The crystal structures of three 3-methyl-1H-1,2,4-triazole-5-thiones, including a second polymorph of 4-[(E)-(5-bromo-2-hydroxybenzylidene)amino]-3-methyl-1H-1,2,4-triazole-5(4H)-thione and a redetermination of 4-amino-3-methyl-1H-1,2,4-triazole-5(4H)-thione

The non-H atoms in the molecules of three closely-related 4-amino-3-methyl-1H-1,2,4-triazole-5-thiones are either exactly or very nearly co-planar, and the compounds exhibit hydrogen-bonded supramolecular assembly in two, one or zero dimensions.

The structure of compound (I) was briefly reported a number of years ago (Escobar-Valderrama et al., 1989): however, there are some unexpected features in the reported ISSN 2056-9890 structure, such as the implausibly wide range of the H-C-H angles in the methyl group, spanning the range 89-135 , and this report does not describe any supramolecular interactions. A second report on this compound (Bigoli et al., 1990) did not include H-atom coordinates, while in a third report (Sarala et al., 2006) the structure was refined in space group Pca2 1 . However, a detailed examination of the atomic coordinates in this latter report using PLATON (Spek, 2009) found a 100% fit to space group Pbcm, indicating that an incorrect space group had probably been selected by these authors. Hence none of the previous reports on compound (I) can be regarded as satisfactory. Accordingly we have now taken the opportunity to re-determine the structure of compound (I) and to analyse in detail the effects of the hydrogen bonding. Compounds (II) and (III) were both prepared by condensation of compound (I) with the appropriate aryl aldehyde: crystallization of compound (III) from acetic acid yields a monoclinic polymorph in space group P2 1 /c, whereas crystallization from ethanol has been reported to provide a triclinic polymorph in space group P1 (Wang et al., 2008). However, the unit-cell dimensions and the space group for (I) together confirm that the form of (I) studied here is the same as that in the original report, despite the use of different crystallization solvents, methanol here as opposed to ethanol in the original report. The molecular structure of compound (I) showing the atom-labelling scheme. The non-H atoms all lie on a mirror plane and the H atom sites in the methyl group all have occupancy 0.5. Displacement ellipsoids are drawn at the 30% probability level.

Figure 2
The molecular structure of compound (II) showing the atom-labelling scheme. Displacement ellipsoids are drawn at the 30% probability level.

Figure 3
The molecular structure of compound (III) in the monoclinic polymorph, showing the atom-labelling scheme and the intramolecular O-HÁ Á ÁN hydrogen bond. Displacement ellipsoids are drawn at the 30% probability level. plane. The reference molecule was selected as one lying on the plane at z = 1/4, and the orientation of the methyl group is such that the H atoms of this group are disordered over two sets of sites, all having occupancy 0.5 (Fig. 1). Although the molecules of compounds (II) and (III) lie in general positions, the non-H atoms are close to co-planar in each case: an intramolecular O-HÁ Á ÁN hydrogen bond in (III) ( Table 2) may contribute to this. Thus in compound (II) the dihedral angle between the two ring planes is 6.31 (10) and, of the atoms in the molecular skeleton, the maximum deviation from the mean plane of the skeletal atoms is 0.097 (2) Å for atom N41, with an r.m.s. deviation of 0.072 Å . In compound (III), the dihedral angle between the two ring planes is just 1.9 (4) , and the maximum deviation of any atom from the mean plane of the molecular skeleton is 0.038 (5) Å for atom C26, with an r.m.s deviation of 0.020 Å .
The methoxy C atoms in compound (II) are almost coplanar with the adjacent aryl ring, as indicated by the relevant torsional angles (Table 1), and the deviations of the two atoms from the plane of the aryl ring (C21-C26) are 0.017 (5) Å for atom C231 and 0.125 (5) Å for atom C241. Consistent with this, the pairs of exocyclic C-C-O angles at atoms C23 and C24 differ by ca 10 , as typically found when methoxy groups are co-planar with an aryl ring (Seip & Seip, 1973;Ferguson et al., 1996). Corresponding bond distances within the triazole rings (Table 1) are very similar for all three compounds, as well as for the two polymorphs of compound (III): the values provide evidence for strong bond localization within the ring, with little or no hint of any aromatic-type delocalization, despite the presence of six -electrons in rings of this type.

Supramolecular interactions
In the crystal structure of compound (I) two independent hydrogen bonds (Table 2) of N-HÁ Á ÁS type (Allen et al., 1997) link the molecules into complex sheets, whose formation is readily analysed in terms of two simple one-dimensional sub-structures (Ferguson et al., 1998a,b;Gregson et al., 2000). In the simpler of these two-sub-structures, molecules related by the 2 1 screw axis along (1/2, y, 1/4) are linked by a hydrogen bond involving the ring N-H unit as the donor, forming a C(4) chain running parallel to the [010] direction (Fig. 4). The H atoms of the amino group also act as hydrogen-bond donors, and the effect is to link molecules related by the 2 1 screw axis along (1/2, 1/2, z) to form a chain of edge-fused R 2 2 (10) rings running parallel to the [001] direction (Fig. 5).
À3.2 (4) Numerical data for the triclinic polymorph of compound (III) have been taken from the original report (Wang et al., 2008), but the atom labels have been adjusted to match the systematic labels used for the structures reported here.

Figure 5
Part of the crystal structure of compound (I) showing the formation of hydrogen-bonded chain of edge-fused R 2 2 (10) rings running parallel to the [001] direction,. For the sake of clarity, the H atoms not involved in the motif shown have been omitted. The atoms marked with an asterisk (*), a hash (#), a dollar sign ($) or an ampersand (&) are at z = 0.75, z = À0.25, z = 1.25 and z = À0.75 respectively.

Figure 6
A stereoview of part of the crystal structure of compound (I) showing the formation of a hydrogen-bonded sheet lying parallel to (100). For the sake of clarity, the H atoms not involved in the motifs shown have been omitted.
The combination of these two chain motifs, along [010] and [001] respectively, gives rise to a sheet lying parallel to (100) ( Fig. 6): just one sheet of this type passes through each unit cell, but there are no direction-specific interactions between adjacent sheets. Hence the supramoleuclar assembly of (I) is two dimensional. The N-H bond in compound (II) participates in the formation of a three-centre (bifurcated) N-HÁ Á Á(O,O) hydrogen-bond system, in which the two acceptors are the O atoms of the methoxy groups (Table 2): this three-centre system is markedly asymmetric, but it is planar within experimental uncertainty. The effect of this interaction is to link molecules related by the 2 1 screw axis along (1/4, 1/2, z) to form a C(10)C(11)[R 2 1 (5) chain of rings running parallel to the [001] direction (Fig. 7). Four chains of this type pass through each unit cell, but there are no direction-specific interactions between the chains: in particular, C-HÁ Á Á(arene) hydrogen bonds and aromaticstacking interactions are both absent from the crystal structure. Hence the supramolecular assembly of (II) is one dimensional.
In addition to the intramolecular hydrogen bond in the molecule of compound (III), noted above, there is a single almost linear N-HÁ Á ÁS hydrogen bond in this structure, which links inversion-related pairs of molecules into a centrosymmetric dimer characterized by an R 2 2 (8) motif (Fig. 8). There are no direction-specific interactions between adjacent dimers: as for compound (II), C-HÁ Á Á(arene) hydrogen bonds and aromaticstacking interactions are both absent from the crystal structure of compound (III). Hence the supramolecular assembly in the monoclinic polymorph of (III) is finite or zero dimensional. The supramolecular assembly in the triclinic polymorph was not analysed in the original report (Wang et al., 2008). In fact, inversion-related pairs of molecules are linked by N-HÁ Á ÁS hydrogen bonds to form centrosymmetric R 2 2 (8) dimers, exactly as in the monoclinic polymorph, but in the triclinic form these dimers are linked by an aromaticstacking interaction to form a -stacked chain of hydrogenbonded dimers running parallel to the [111] direction.
Thus for the three structures reported here, the supramolecular assembly in compounds (I), (II) and the monoclinic polymorph of (III) is, respectively, two one and zero dimensional, while for the triclinic polymorph of (III) it is one dimensional.

Database survey
Here we briefly compare the supramolecular assembly in compounds (IV)-(VIII) (see Scheme 2), which all have molecular constitutions which are similar to those of compounds (II) and (III) reported here.
Compounds (IV)  and (V) (Sarojini, Manjula, Kaur et al., 2014) both crystallize in the triclinic space group P1, but they are not isostructural, as they crystallize with Z 0 values of 2 and 1, respectively. However, their supramolecular assembly is rather similar: in the structure of compound (IV), two independent N-HÁ Á ÁS hydrogen bonds link the two molecules of the selected asymmetric unit into a cyclic dimeric aggregate, while in compound (V) inversion-related pairs of molecules are linked by N-HÁ Á ÁS hydrogen bonds to form a cyclic centrosymmetric R 2 2 (8) dimer, analogous to those found in both polymorphs of compound (III). A similar centrosymmetric dimer is observed for compound (VI) ( Part of the crystal structure of the monoclinic polymorph of compound (III) showing the formation of a hydrogen-bonded R 2 2 (8) dimer. For the sake of clarity, the H atoms bonded to C atoms have been omitted. The atoms marked with an asterisk are at the symmetry position (1 À x, 1 À y, 1 À z). (Sarojini, Manjula, Narayana et al., 2014), motifs of this type form part of a ribbon containing alternating edge-fused R 2 2 (8) and R 4 4 (28) rings running parallel to the [210] direction and in which both ring types are centrosymmetric. Finally, compound (VIII), which differs from (IV) in containing an ethyl substituent rather than a methyl substituent, but which crystallizes with Z 0 = 1 in P2 1 /c. rather than with Z 0 = 2 in P1 as for (IV), also contains a centrosymmetric R 2 2 (8) dimeric aggregate .

Synthesis and crystallization
Colourless blocks of compound (I) were grown by slow evaporation, at ambient temperature and in the presence of air, of a solution in methanol. For the synthesis of compounds (II) and (III), to mixtures of 4-amino-3-methyl-1H-1,2,4-triazole-5(4H)-thione (0.01 mol) with either 3,4-dimethoxybenzaldehyde (0.01 mol), for (II), or 5-bromo-2hydroxybenzaldehyde (0.01 mol), for (III), in hot ethanol (15 ml) was added a catalytic quantity of concentrated sulfuric acid, and each mixture was then heated under reflux for 36 h. The mixtures were cooled to ambient temperature and the resulting solid products (II) and (III) were collected by filtration. For (II) and (III), colourless blocks were grown by slow evaporation, at ambient temperature and in the presence of air of solutions in either dichloromethane-methanol (1:1,

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 3. All H atoms, including the disordered methyl H atoms in (I), were located in difference maps. The H atoms bonded to C atoms were then treated as riding atoms in geometrically idealized positions with C-H distances 0.93 Å (alkenyl and aromatic) or 0.96 Å (methyl) and with U iso (H) = kU eq (C), where k = 1.5 for the methyl groups, which were permitted to rotate but not to tilt, and 1.2 for all other H atoms bonded to C atoms. For the H atoms bonded to N atoms in compounds (I) and (II), the atomic coordinates were refined with U iso (H) = 1.2U eq (N), giving the N-H distances shown in Table 2  Computer programs: APEX2 (Bruker, 2007), SAINT (Bruker, 2007), SHELXS97 (Sheldrick, 2008), SHELXL2014 (Sheldrick, 2015) and PLATON (Spek, 2009

Special details
Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic)  where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.60 e Å −3 Δρ min = −0.57 e Å −3 Special details Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.