Cooperativity of Halogen- and Chalcogen-Bonding Interactions in the Self-Assembly of 4 ‑ Iodoethynyl- and 4,7-Bis(iodoethynyl)benzo-2,1,3-chalcogenadiazoles: Crystal Structures, Hirshfeld Surface Analyses, and Crystal Lattice Energy Calculations

Information ABSTRACT: Several new 4-iodoethynyl- and 4,7-bis-(iodoethynyl)benzo-2,1,3-chalcogenadiazoles were prepared, and a comprehensive analysis of the most prominent secondary bonding interactions responsible for the crystal self-assembly was performed using X-ray di ﬀ raction. The presence of both the iodoethynyl and chalcogenadiazole moieties allows an evaluation of the preference of these molecules to aggregate through either chalcogen- or halogen-bonding interactions in the solid state. The crystal structures of the compounds revealed that their solid-state arrangements are in ﬂ uenced by the nature of the chalcogen atom: for the crystals of the thiadiazoles studied, the  C − I ··· N halogen bonds were preferred, whereas in the corresponding 2,1,3-selenadiazole derivatives, the self-complementary [Se ··· N] 2 supramolecular synthons together with the  C − I ··· N halogen-bonding interactions determined the molecular self-assembly. Furthermore, in the case of the bis(iodoethynyl) derivative the crystal structure was additionally in ﬂ uenced by the  C − I ··· π (ethynyl) halogen bond. Hirshfeld surface and 2D ﬁ ngerprint plot analyses were used to demonstrate the intermolecular interactions and intercontact distributions. Also, the total lattice energies were calculated using the CRYSTAL09 and CrystalExplorer programs. They both indicated intermolecular π ··· π interactions as the forces of substantial contribution to energies were also determined using the Hartree − Fock method 72 and the 3-21G basis set 61,73 implemented in the CrystalExplorer17 program. The atomic coordinates used in the calculations were obtained from the crystallographic data.


■ INTRODUCTION
A rational design of functional materials is not possible without a deep understanding of forces responsible for the aggregation of molecules in the solid state. 1,2 Therefore, different classes of noncovalent intermolecular interactions, ranging from very weak contacts, such as van der Waals interactions, through middle-strength π···π interactions, to strong forces, such as hydrogen bonds, are currently receiving a great deal of attention. 3−5 Among them, σ-hole bonds 6 are of special interest, as they are relatively new and have not yet been fully explored. They are schematically represented as X−E···Y, where the positive electrostatic potential occurring in a molecular entity X−E interacts with a nucleophilic region Y in another or the same molecule. 7,8 When the central polarizable atom E represents elements of groups 15−17 of the periodic table, the interactions are referred to as pnictogen (PnB), 9,10 chalcogen (ChB), 11,12 and halogen bonds (XB), 13−15 respectively. Their features and properties are explained in terms of electrostatics, polarization, and dispersion. 16 Without a doubt, halogen bonds have been the most extensively studied, and their importance has been evaluated in materials science, supramolecular chemistry, and biological systems. 1,17−19 Chalcogen bonding, although defined by IUPAC only in 2019, 20 has already found many applications: namely, in crystal engineering, 21 materials chemistry, 22,23 optoelectronics, 24 medicinal chemistry, 25 molecular recognition, 26 and organocatalysis. 27−30 One of the attributes of XB and ChB is their high directionality and strength tunability, which are essential in programing new noncovalent three-dimensional architectures. The main difference between XB and ChB results from the valence electronic structure. On a neutral halogen atom there is at most one σ-hole, 13,31 whereas a chalcogen atom is able to form two σ-hole interactions at the extension of the covalent bonds in which it is involved. 32,33 Representative examples of ChBs are the structures of 1,2,5telluradiazole 3 4 , 3 5 and phenanthro [9,10-c] [1,2,5]telluradiazole, 36 whose molecules organize through cyclic four-membered [Te···N] 2 supramolecular synthons into ribbons in the solid state. In this arrangement, each Te atom forms two bifurcated chalcogen bonds. Other examples include cocrystals of 3,4-dicyano-1,2,5-chalcogenadiazoles with substituted pyridine N-oxides 37 and other Lewis bases. 38 However, double-chalcogen-bond systems are not always observed. This is mainly due to the proximity of both σ-holes, which prevents the coordination of two sterically crowded nucleophiles. 39 The formation of a single X−E···Y interaction during the selfassembly process is more common for lighter chalcogen atoms such as sulfur. A detailed classification of molecular motifs based on chalcogen-bonding interactions is described in the work published by Bonifazi and co-workers. 40 Among the elements recognized as donors of chalcogen bonds, namely sulfur, selenium, and tellurium atoms, the last element forms the strongest ChBs. 41 Nonetheless, compounds containing sulfur and selenium atoms have been the most frequently studied. The main reason for this is that sulfur can participate in all kinds of different interactions: for example, as an acceptor of hydrogen or halogen bonds. Second, sulfur-and selenium-containing molecules are present in living organisms and are of interest for pharmaceutical applications. In turn, tellurium compounds are usually prone to decomposition due to their air and moisture sensitivity. Moreover, their synthesis is often much more challenging.
Most of the XB and ChB molecular motifs reproduced by self-aggregating molecules with high recognition fidelity have been provided by crystal engineering. Molecules in which XB or ChB donors and acceptors are incorporated in the rigid skeleton close to each other, such as in chalcogenazolo [5,4β]pyridines 42 or benzo-2,1,3-chalcogenadiazoles and their derivatives, 43 constitute a highly desirable model for this kind of study. In the absence of any additional functional group, the heterocycle moieties of benzo-2,1,3-chalcogenadiazoles assemble through single-or double-chalcogen-bonding E···N interactions to form dimeric or polymeric structures, and their geometric arrangements can be modulated by the size and position of the substituents.
Continuing our interest in the construction of supramolecular assemblies with the use of ChBs, 44,45 we decided to exploit a group of benzo-2,1,3-chalcogenadiazoles bearing one or two iodoethynyl substituents at the 4-or 4,7-positions (Scheme 1). The self-assembly process of these compounds can be driven by the formation of halogen or chalcogen bonds or both of these interactions.
It is worth mentioning that the simultaneous use of XB and ChB interactions in supramolecular assemblies has been marginally investigated. 45,46 However, during the review procedure of this paper, a closely related paper dealing with the competition between different σ-hole interactions in the crystals of 4,7-dihalogeno-and 4,7-bis(halogenoethynyl)benzo-2,1,3-chalcogenadiazoles was published by Aakeroÿ and coworkers. 47 The iodoethynyl moiety in our model compounds was chosen as a substituent because the iodine atom attached to sp-hybridized carbon tends to form strong halogen bonds. 48 Moreover, due to the location of the substituent close to the nitrogen atom of the chalcogenadiazole ring, the molecules of the title compounds are able to form a self-complementary XB interaction.
The molecules of the designed compounds contain one S or Se atom acting as the ChB donor, one or two iodine atoms as the XB donor, and two nitrogen atoms as acceptors of either ChB or XB (Scheme 2). Thus, upon crystallization, these compounds could aggregate through one of the arrangements shown in Scheme 3.
The chalcogenadiazoles substituted with one iodoethynyl moiety and in the absence of ChB interactions are expected to form the infinite-chain polymers I with the use of XB interactions or dimers II through self-complementary XB bonding. However, the analogous chalcogenadiazoles able to form strong ChB interactions could self-assemble into chain III or double-chain IV polymers held together by both XBs and self-complementary [E···N] 2 interactions. On the other hand, in the case of the chalcogenadiazole molecule bearing two iodoethynyl units in the absence of the ChB it is expected to create the chain polymer V through self-complementary XB bonds or the more complex 2D structure VI held together by XB interactions, whereas cooperation of the XB with ChB interactions would result in the chain structure VII, closely related to that of III, composed of chalcogenazole dimers connected by XB interactions. In order to explore the effect of a simultaneous contribution of XB and ChB interactions on the supramolecular structures of the title benzo-2,1,3-chalcogenadiazoles, we synthesized the compounds shown in Scheme 1 and determined their crystal structures. In the next step, we used Hirshfeld surface and 2D fingerprint plot analyses 49,50 to demonstrate the intermolecular interactions and intercontact distributions. Finally, we calculated the total lattice energies to assess the contributions from individual intermolecular forces to the stability of the crystal lattices. X-ray Diffraction (XRD). Reflection intensities were measured at 130 K using an Oxford Diffraction SuperNova diffractometer equipped with high-flux microfocus Nova CuKα radiation. The data were processed with CrysAlis PRO software. 51 Absorption corrections were made using the implemented methods: analytical 52 or Gaussian in case of [S](CC-I) 2 . The structures were solved using a dual-space algorithm with SHELXT 2018/2. 53 Using Olex2 1.3 54 the structures were refined by full-matrix least squares based on F 2 with SHELXL 2018/3. 53 All of the non-H atoms were refined anisotropically. The Cbound H atoms were constrained to their calculated positions and refined as riding on their pivot atoms with U iso (H) = 1.2U eq (C) or 1.5U eq (methyl C). The ORTEP representations of the asymmetric units of the presented crystals are shown in Figure S1 in the Supporting Information.

■ EXPERIMENTAL METHODS
Powder X-ray Diffraction (PXRD). The XRD patterns of the samples were recorded on a Bruker AXS D2 Phaser diffractometer with Cu Kα radiation (λ = 1.5418 Å). The operating voltage and current were maintained at 30 kV and 10 mA, respectively. The samples were scanned from 5 to 45°2θ. The scans were made with a step size of 0.02°, a counting rate of 2 s/step, and spinning of the sample. The acquired data were analyzed using SigmaPlot for Windows 11.0, Build 11.0.0.77 (Systat Software, Inc., 2008) and are shown in Figure S2 in the Supporting Information.
Theoretical Calculations. The evaluation of the molecular electrostatic potential was conducted using the Gaussian09 package. 55 The geometry optimizations of isolated molecules were carried out at the DFT(B3LYP)/6-31G** 56−60 (and DFT(B3LYP)/3-21G* for the iodine atom 61 ) level of theory. The B3LYP functional is known to produce reliable results for the thermodynamic data and has been proven to provide accurate qualitative results for the molecular electrostatic potential of similar molecules. 62−64 Molecular electrostatic potential (MEP) maps were visualized with GaussView 5.0.8, 65 and then the maximum and minimum values of the electrostatic potential (V S,max and V S,max ) on individual atoms were calculated.
The geometry of all the crystal structures was optimized at the DFT(B3LYP)/6-31G** 56−60 (and DFT(B3LYP)/3-21G* for the iodine atom 61 ) level of theory in the Crystal program (CRYSTAL09 version) 66 prior to further computational analyses. During the optimization procedure the cell parameters were kept fixed while the atom positions were varied. Crystal cohesive energies for the studied crystal systems were calculated at the same level of theory. The results were corrected for dispersion 67−69 and the basis set superposition error (BSSE). 70 Ghost atoms used for the BSSE estimation were selected up to a distance of 5 Å from the considered molecule in the crystal lattice.
The Hirshfeld surfaces and fingerprint plots were generated using the CrystalExplorer17 program. 71 The intermolecular interaction energies were also determined using the Hartree−Fock method 72 and the 3-21G basis set 61,73 implemented in the CrystalExplorer17 program. The atomic coordinates used in the calculations were obtained from the crystallographic data.

■ RESULTS AND DISCUSSION
Electrostatic Potential Maps. In order to acquire a deeper insight into the regions susceptible to potential interaction, molecular electrostatic potential (MEP) calculations were performed for the six investigated compounds ( Figure 1).
The 3D contour maps of the MEP show the expected anisotropic distribution of electron density around the iodine atom with an electron-deficient region at the tip of the C−I bond, a so-called σ-hole. The magnitude of the local maximum of the electrostatic potential (V S,max ) at the iodine atom is slightly lower for CH 3 [S]CC-I and CH 3 [Se]CC-I than for the remaining chalcogenadiazoles. The MEP surfaces additionally reveal the presence of two regions with positive electrostatic potential at the chalcogen atoms. Both σ-holes are more pronounced on selenium atoms than on sulfur atoms, and the local MEP values for Se are approximately 30 kJ mol −1 higher (about one and a half times higher) than for S. Moreover, the positive charge is not concentrated symmetrically on both sides of the chalcogen atom for derivatives bearing one (iodoethynyl) or two (iodoethynyl and methyl) different substituents attached to the benzene ring. The results are in line with previously published data indicating that the V S,max values on the chalcogen atom in chalcogenadiazoles can be tuned by the electronegativity of the substituent and the substitution pattern of chalcogenadiazoles. 16,74 For the compounds studied, the higher energy value of the electrophilic region is located opposite the −CC−I substituent. This suggests that the σ-hole is slightly more prone to interact with a nucleophile. On the other hand, other energy components not included in the MEP may also play a pivotal role.
Crystal Structure Descriptions. 4-Iodoethynylbenzo-2,1,3-thiadiazole ([S]CC-I) crystallizes in the monoclinic space group P2 1 /m (Z′ = 0.5), with half of the molecule in an asymmetric unit located on a mirror plane. The [S]CC-I molecules interact through C−I···N XBs, forming infinite chains (motif I, Figure 2a). The distance between the I and N atoms is 2.907(6) Å, which is 17% shorter than the sum of their van der Waals radii (δ % = 83%), while the valence angle of this bond has an advantageous value of 178.5(2)° (Table 1). Through weak C Ar −H···I hydrogen bonds (Table 1) the chains aggregate into (010) polar sheets where the −CC−I substituents are oriented in the same direction. However, the symmetry center relates the adjacent layers, which leads to the cancellation of the bulk polarity (Figure 2b). The 3D structure is stabilized by π···π stacking interactions between the benzene rings of the [S]CC-I molecules ( Table 2 and Figure S3 in the Supporting Information).
An additional iodoethynyl substituent at the 7-position causes a complete change of the crystal structure of [S](CC-I) 2 . Its crystals belong to the monoclinic crystal system, space group P2 1 /c (Z′ = 1). The molecules arrange themselves in centrosymmetric dimers stabilized by XBs (motif II), forming R 2 2 (12) rings ( Figure 3a). The N···I length is 3.064(8) Å (δ % = 87%), and the valence angle is 165.9(2)°. The geometric parameters of this interaction suggest that it should have  Table 3). This structure has been also reported by Aakeroÿ and co-workers. 47 Surprisingly, the crystal structure of CH 3 [S]CC-I contains none of the predicted supramolecular motifs. Still, as expected, the crystal structure is governed by XB interactions. The C− I···N XBs link CH 3 [S]CC-I molecules in helices extended along [010] with iodoethynyl substituents directed to the interior (Figure 4a). Despite the molecules being twisted in the helices, geometrical parameters of the XB interactions have advantageous values. The I···N distance is ca. 0.5 Å shorter than the sum of van der Waals radii of I and N atoms, and the C−I···N angle has a value of 173.7(1)°( Table 1). The adjacent helices interact through two kinds of stacking interactions. In the first, the benzene ring is placed above the thiadiazole ring ( Figure 4b); in the second, two benzene rings overlap. In such a way, a 3D structure is formed ( Figure 4c and Table 2).   The parameter δ % corresponds to the ratio of the length of secondary interaction and the sum of van der Waals radii of two interacting atoms according to the following equation: δ = × . When E is an H atom with a fixed distance C−H (X−E), the δ % values are not given.   As expected, replacing the sulfur atom with selenium enhances the formation of ChB, and the [Se···N] 2 synthon is formed in the crystal structure of [Se]CC-I. Z-shaped dimers (motif IV) are connected by XBs into [1−10] chains ( Figure  5a). These 1D motifs have corrugated edges, and adjacent chains are connected by following the groove and ridges principle. The benzene ring of one chain enters the groove of the other, thus forming a (11−1) layer. This arrangement is supported by weak C−H···I and C Ar −H···π(ethynyl) interactions (Figure 5b and Tables 1 and 3). π-stacking interactions between the benzene and selenadiazole rings from the neighboring sheets (Table 2)  The CH 3 [Se]CC-I crystal also belongs to space group P1̅ but has two molecules, A and B, in the asymmetric unit (Z′ = 2). These molecules, through the ChBs, form the [Se···N] 2 synthons. However, in this case, this motif is noncentrosym-    (Figure 6b). Still, weak C−H···I and C Ar − H···π(ethynyl) interactions are observed to support the creation of (100) sheets (Tables 1 and 3). In this case, adjacent layers are not only shifted as in [Se]CC-I but also inverted through the symmetry center. This allows the overlapping of whole molecules (Table 2 and Figure S5 in the Supporting Information). The main difference in the crystal structures of [Se]CC-I and CH 3 [Se]CC-I arises from the use of different σ-holes located on the selenium atoms. The molecules of the methyl derivative form [Se···N] 2 synthons using the electrophilic region, with a slightly higher positive electrostatic potential being found opposite the ethynyl moiety (Figure 1, motif III), whereas [Se]CC-I molecules utilize a σ-hole with a lower value of V S,max on the same side as the iodoethynyl group (Figure 1, motif IV).
Despite many attempts to crystallize [Se](CC-I) 2 , we were unable to obtain good-quality crystals suitable for X-ray diffraction. Table 4 provides crystal data for the studied compounds.
Hirshfeld Surface Analysis and Fingerprint Plots. Intermolecular interactions influencing the packing of the molecules in crystals were inspected using a Hirshfeld surface (HS) analysis. 49,50 The molecular HS for each chalcogenadiazole studied generated from the CrystalExplorer17 program are presented in Figure 7. We used normalized contact distances (d norm ) parameters 50 and 2D fingerprint plots to identify noncovalent interactions. In this concept, the color scale reflects intermolecular distances. Spots on the Hirshfeld surface correspond to distances between two considered atoms: bright red denotes shorter than, white equal to, and blue longer than the sum of their van der Waals (vdW) radii.
In     Table S1 in the Supporting Information.
It is worth pointing out that there is no correlation between the percentage contribution of specific contacts to the Hirshfeld surface and their strength. Taking into account the fact that supramolecular motifs are generally formed as a result of secondary bonding interactions, in analyzing the HS we particularly focused our attention on N···I contacts, which represent halogen interactions, and N···S or N···Se contacts, which represent ChBs (Figure 8). Both of them are directional and often play a significant structure-directing role in the selfassembly of molecules. From the plots we can see that the N···I contacts comprise no more than 14% of the total surface area for each compound and reach a maximum value of 13.8% for [S](CC-I) 2 . It is important to note that in the crystal network of [S]CC-I out of the two possible N···I or N···S contacts only halogen interactions appear and comprise 7.5% of the whole surface. In all five compounds, the sum of the internal and external distances (d i + d e ) for N···I is located at around 3.0 Å, which is less than the sum of the vdW radii of N (1.55 Å) and I (1.98 Å) atoms, indicating the strong nature of these contacts. Lattice Energies and Energy Frameworks. The magnitude of the forces that hold the molecules in a crystal together is one of the most important thermodynamic parameters of a crystalline solid (having an effect on many of its physicochemical properties). Thus, we decided to investigate the overall interactions that contribute to the energy of a crystal lattice. The total lattice energies (E L ) for all five studied crystals were calculated using the CRYSTAL09 and CrystalExplorer17 programs, and the results are shown in Table 5.     Table 5).
The contributions of the electrostatic (Coulombic) and dispersion components to the total energy within the studied crystals are visualized in the energy frameworks ( Figure 9). The strongest electrostatic forces are depicted as wide cylindrical red tubes. In [S]CC-I, they are located between the adjacent molecules, which contact via C−I···N bonds, generating infinite straight chains. The dispersion energy represented by the green bars forms a more complex structure. The larger bars are arranged in a zigzag pattern along the b axis connecting the neighboring layers. They mainly reflect the π···π stacking interactions of aromatic rings that are 3.545 Å apart ( Table 2).
The topology of the electrostatic energy frameworks for [Se]CC-I is a parallelogram formed by two intermolecular Se···N ChBs (stronger, wider cylinders) and two C−I···N XBs (weaker, thinner tubes), which are slightly shifted in the same direction between the neighboring layers. The dispersion energy framework shows that the most robust interactions are located between the overlapping aromatic rings of the molecules in the adjacent layers spreading along the a axis. Larger red tubes in the crystal lattice of [S](CC-I) 2 are visible in the dimers. They correspond to two C−I···N XBs stabilizing the R 2 2 (12) supramolecular motif. Smaller tubes present in the single layer as well as between them form a Figure 9. Energy frameworks for all benzo-2,1,3-chalcogenadiazole derivatives. Electrostatic (red), dispersion (green), and total (blue) interaction energies are shown as cylindrical tubes whose diameters are proportional to the magnitude of the energies. The tubes were adjusted to the same scale factor of 75 with a cutoff value of 5 kJ mol −1 .
fused-triangle topology, which represents the C−I···π XB. In the dispersion energy, frameworks with the strongest interactions are visible between the aromatic rings of the adjacent dimers and they form a zigzag pattern along the a axis. In CH 3 [S]CC-I, the dispersion component forms zigzag chains in which cylindrical green tubes connect the aromatic rings of the molecules belonging to two interpenetrating helices. The Coulombic forces also generate a zigzag that runs along the helix axis. They reflect the C−I···N interactions between the adjacent molecules within the helix. The energy frameworks for the crystal structure of CH 3 [Se]CC-I are entirely different from those for CH 3

■ CONCLUSIONS
In summary, we presented the results of experimental crystallographic studies in conjunction with quantum-chemical calculations for a series of novel iodoethynyl-substituted benzo-2,1,3-chalcogenadiazoles, which have the potential to aggregate during the crystallization process through either halogen-or chalcogen-bonding interactions or both. Considering the obtained results, we came to the conclusion that the chalcogen atom in these compounds is a primary factor determining the type of the observed interactions and consequently is responsible for the organization of the molecules in the solid state. In the thiadiazoles studied, the C−I···N halogen bonds are the leading directional interactions that determine the emergence of specific motifs in the crystal: chains, dimers, and helices for [S]CC-I, [S](CC-I) 2 and CH 3 [S]CC-I, respectively. However, in the case of the bis(iodoethynyl) derivative the C−I···π(ethynyl) halogen bonding additionally contributes to the self-assembly of the crystal. Furthermore, in the crystal networks of the analogous selenadiazoles, apart from the C−I···N halogen interactions, self-complementary [Se···N] 2 chalcogen interactions were also observed. The cooperation of these forces organizes the molecules into an array of chalcogen-bonded dimers forming polymeric chain structures with the use of halogen bonds. Their architecture depends on the different σ-holes on the selenium atom used for the formation of the [Se···N] 2 synthon.
The calculated total lattice energies indicate that, although the cooperativity of halogen and chalcogen bonds determine short-range supramolecular order, the contribution of the aromatic π···π interactions is also an important factor that determines the molecular packing of the crystals. For all of the studied compounds, the dispersive components make a higher contribution to the total lattice energies in comparison to the cohesive terms. Furthermore, an analysis of the fingerprint plots showed that the changes in the intercontact distribution observed in the crystal structures depend on the substituent (H, CH 3 , or iodoethynyl) in the benzene ring of the title benzo-2,1,3-chalcogenadiazoles.  The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.