspondence should be addressed. E-mail: daniel_kaminski3@ wp.pl. pubs.acs.org/crystal

Published on Web 06/24/2010

In this work, we want to study intermolecular interactions in FABTHþCl- alcohol free crystals (crystallized from butanol), containing water molecules incorporated into a crystal lattice (and hereafter denoted as FABTHþCl-w) and solvated crystals of FABTHþCl- crystallized from methanol and propan-2-ol (hereafter abbreviated FABTHþCl-m and FABTHþCl-pw, respectively). Our aim is to rationalize the physicochemical properties of FABTHþCl-, in particular, to look for factors controlling the conformation of the FABTHþ cation. We will present details of the crystal structures of the solvates and expect that intermolecular interactions modified by different solvent molecules should also influence the arrangement of molecules in 3D crystal lattices. Because the studied compounds are very similar, we want to use the Hirshfeld surfaces7-9 to compare intermolecular interactions in the crystal lattices. The Hirshfeld surface analysis should be very helpful in finding small differences between the solvates,10 and to our knowledge, this is one of the very first applications of Hirshfeld analysis in such studies. Materials and Methods Materials. 2-(4-Fluorophenylamino)-5-(2,4-dihydroxybenzeno)1,3,4-thiadiazole chloride (FABTHþCl-)11,12 (empirical formula C14N3O2FSH10Cl, weight 309 g/mol, and melting temperature of 279-280 C) consists of three parts: resorcin, thiadiazole, and fluorobenzene rings (see Figure 1). Details of the synthetic procedures are described elsewhere.13 Compounds were purified by means of HPLC (YMC C-30 column with a length of 250 mm and internal diameter 4.6 mm). The solvent mixture of acetonitrile/ CH3OH/H2O (72:8:3 by volume) was applied as a moving phase. Then, FABTHþCl - was recrystallized from 96% methanol. r 2010 American Chemical Society

Article

Crystal Growth & Design, Vol. 10, No. 8, 2010

3481

Figure 1. Atom labeling and ORTEP representation of the anisotropic displacement parameters at the 50% probability level for (a) FABTHþCl-w, (b) FABTHþCl-m, and (c) FABTHþCl-pw. In order to remove residuals of the solvents (after evaporation in an N2 atmosphere), samples were placed in a vacuum for 1.5 h. The solvents used were purchased from Sigma-Aldrich Co. All measured FABTHþCl- crystals were obtained from a mixture of different alcohols, such as methanol, propan-2-ol, or butanol, with 2 M HCl as the protonating agent in the ratio of 4:1. Then, FABT was dissolved in such a prepared mixture at 35 C. Crystals were grown at room temperature for 3 weeks. The ratio of alcohol to water in the crystallization mixture (during alcohol evaporation) was no lower than 1:1. Too low a concentration of alcohol leads to crystallization of alcohol free crystals. A small amount of alcohol free FABTHþCl-w crystals with the characteristic dark yellow color also for crystallization from methanol and propan-2-ol mixtures was observed on the bottom of a crystallization flask (effect of water absorption from the atmosphere). The solvates with methanol and propan-2-ol have a light yellow color. In the crystals grown from the butanol/water mixture (even in the ratio 6:1), still only water molecules enter the crystal net. We attempted several times to crystallize the bare salt single crystals, unfortunately obtaining in each case a powder of FABTHþCl-. The same result was observed when FABTHþCl-m crystals were taken out from the crystallization mixture and left under air. X-ray Diffraction. Data collection for single crystals of FABTHþCl-m and FABTHþCl-pw (T=100 K) was carried out on a Bruker AXS KAPPA APEX II ULTRA diffractometer with a TXS rotating molybdenum anode and multilayer optics. Data sets were collected using the omega scan method, with an angular scan width of 0.5 for FABTHþCl-m and 0.3 for FABTHþCl-pw. In both cases the exposure time was 20 s per frame. The data were corrected for Lorentz and polarization effects. Indexing, integration, and scaling were performed with the original Bruker Apex II software.14,15 Multiscan absorption correction was applied using SADABS.16 Data collection for a single crystal of FABTHþCl-w (T = 100 K) was carried out on a single crystal X-ray κ-axis KM4CCD diffractometer17 with Mo KR radiation monochromated by graphite with the use of the omega scan technique. The crystal was positioned 65 mm from a CCD 10241024 pixel camera. The 2θ angle range was extended from ca. 2 up to 57. Each frame was measured at an

0.8 angle interval and a counting time of 40 s. A multiscan absorption correction was applied.18 Data reduction and analysis were carried out with CRYSALIS RED.17 All structures were solved by direct methods19 and refined using SHELXL.20 The refinement was based on squared structure factors (F2) for all reflections except those with very negative F2. Most of the hydrogen atoms were located in idealized averaged geometrical positions (those at the thiadiazole ring); however, the other hydrogens were found from difference electron density maps. Scattering factors were taken from Tables 6.1.1.4 and 4.2.4.2 in ref 21. Table 1 includes experimental details for all measured crystals. Entries 768785-768787 of the CCDC contain the supplementary crystallographic data for FABTH þCl -w, FABTHþCl -m, and FABTHþCl-pw crystals. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc. cam.ac.uk/data_request/cif. Hirshfeld Surfaces. To compare intermolecular interactions in investigated solvates, we used fingerprint plots generated from Hirshfeld surfaces.22 A weighing function is used to define the Hirshfeld surface: P Fi ðr - ri Þ i∈moleculeA P wA ðrÞ ¼ Fk ðr - rk Þ i∈crystal

where Fi is the spherically averaged atomic electron density of the i-th atom in the molecule (centered at point ri) and Fk is the electron density of the k-th atom surrounding a particular molecule in the crystal. This weighing function defines the so-called Hirshfeld surface for molecule A when wA(r) = 0.5 for every point r at the surface. Within the Hirshfeld surface, the promolecule electron density dominates over the procrystal electron density. It is possible to map different properties on Hirshfeld surfaces: properties related to the shape of the surface (e.g., curvedness) and also those connected with distances: de, external distance from the Hirshfeld surface to an atom belonging to the closest molecules outside the surface; di, internal distance from the surface to an atom inside the surface; and

3482

Crystal Growth & Design, Vol. 10, No. 8, 2010

Kami nski et al.

Table 1. Data Collection and Refinement Details for the Measured Structures FABTHþCl-w formula system space group unit cell dimensions a/A˚ b/A˚ c/A˚ R/deg β/deg γ/deg volume/A˚3 density absorption coefficient crystal size theta range no. of reflections no. of independent refls/parameters goodness-of-fit S R (all data)/R [F2o>4σ(F2o)] wR2 (all data)/wR2 [F2o>4σ(F2o)] largest diff. peak and hole

FABTHþCl-m

FABTHþCl-pw

C14H11N3F1S1O3 3 Cl 3 H2O monoclinic P21/c

C14H11N3F1S1O3þ 3 Cl- 3 CH3OH

monoclinic P21/n

C14H11N3F1S1O3þ 3 Cl- 3 H2O 3 C3H7OH monoclinic P21/c

9.0617(9) 12.8358(7) 13.1915(10) 90.00 103.558(9) 90.00 1491.6(2) 1.593 0.426 0.25  0.12  0.10 2.94-28.34 10919 3446/232 0.731 0.0992/0.0363 0.0575/0.0514 0.32; -0.26

7.8639(7) 13.2352(11) 19.5416(17) 90.00 92.082(5) 90.00 2032.6(3) 1.424 0.333 0.31  0.15  0.12 1.86-25.05 19930 3609/287 1.062 0.0935/0.0644 0.1880/0.1724 0.79; -0.40

10.1887(6) 6.4533(4) 29.2134(19) 90.00 91.939(4) 90.00 1919.7(2) 1.446 0.346 0.22  0.14  0.11 2.14-27.32 10302 4316/328 1.038 0.0764/0.0493 0.1081/0.0993 0.56; -0.30

þ

-

Table 2. Bond Distances and Angles for FABTHþ Cations in Different Crystals FABT crystallized from

S1-C8 S1-C7 F1-C1 O1-C12 O2-C14 N1-C7 N1-C4 N2-C7 N2-N3 N3-C8 C1-C2 C1-C6 C3-C2 C4-C3 C4-C5 C6-C5 C8-C9 C9-C14 C9-C10 C10-C11 C11-C12 C12-C13 C13-C14

methanol

propanol

butanol

1.738(4) 1.749(4) 1.368(5) 1.341(5) 1.351(5) 1.347(5) 1.403(5) 1.312(5) 1.372(5) 1.314(5) 1.362(6) 1.377(6) 1.381(6) 1.386(6) 1.398(6) 1.383(6) 1.438(6) 1.404(6) 1.405(6) 1.360(6) 1.402(6) 1.380(6) 1.378(6)

1.718(2) 1.757(2) 1.359(3) 1.355(3) 1.345(3) 1.353(3) 1.403(3) 1.300(3) 1.371(3) 1.318(3) 1.369(4) 1.377(4) 1.389(4) 1.390(3) 1.399(3) 1.381(4) 1.442(3) 1.401(3) 1.406(3) 1.366(3) 1.396(4) 1.384(3) 1.384(3)

1.713(2) 1.751(2) 1.359(2) 1.343(3) 1.341(3) 1.340(3) 1.398(3) 1.297(3) 1.379(2) 1.307(3) 1.356(3) 1.374(3) 1.383(3) 1.379(3) 1.386(3) 1.374(3) 1.425(3) 1.408(3) 1.407(3) 1.360(3) 1.397(3) 1.367(3) 1.373(3)

dnorm, which combines both de and di, each normalized by the van der Waals (vdW) radius for the particular atoms involved in close proximity to the surface. When de and di are calculated for each point of the Hirshfeld surface, a 2D (de vs di) plot called a fingerprint is created.22 As on Hirshfeld surfaces, the closest contacts from a point belonging to the surfaces to a particular atom, both inside and outside the surface, can be illustrated, and one can easily compute relative contributions to the Hirshfeld surface area for the various close intermolecular interactions. All the interactions sum up to 100%, so a percentage of interaction can be estimated.23 The presented figures of Hirshfeld surface and fingerprint plots were prepared in the CrystalExplorer program24, which is using standardized hydrogen atom positions. Computational Details. To check the stability of the observed FABTHþ conformations and the rotational barrier between them, we performed the density functional calculations of the total energy for the isolated molecule of FABTHþ in a vacuum. All geometry and energy calculations were carried out using the Gaussian03 program.25 As the first step, the geometry optimization was done by applying the AM1 semiempirical method. The second optimization step was performed at the B3LYP level of theory with the 6-31þþG (2d, 2p) basis set and standard convergence criteria.

Results and Discussion Geometry of FABTHþ. The bond lengths and valence angles are very similar in all FABTHþ cations crystallized from different solvents (see Table 2). In fact, the most spectacular difference in the geometric parameters of the FABTHþ cations regards its conformation. It appears that in the FABTHþCl-w hydrate the resorcin ring is rotated in the opposite direction with respect to the thiadiazole ring compared to the other solvates. The S1C8C9C14 torsion is equal to 176.4 in FABTHþCl-w and -3.7 and -1.7 in FABTHþCl-m and FABTHþCl-pw, respectively. This is caused by hydrogen interactions of the o-OH group with the surrounding solvent molecules. Additionally, the fluorobenzene ring is slightly rotated with respect to the thiadiazole ring with the dihedral angle C7-N1-C4-C3 of 2.5 in FABTHþCl-w and -1.6 and -9.9 in FABTHþCl-m and FABTHþCl-pw, respectively. The larger value of this torsion angle in FABTHþCl-m could be due to asymmetric interactions in the stacks. The N2-N3 bond length (on average 1.374 A˚) in the thiadiazole ring cation confirms the single character of this bond, with a similar bond length as that of the singular N-N bond in pyrazole (1.366 A˚26). The C-S bonds are in the range from 1.713 A˚ to 1.757 A˚, as can be expected for single bonds of this type (ref 1.751 A˚ in ref 26);they are asymmetric, as can be seen in the values included in Table 2. The primary reason for this asymmetry in the C-S bond lengths of the thiadiazole ring is the localization of charge resulting from protonation of the ring nitrogen atom. This protonation is a cause of the redistribution of electron density in the ring. This effect is smaller in the case of the FABTHþCl-m due to larger errors of structural parameters;in this case it is within the level of errors. The aromatic C-C bond lengths are typical (in the range from 1.36 A˚ to 1.41 A˚), although on average, they are slightly shorter in the fluorobenzene ring than in the resorcin fragment. The full table of the valence angles is attached in the Supporting Information (Table S1). Packing and Intermolecular Interactions. FABTHþCl-w. Packing of molecules and their weak interactions in the crystal structure FABTHþCl-w are illustrated in Figures 2 and 3. The crystal lattice of this compound consists of the FABTHþ cations arranged in separate columns (see Figure 2).

Article

Crystal Growth & Design, Vol. 10, No. 8, 2010

3483

Figure 2. Packing of molecules in the FABTHþCl-w crystal lattice: projection (a) along the a axis or (b) along the c axis. Bigger green dots denote chloride anions, and smaller red ones denote water molecules in the channels.

Figure 3. FABTHþCl-w crystal structure: (a) short intermolecular contacts and HBs in the FABTHþ plane; (b) stacking interactions in FABTHþ columns.

Two types of FABTHþ columns parallel to the a direction are present in the crystal lattice, with molecules from each of them oriented perpendicularly to each other. They are related by the glide c planes. The columns are arranged in such a way that they form channels going through the crystal lattice filled with water molecules and with chloride anions located around the centers of symmetry. The FABTHþ cations in a single column have sandwich-like packing with the opposite orientation of the neighboring molecules, as is the common case for the molecules carrying a nonzero dipole moment. They are located in a given column in a slightly asymmetric manner (see Figure 3), with the interplanar distances between the neighboring molecules equal to 3.2 A˚ and 3.3 A˚. The shortest intermolecular distance between the atoms belonging to the different neighboring molecules is 3.240 A˚ (the N2 3 3 3 C13 close contact). The FABTHþ cations are also shifted along the longest molecular axis in such a manner that the π-electrons from the resorcin rings interact with the π-electrons from the thiadiazole fragment. The oxygen O1 atom from the resorcin ring is hydrogen bonded with the H3N hydrogen from the thiadiazole ring (1.99 A˚ ; see Figure 2). The details of all hydrogen bonds and other close contacts present in this crystal structure are shown in Table 3. The hydrogen atom from the o-OH group in the resorcin ring forms a hydrogen bond with the oxygen O1S atom from the water molecule (2.57 A˚; see Figure 2a). The hydrogen atom from the thiadiazole ring and the oxygen atom from the o-OH group form a weak H-bond with the F atom from the

next nearest column of the FABTHþ cations (H3N 3 3 3 F1 2.19 A˚). The hydrogen from the p-OH group, the H1O in the resorcin ring, and the hydrogen H1N from the amine group form bonds with two chloride anions Cl1 (2.24 A˚ and 2.30 A˚, respectively). Additionally, the chloride anion Cl1 interacts with two surrounding water molecules O1S and O1S, forming H1OS 3 3 3 Cl1 and H2OS 3 3 3 Cl1 H-bonds (2.41 A˚ and 2.24 A˚, respectively). The structure is mainly stabilized by ionic interactions between the chloride anions and FABTHþ cations and, additionally, fine-tuned by a network of hydrogen bonds. FABTm. The crystal structure of FABTHþCl-m is built of the FABTHþ cations arranged in layers and separated by methanol and chloride anions (see Figure 4). The FABTHþ cations in a given layer have the opposite tilt compared to the neighboring FABTHþ layers. The neighboring FABTHþ cations in each layer are oppositely oriented in such a manner that the resorcin ring is over the fluorobenzene ring from the previous molecule. Each FABTHþ cation in such a layer is shifted with respect to the previous one by ∼1.5 A˚. The FABTHþ cations in a column interact via π 3 3 3 π interactions between the aromatic parts with an interplanar spacing of ∼3.42 A˚ and 3.38 A˚ (see Figure 5b). The H1O atom from the p-OH group in the resorcin ring forms a hydrogen bond (1.77 A˚) with a methanol molecule through the O2S atom (see Table 4). Similarly, H2O from the o-OH group is H-bonded to the methanol O3S atom (1.53 A˚). The H3N from the thiadiazole fragment forms an H-bond to the third methanol

3484

Crystal Growth & Design, Vol. 10, No. 8, 2010

Kami nski et al.

Table 3. Details of H-Bonding and Other Intermolecular Interactions in the FABTHþCl-w Crystal Lattice D-H 3 3 3 A N3-H3N 3 3 3 O1 N3-H3N 3 3 3 F1 N1-H1N 3 3 3 Cl1 O1-H1O 3 3 3 Cl1 O2-H2O 3 3 3 O1S O1S-H1OS 3 3 3 Cl1 O1S-H2OS 3 3 3 Cl1

symm

d(D-H) [A˚]

x, y, z -x þ 1, y - 1/2, -z þ 3/2 x, y, z þ 1 -x, y - 1/2, -z þ 3/2 x, y, z x, y - 1, z þ 1 -x, -y, -z þ 1

0.86(2) 0.86(2) 0.86(2) 0.87(2) 0.75(3) 0.74(3) 0.89(3)

C11-H11 3 3 3 O1S C13-H13 3 3 3 O1S

-x, y þ 1/2, 3/2 - z x, y, z

C8 3 3 C7 3 3 C13 3

-x, -y, 2 - z -x, -y, 2 - z -x, -y, 2 - z

3 C14 3 C13 3 3 S1

d(H 3 3 3 A) [A˚] 2.00(2) 2.19(2) 2.30(2) 2.24(2) 1.83(3) 2.41(3) 2.24(3)

Weak Hydrogen Bonds 0.93 0.93

2.69 2.55

d(D 3 3 3 A) [A˚] 2.575(2) 2.999(2) 3.152(2) 3.079(2) 2.568(3) 3.116(2) 3.129(3) 3.606(3) 3.203(3)

— DHA [deg] 123(2) 156(2) 170(2) 163(2) 168(2) 159(2) 172(2) 166 127

Distances in Face-to-Face Stacking 3.363(3) 3.328(3) 3.484(2)

Figure 4. Crystal packing of FABTHþCl-m: (a) view along the a axis; (b) projection along the c axis.

Figure 5. FABTHþCl-m crystal structure: (a) short contacts and HBs in the FABTHþ plane and (b) the stacking interactions in FABTHþ columns.

moiety (O1S; 1.84 A˚; see Figure 5). This hydrogen bond is accompanied by weaker H10 3 3 3 O1S interactions (2.55 A˚). All hydrogen atoms from the hydroxyl groups in the methanol moieties are interacting with the chloride anions with distances of H1SO 3 3 3 Cl1 = 2.29 A˚, H2SO 3 3 3 Cl1 = 2.34 A˚, and H3SO 3 3 3 Cl1 = 2.13 A˚. Similarly to the previous structures, the chloride anion interacts with the H1N hydrogen (2.32 A˚). All these interactions are illustrated in Figure 5a. FABTHþCl-pw. The asymmetric part of the unit cell of the FABTHþCl-pw crystals contains one independent FABTHþ cation (Figure 1), one chloride anion, and one

water molecule, all in general positions. The 3D crystal structure of this solvate consists of layers of the FABTHþ cations accompanied by chloride anions (see Figure 6) located between the cations. The FABTHþ layers are arranged in such a manner that the hydrophobic groups (fluorobenzene fragments) are pointing toward the layer of propan-2-ol molecules, whereas the resorcin rings of the FABTHþ moieties are interacting with the layer of water molecules. This structure is an interesting example of segregation of two different solvent moieties into their separate layers in a 3D crystal structure. All important hydrogen bonds and weak interactions in this

Article

Crystal Growth & Design, Vol. 10, No. 8, 2010

3485

Table 4. Details of H-Bonding and Other Intermolecular Interactions in the FABTHþCl-m Crystal Structure D-H 3 3 3 A N1-H1N 3 3 3 Cl1 N3-H3N 3 3 3 O1S O1S-H1SO 3 3 3 Cl1 O1-H1O 3 3 3 O2S O2S-H2SO 3 3 3 Cl1 O2-H2O 3 3 3 O3S O3S-H3SO 3 3 3 Cl1

symm

d(D-H) [A˚]

x, y, z -x þ 1, -y þ 2, -z þ 2 x, y, z -x þ 2, y þ 1/2, -z þ 3/2 x, y, z x, y, z -x þ 2, y þ 1/2, -z þ 3/2

0.88 0.88 0.83(6) 0.89(6) 0.77(5) 1.06(7) 0.97(6)

Weak Hydrogen Bonds C6-H6 3 3 3 O1 x- 1, y - 1, z C10-H10 3 3 3 O1S -x þ 1, -y þ 2, -z þ 2

0.95 0.95

Distances in Face-to-Face Stacking C3 3 3 3 C13 -x þ 2, -y þ 2, -z þ 2 C6 3 3 3 C10 -x þ 1, -y þ 2, -z þ 2

3.345(5) 3.312(5)

d(H 3 3 3 A) [A˚] 2.32 1.85 2.29(6) 1.77(6) 2.34(5) 1.53(7) 2.13(7) 2.6 2.6

d(D 3 3 3 A) [A˚] 3.173(4) 2.695(4) 3.111(4) 2.638(4) 3.101(4) 2.565(4) 3.075(4)

— DHA [deg] 163 161 174(6) 166(5) 176(5) 164(5) 163(6)

3.467(5) 3.458(5)

152.7 160.4

Figure 6. Packing of FABTHþCl-pw: (a) view along the a axis (H-atoms omitted for clarity, the red dots denote water molecules, whereas the green ones denote the chloride anions); (b) view along the b axis.

Figure 7. FABTHþCl-pw: (a) short contacts and HBs in the FABTHþ plane and (b) stacking interactions in FABTHþ columns. Table 5. Details of H-Bonding and Other Intermolecular Interactions for FABTHþCl-pw d(D 3 3 3 A) [A˚] 3.099(2) 2.755(3) 2.597(3) 3.158(2) 3.160(2) 3.162(2) 3.335(3)

— DHA [deg]

0.78(4) 0.90(4) 0.82(3) 0.80(3) 0.77(4) 0.76(4) 0.85(4)

d(H 3 3 3 A) [A˚] 2.322(4) 1.854(4) 1.796(3) 2.367(4) 2.416(4) 2.409(4) 2.525(4)

Weak Hydrogen Bonds C13-H13 3 3 3 O1 1 - x, 3 - y, -z C10-H10 3 3 3 O1S x, y, z C11-H11 3 3 3 Cl1 x, y þ 1, z

0.91(3) 0.92(3) 0.96(3)

2.63(3) 2.40(3) 2.81(3)

3.437(3) 3.298(3) 3.527(3)

148(2) 167(2) 133(2)

Distances in Stacking Face-to-Face C13 3 3 3 C13 1 - x, 2 - y, -z C12 3 3 3 C14 1 - x, 2 - y, -z C1 3 3 3 C7 x, y - 1, z

3.387(5) 3.159(3) 3.345(3)

D-H 3 3 3 A O1-H1O 3 3 3 Cl1 N3-H3N 3 3 3 O1S O2-H2O 3 3 3 O2S N1-H1N 3 3 3 Cl1 O1S-H1OS 3 3 3 Cl1 O2S-H1SO 3 3 3 Cl1 O2S-H2SO 3 3 3 Cl1

symm

d(D-H) [A˚]

x, y þ 1, z x, y, z x, y þ 1, z - 1 x þ 1, y, z x, y, z -x þ 1, -y þ 1, -z þ 1 x þ 1, y, z þ 1

172(3) 176(3) 164(2) 171(3) 163(2) 169(3) 159(3)

3486

Crystal Growth & Design, Vol. 10, No. 8, 2010

structure are illustrated in Figure 7. Again, the FABTHþ cations in a column interact via π 3 3 3 π interactions between the aromatic fragments with a symmetric spacing of ∼3.32 A˚ (see Figure 7b). All FABTHþ cations in a given column have the same orientation (fluorobenzene rings are pointing in the same direction). Each moiety in a column is shifted with respect to the neighboring one along the longest molecular axis by about ∼5.5 A˚ and along the shorter by ∼1 A˚. For this reason, the fluorobenzene ring from the first considered molecule is only partly overlapping with the thiadiazole ring from the nearest one. The H-bonds and other closest contacts for this structure are summarized in Table 5 and Figure 6. Two neighboring cations of FABTHþ interact via C13-H13 3 3 3 O1 H-bonds and form dimers related by the center of symmetry (see Figure 7). The H1N hydrogen from the FABT molecule interacts with the Cl1 (2.37 A˚), and H1O from the p-OH group in the resorcin ring reacts with the other chloride anion (2.33 A˚). The molecular interactions are strengthened by a dense network of hydrogen bonds introduced by water molecules located close to the chloride anions. The H3N hydrogen atom forms a strong H-bond with the O1S oxygen atom from propan-2-ol (1.86 A˚). Also, H1SO from the water molecule interacts with Cl1 (2.41 A˚). The hydrophobic part of propan-2-ol is located close to the hydrophobic part of the FABTHþ cations with its H1OS hydrogen atom from the hydroxyl group pointing toward the chloride anion (2.42 A˚). Hirshfeld Surfaces. The Hirshfeld surfaces were calculated for the FABTHþ cations in each structure. Although in general these surfaces are quite similar, they also exhibit some specific differences induced by interactions with different solvent moieties (see Figure 8). Closer inspection of the Hirshfeld

Figure 8. dnorm (a-c) and curvedness (d-f) mapped on a Hirshfeld surface for the FABTHþ cations in (a and d) FABTHþCl-w, (b and e) FABTHþCl-pw, and (c and f) FABTHþCl-pw.

Kami nski et al.

surfaces also reveals small differences resulting from different conformations of the resorcin ring. The analysis of curvedness mapped on the Hirshfeld surface shows the following differences (Figure 8d-f): the largest region of the flat curvedness appears for FABTHþCl-m. This is due to a complementary overlap of the FABTHþ cations arranged in stacks. From analysis of dnorm mapped on a Hirshfeld surface, one can find that the FABTHþ cation interacts with molecules of different solvents via the same active groups: two OH and two NH groups. One can observe the following similarities for the three studied compounds: first, the o-OH forms hydrogen bonds with the oxygen atoms from surrounding water or methanol molecules, and second, the linking amine NH (N1) always interacts with the chloride anions. However, there are also some differences: the p-OH group of FABTHþCl-pw and FABTHþCl-w interacts with the chloride anion while, in the case of the FABTHþCl-m crystallized from methanol, the O-H 3 3 3 O H-bond is formed to the solvent molecule. In crystals grown from butanol, the N3 nitrogen from the thiadiazole ring interacts with the F atom from the other FABTHþ cation, whereas, for both other solvates, it forms the N-H 3 3 3 O hydrogen bond (with oxygen from methanol or propan-2-ol). Due to these differences, the fingerprint plots for FABTHþ from the above three structures also differ quite significantly (see Figure 9). The 1 in this figure denotes those interactions for which the hydrogen atom is inside the surfaces and the oxygen atoms outside. For FABTHþCl-w, only one hydrogen bond between the o-OH group from the resorcin ring and a water molecule contributes only about 5% to the total amount of interactions in this structure (see Figure 10). For the FABTHþCl-m and FABTHþCl-pw crystals, additionally, the N-H 3 3 3 O interactions are present, and this is why they together constitute ca. 9.5% and 8% of the total interactions, respectively. Number 2 on the fingerprint plots (see Figure 9) points to the H 3 3 3 Cl interactions (H1N inside the Hirshfeld surface). For FABTHþCl-w and FABTHþCl-pw, additional O-H 3 3 3 Cl interactions have to be taken into account. As a consequence, the H 3 3 3 Cl interactions constitute ca. 6% and 5% FABTHþCl-w and FABTHþCl-pw, and only 3% for FABTHþCl-m, respectively. Number 3 corresponds to the O 3 3 3 H interactions (O inside the Hirshfeld surface). Number 4 appears for FABTHþCl-w and denotes the F 3 3 3 H interactions between the H3N and fluoride atom from the next neighboring FABTHþ cation. Number 5 appears only for the FABTHþCl-pw, and it indicates the C-H 3 3 3 C interactions between the solvent and the C6 atom. In general, the H 3 3 3 H interactions dominate in all structures, with their contributions ranging from 30% to 35% of the total amount.

Figure 9. Fingerprint plots for (a) FABTHþCl-w, (b) FABTHþCl-m, and (c) FABTHþCl-pw.

Article

Crystal Growth & Design, Vol. 10, No. 8, 2010

Figure 10. Relative contributions of chosen intermolecular interactions to the Hirshfeld surface area for three solvates of FABTHþ.

Computational Confirmation. The main results of this work are confirmed by DFT calculations for the protonated FABTHþ cation in the vacuum. It appears that the conformation with the o-OH group located on the nitrogen side of the thiadiazole fragment is more stable than the other conformer (with the o-OH group on the sulfur side of the thiadiazole ring; see Table 2S in the Supporting Information). The energy difference between both conformations is equal to 3.2 kcal/ mol. The small energy difference between the two conformers can be easily overcome by strong hydrogen bonds. However, the rotational barrier between the two conformers around the single C8-C9 bond amounts to 12.5 kcal/mol, which is a significantly greater amount. This means that solid state rotational transformation around this bond is very unlikely. Conclusions In this work we present details of three different X-ray structures of solvates of biologically active 2-(4-fluorophenylamino)-5-(2,4-dihydroxybenzeno)-1,3,4-thiadiazole (FABT) chloride (FABTHþCl-) crystallized from butanol, metanol, and propan-2-ol (FABTHþCl-w, FABTHþCl-m, and FABTHþCl-pw, respectively). The main structural difference between the alcohol free structure (FABTHþCl-w) compared to the structures with methanol or propan-2-ol is the orientation of the resorcin ring with respect to the thiadiazole molecular fragment. For crystals with methanol and propan-2-ol, the o-OH group is located on the same side of the molecule as the S1 atom. This conformation seems to be promoted by interactions of the o-OH group with the nearest solvent moieties. However, both conformations are stable and dependent on the presence of hydrogen bonds between the hydroxyl groups, the N-H dipoles, and surrounding chloride anions and methanol and water moieties. In the case of the FABTHþCl-w, the crystal structure consists of columns of FABTHþ cations forming intermolecular channels containing two water molecules and two chloride anions related by the centers of symmetry. The FABTHþCl-m crystal structure is built of two separate layers consisting of the FABTHþ cations (the first layer) and methanol and chloride anions (the second layer). The third crystal structure, FABTHþCl-pw, forms a quite unusual 3D structure with the separate layers of water and propan-2-ol molecules glued by chloride anions. The FABTHþ cations in all structures tend to form stacks. The Hirshfeld surfaces appear to be a very useful tool in the analysis of similar structures. The curvedness mapped on the Hirshfeld surface shows that the largest region of flat curvedness

3487

appears for the FABTHþCl-m. This analysis allows us to estimate the percentage of specific interactions: It appears that in all structures the H 3 3 3 H interactions constitute ca. 30% of all interactions. In the case of the FABTw, the other types of interactions constitute ca. 5-6% of the total amount each— these are H 3 3 3 O, H 3 3 3 Cl, F 3 3 3 H, and C 3 3 3 C interactions. For FABTm, these proportions are different: 11.6% of C 3 3 3 C, 9.5% of H 3 3 3 O, and 6% of O 3 3 3 H, 6% of C 3 3 3 H, and only 3% of H 3 3 3 Cl interactions. In the third structure, FABTpw, one can find 9% of C 3 3 3 H, 8% of H 3 3 3 O, 6.6% of C 3 3 3 C, 5% of H 3 3 3 Cl, and 3.5% of O 3 3 3 H interactions. DFT calculations confirm experimental results, enabling estimation of the energy difference between the conformers of the FABTHþ cation (3.2 kcal/mol) and of the rotational energy barrier (12.6 kcal/mol). Acknowledgment. X-ray single crystal measurements were accomplished at the Structural Research Laboratory of the Chemistry Department, Warsaw University, Poland, established with financial support from the European Regional Development Foundation in the Sectoral Operational Programme ‘‘Improvement of the Competitiveness of Enterprises, years 2004-2006’’, project no. WKP_1/1.4.3./1/2004/ 72/72/165/2005/U. Support from the Foundation for Polish Science for K.W. and A.H. (Mistrz professorship) is greatly acknowledged. Supporting Information Available: Table 1S, showing valence angles in FABTHþCl- solvates; Table 2S, showing results of theoretical calculations for the FABTHþ cation; and submission details to the Cambridge Crystallographic Data Centre. This information is available free of charge via the Internet at http://pubs.acs.org/.

References (1) Brittain, H. G. Polymorphism and solvatomorphism. J. Pharm. Sci. 2009, 98 (5), 1617–1642. (2) Byrn, S. R.; Pfeiffer, R. R.; Stowell, J. G., Solid-State Chemistry of Drugs; SSCI: West Lafayette, IN, 1999. (3) Palekar, V. S.; Damle, A. J.; Shukla, S. R. Synthesis and antibacterial activity of some novel bis-1,2,4-triazolo[3,4-b]-1,3,4-thiadiazoles and bis-4-thiazolidinone derivatives from terephthalic dihydrazide. Eur. J. Med. Chem. 2009, 44 (12), 5112-6. (4) Rajak, H.; Deshmukh, R.; Aggarwal, N.; Kashaw, S.; Kharya, M. D.; Mishra, P. Synthesis of novel 2,5-disubstituted 1,3,4thiadiazoles for their potential anticonvulsant activity: pharmacophoric model studies. Arch. Pharm. (Weinheim) 2009, 342 (8), 453–61. (5) Gagos, M.; Matwijczuk, A.; Kami nski, D.; Niewiadomy, A.; Karwasz, G. P. Spectroscopic studies of 1,3,4-thiadiazoles: intramolecular proton transfer induced by solvent polarizability. J. Phosphoresc., accepted. (6) Rog, T.; Stimson, L. M.; Pasenkiewicz-Gierula, M.; Vattulainen, I.; Karttunen, M. Replacing the cholesterol hydroxyl group with the ketone group facilitates sterol flip-flop and promotes membrane fluidity. J. Phys. Chem. B 2008, 112 (7), 1946–52. (7) Spackman, M. A.; Byrom, P. G. A novel definition of a molecule in a crystal. Chem. Phys. Lett. 1997, 267 (3-4), 215–220. (8) McKinnon, J. J.; Spackman, M. A.; Mitchell, A. S. Novel tools for visualizing and exploring intermolecular interactions in molecular crystals. Acta Crystallogr., B 2004, B60, 627–668. (9) McKinnon, J. J.; Fabbiani, F. P. A.; Spackman, M. A. Comparison of Polymorphic Molecular Crystal Structures through Hirshfeld Surface Analysis. Cryst. Growth Des. 2007, 7 (4), 755–769. (10) Spackman, M. A.; Jayatilaka, D. Hirshfeld surface analysis. CrystEngComm 2009, No. 11, 19–32. (11) Rzeski, W.; Matysiak, J.; Kandefer-Szerszen, M. Anticancer, neuroprotective activities and computational studies of 2-amino1,3,4-thiadiazole based compound. Bioorg. Med. Chem. 2007, 15 (9), 3201–7. (12) Matysiak, J.; Opolski, A. Synthesis and antiproliferative activity of N-substituted 2-amino-5-(2,4-dihydroxyphenyl)-1,3,4-thiadiazoles Bioorg. Med. Chem. 2006, 14 (13), 4483–9.

3488

Crystal Growth & Design, Vol. 10, No. 8, 2010

(13) Niewiadomy, A.; Matysiak, J. The method of synthesis of 2-aryl(alkyl, alkenyl)amino-5-(2,4-dihydroxybenzene)-1,3,4-thiadiazoles; Patent pending P362805; Poland, 2003. (14) Bruker Nonius (2007). APEXII-2008v1.0. (15) Bruker Nonius (2007). SAINT V7.34A. (16) Sheldrick, G. M. SADABS; University of G€ ottingen, Germany, 1996. (17) Oxford Diffraction. CrysAlis CCD and CrysAlis RED; Oxford Diffraction Poland: Wroczaw, Poland, 2001. (18) CrysAlis RED, Version 1.171.33.41 (release 06-05-2009 CrysAlis171 .NET) (compiled May 6 2009,17:20:42) Empirical absorption correction using spherical harmonics, implemented in SCALE3 ABSPACK scaling algorithm; Oxford Diffraction Ltd. (19) Sheldrick, G. M. Phase annealing in SHELX-90: direct methods for larger structures. Acta Crystallogr., A 1990, A46 (6), 473–478. (20) Sheldrick, G. M. SHELXL93, Program for the Refinement of Crystal Structures; University of G€ ottingen: Germany, 2003. (21) International Tables for Crystallography; Wilson, A. J. C., Ed.; Kluwer: Dordrecht, 1992; Vol. C. (22) Spackman, M. A.; McKinnon, J. J. Fingerprinting intermolecular interactions in molecular crystals. CrystEngComm. 2002, 4, 378–392. (23) McKinnon, J. J.; Jayatilaka, D.; Spackman, M. A. Towards quantitative analysis of intermolecular interactions with Hirshfeld surfaces. Chem. Commun. 2007, 3814–3816.

Kami nski et al. (24) Wolff, S. K.; Grimwood, D. J.; McKinnon, J. J.; Jayatilaka, D.; Spackman, M. A. Crystal Explorer, 2.0 ed.; University of Western Australia: Perth, 2005-2007 (http://hirshfeldsurface.net/CrystalExplorer). (25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, J., T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision C.02; Wallingford, CT, 2004. (26) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1–S19.