Contrasting the Noncovalent Interactions of Aromatic Sulfonyl Fluoride and Sulfonyl Chloride Motifs via Crystallography and Hirshfeld Surfaces

Abstract

A heteroaryl sulfonyl(VI) fluoride, 4-chloro-7-fluorosulfonyl-2,1,3-benzoxadiazole, was synthesized from its chloride counterpart (4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole) and the X-ray structure analysis of these compounds and the interactions in the solid-state were thoroughly examined. Hirshfeld surface analysis is used to provide a thorough and complete picture of the changes arising from the different halides in the functional groups. Surface analysis reveals that the fluoride does not participate in any hydrogen interactions as opposed to the chloride. However, the fluorine atom is observed to form close interactions with several π bonds. For both moieties, however, the sulfonyl oxygens show comparable interactions with respect to both magnitude and interatomic distances. The Hirshfeld surface analysis is coupled with computational studies to help elucidate the observed interactions that are found from the distinct nitrogen, chlorine, and oxygen atoms present in the molecules, providing new physical insights to the correlation between their structures and properties

Introduction

Aryl sulfonyl(VI) fluorides and aryl sulfonyl(VI) chlorides are the most commonly used S^VI electrophiles, which organic chemists employed to form inorganic sulfate, sulfamate, and sulfamide connective units.^[1] With their unique stability–reactivity pattern due to the strong S–F bond,^[2] sulfonyl fluorides are resistant to hydrolysis and thermolysis, demonstrate selective reactivity with nucleophiles, and are more thermally and chemically robust when contrasted with sulfonyl chlorides.^[3] In contrast to the better-known chloride congeners, sulfonyl fluorides have recently sparked a great deal of interest in the areas of synthetic organic,^[4] biological,^[5] medicinal,^[6,7] and materials chemistry research.^[8,9] For instance, Xue et al. recently reported using Me₃N–SO₂F as an improved electrolyte for high-voltage rechargeable lithium-metal batteries.^[10]

Specifically, as SuFExable compounds, heteroaryl sulfonyl fluorides are particularly valuable in the field of chemical biology.^[11] For instance, leveraging the targeted reactivity of SO₂F moiety, Gu et al. described the use of sulfonyl fluoride probes that label specific proteins in both plant and mouse proteomes.^[12] Furthermore, aryl sulfonyl fluorides were employed in the selective binding of proteins with the additional benefit of serving as fluorescent labels.^[13] Likewise, sulfonyl chlorides are of fundamental importance in chemical synthesis. Despite the many decades of research into the synthesis, of sulfonyl chlorides, there remain potential issues regarding the stability and the late-stage synthesis of the functional groups due to their facile reductive failure of the S^VI–Cl bond, yielding S^IV species and Cl⁻.^[14] Despite the advances in synthesizing aryl sulfur fluorides and their successful applications, fundamental studies on the correlation between their structures and properties have remained unexplored. We recently reported the development of novel sulfonyl fluorides as novel functional ionic liquids^[8] and their structure-property relationships were thoroughly studies via the crystallographic approach.

Although (hetero)aryl–S^VI compounds hubs offer immense scope as a platform for structural diversification via the Sulfur(VI)-Fluoride Exchange (SuFEx) click reactions, their empirical structure-property relationships are vastly underexplored. Therefore, in our ongoing pursuit of the rational design of biologically active compounds, herein we report the first analysis of the crystal structures of 4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole (1) and 4-chloro-7-fluorosulfonyl-2,1,3-benzoxadiazole (2) (Figure 1). The crystallographic analysis provides a direct insight into the intermolecular spatial relationship, providing a basis for an enhanced understanding (and predict) of the physicochemical relationship of these compounds, illustrating the properties of the sulfonyl fluoride vs. the sulfonyl chloride. Despite their vast potentials, benzoxadiazoles have remained a less-studied class of heterocycles, particularly when compared to structurally related benzimidazoles.^[15] The crystal structure of the unsubstituted 2,1,3-benzofurazan was reported by Luzzati in the 1950s.^[16] Later, an in-depth analysis by Ams et al. reported the intermolecular interactions and subsequent properties of the sulfur derivative.^[17] In their studies, they found that the diazole nitrogen atoms could act as Lewis bases, donating electrons to form chalcogen bonds to the central sulfur in the ring making distinctive square-shaped interactions. Additional structure-property studies by Saha included several derivatives of nitrobenzoxadiazoles bearing cycloalkanes of increasing size which were synthesized and characterized, examining how sterics play a role in the structures.^[18]

Figure 1.

The sketches and the asymmetric unit of 1 (left) and 2 (right) shown with 50% probability ellipsoids. Atom labels are shown for clarity.

Having now returned to the subject, we thoroughly examined the solid-state structure of compounds 1 and 2 to gain an understanding of the noncovalent interactions (NCIs) via Hirshfeld surface analysis.^[19] Surface analysis of the compounds reveals multiple unique interactions arising from the electronegative atoms in the ring (N, O, Cl). Further, interactions with the π system of the heterocycle as well as interactions with the halogens are also found to play a role in the long-range ordering of the structures, with halogen⋯π bonding observed for the SO₂F moieties. The observations drawn from the crystal structures help rationalize the changes imparted by the change of a SO₂Cl to a SO₂F moiety. The crystallographic and surface analysis work is supplemented with computational studies, and the results of the theoretical studies correlate to the compounds solidstate structure. In fact, understanding the structural factors that underpin the formation and stability of sulfonyl fluorides is critical for the design and development of the new members of this important family of biologically relevant compounds. Thus, a thorough understanding of all of the interactions within the lattice, through detailed crystallographic analysis, leads to the identification of the dominant features and, thus, assists in the target-specific design of novel compounds with desired physical properties.

Results and Discussion

Synthesis and Thermal Stability Evaluation

Compound 2 is readily synthesized via the aqueous potassium bifluoride-mediated chloride–fluoride exchange (halex) process in 100% yield, using its chloride congener (1) and saturated KHF₂ solution (pH~3.0) as the fluoride source (Scheme 1). As reported previously by Sharpless and coworkers, this process is profoundly effective for converting sulfonyl chlorides to the corresponding sulfonyl fluorides.^[20] Full conversion was achieved within two hours the final product 2 was purified by crystallization in isopropyl alcohol. The formation of the compound 2 was confirmed by ¹H, ¹³C, and ¹⁹F NMR, and by single crystal X-ray diffraction (SC-XRD). Detailed synthetic procedure and NMR spectra can be found in the Supplementary Information (SI). The rationale for selecting the 2,1,3-benzoxadiazole backbone for this study is that ArSO₂F are considerably more resistant to hydrolysis than alkyl derivatives and an electron withdrawing group (chloride) on the aromatic ring increases the electrophilicity of S^VI site, making it more reactive to desirable SuFEx click reaction.^[21]

Scheme 1.

Synthetic route to compound 2 via the on-water sulfonyl chloride-fluoride exchange, employing potassium bifluoride as the fluoride source. Much greater thermostability of sulfonyl fluoride towards thermolysis is observed.

As expected, the compound 2 shows robust thermodynamic stability towards hydrolysis and thermal decomposition (i.e. it was stable at 130°C for three hours, but the chloride analog rapidly decomposes under this condition). However, irreversible reduction to corresponding sulfinic acid (ArS(O)OH) can occur rapidly for 1 in the presence of water. These observations are consistent with the measured bond lengths of SO₂–F compared to SO₂–Cl in their crystal structures, which is inversely related to the bond strength: S–Cl bond (2.0190 (4) Å) is considerably longer than S–F bond (1.5458 (18) Å), indicating chemical and thermal fragility of 1 relative to 2. Our comparative observations are in accordance with the gas chromatography analysis of sulfonyl halides, provided more than 50 years ago.^[22]

X-ray Crystallography

Compound 1 crystallizes in the Pca2₁ space group with a single molecule in the asymmetric unit (Figure 1). Both rings, that is the 5-membered oxadiazole and the benzene ring, are planar due to conjugation. The sulfonyl chloride group is aligned such that one of the oxygen atoms (O3) is approximately coplanar with one of the hydrogen atoms (H3) with a C3–C2–S1–O3 torsion angle of 6.17(16)°. The O3⋯H3 distance of 2.54 Å indicates some amount of intramolecular hydrogen interaction between the methylene hydrogens and substituents on the rings. Similar interactions have been observed in related structures.^[23] The chlorine of the sulfonyl chloride moiety is sitting above the plane of the rings, with a C1–C2–S1–Cl2 torsion angle of 69.63(13)°. The electronic structure of the rings, specifically the location of double bonds, is readily observed in the crystal structure with C2–C3 and C4–C5 distances of 1.359(2) Å and 1.368(2) Å being shorter than the C3–C4 distance of 1.437(2) Å. The other bonds between the nitrogen and carbons, N1–C1 and N2–C6, are 1.318(2) Å and 1.315(2) Å, respectively.

Compound 2 crystallizes in the P2₁2₁2₁ space group with a single molecule in the asymmetric unit (Figure 1). As expected, the overall structure of 2 is very similar to that of 1, with most distances and angles being comparable (see table S1). For example, the fluorine atom also sits above the plane of the rings with a C1–C2–S1–F1 torsion angle of 62.80(2)°. This torsion angle also allows for the intramolecular interaction between the sulfonyl oxygen O3 and the hydrogen H3 at a distance of 2.56 Å, which is nearly the same as that observed in 1. The carbon-carbon and carbon-nitrogen bond distances also follow the same trend as with 1, with C2–C3 and C4–C5 being shorter (1.363(4) and 1.348(4) Å respectively) than C3–C4 (1.438(4) Å). Thus, broadly speaking, there appears to be little influence in the core ring structure of the molecule by changing from a sulfonyl chloride to a sulfonyl fluoride group.

Chemical and Molecular Descriptors

A list of relevant chemical descriptors is provided in table S2. Examining the chemical descriptors for both molecules reveals noteworthy similarities. For example, both molecules’ globularity, asphericity, surface area, dipole moment, and total void space are nearly identical. Close inspection of the location of the calculated voids, however, does lead to important observations (see Figure 2). Compound 1 has two areas of void space: one existing near one face of the N–O–N region of the oxadiazole ring, with the void space residing on the same face as where the Cl2 resides. This region of void space is not present in 2 and could be due to the larger chlorine atom in the sulfonyl chloride moiety occupying more space when compared with the smaller fluorine. The second region of void space is a small region found between two stacked benzene rings. Likely this void region is an artifact present due to observed Cl⋯π interactions being favored over other interactions (vide infra).

Figure 2.

Depictions of the void spaces in the crystal structure of 1 (top) and 2 (bottom). A probe radius of 0.7 Å with a grid spacing of 0.1 Å was used to calculate the voids.

Compound 2 has distinctive void regions. The first region (near N1) could be explained by steric blocking via the SO₂F group. However, N1 in compound 1 participates in specific hydrogen interactions (vide infra). The second void region is located between two symmetry adjacent oxygen atoms (O1 & O2) and is roughly co-planar with the hydrogen atoms. The presence of this void space likely points to repulsive or destabilizing interactions between these oxygen atoms. However, this void region is near the hydrogen atoms which are typically observed forming interactions with electronegative atoms. Thus, it could be rationalized that interactions with the hydrogen atoms are disfavored over other interactions observed in 2 but not in 1 (e.g., F⋯π). It should be noted that the expected hydrogen interactions are observed in structure 1, with both oxygen and nitrogen atoms forming hydrogen interactions in the described region.

Figure S1 helps to visualize the unique hydrogen interactions for both molecules. The presence of distinctive sets of H-interactions in 1 as compared to 2, in addition to N1 not displaying any interactions in 2, could point to the existence of polymorphs for both compounds as the atoms interacting with the hydrogen atoms are present in both molecules leading to a potential for the formation of different sets of interactions. Recent work by our group has shown that these heterocyclic compounds are capable of forming polymorphs which would likely have different sets of interactions.^[24]

Figure S2 visualizes the numeric percentage of atomic interactions from each atom inside and outside of the calculated surfaces for both compounds, providing quantitative data on the interactions arising from specific groups of atoms in each structure. Examining Figure S2, it is clear that the halogens (both chlorine and fluorine) and oxygen atoms are responsible for over 50% of the intermolecular interactions in both compounds. Carbon, nitrogen, and hydrogen are approximately equal in their contributions. The fluorine atom in 2 replaces a portion of the chlorine interactions found in 1 with the total chlorine interactions in 1 (29.9% & 25.3%) being approximately equal to the sum of the fluorine & chlorine interactions in 2 (25.9% & 23.9%) in reference to both internal and external interactions. Thus, by simply examining the total numerical percentage of interactions, it would appear that the two structures would have similar oxygen and halogen-based interactions. However, there are marked differences in the two structures which can be examined in further detail through the surface interactions and fingerprint plots.

Surface and Fingerprint Analysis

To further evaluate any changes imparted by the functional groups, Hirshfeld surface analysis^[25] was completed and the results deconstructed to a per-atom basis into fingerprint plots.^[26] The Hirshfeld surface mapped with d_norm and the shape index are shown in Figure 3. Both the mapped d_norm and shape index functions provide a useful visualization of the interactions discussed herein. While Hirshfeld surface analysis is a powerful tool useful in the analysis of interactions within crystal structures, the results of the analysis should be carefully scrutinized to check if the results make sense within the context of the system examined.^[27] Specifically, simply because two atoms are near each other within a crystal does not necessarily imply a stabilizing interaction. For example, hydrogen bonds have requirements of both distance and angle.^[28] Thus, each interaction should be carefully examined and contextualized to draw out the appropriate implications.

Figure 3.

(top row) The Hirshfeld surface mapped with the d_norm function (left) and the shape index (right) for compound 1. (bottom row) The Hirshfeld surface mapped with the d_norm function (left) and the shape index (right) for compound 2. The molecules are shown from both faces, flipped 180° with the sulfonyl halogen being arbitrarily shown as ‘up’ on the left.

Hydrogen interactions

Specific interactions are visualized in the fingerprint plots, manifesting as characteristic features. For example, both compounds display a set of six spikes present in the lower region of the fingerprint corresponding to, predominantly, H⋯Cl, H⋯N, and H⋯O interactions (and the reciprocal interactions, see Figure S3). Looking at the shape of these fingerprints reveals key distinctions between the two molecules. For example, the wings in 1, corresponding to the H⋯Cl | Cl⋯H interactions (d_i≈1.7 Å, d_e≈1.1 Å | d_i≈1.1 Å, d_e≈1.7 Å), are more defined and account for a much larger portion of the total interactions when compared with 2. 1 also has shorter H⋯Cl and H⋯N interactions, observed by the location of the ends of the spikes in the plots. The shortest H⋯Cl interaction in 2 is between H3 and Cl1 and a distance of 3.35 Å (d(H3ⁱ⋯Cl1)) while in 1 the shortest is 2.98 Å (d(H4^j⋯Cl2)). Of particular relevance, the sulfonyl chloride Cl atom (Cl2) displays more interactions and also shorter interactions with hydrogens when compared with the aromatic chlorine Cl1. The fluorine atom in 2 does not show any appreciable amount of interactions with any hydrogen atoms. However, the sometimes weaker nature of these H⋯F interactions could be traded for other more favorable interactions such as H⋯O, helping rationalize the relative absence of any H⋯F interactions.^[29] In line with this conjecture, it is observed that 2 has a notably higher percentage of H⋯O interactions as compared with 1 as observed in the deconstructed fingerprints.

Concerning the H⋯O | O⋯H interactions, both compounds display similarities and differences. For example, in both compounds, the shortest H⋯O interactions are from the sulfonyl groups. Specifically, in compound 1, O2 shows the shortest interactions with hydrogen atoms at 2.54 Å (d(H4^k⋯O2)) while for compound 2, the shortest interaction is 2.73 Å (d(H3^l⋯O2)), as seen in Figure S1. To move to contrasting interactions, intermolecular interactions for the sulfonyl group in 2 exhibits distinctive bridging interactions as shown in Figure S4. In particular, O3 is interacting with H4 at a distance of 2.76 Å (d(H4^m⋯O3)) and O2 with H3 at a distance of 2.73 Å (d(H3^l⋯O2)). As discussed for 1, while O2 is interacting with hydrogen atoms, O3 does not appear to make any interactions with any hydrogens and thus is not bridging. Further, it should be noted that for both compounds O1, the central oxygen in the benzoxdiazole ring, is also interacting with hydrogens, albeit at notably longer distances ranging from 3.26–4.12 Å, with angles ranging from 95–115°. Given the distances and angles, the interactions observed for O1 are likely not as significant when compared with the sulfonyl oxygen interactions.

Turning to the H⋯N interactions, compound 1 displays sharp, defined spikes at d_i≈1.5 Å, d_e≈1.1 Å | d_i≈1.1 Å, d_e≈1.5 Å. These spikes arise from H⋯N interactions from both N1 and N2 with adjacent hydrogen atoms. N1 shows the shortest interaction with H3 at a distance of 2.69 Å (d(H3^k⋯N1)) with a N1⋯H3–C3 angle of 165.94°. N2 has a longer interaction at 3.23 Å (d(H4ⁿ⋯N2)) with an angle of 120.55°. Longer interactions do exist, visualized by the disperse spots at longer d_i/d_e distances on the fingerprints (see Figure S3). These interactions, however, are likely not as relevant given the distances and angles involved. Contrasting 1 with 2 we see a more disperse set of interactions, manifesting as blunted peaks in the fingerprint for 2. This points to both the less linear nature of these H⋯N interactions in addition to longer interactions overall. The shortest interaction in 2 is between N2 and H3 at a distance of 2.83 Å (d(H3°⋯N2)) with an N2⋯H3–C3 angle of 116.40°. There are two noteworthy curiosities regarding the H⋯N interactions in 2. The first is that the percentage of interactions is identical to 1 (8.8% for both). The second is the lack of any hydrogen interactions with N1. These two observations lead to the conclusion that H⋯N interactions in these systems are varied in their strength and can involve either nitrogen in the ring. However, it appears that N2 is more likely to participate in H⋯N interactions. Part of this conclusion is rationalized through computational studies (vide infra).

Halogen Interactions

The central, diagonal region in fingerprint plots is typically the location for halogen interactions (Figure S5). Both structures display a region of Cl⋯Cl interactions with 1 showing a higher percentage as compared with 2, as expected. With respect to halogen bonding, these interactions do not appear to fall into the defined type I, II, or III halogen bonds as the distances and angles observed in either compound do not align with established measurements.^[30] There is a percentage of Cl⋯F interactions which are observed in 2. However, these also do not appear to fall into the expected halogen bonding criteria either.

The central green/light blue line in the fingerprints of both compounds is from a combination of Cl⋯Cl and Cl⋯C interactions, the latter of which is indicative of Cl⋯π interactions. Examining a packing diagram for 1 (Figure S6), the layered, face-on arrangement of the molecules allows for the aromatic chlorine (Cl2) to reside above a symmetry adjacent benzene ring, giving rise to the observed Cl2⋅⋅⋅π interactions depicted in Figure S7 (d=3.5774(4) Å, (d(Cl2⋯centroid). The sulfonyl chlorine (Cl1) shows no interactions with π systems though sulfonyl chloride groups have shown Cl⋯π interactions in previously reported structures.^[31] Of particular note, compound 2 shows no Cl⋯π interactions. These differences are readily observed in the respective fingerprint plots shown in Figure S5 when examining the shape of the specific interaction regions.

There are a set of F⋯C interactions that are observed in 2. The fluorine atom makes a close contact with the π bonding region between N2–C6 at a distance of 2.950(3) Å from the fluorine to the bond center. F⋯π interactions have been shown to be particularly stabilizing especially in the case of electron-deficient aromatic systems. The distances observed in 2 are similar to those which have been previously reported, which are approximately 3.0 Å. Another longer interaction is seen between F1 and the π region between C2–C3 at an F⋯π distance of 3.511(3) Å. These interactions account for a portion of F⋯C and F⋯N fingerprint plots.

The sulfonyl oxygens and the benzoxadiazole oxygen are seen interacting with the chlorine atoms in both of the molecules. The sulfonyl oxygen interactions (O2 & O3) are similar in distance when compared with O1. For both structures, the Cl⋯O interactions from the sulfonyl groups range from 3.27 Å to 4.00 Å. The interactions from O1 range from 3.24 Å to 3.72 Å. These Cl⋯O interactions appear to play a notable role in the packing of the crystal given the relatively high percentage seen in both systems: 15.6% in 1 and 13.0% in 2.

Of note, the fluorine atom in 2 is interacting with O1 at a distance of 2.963(3) Å with longer interactions to the sulfonyl oxygen atoms. The arrangement of the molecule is so as to facilitate these O⋯F interactions, aligning the molecules in a way that appears to prevent any favorable interactions with N1 leaving the void space near N1 as previously discussed. Figure S8 is provided to help visualize the arrangements and interactions.

Computational Studies

Electrostatic Potential

To further evaluate the interactions and potential interactions of the benzoxadiazole rings, the structure was analyzed in silico using the Spartan software suite (Spartan’20, Wavefunction, Inc., Irvine, CA USA). The molecular structures derived from the crystallographic studies were optimized and the resultant discussion based on the optimized structures. Images showing the structures and the electrostatic potential maps are seen in Figures S9 and S10. Table S1 compares the calculated vs. experimental bonds lengths.

The electrostatic potential (ESP) surface of a compound can help to explain several of the intermolecular interactions and crystal packing. For example, both of the chlorine atoms display a region of slightly positive charge (light blue) helping to rationalize the observed interactions between both chlorine atoms and the other electronegative atoms in the structure. With respect to compound 1, Cl1 has a more positive potential (V_s,max =+134.6 kJ/mol) than Cl2 (V_s,max =+97.1 kJ/mol). Both the experimental and theoretical observations thus point to the presence of a σ-hole in the chorine atoms.^[32] This is not observed, however, in F1 of compound 2 wherein no notable electron deficiencies are observed on the ESP surface.

The two benzoxadiazole nitrogen atoms (N1 & N2) also exhibit distinctive interactions in the solid-state. Close examination of the electrostatic potential maps helps to elucidate some characteristics of these two atoms. Both compounds display the same trend wherein N1 has a higher electrostatic potential (≈131 kJ/mol) vs. N2 (≈98 kJ/mol). In a similar vein, the sulfonyl oxygens (O2 & O3) are more negative than the benzoxadiazole oxygen (O1). This result is in line with our expectations given the distinct electronic environment of the different oxygen atoms (ring vs ancillary functional group). This observation is also in line with the observed H⋯O | O⋯H interactions predominantly arising from the sulfonyl oxygen atoms rather than O1.

Finally, to bring the concepts of the ESP maps and the crystal packing together, the ESP mapped on the Hirshfeld surface are used to visualize the packing of both structures (see Figure 4). The interactions between the positive and negative regions are evident when presented in this manner, offering an additional view of the impacts of the differing interactions in 1 vs. 2 arising from their respective sulfonyl moieties.

Figure 4.

A depiction of the packing for compounds 1 (left) and 2 (right) shown with the electrostatic potential mapped on the Hirshfeld surface. Alternating regions of negative potential (red) and positive potential (blue) are observed to preferentially interact.

Energy Frameworks

To better understand how the overall interactions and changes affect the two structures, the energy frameworks of the two molecules were calculated. Figures 5 & 6 and table S3 provide a visual and quantitative summary of the individual interactions leading to the formation of the frameworks. The size of the cylinders is a relative representation of the magnitude of the interaction energies, i.e., the larger the cylinder the greater the energy. The frameworks offer a clear distinction between the two molecules and the interactions leading to the formation of the crystalline structure. The packing in compound 1 is dominated by the dispersion forces. The coulombic forces run, predominantly, in a planar manner between molecules perpendicular to the 001 plane. The dispersion forces, however, show both vertical and planar interactions, linking distinct ‘sheets’ of molecules together. The dispersion and coulombic frameworks in 2, however, show only subtle distinctions wherein the dispersion framework has a slightly higher contribution when compared with the coulombic framework. Thus, for both molecules the dispersive interactions dominate the formation of the crystalline state. Examining table S3 shows that while polarization (E_pol) does contribute to the stabilization of the crystal, repulsion has a higher destabilizing effect (E_rep) often negating any contributions from polarization. The frameworks along with the electrostatic potential surfaces combine to provide a practical visualization for the distinctive structures and crystalline packing for both compounds.

Figure 5.

Compound 1: (A) Color coded grouping of molecules corresponding to calculated interaction energies used for depiction of the frameworks (see table S3); (B) Coulombic framework; (C) Dispersion framework; (D) Total energy framework.

Figure 6.

Compound 2: (A) Color coded grouping of molecules corresponding to calculated interaction energies used for depiction of the frameworks (see table S3); (B) Coulombic framework; (C) Dispersion framework; (D) Total energy framework.

Conclusion

The structures of two aromatic sulfur(VI) oxyhalides containing 2,1,3-benzoxadiazole heterocycle are reported, and the non-covalent interactions are examined via Hirshfeld surface analysis. Theoretical calculations were used to rationalize and gain a better understanding of these interactions. Several key observations emerge regarding the interactions present in these molecules:

• Experimentally and computationally, both nitrogens of the benzoxadiazole ring can participate in any number of interactions, N1 and N2 have distinct electronic structure that accounts for their unique intermolecular interactions. Additionally, these nitrogens participate in both intra- and intermolecular interactions.

• The dominant interactions from the sulfonyl groups arise from the oxygen atoms, not the halogens, whether SO₂F or SO₂Cl.

• When specifically contrasting the chlorine vs. fluorine moieties, the fluorine atom did not exhibit any hydrogen interactions within this system whereas the chlorine does.

• The sulfonyl fluoride moiety does not appear to influence the electronic structure of the benzoxadiazole ring directly given the similarities in bond distances observed in the heterocycle C–C bonds.

• Based observations discussed herein, and following on previously published results, we do expect that polymorphs of the sulfonyl fluoride compound are likely to exist though we were not able to see any with the samples we had grown.

With the recent interest in click chemistry,^[33] sulfonyl fluorides will continue to draw considerable attention, particularly due to their application in this field.^[1] As such, we expect that sulfur(VI) fluorides will be applied to crystal engineering and materials design fields.^[34] Specifically, therefore, understanding the interactions driving the formation of long-range ordering will be of paramount importance to efficiently advance this field. The results presented herein provide a solid foundation as a jumping-off point for the development of these classes of materials. Further, sulfur(VI) fluorides will inevitably be applied to pharmaceuticals and biological applications. Thus, our work provides a fundamental understanding of how these moieties fit into the larger knowledge of molecular design (e.g., Topliss trees^[35] and Hansch analysis^[36]).

Experimental Section

Chemicals

4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole was purchased from Sigma-Aldrich and used as received without further purification. All other chemicals were purchased from Fisher Scientific.

Synthesis

Synthesis of 4-chloro-7-fluorosulfonyl-2,1,3-benzoxadiazole (2): Compound 2 is readily prepared via the neat reaction between a saturated KHF₂ solution (Warning! Potassium bifluoride solutions are hazardous and will etch glassware) and commercially available 4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole (2). In a 50 mL round bottom flask, a Teflon-coated egg-shaped stir bar, deionized water (890 uL) and KHF₂ (390 mg, 4.94 mmol, 98% purity, Alfa Aesar^™) were added. The solution was vigorously stirred for 2 hours at ambient temperature to form a near saturated KHF₂ solution. Then, 4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole (500 mg, 1.976 mmol) was added in one portion to the solution while stirring. The mixture was stirred overnight in the sealed flask. Dichloromethane was then used to extract the reaction mixture (3×20 mL). The organic layers were combined and washed with water (3×50 mL) and saturated sodium chloride (1×25 mL). The solvent was removed under reduced pressure. Isopropanol was used to recrystallize the compound yielding a white powder (0.387 g, 83%). ¹H-NMR (400 MHz; acetone-d₆): δ 8.57 (dd, J=7.5, 0.8 Hz, 1H), 8.07 (dd, J=7.5, 1.1 Hz, 1H); ¹³C-NMR (101 MHz; acetone-d₆): δ 150.20 (s), 145.85 (s), 140.28 (s), 137.78 (s), 132.10 (s), 131.03 (s).; ¹⁹F-NMR (376 MHz, acetone-d6): δ 62.4 (s)

Single crystals of the starting material, 4-chloro-7-chlorosulfonyl-2,1,3-benzoxadiazole, were grown from slow diffusion of hexanes into a saturated solution of the compound in methyl ethyl ketone.

Single crystals of 4-chloro-7-fluorosulfonyl-2,1,3-benzoxadiazole suitable for diffraction were grown from slow evaporation of ethanol.

Spectroscopy

¹H, ¹³C and ¹⁹F NMR spectroscopies were performed on a JEOL 400 MHz NMR. Deuterated NMR solvents were purchased from Cambridge Isotope Labs. Chemical shifts were referenced to the residual solvent peaks in the NMR spectra.

Single Crystal Diffraction

Single crystals of both compounds were coated with Parabar 10312 oil and transferred to the goniometer of a Bruker D8 Quest Eco diffractometer with Mo Kα wavelength (λ=0.71073 Å) and a Photon II area detector. Examination and data collection were performed at 100 K for 1 and 150 K for 2.

For both compounds, data were collected, reflections were indexed and processed, and the files scaled and corrected for absorption using APEX3^[37] and SADABS.^[38] For all compounds, the space groups were assigned using XPREP within the SHELXTL suite of programs,^[39,40] the structures were solved by direct methods using ShelXS or ShelXT^[41] and refined by full matrix least squares against F² with all reflections using Shelxl2018^[42] using the graphical interfaces Shelxle^[43] and/or Olex2.^[44] H atoms were positioned geometrically and constrained to ride on their parent atoms. C–H bond distances were constrained to 0.95 Å for aromatic C–H moieties. U_iso(H) values were set to a multiple of U_eq(C) with 1.2 for C–H units.

Deposition Number(s) 2179436 (for 1), 2179437 (for 2) contains the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe http://www.ccdc.cam.ac.uk/structures.

Hirshfeld Analysis

Hirshfeld surfaces, the resultant images, energy frameworks, and fingerprint plots were calculated and produced using CrystalExplorer21.^[45] Distance analysis of the structures was accomplished using Olex2 and Mercury.^[46]

Energetic frameworks were calculated following the established procedures.^[47] Interactions energies were calculated using the B3LYP/6-31G(d,p) model imbedded within CrystalExplorer using Tonto.^[48]

Relevant references for molecular descriptors such as globularity,^[49] ovality,^[50] asphericity,^[50] void space,^[51] polar surface area,^[52] and dipole moment,^[53] are provided here for clarity.

Computational Studies

All computations and resultant data were obtained using the Spartan software suite (Spartan’20, Wavefunction, Inc., Irvine, CA USA). The initial geometry of both compounds was loaded into the software from the cif files and optimized employing RI-MP2 methods^[54] with a 6-311++G(2d,2p) basis set. Vibrational frequencies were checked for imaginary values to ensure the resultant structures were at a minimum.

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.