(N,N-Diallyldithiocarbamato-κ2 S,S′)triphenyltin(IV) and bis(N,N-diallyldithiocarbamato-κ2 S,S′)diphenyltin(IV): crystal structure, Hirshfeld surface analysis and computational study

Distinct tin coordination geometries are found in the title compounds, i.e. (C6H5)3Sn[S2CN(CH2C(H)=CH2)2] has a geometry intermediate between square-pyramidal and trigonal–bipyramidal, while the geometry for (C6H5)2Sn[S2CN(CH2C(H)=CH2)2]2 is based on an octahedron with cis-phenyl groups.


Chemical context
Dithiocarbamate anions of general formula À S 2 CNRR 0 , R/R 0 = H, alkyl and aryl, are readily prepared from the facile reaction of an amine with CS 2 in the presence of base. Thus, the number of derivatives which can be prepared is largely dictated by the availability of amines and hence, an enormous range of dithiocarbamate anions are available for complexation to metals/heavy elements. A key interest in developing metal/heavy element compounds of dithiocarbamates relates to their biological potential (Hogarth, 2012). In the context of ISSN 2056-9890 anti-cancer properties, a number of recent reports have described the efficacy of phosphanegold (Jamaludin et al., 2013), zinc (Tan et al., 2015) and bismuth (Ishak et al., 2014) dithiocarbamates, buoyed by the observation that many of these species promote cancer cell death by apoptosis; bismuth derivatives exhibit in vivo anti-tumour activity (Li et al., 2007). Organotin compounds are well known for their anti-cancer potential (Gielen & Tiekink, 2005) and there is a strong body of literature on organotin dithiocarbamates in this context (Tiekink, 2008).
In the past few years, there has been a resurgence of interest in the anti-cancer activity of organotin dithiocarbamates (Khan et al., 2015;Mohamad, Awang, Kamaludin et al., 2016) and very recently, a report on the in vitro cytotoxicity trial of several tin diallyldithiocarbamate compounds was described as well as a preliminary assessment of anti-microbial activity (Adeyemi et al., 2019); some phosphane-gold(I) and phosphane-silver(I) dithiocarbamates are known to be bactericidal based on pharmacokinetic studies (Sim et al., 2014;. The aforementioned report on tin diallyldithiocarbamate compounds (Adeyemi et al., 2019) also presented the first crystal-structure determinations for tin compounds of diallyldithiocarbamate. In a continuation of recent structural studies in this area (Mohamad et al., 2017(Mohamad et al., , 2018aHaezam et al., 2019), herein, two organotin compounds of diallyldithiocarbamate, (C 6 H 5 ) 3 Sn[S 2 CN(CH 2 C(H) CH 2 ) 2 ], (I), and (C 6 H 5 ) 2 Sn[S 2 CN(CH 2 C(H) CH 2 ) 2 ] 2 , (II), have been synth-esized and studied by X-ray crystallography. In addition, the supramolecular associations in their crystals have been evaluated by Hirshfeld surface analyses and computational chemistry.

Structural commentary
The tin atom in (I), Fig. 1, is coordinated by three ipso-carbon atoms of the phenyl groups as well as by an asymmetrically bound dithiocarbamate anion, Table 1. There is a relatively large disparity in the Sn-S separations, i.e. Á(Sn-S) = [(Sn-S long ) -(Sn-S short )] = 0.47 Å , indicating that the Sn-S2 interaction is weak. Evidence in support of this conclusion is seen in the pattern of C-S bond lengths. Thus, the C1-S2 bond involving the less tightly bound S2 atom is about 0.07 Å shorter than the analogous bond with the tightly bound S1 atom. Nevertheless, there is a clear influence exerted by the S2 atom upon the Sn-C bond lengths with the Sn-C31 bond being appreciably longer than the other Sn-C bonds. This is traced to the trans effect exerted by the S2 atom as this forms a S2-Sn-C31 angle 156.01 (5) . It is noted that there is no other (approximate) trans angle subtended at the tin atom in (I). Assuming a five-coordinate, C 3 S 2 , geometry, the range of 168 Haezam et al. [Sn(C 6 H 5 ) 3 (C 7 H 10 NS 2 )] and [Sn(C 6 H 5 ) 2 (C 7 H 10 NS 2 ) 2 ] Acta Cryst. (2020). E76, 167-176 research communications Figure 1 The molecular structure of (I) showing the atom-labelling scheme and displacement ellipsoids at the 70% probability level. 104.14 (7) C21-Sn-C31 107.13 (7) S1-Sn-C11 128.76 (5) S2-Sn-C31 156.01 (5) angles subtended at the tin atom is 65.470 (14) , for the S1-Sn-S2 chelate angle, to the aforementioned trans angle. The value of is a convenient descriptor for the assignment of a five-coordinate geometry, which ranges in value from 0.0 for an ideal square pyramid to 1.0 for an ideal trigonal bipyramid (Addison et al., 1984). The value of the case of (I) is 0.45, which is indicative of an intermediate geometry with a slight tendency towards square pyramidal. On the other hand, should the coordination geometry be considered C 3 S tetrahedral, i.e. the weak Sn-S2 bond was ignored, the range of tetrahedral angles would be 91.01 (5) , for S1-Sn-C31, to 128.76 (5) , for S1-Sn-C11. Finally, it is noted the C1-N1 bond length of 1.330 (3) Å is consistent with significant double-bond character in this bond, which arises from a major contribution of the 2À S 2 C N + (CH 2 C(H) CH 2 ) 2 canonical form to the electronic structure of the dithiocarbamate ligand. A distinct coordination geometry for the tin atoms is noted for (II), Fig. 2, for which two independent molecules comprise the crystallographic asymmetric unit. The tin atom in each molecule is coordinated by two ipso-carbon atoms of the phenyl groups as well as by two asymmetrically bound dithiocarbamate anions, Table 2. There is a disparity in the Sn-S separations, i.e. Á(Sn-S) = 0.19 and 0.11 Å , for the S1-and S3-dithiocarbamate anions of the first independent molecule; the comparable values for the second molecule are similar at 0.21 and 0.11 Å . The disparities in Á(Sn-S) are reflected in the associated C-S bond distances, Table 2. Gratifyingly, the greater differences in C-S bonds, i.e. 0.03 and 0.04 Å for the S1-dithiocarbamate anions of each independent molecule, are correlated with the greater values in Á(Sn-S). The C1-N1 and C8-N2 bond lengths in both molecules are short for the reasons mentioned for (I) above. The C 2 S 4 coordination geometry is based on an octahedron and has a cis-disposition of the ipso-carbon atoms with the more tightly bound sulfur atoms close to being trans. A partial explanation of the lengthening of the Sn-S2 and Sn-S4 bonds relates to the trans-influence exerted by the phenyl substituents approximately opposite the S2 and S4 atoms.
A view of the superimposition of the two molecules comprising the asymmetric unit in (II) is shown in Fig. 3 whereby the Sn1-and inverted-Sn2-molecules are overlapped so that two chelate rings, i.e. (Sn1,S1,S2,C1) and (Sn2,S3,S4,C8), are coincident. This shows there are nontrivial conformational differences between the molecules. While the dihedral angles between the two phenyl substituents are equal within experimental error in the two molecules, i.e. 81.28 (13) and 81.63 (14) , more telling are the angles they form with the respective, cis-disposed chelate rings, i.e. 81.06 (10) and 35.93 (10) for the  and 74.71 (6) for the Sn2-molecule. Differences are also noted in the relative orientations of the allyl substituents. Thus, for the overlapped dithiocarbamate ligands, the N1- The molecular structures of the two independent molecules comprising the asymmetric unit of (II) showing the atom-labelling scheme and displacement ellipsoids at the 70% probability level.  (7) 161.40 (7) S4-Sn-C15 160.59 (7) 159.91 (7) C15-Sn-C21 99.84 (9) 103.34 (9) N1-C2-C3-C4 12.5 (4) 110.9 (4) 114.6 (4) a C5-C6-C7 torsion angle of À122.3 (3) is an outlier with respect to the other torsion angles with the direct equivalent angle for the inverted Sn2-molecule being 13.3 (4) . While the N-C-C-C torsion angles for the second pair of dithiocarbamate ligands are similar, Table 2, there is a misalignment of these ligands as seen in the dihedral angle formed between the chelate rings of 80.98 (5) and 76.55 (6) for the Sn1-and Sn2-molecules, respectively. The difference in coordination modes of the dithiocarbamate ligands and coordination geometries are related, at least in part, to the different Lewis acid strength of the tin atoms, with the Lewis acidity in the triphenyltin species being significantly less than that in the diphenyltin species.

Supramolecular features
The only directional point of contact between molecules based on the distance criteria in PLATON (Spek, 2020) are phenyl-C-HÁ Á Á(phenyl) interactions, Table 3. Here, the (C21-C26) ring is pivotal by donating a C-H atom to a symmetry-related (C31-C36) ring and the same time accepting a phenyl-C-HÁ Á Á(phenyl) interaction from a (C11-C16) ring to construct a linear, supramolecular chain aligned along the a-axis direction, Fig. 4(a). The chains assemble in the crystal without directional interactions between them, Fig. 4(b).
The molecular packing in (II) is also largely devoid of directional interactions. Indeed, the only connections evident are vinylidene-C-HÁ Á Á(phenyl) interactions, Table 4, which serve to link the independent molecules comprising the asymmetric unit into a supramolecular chain aligned along the a-axis direction. In essence, the vinylidene-hydrogen atoms of the Sn1-molecule bridge translationally related Sn2-molecules into a linear chain, Fig. 5(a). The chains pack without directional interactions between them, Fig. 5

Figure 4
Molecular packing in the crystal of (I): (a) supramolecular chain along the a-axis direction sustained by phenyl-C-HÁ Á Á(phenyl) interactions shown as purple dashed lines (non-participating hydrogen atoms have been removed) and (b) a view of the unit-cell contents in projection down the a axis with one chain highlighted in space-filling mode. Table 4 Hydrogen-bond geometry (Å , ) for (II).

D-HÁ
performed with Crystal Explorer 17 (Turner et al., 2017) following literature protocols (Tan, Jotani et al., 2019). The calculations highlight the influence of the discussed C-HÁ Á Á interactions (Tables 3 and 4) as well the short interatomic contacts collated in Table 5. The short interatomic contacts are indicated as diminutive or faint-red spots near the participating atoms on the Hirshfeld surfaces mapped over d norm for (I) and (II) in Figs. 6 and 7, respectively. Further, the donors and acceptors of the intermolecular C-HÁ Á Á contacts for both (I) and (II) are evident as the blue bumps and red concave regions, respectively, on the Hirshfeld surfaces mapped with shape-index property shown in Fig. 8. In the absence of potential hydrogen bonds in (I) and (II), both the blue and red regions corresponding to positive and negative electrostatic potential, respectively, on Hirshfeld surfaces mapped over electrostatic potential in Fig. 9 and arise owing to the polarization of charges towards the participating residues. The overall two-dimensional fingerprint plots for (I) and the individual molecules of (II) are illustrated in Fig. 10(a), and those delineated into HÁ Á ÁH, CÁ Á ÁH/HÁ Á ÁC and SÁ Á ÁH/HÁ Á ÁS contacts are illustrated in Fig. 10(b)-(d), respectively. The percentage contributions from different atom-atom contacts to the Hirshfeld surfaces of (I), Sn1-and Sn2-molecules of (II) are quantitatively summarized in  Molecular packing in the crystal of (II): (a) supramolecular chain along the a-axis direction sustained by methylene-C-HÁ Á Á(phenyl) interactions shown as purple dashed lines (non-participating hydrogen atoms have been removed) and (b) a view of the unit-cell contents in projection down the a axis with one chain highlighted in space-filling mode.

Figure 6
Two views of Hirshfeld surface for (I) mapped over d norm in the range À0.028 to +1.257 arbitrary units.

Figure 7
Views of the Hirshfeld surfaces for (II) mapped over d norm for the (a) and (b) Sn1-molecule in the range À0.026 to +1.372 arbitrary units, and (c) and (d) Sn2-molecule in the range À0.027 to +1.383 arbitrary units. plot delineated into HÁ Á ÁH contacts for (I), Fig. 10(b), a pair of small and proximate peaks at d e + d i $2.2 Å results from the presence of a short interatomic contact between the phenyl-H15 and H24 atoms, Table 5. The presence of a single peak at d e + d i $2.2 Å in the analogous plot for the Sn1-molecule of (II) is due to the short HÁ Á ÁH contact between the phenyl-H17 and H26 atoms. Another short interatomic HÁ Á ÁH contact Views of Hirshfeld surfaces mapped over the electrostatic potential (the red and blue regions represent negative and positive electrostatic potentials, respectively) for (a) (I) in the range À0.032 to +0.043 atomic units (a.u.), (b) the Sn1-molecule in (II) in the range À0.039 to +0.040 a.u. and (c) the Sn2-molecule in (II) in the range À0.038 to +0.047 a.u.

Table 5
Summary of short interatomic contacts (Å ) in (I) and (II) a .

Contact
Distance Symmetry operation     involving the H17 and H18A atoms of the Sn1-molecule and the H9A2 and H13A atoms of the Sn2-molecule, Table 5, are evident as the pair of peaks at d e + d i $2.2 Å and at d e + d i $2.3 Å in the corresponding delineated plot for the Sn2molecule.
The presence of short interatomic CÁ Á ÁH/HÁ Á ÁC contacts in each of (I) and (II), summarized in Table 5, are evident as the forceps-like tips at d e + d i $2.8 Å in Fig. 10(a). Also, the intermolecular C-HÁ Á Á contacts are characterized as a pair of wings in their respective delineated plots shown in Fig. 10(c). The short interatomic CÁ Á ÁH/HÁ Á ÁC contacts in the crystal of (II) appear as a pair of forceps-like tips at d e + d i $2.7 Å for the Sn1-molecule and as two pairs of similar adjoining tips at the same distances d e + d i $2.8 Å for the Sn2molecule in the plots of Fig. 10(c). For (I), in the fingerprint plot delineated into SÁ Á ÁH/HÁ Á ÁS contacts of Fig. 10(d), the short interatomic contacts involving thiocarbamate-S1 and the H2A and H7B atoms are evident as the pair of conical tips at d e + d i $2.9 Å . Similar contacts in the crystal of (II) are also evident as the conical tips at d e + d i $2.9 Å in Fig. 10(d) in the upper and lower regions of the plots for the Sn1-and Sn2molecules, respectively.

Computational chemistry
The pairwise interaction energies between the molecules within the crystals of (I) and (II) were calculated by summing up four energy components, namely the electrostatic (E ele ), polarization (E pol ), dispersion (E dis ) and exchange-repulsion (E rep ) energies, in accord with literature protocols (Turner et al., 2017). In the present analysis, these energies were obtained by using the wave function calculated at the HF/3-21G level of theory. The specific contacts and associated energies are quantitatively summarized in Table 7. An analysis of these energies for (I) and (II) reveals that the dispersive component makes the major contribution to all the specified intermolecular interactions in the crystals of (I) and (II). However, as clearly evident from the relevant interaction energies listed in Table 7 and in the Hirshfeld surfaces mapped over the electrostatic potential of Fig. 9, where intense blue and red regions are apparent around the donors and acceptors, the C-HÁ Á Á contacts in (II) have more significant contributions from the E ele component, in contrast to mainly dispersive contributions in the case of (I).
A further noticeable observation about the strength of the intermolecular interactions from Table 7 is that those intermolecular contacts arising from the same pair of symmetryrelated molecules have the greater interaction energies. The magnitudes of intermolecular energies were also represented graphically by energy frameworks to view the supramolecular architecture of both the crystals through cylinders joining the centroids of molecular pairs using red, green and blue colour codes for the E ele , E disp and E tot terms, respectively. In summary, the images of Fig. 11 highlight the importance of dispersion forces in the crystals of (I) and (II).  Table 7 Summary of interaction energies (kJ mol À1 ) calculated for (I) and (II).

Database survey
relatively large number of organotin dithiocarbamates have been synthesized and investigated by X-ray crystallography (Tiekink, 2008). The coordination geometry described for (I) conforms with literature expectations in that all R 3 Sn(S 2 CNR 0 R 00 ) molecules conform to this structural motif (Tiekink, 2008;Mohamad et al., 2018a). The Sn-S1 bond length in (I) of 2.4749 (4) Å is slightly longer that the average Sn-S short bond of 2.47 Å in all Ph 3 Sn(S 2 CNR 0 R 00 ) structures, while the Sn-S2 bond of 2.9456 (5) Å in (I) is about 0.10 Å shorter than the average Sn-S long of 3.04 Å in these structures. Consistent with these trends, Á(Sn-S) in (I) of 0.47 Å is less than the average Á(Sn-S) value of 0.57 Å calculated from all Ph 3 Sn(S 2 CNR 0 R 00 ) structures. Greater structural diversity is noted for R 2 Sn(S 2 CNR 0 R 00 ) 2 (Tiekink, 2008), including differences in coordination numbers and geometries (Mohamad, Awang, Jotani et al., 2016). Of the now, 17 structures of the general formula Ph 2 Sn(S 2 CNR 0 R 00 ) 2 , nine adopt the cis-C 2 S 4 structural motif exemplified by (II), including the two polymorphs of Ph 2 Sn(S 2 CNEt 2 ) 2 (Lindley & Carr, 1974;Hook et al., 1994). The remaining structures adopt the usual motif for R 2 Sn(S 2 CNR 0 R 00 ) 2 , namely a geometry based on a bipyramidal skewed-bipyramid. Here, the dithiocarbamate ligands coordinate in an asymmetric fashion with the tin-bound phenyl substituents disposed to lie over the weaker Sn-S bonds, exemplified by the two independent molecules comprising the asymmetric unit of Ph 2 Sn[S 2 CN(Me)Hex] 2 (Hex = n-hexyl, -C 7 H 15 ) (Ramasamy et al., 2013). Clearly there is a subtle interplay between the electronic and steric characteristics of the dithiocarbamate ligands and molecular packing effects in determining the structural motif adopted by Ph 2 Sn(S 2 CNR 0 R 00 ) 2 in their respective crystals.

Synthesis and crystallization
All chemicals and solvents were used as purchased without purification. The melting point was determined using an automated melting point apparatus (MPA 120 EZ-Melt). Carbon, hydrogen and nitrogen analyses were performed on a Leco CHNS-932 Elemental Analyzer.
The synthesis of (I) and (II) followed established literature procedures (Awang et al., 2011;Ajibade et al., 2011). For each synthesis, initially, diallylamine (Aldrich; 1.27 ml, 10 mmol) dissolved in ethanol (30 ml) was stirred under ice-bath conditions for 20 mins. A 25% ammonia solution (1 to 2 ml) was added followed by stirring for 30 mins to establish basic conditions. Then, a cold ethanolic solution of carbon disulfide (0.60 ml, 10 mmol) was added dropwise into the solution and stirred for about 2 h.
For (I), triphenyltin(IV) dichloride (Merck; 3.85 g, 10 mmol) dissolved in ethanol (20-30 ml) was added dropwise into the diallyldithiocarbamate solution and further stirred for 2 to 3 h. Next, the precipitate that formed was filtered off and washed with cold ethanol a few times to remove any impurities. Finally, the filtered precipitate was dried in a desiccator overnight. Recrystallization was carried out by dissolving the The energy frameworks calculated for (I) viewed down the a-axis direction showing the (a) electrostatic potential force, (b) dispersion force and (c) total energy. The corresponding plots for (II), viewed down the a-axis direction are shown in (d)-(f), respectively. The energy frameworks were adjusted to the same scale factor of 50 with a cut-off value of 5 kJ mol À1 within 2 Â 2 Â 2 unit cells.
compound in a chloroform and ethanol solvent mixture (5 ml; 1:1 v/v), which was allowed to slowly evaporate at room temperature yielding colourless crystals.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 8. Carbon-bound H atoms were placed in calculated positions (C-H = 0.95-0.99 Å ) and were included in the refinement in the riding-model approximation, with U iso (H) set to 1.2U eq (C). In (II), the maximum and minimum residual electron density peaks of 1.23 and 0.73 e Å À3 , respectively, were located 0.80 and 0.74 Å from the S2 and Sn1 atoms, respectively.  Windows (Farrugia, 2012), DIAMOND (Brandenburg, 2006); software used to prepare material for publication: publCIF (Westrip, 2010). where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.001 Δρ max = 0.82 e Å −3 Δρ min = −0.50 e Å −3 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Bis(N,N-diallyldithiocarbamato-κ 2 S,S′)diphenyltin(IV) (II)
Crystal data where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.003 Δρ max = 1.23 e Å −3 Δρ min = −0.73 e Å −3 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.