Direct investigation of chalcogen bonds by multinuclear solid-state magnetic resonance and vibrational spectroscopy

Vijith Kumar , Yijue Xu , César Leroy and David L. Bryce *
Department of Chemistry and Biomolecular Sciences, University of Ottawa, 10 Marie Curie Private, Ottawa, Ontario K1N 6N5, Canada. E-mail: dbryce@uottawa.ca; Fax: +1-613-562-5170; Tel: +1-613-562-5800 ext. 2018

Received 19th November 2019 , Accepted 21st January 2020

First published on 21st January 2020


We report a multifaceted experimental and computational study of three self-complementary chalcogen-bond donors as well as a series of seven chalcogen bonded cocrystals. Bis(selenocyanatomethyl)benzene derivatives were cocrystallized with various halide salts (Bu4NCl, Bu4NBr, Bu4NI) and nitrogen-containing Lewis bases (4,4′-bipyridine and 1,2-di(4-pyridyl)ethylene). Three new single-crystal X-ray structures are reported. 77Se solid-state nuclear magnetic resonance spectroscopic study of a series of cocrystals establishes correlations between the NMR parameters of selenium and the local ChB geometry. For example, the 77Se isotropic chemical shift generally decreases on cocrystal formation. Diagnostic 13C chemical shifts are also described. In addition, all the chalcogen bonded cocrystals and pure tectons are investigated by Raman and IR spectroscopy techniques. Characteristic red shifts of the NC–Se stretching band upon cocrystal formation on the order of 10 to 20 cm−1 are observed, which provides a distinct signature of the chalcogen bond involving selenocyanates. The 125Te chemical shift tensor and X-ray structure of chalcogen-bonded tellurocyanatomethylbenzene are also reported. Insights into the connection between the electronic structure of the chalcogen bond and the experimentally measured 77Se chemical shift tensors are afforded through a natural localized molecular orbital density functional theory analysis. For the systems studied here, the lack of a very strong a correlation between experimental and DFT-computed 77Se chemical shift tensors leads to the conclusion that many structural features likely influence their ultimate values; however, computations on model systems reveal that the ChB alone produces consistent and predictable effects (e.g., the chalcogen chemical shift decreases as the chalcogen bond is shortened).


Introduction

Chalcogen bonding (ChB) is a non-covalent interaction occurring between the electrophilic region associated with a group 16 chalcogen atom (e.g., S, Se, Te) and a nucleophilic region such as a Lewis base (LB) in the same or another molecular entity.1 Similar to halogen bonding (XB),2 chalcogen bonding is a highly directional secondary bonding interaction with an R–Ch⋯LB angle of approximately 180°.3 Even though the ChB has been less explored than hydrogen bonds (HB)4 or XB, recent studies demonstrate the ChB has a tremendous potential, with several applications in various fields such as supramolecular chemistry,5 crystal engineering,6 anion sensing and transport,7 organocatalysis,8 pharmaceutics,9 and biological processes.10 Despite the promising applications in a myriad of different fields, the nature of the ChB continues to be examined in the literature from experimental and computational perspectives.11–13 For example, the directional deviations between the two regions of depleted electron density (σ-holes), the maxima of the electrostatic surface potential (ESP), and the Ch⋯LB axes, have recently been examined in some detail.14,15 Some open questions include how the NMR response is affected by chalcogen bonding, how general is this response, whether such a response can be used in a predictive capacity, and what parallels can be drawn with the NMR response to related non-covalent interactions including hydrogen bonds and halogen bonds. Due to their direct relationship with the molecular orbitals centred on the nucleus of interest, chemical shift tensor measurements, in concert with computational chemistry, can provide direct insight into the electronic structure of the chalcogen bond (vide infra).

Theoretical investigations such as quantum chemical calculations based on perturbation theory and density functional theory (DFT) have been carried out to understand the nature of ChB interactions in the gas phase.16 UV-vis absorbance and emission spectroscopy as well as NMR spectroscopy have been recently utilised to examine the ChB in solution.17 X-ray diffraction has been widely used to investigate and quantify the ChB geometry in the solid state. However, complex chalcogenated systems such as metal organic frameworks, polymers, peptides, and drug molecules etc., often do not crystallize easily and may exist as multicrystalline or amorphous solids; in these cases, the use of complementary techniques like solid-state nuclear magnetic resonance (SSNMR) and vibrational spectroscopy are crucial and rapid tools for structural characterization.

In recent years, SSNMR spectroscopy has been well acknowledged as a powerful tool to characterize various non-covalent bonds including σ-hole interactions.18 Our group has reported extensively on the application of SSNMR and nuclear quadrupole resonance (NQR) spectroscopies to analyse and provide new structural insights into σ-hole interactions such as halogen,19 tetrel20 and pnictogen21 bonds and to contextualize them in the broader setting of established chemical bonding paradigms. These studies provide a distinctive way to characterize σ-hole interactions, and establish relationships between local bonding geometry and various NMR and NQR parameters. Selenium-77 and tellurium-125 possess favourable NMR properties (spin-1/2 nuclei) including moderate natural abundances (7.63% for 77Se and 7.07% for 125Te) and gyromagnetic ratios (γ = 5.12 × 107 rad T−1 s−1 for 77Se and γ = −8.51 × 107 rad T−1 s−1 for 125Te, respectively).22 Selenium and tellurium SSNMR provides a particularly sensitive probe of molecular structure in part due to broad chemical shift ranges spanning over 3000 and 6000 ppm, respectively.23 While 33S is present in the most chemically and biologically active organosulfur compounds,24 its low natural abundance of 0.76%, rather small gyromagnetic ratio (γ = 2.05 × 107 rad T−1 s−1), and moderate quadrupole moment of −69.4 mb conspire to make 33S SSNMR quite challenging.25

Recent literature reveals the seleno and tellurocyanate derivatives are plausible ChB donor motifs, due to the strong electron-withdrawing power of the cyano group, which makes the σ-hole potentials at chalcogen atoms remarkably positive.26 Fourmigué and co-workers reported that benzylic selenocyanates are strong and directional ChB donors for crystal engineering purposes, where the chain-like motifs can be tuned into 1D extended structures upon cocrystallization with ditopic ChB acceptors.27 Recently, we have demonstrated the strong aptitude of benzylic selenocyanates for the ChB driven recognition of onium halides in the solid state and in solution by using a combination of X-ray crystallography, multinuclear solution magnetic resonance spectroscopy and quantum chemical calculations.15 Herein, seven chalcogen-bonded anionic and neutral cocrystals with benzylic selenocyanates, and the self-complementary benzylic seleno and tellurocyanates (1, 2, and 3) are studied. These simple ChB donors were chosen to be devoid of functional groups that could interfere with the occurrence of the chalcogen bonds or modify their features (Fig. 1). As far as the ChB acceptors are concerned, the halides and neutral pyridine based ditopic acceptors were chosen to examine the changes in the spectral parameters with various chalcogen-bond synthons. All cocrystals are characterized by single crystal and powder X-ray diffraction (SCXRD and PXRD) and multinuclear solid-state magnetic resonance spectroscopy. Natural localized molecular orbital (NLMO) DFT analysis is used to reveal the origins of the observed changes in the 77Se chemical shift tensors. We also establish Raman and FT-IR spectroscopic techniques to rapidly probe for the presence of chalcogen bonds in these powdered materials. This systematic study combines different solid-state characterization methods to provide both experimental and theoretical insights into the correlation between the selenium chemical shift tensors and the ChB electronic structure.


image file: c9cp06267j-f1.tif
Fig. 1 Left: Molecular structures of the benzylic selenocyanates 1, 2, and benzylic tellurocyanate 3. Top right: Pair of σ-holes on Se atoms visualized as the electropositive regions on ESP surfaces for the ChB donor 2 with an isodensity of 0.02 a.u. (B3LYP/Def2TZVP). Red indicates negative charge density, and blue is positive charge density. Bottom right: Schematic representation of the observed chalcogen bond synthon in the series of cocrystals (Ch = Se, Te; X = N, I, Br and Cl; R = aromatic systems).

Results and discussion

Structural studies of the ChB donors and cocrystals

The ChB donors 1,3-bis(selenocyanatomethyl)benzene (1) and 1,4-bis(selenocyanatomethyl)benzene (2) are prepared from their respective benzyl bromides and potassium selenocyanates in DMF under an argon atmosphere as described earlier (ESI, paragraph S.2).27 Benzylic tellurocyanate (3) was prepared by treating benzyl bromide with in situ generated potassium tellurocyanate in dry DMSO under argon atmosphere according to the previously reported protocol (ESI, paragraph S.2).28 Solid-state structural studies reveal the crystal packing of 1 (FEDHUU) and 2 (POXYEH) are mainly driven by intramolecular Se⋯N ChB interactions, resulting in extended chain-like structures (ESI, Fig. S5). The pure ChB donor 3 crystallizes in the monoclinic crystal system with space group P21/c. As expected, molecules are associated through short and linear Te⋯N chalcogen bonds forming chains running along the crystallographic a axis (C–Te⋯N distance 2.814 Å, which corresponds to normalized contact (Nc) values of 0.77, and a C–Te⋯N angle of 172.1°). The normalized contact is defined as the distance between the interacting atoms divided by the sum of their van der Waals radii. The MESP maps of compounds 1, 2, and 3 are presented in the ESI (Fig. 1 and Fig. S18) and visualized with an isodensity of 0.02 a.u. (electron per Bohr3). MESP surfaces of the Ch atoms clearly show the presence of two electropositive regions (σ1 and σ2 in dark blue). The region (σ1) along the prolongation of the electron withdrawing nitrile group is more electropositive than the region (σ2) opposite the Ch–CH2 bond. (On the 0.02 a.u. surface: 1: σ1: 608.59 kJ mol−1, σ2: 522.47 kJ mol−1; 2: σ1: 629.86 kJ mol−1, σ2: 522.47 kJ mol−1; 3: σ1: 705.73 kJ mol−1, σ2: 628.81 kJ mol−1. On the 0.001 a.u. surface: 1: σ1: 161.7 kJ mol−1, σ2: 80.9 kJ mol−1; 2: σ1: 162.0 kJ mol−1, σ2: 82.2 kJ mol−1; 3: σ1: 180.1 kJ mol−1, σ2: 94.5 kJ mol−1). These calculated values are consistent with related literature reports on selenocyanate derivatives.29 These positive regions interact with incoming electron-rich areas of the halide ions and/or neutral pyridine-based ditopic acceptor moieties.

The cocrystals of 1 and 2 with 1,2-di(4-pyridyl)ethylene (DPE) were obtained by dissolving an equimolar ratio of both tectons in acetone and slow solvent evaporation at room temperature. Cocrystal formation is preliminarily confirmed by PXRD analysis (ESI, Fig. S1–S4) and changes in the melting point of the complex (64 °C for 1·DPE, 110 °C for 2·DPE) in comparison with pure individual tectons. The X-ray structure reveals the 1[thin space (1/6-em)]:[thin space (1/6-em)]1 adduct of new chalcogen bonded systems 1·DPE and 2·DPE crystallized in a monoclinic system with the C2/c space group and a triclinic crystal system with the P[1 with combining macron] space group, respectively. In cocrystal 2·DPE, para-bis(selenocyanatomethyl)benzene and DPE are located on inversion centers, while in 1·DPE, meta-bis(selenocyanatomethyl)benzene is located on a two-fold axis, and the DPE is positioned on an inversion centre. As detailed in Fig. 2 and the ESI, Fig. S6 and S7, ChB donors 1–3 are self-complementary systems, therefore the ditopic ChB acceptor pyridine moiety competes efficiently with the nitrogen atom of the nitrile group to result in short and directional intermolecular ChBs favouring the formation of chain-like structures. In particular, ChB brings the value of Nc down to 0.80 in the DPE cocrystals. Various ChB geometries and structural characteristics are listed in Table 1. The cocrystals of 2 with tetrabutylammonium iodide (GIHCOS), bromide (GIHDAF), and chloride (GIHGIQ) were prepared based on our previous work,15 and the cocrystals of 1 and 2 with 4, 4-bipyridine (FEDJAC, FEDJEG) were reproduced according to a literature report.27a


image file: c9cp06267j-f2.tif
Fig. 2 Partial ball and stick representations (Mercury 4.10.2) show the intra- and inter-molecular chalcogen bonds with series of monotopic and ditopic linkers containing halide and nitrogen acceptors. (a) Pure ChB donor 3; (b) cocrystal 2·(Bu4N)Cl; (c) 2·(Bu4N)I; (d) 2·(Bu4N)Br; (e) 2·DPE; (f) 1·DPE. Colour codes: grey, carbon; blue, nitrogen; white, hydrogen; yellow, selenium; dark brown, tellurium; green, chloride; brown, bromide; iodide, magenta (dashed black lines denotes ChB).
Table 1 Summary of the C–Ch⋯A chalcogen bonding geometries observed via SCXRD. (C = carbon, Ch = Se or Te, A = N, I, Br, or Cl)
Compound C–Ch⋯A (Å) C–Ch⋯A (°) N c Ref code
a N c = dCh⋯ArvdW.
1 3.010(2) 172.9(6) 0.86 FEDHUU
3.015(2) 174.1(6) 0.87
2.965(2) 175.9(6) 0.85
3.017(2) 175.7(6) 0.87
2 2.997(2) 174.4(8) 0.86 POXYEH
3.020(2) 171.8(6) 0.87
3 2.814(9) 172.1(3) 0.77 This work
1·BP 2.830(2) 177.2(9) 0.81 FEDJAC
1·DPE 2.792(4) 173.5(2) 0.80 This work
2·(Bu4N)I 3.506(1) 174.9(9) 0.88 GIHCOS
2·(Bu4N)Br 3.249(1) 174.4(2) 0.86 GIHDAF
2·(Bu4N)Cl 3.191(2) 174.3(3) 0.88 GIHGIQ
2·BP 2.897(4) 176.7(1) 0.83 FEDJEG
2·DPE 2.865(2) 167.8(9) 0.82 This work


An analysis of relevant chalcogen-bonded structures available in the Cambridge Structural Database (CSD) was carried out to provide some context for the structural features presented in Table 1. In total, 511 crystal structures were found in the CSD exhibiting Se⋯N ChB synthon motifs, and among these 120 entries correspond to selenocyanate motifs. Similarly, there are few structures reported corresponding to Se⋯A chalcogen bonding (A = I (Se⋯I: 44 hits), Br(Se⋯Br: 52 hits), Cl(Se⋯Cl: 55 hits) and F(Se⋯F: 2 hits)). The distribution of Se⋯N distances in these ChB synthon motifs were found to exhibit maxima in the range 2.50–3.50 Å. The distribution of Se⋯A distances falls in the range of 2.65 to 3.87 Å (Se⋯Cl = 2.65–3.62 Å, Se⋯Br = 2.78–3.75 Å, Se⋯I = 3.10–3.87 Å) and Te⋯A distances fall in the range of 2.70 to 4.05 Å (Te⋯Cl = 2.70–3.3.80 Å, Te⋯Br = 3.08–3.91 Å, Te⋯I = 3.19–4.05 Å). Histograms summarizing these data may be found in the ESI.

Multinuclear solid-state magnetic resonance spectroscopic studies and NLMO analysis

The new crystal structures described above and the previous literature demonstrate the occurrence of strong and linear ChBs with different electron donor systems. We utilized 13C, 77Se, and 125Te magic-angle spinning (MAS) SSNMR experiments to investigate the electronic environment of chalcogen atoms in the cocrystals and compare these data with those for pure benzylic selenocyanates 1 and 2. 77Se and 125Te SSNMR spectra of the ChB donors and their respective cocrystals are shown in Fig. 3 and in the ESI (paragraph S.6, Fig. S8–S16).
image file: c9cp06267j-f3.tif
Fig. 3 Examples of NMR spectra of chalcogen-bonded systems. (A) Experimental 77Se CP/MAS NMR spectra (blue) obtained at 9.4 T for pure ChB donor 2 and selected chalcogen-bonded cocrystals 2·DPE and 2·(Bu4N)Br. The corresponding simulated spectra are shown at the bottom (red). (B) Experimental 125Te CP/MAS NMR spectrum (blue lines) obtained at 9.4 T and simulated spectrum (red lines) for pure ChB donor 3. Orange asterisks indicate the isotropic chemical shifts.

Spectra of all the compounds were recorded with a minimum of two different spinning speeds in order to distinguish the isotropic peaks from the spinning sidebands and to increase the precision of the fitted parameters. The isotropic chemical shifts (δiso), spans (Ω = δ11δ33), skews (κ = 3(δ22δiso)/Ω) and the principal components (δ11, δ22, δ33) are detailed in Table 2. The 77Se NMR spectrum of pure ChB donor 2 reveals two chemical shifts at 343.0 and 357.2 ppm, indicating the existence of two crystallographically distinct selenium sites (Table 2, ESI Fig. S8–S115). Similarly, the presence of four crystallographically distinct selenium sites in donor 1 gives rise to four different 77Se isotropic chemical shifts, ranging from 321.9 to 365.3 ppm (Table 2). Multiple attempts were carried out to measure 77Se NMR spectra for cocrystal 1·DPE; unfortunately the material was pressure and temperature sensitive and starts to decompose in the NMR rotor, even spinning at a moderately low speed of 3 kHz.

Table 2 Experimental 77Se and 125Te chemical shift tensors for pure ChB donors and cocrystals
Compound δ iso (ppm) Ω (ppm) κ δ 11 (ppm) δ 22 (ppm) δ 33 (ppm)
1 365.3 (0.2) (site 1) 600 (3) −0.08 (0.02) 673 (5) 349 (4) 73 (4)
352.3 (0.2) (site 2) 518 (6) 0.59 (0.02) 560 (6) 454 (4) 42 (4)
336.4 (0.2) (site 3) 569 (4) 0.32 (0.02) 591 (6) 397 (4) 22 (5)
321.9 (0.2) (site 4) 686 (7) −0.20 (0.03) 688 (7) 276 (6) 2 (6)
1·BP 342.5 (0.1) 702 (3) −0.17 (0.02) 713 (4) 301 (6) 11 (6)
2 343.0 (0.1) (site 1) 617 (6) −0.13 (0.02) 665 (4) 316 (3) 48 (4)
357.2 (0.1) (site 2) 629 (4) −0.23 (0.02) 696 (6) 309 (4) 67 (4)
2·BP 331.7 (0.2) 713 (3) −0.18 (0.02) 709 (5) 289 (5) −3 (5)
2·DPE 339.4 (0.1) 688 (3) −0.11 (0.02) 696 (6) 314 (4) 8 (4)
2·(Bu4N)Cl 328.7 (0.1) (site 1) 598 (1) −0.33 (0.01) 660 (1) 263 (1) 63 (1)
318.1 (0.1) (site 2) 615 (2) −0.31 (0.01) 657 (2) 255 (1) 42 (1)
2·(Bu4N)Br 314.3 (0.2) 613 (3) −0.25 (0.02) 646 (5) 263 (3) 33 (4)
2·(Bu4N)I 319.0 (0.1) 606 (5) −0.28 (0.01) 650 (4) 262 (3) 44 (3)
3 679.1 (0.3) 1635 (8) −0.85 (0.04) 1728 (3) 216 (4) 93 (5)


Inspection of the 77Se CS tensor data in Table 2 reveals several interesting points. Firstly, an increase in the value of δ11 relative to pure 1 or 2 is noted upon cocrystal formation with nitrogen-based electron donors. Conversely, δ11 decreases upon cocrystal formation with halide ion electron donors. This highlights the point that the pure ChB donor compounds feature self-complementary chalcogen bonds and thus their NMR properties may not necessarily fall at one extreme of the ranges observed. Secondly, the value of δ22 decreases upon cocrystal formation. Thirdly, the value of δ33 generally decreases upon cocrystal formation with all electron donors. All of these changes combine to result generally in a decreased isotropic 77Se chemical shift upon cocrystal formation. The span of the CS tensor increases upon cocrystal formation with nitrogen-based electron donors, but decreases upon cocrystal formation with halide ion electron donors; this is again reflective of the fact that the pure ChB donors feature chalcogen bonds of intermediate strength relative to the two cocrystal subsets (see Table 1). A comparison between the DFT calculated 77Se chemical shift values and the experimental values (ESI; paragraph S.8, Fig. S19) reveals a poor correlation likely due to the lack of crystal packing in the cluster-based DFT-GIAO methods and the high sensitivity of the selenium atom to the local environment.

Analogous to our previous work on halogen bonds,30 an NLMO analysis of the 77Se isotropic magnetic shielding values was conducted. The main contributions are: the sum of the selenium core orbitals (CR), two selenium lone pair orbitals (LPs), bonding orbitals between selenium and nitrile carbon (BD Se–CN) and bonding orbitals between selenium and methyl carbon (BD Se–CH2). The values are provided in the ESI (Table S3). While the core orbitals have a consistent positive contribution to the isotropic magnetic shielding, contributions from the other orbitals vary with the ChB geometry. The main NLMO contributions to the isotropic magnetic shielding values are plotted as a function of ChB geometrical features: ChB bond distance, angle and Nc values (ESI). While there is a degree of scatter, the magnitude of the contribution from the bonding orbital between Se and C[triple bond, length as m-dash]N tends to decrease as the ChB bond shortens. Therefore, the Se–C bond can be a sensitive probe for the occurrence of ChB formation.

The dependencies of the 77Se chemical shift tensors on Se⋯Br/N bond length were also calculated using models built based on the experimental structures of 2·(Bu4N)Br and 2·BP. The Se⋯Br distance was varied between 2.70 Å and 3.70 Å in 0.2 Å increments with the C–Se⋯Br angle of fixed at the experimental value of 174.4°. The Se⋯N distance was varied between 2.50 Å and 3.50 Å in 0.2 Å increments with the C–Se⋯N angle set to the experimental value of 177.2°. A plot of the principal components of the 77Se chemical shift tensors as a function of ChB distance is provided in Fig. 4. The isotropic chemical shifts decrease as the ChB is shortened. There is also a second order polynomial decrease in the principal components as the ChB distance reduces. The change in the most deshielded tensor component, δ11, is more significant than that for the most shielded tensor component δ33, resulting in a decrease in span when shortening the ChB.


image file: c9cp06267j-f4.tif
Fig. 4 Plot of calculated 77Se isotropic chemical shift (a), span (b) and skew (c) as a function of ChB distance for the models of 2·(Bu4N)Br (blue circles) and 2·BP (orange triangles). (d) A plot of calculated 77Se principal components of chemical shift tensors: δ11 (blue circles), δ22 (orange triangles) and δ33 (green diamonds) as a function of ChB distance for the models of 2·(Bu4N)Br.

The main NLMO contributions to the computed 77Se isotropic magnetic shielding changes relating to the ChB distance are similar to what was noted above: the sum of the selenium core orbitals (CR), two selenium lone pair orbitals (LPs), bonding orbitals between selenium and nitrile carbon (BD Se–CN) and bonding orbitals between selenium and methyl carbon (BD Se–CH2). A figure summarizing the main NLMO contributions and a plot of the main NLMO contributions to the magnetic shielding tensors as a function of ChB distance is provided in the ESI, Fig. S21. Similar to the experimentally-based NLMO contributions, the sum of the selenium core orbitals makes a positive contribution to the isotropic magnetic shielding values. In both cases, one of the selenium lone pair orbitals (π orbitals) makes a negative contribution, which was observed experimentally in most cases, whereas the sign of the contributions from other lone pair orbitals (σ orbitals) varies. In both Br and BP systems, there are decreasing contributions from CR and BD Se–CN orbitals as the bond distance becomes shorter. The contributions from Se π lone pair and BD Se–CH2 orbitals increase as the ChB bond gets shorter.

The experimental 125Te CP/MAS NMR spectrum of pure donor 3 obtained at 9.4 T is shown in Fig. 3. Spectral simulation provides an isotropic chemical shift of 679.1 ppm and a span of 1635 ppm (Table 2). These data may be compared, for example, with those available from Collins et al. for some salts of the trimethyltelluronium (TMT) and triphenyltelluronium (TPT) cations.31 Isotropic chemical shifts ranging from 404 to 472 ppm are reported for various TMT salts and from 731 to 802 ppm for TPT salts. The 125Te CS tensor spans are only on the order of 100 ppm for the TMT salts. The data for 3 are largely reflective of the two-coordinate nature of Te in this sample, as opposed to the three-coordinate geometries found in TMT and TPT. Notably, in their work, evidence for chalcogen bonding of halides to tellurium is also obtained through J coupling measurements.31 Due to the poor thermal stability of 3, even after several attempts we were unable to prepare cocrystals. The pure tellurium donor material also slowly starts to decompose while spinning the sample in the NMR rotor. Thus, no further 125Te studies were pursued as part of this study. However, the CS data reported for 3 provide a valuable benchmark for the chalcogen-bonded benzylic tellurocyanate motif.

13C CP/MAS SSNMR spectroscopy was carried out on pure ChB donors and cocrystals to characterize the interactions in terms of their 13C chemical shifts. As shown in Fig. 5, the isotropic 13C resonances of the nitrile carbon exhibit an asymmetric splitting. Additional 13C CP/MAS SSNMR data were also acquired at 4.7 T; this field-dependent splitting is due to the presence of residual dipolar coupling between 13C and 14N instead of the presence of impurity or different sites as shown in ESI, (Fig. S17). Simulations in Fig. 5 were adjusted by including the direct dipolar coupling (RDD) and J coupling between 13C and 14N, and quadrupolar coupling constant (CQ) and asymmetry parameter (ηQ) of 14N. The RDD value, image file: c9cp06267j-t1.tif, which depends on the inverse cube of the motionally averaged internuclear distance, may be calculated directly from the single-crystal X-ray structure. J coupling and ηQ were not found to notably influence the simulated peak shapes; therefore, these two values were fixed to the DFT calculated values during the fitting process. The only adjusted parameters during the fitting process were therefore δiso and CQ. It may be noted that as there are two crystallographically distinct sites in pure 2, there are two sets of NMR parameters used to fit the spectrum for this compound (see ESI). The effect from J coupling between 13C and 77Se is omitted due to the relatively low natural abundance of the latter; nevertheless this could contribute to spectral broadening near the baseline. The results are summarized in the ESI (Table S2). Formation of a chalcogen bond to 2 causes a decrease in the chemical shift (Δδ(13C) ranging from −1.6 to −2.5 ppm) upon the formation of the NC–Se⋯A motif.


image file: c9cp06267j-f5.tif
Fig. 5 Nitrile carbon regions of the 13C CP MAS SSNMR spectra (blue lines) obtained at 9.4 T for pure ChB donor 2 and selected chalcogen-bonded cocrystals 2·DPE and 2·(Bu4N)I. The corresponding simulated spectra (see text for discussion) are shown in red.

Raman and Fourier transform infrared spectroscopy

FTIR and Raman spectroscopies have been applied as simple and fast screening methods to detect the occurrence of ChB between the benzylic selenocyanates 1 and 2, and a series of electron donors. It is well-known that the intermolecular interaction between an electron-donor species and an electrophile affects the vibrational motions as assessed by intensity variations and frequency shifts. Due to the limited literature32 on the vibrational behavior of chalcogen-bonded systems, we carried out a detailed vibrational analysis by density functional theory (DFT) calculations (gas phase) for all the structures studied herein to accurately assign the molecular vibration bands and to predict detectable changes upon formation of the chalcogen bond. Selected experimental and calculated FTIR and Raman absorption frequencies for all the cocrystals and the individual ChB donors are reported in Table 3. These data show a good agreement between the calculated and experimental C–Se absorption frequency, with a variation of ∼20 cm−1. Experimental Raman spectroscopy reveals the C–Se in plane stretching vibrations of benzylic selenocyanate 2 appear at 510 cm−1 and this is in agreement with the calculated value, i.e., 510 cm−1 (Table 3 and Fig. 6).
Table 3 Selected experimental and calculated IR and Raman stretching vibration bands (in cm−1) associated with the N[triple bond, length as m-dash]C and N[triple bond, length as m-dash]C–Se bonds in pure ChB donors and related cocrystals
Compound Experimental Calculated

image file: c9cp06267j-t2.tif

image file: c9cp06267j-t3.tif

image file: c9cp06267j-t4.tif

image file: c9cp06267j-t5.tif

a Not detected, presumably due to spectral broadening. b An additional peak is observed at 504 cm−1.
1 2147 2149 512 513
1·BP 2139 2138 496 495
2 2147 2150 510 510
2·BP 2139 2137 495 501
2·BPE 2138 2137 496 493
2·(Bu4N)Cl 2144 2145 n.d.a 546
2·(Bu4N)Br 2141 2140 488b 474
2·(Bu4N)I 2143 2143 490 471



image file: c9cp06267j-f6.tif
Fig. 6 Experimental Raman spectra of ChB donor 2 (blue, bottom) and related cocrystals 2·BP (red, middle) and 2·(Bu4N)Br (green, top). A clear red shift and an intensity increase are observed for C–Se along the extension of the covalently bonded C[triple bond, length as m-dash]N atom, as well as a red shift and decrease in intensity of C[triple bond, length as m-dash]N stretching band upon occurrence of the chalcogen bond highlighted in light blue.

Interestingly, in benzylic selenocyanate-electron donor complexes, we find that the occurrence of ChB produces a red shift in the C–Se stretching region of 488–510 cm−1 (Table 3). A similar red shift and intensity variation of the C–X stretching band (X = Br and I), which is the distinct signature of the halogen bond has been reported by Resnati and co-workers.33 The observed red shift is likely attributable to the weakening and lengthening of the C–Se bond upon forming a ChB with electron donor systems. In addition to the C–Se stretching, vibrations related to nitrile (C[triple bond, length as m-dash]N) covalently bonded to selenium in the benzylic selenocyanates could be an indirect indicator for the occurrence of chalcogen bond.

Conclusions

A series of cocrystals has been prepared based on NC–Se⋯X chalcogen bonds and characterized with X-ray crystallography, multinuclear solid-state magnetic resonance, and vibrational spectroscopies. The σ-hole present on the prolongation of the NC–Se covalent bond interacts with anionic and neutral Lewis bases to form isolated or extended neutral 1D networks. From 77Se CP/MAS SSNMR spectroscopy, we observe a general decrease in the isotropic chemical shift of selenium for the chalcogen bonded cocrystals in comparison to the pure benzylic selenocyanates. Changes in the principal components of the 77Se chemical shift tensors were correlated to the local ChB geometry. Through NLMO calculations it was found that these changes are mainly associated with the sum of the selenium core orbitals, with further important contributions from two selenium lone pair orbitals, bonding orbitals between selenium and nitrile carbon, and bonding orbitals between selenium and methylene carbon. A lack of strong correlation between experimental and computed 77Se chemical shift tensors implies that many structural features likely influence their ultimate values; however, computations on model systems reveal that the ChB alone produces consistent and predictable effects.

13C CP/MAS SSNMR revealed that the NC–Se carbon chemical shifts decrease upon the formation of chalcogen-bonded cocrystals, compared to the pure ChB donors. We have also demonstrated that the NC–Se vibration band undergoes a clear red shift and intensity change upon the formation of a chalcogen bond in the cocrystals. This Raman and IR spectroscopic fingerprint of the selenocyanate-based chalcogen bonded cocrystals could prove to be a simple and rapid tool to detect the occurrence of ChB when alternative methods are not apposite.

The combination of solid-state NMR and vibrational spectroscopy techniques employed here, as well the insights gained from computational chemistry, affords new directions for the characterization of the chalcogen bonds in more complex systems, in excellent complementarity with X-ray diffraction methods.

Experimental

Preparation of cocrystals

ChB donor 1, 2, and acceptor systems such as tetrabutylammonium halides ((Bu4N)I, (Bu4N)Br, (Bu4N)Cl), 4,4′-bipyridine (BP) and 1,2-di(4-pyridyl)ethylene (DPE) were dissolved separately in acetone (1 equivalent of onium halides and 0.5 eq. of BP and DPE for ChB donors). These two solutions were mixed and allowed to evaporate slowly at room temperature. The cocrystals of 1 and 2 with DPE are obtained by dissolving the two tectons in acetone and diethyl ether was diffused as a second less efficient solvent.

Powder X-ray diffraction

Pure ChB donors, ChB acceptors and cocrystals were individually packed in an aluminium or glass sample holder and data sets were collected on a Rigaku Ultima IV powder diffractometer at 293 K (±2) (CuKα1 radiation with a wavelength of λ = 1.54056 Å). The measurements were carried out in focused beam geometry with a step-scan technique in 2θ range of 5–50°. Data were acquired by scintillation counter detector in continuous scanning mode with a step size of 0.02°. The experimental PXRD patterns of pure ChB donors, cocrystals and simulated patterns from the single crystal data are provided in ESI S.4.

Single crystal X-ray studies

The crystals were mounted on glass fibres with glue prior to data collection. Crystals were cooled to 200 ± 2 K. The data were collected on a Bruker AXS diffractometer equipped with MoKα radiation (wavelength of λ = 0.7103 Å) with an APEX II CCD detector. The raw data collection and processing were performed with the Bruker APEX II software package. Structure solution and the refinement details are provided in ESI S.5.

Solid-state NMR spectroscopy

SSNMR experiments were conducted using a Bruker Avance III NMR spectrometer (B0 = 9.4 T, νL(77Se) = 76.311 MHz and νL(125Te) = 126.240 MHz) equipped with a triple-resonance 4 mm MAS probe. Samples were ground into fine powder and packed in 4 mm o.d. zirconium oxide rotors. 77Se SSNMR chemical shifts were referenced to solid (NH4)2SeO4, (δiso = 1040.2 ppm) and 125Te chemical shifts were referenced to solid telluric acid Te(OH)6 (δiso = 685.5 ppm and 692.2 ppm).31 The MAS frequency was varied from 3 to 12.5 kHz to obtain spectra with a sufficient number of sidebands for spectral fitting purposes (Fig. S9–S17, ESI). Spectra were obtained at 298 K for pure ChB donors 1, 2, and their respective cocrystals; in a few cases the temperature was set to 288 K or 278 K due to the low melting point of the cocrystals and to avoid possible phase transition (Fig. S9–S17, ESI). A standard 1H → 77Se/125Te CP pulse sequence was employed for all cocrystals. The π/2 pulse length was optimized to be 3.2 μs for 77Se and 1.60 μs for 125Te. The CP contact time used was 7 ms for 77Se and 2 ms for 125Te and the recycle delays varied from 10 s to 4 min. The total number of transients varied from 512 to 16[thin space (1/6-em)]000. The 77Se NMR spectra were simulated using a Herzfeld-Berger analysis34 and dmfit.35 Residual dipolar coupling between 13C and 14N was analyzed using the WSolids program.36 The experimental and simulated spectra for the pure ChB donors and the chalcogen-bonded cocrystals are provided in the ESI, Section S.6.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We thank Dr Jeffrey Ovens, Dr Glenn Facey, Dr Peter Pallister and Dr Yun Liu for technical support and useful discussions. D. L. B. thanks the Natural Sciences and Engineering Research Council of Canada for funding.

References

  1. C. B. Aakeroy, D. L. Bryce, G. R. Desiraju, A. Frontera, A. C. Legon, F. Nicotra, K. Rissanen, S. Scheiner, G. Terraneo, P. Metrangolo and G. Resnati, Pure Appl. Chem., 2019, 91, 1889–1892 CAS.
  2. G. Cavallo, P. Metrangolo, R. Milani, T. Pilati, A. Priimagi, G. Resnati and G. Terraneo, Chem. Rev., 2016, 116, 2478–2601 CrossRef CAS.
  3. (a) L. Vogel, P. Wonner and S. M. Huber, Angew. Chem., Int. Ed., 2019, 58, 1880–1891 CrossRef CAS PubMed; (b) P. Scilabra, G. Terraneo and G. Resnati, Acc. Chem. Res., 2019, 52, 1313–1324 CAS.
  4. G. R. Desiraju, Acc. Chem. Res., 2002, 35, 565–573 CrossRef CAS.
  5. (a) L. Chen, J. Xiang, Y. Zhao and Q. Yan, J. Am. Chem. Soc., 2018, 140, 7079–7082 CrossRef CAS PubMed; (b) P. C. Ho, P. Szydlowski, J. Sinclair, P. J. W. Elder, J. Kubel, C. Gendy, L. M. Lee, H. Jenkins, J. F. Britten, D. R. Morim and I. Vargas-Baca, Nat. Commun., 2016, 7, 1–10 Search PubMed.
  6. Y. Zhang and W. Wang, Crystals, 2018, 8(163), 1–9 Search PubMed.
  7. (a) J. Y. C. Lim, I. Marques, A. L. Thompson, K. E. Christensen, V. Felix and P. D. Beer, J. Am. Chem. Soc., 2017, 139, 3122–3133 CrossRef CAS PubMed; (b) S. Benz, M. Macchione, Q. Verolet, J. Mareda, N. Sakai and S. Matile, J. Am. Chem. Soc., 2016, 138, 9093–9096 CrossRef CAS PubMed.
  8. (a) K. T. Mahmudov, M. N. Kopylovich, M. F. C. Guedes da Silva and A. J. L. Pombeiro, Dalton Trans., 2017, 46, 10121–10138 RSC; (b) W. Wang, H. Zhu, S. Liu, Z. Zhao, L. Zhang, J. Hao and Y. Wang, J. Am. Chem. Soc., 2019, 141, 9175–9179 CrossRef CAS PubMed.
  9. S. P. Thomas, V. Kumar, K. Alhameedi and T. N. G. Row, Chem. – Eur. J., 2019, 25, 3591–3597 CrossRef CAS PubMed.
  10. (a) K. Křiž, J. Fanfrlik and M. Lepšik, ChemPhysChem, 2018, 19, 2540–2548 CrossRef PubMed; (b) K. Raja and G. Mugesh, Angew. Chem., Int. Ed., 2015, 54, 7674–7678 CrossRef CAS PubMed; (c) A. Lange, M. Gunther, F. M. Buttner, M. O. Zimmermann, J. Heidrich, S. Hennig, S. Zahn, C. Schall, A. Sievers- Engler, F. Ansideri, P. Koch, M. Laemmerhofer, T. Stehle, S. A. Laufer and F. M. Boeckler, J. Am. Chem. Soc., 2015, 137, 14640–14652 CrossRef CAS PubMed.
  11. M. Fourmigué and A. Dhaka, Coord. Chem. Rev., 2020, 403, 213084 CrossRef.
  12. A. Bauzá and A. Frontera, ChemPhysChem, 2020, 21, 26–31 CrossRef PubMed.
  13. S. Scheiner, M. Michalczyk, R. Wysokiński and W. Zierkiewicz, Chem. Phys., 2020, 530, 110590 CrossRef.
  14. (a) M. E. Brezgunova, J. Lieffrig, E. Aubert, S. Dahaoui, P. Fertey, S. Lebegue, J. G. Angyan, M. Fourmigué and E. Espinosa, Cryst. Growth Des., 2013, 13, 3283–3289 CrossRef CAS; (b) H. Wang, J. Liua and W. Wang, Phys. Chem. Chem. Phys., 2018, 20, 5227–5234 RSC; (c) T. Clark, M. Hennemann, J. S. Murray and P. Politzer, J. Mol. Model., 2007, 13, 291–296 CrossRef CAS PubMed; (d) P. Politzer, J. S. Murray, T. Clark and G. Resnati, Phys. Chem. Chem. Phys., 2017, 19, 32166–32178 RSC.
  15. V. Kumar, C. Leroy and D. L. Bryce, CrystEngComm, 2018, 20, 6406–6411 RSC.
  16. (a) A. C. Legon, Phys. Chem. Chem. Phys., 2017, 19, 14884–14896 RSC; (b) W. Dong, Q. Li and S. Scheiner, Molecules, 2018, 23(1681), 1–7 Search PubMed.
  17. (a) G. E. Garrett, G. L. Gibson, R. N. Straus, D. S. Seferos and M. S. Taylor, J. Am. Chem. Soc., 2015, 137, 4126–4133 CrossRef CAS PubMed; (b) G. E. Garrett, E. I. Carrera, D. S. Seferos and M. S. Taylor, Chem. Commun., 2016, 52, 9881–9884 RSC.
  18. P. C. Vioglio, M. R. Chierotti and R. Gobetto, CrystEngComm, 2016, 18, 9173–9184 RSC.
  19. (a) J. Viger-Gravel, S. Leclerc, I. Korobkov and D. L. Bryce, J. Am. Chem. Soc., 2014, 136, 6929–6942 CrossRef CAS PubMed; (b) D. L. Bryce and J. Viger-Gravel, Top. Curr. Chem., 2015, 358, 183–204 CrossRef CAS PubMed; (c) P. C. Vioglio, P. M. J. Szell, M. R. Chierotti, R. Gobetto and D. L. Bryce, Chem. Sci., 2018, 9, 4555–4561 RSC; (d) P. M. J. Szell and D. L. Bryce, J. Phys. Chem. C, 2016, 120, 11121–11130 CrossRef CAS; (e) P. M. J. Szell, S. A. Gabriel, R. D. D. Gill, S. Y. H. Wan, B. Gabidullin and D. L. Bryce, Acta Crystallogr., Sect. C: Struct. Chem., 2017, 73, 157–167 CrossRef CAS; (f) P. M. J. Szell, B. Gabidullin and D. L. Bryce, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 2017, 73, 153–162 CrossRef CAS; (g) P. M. J. Szell, G. Cavallo, G. Terraneo, P. Metrangolo, B. Gabidullin and D. L. Bryce, Chem. – Eur. J., 2018, 24, 11364–11376 CrossRef CAS PubMed.
  20. (a) S. A. Southern and D. L. Bryce, J. Phys. Chem. A, 2015, 119, 11891–11899 CrossRef CAS; (b) S. A. Southern, D. Errulat, J. M. Frost, B. Gabidullin and D. L. Bryce, Faraday Discuss., 2017, 203, 165–186 RSC.
  21. C. Leroy, R. Johannson and D. L. Bryce, J. Phys. Chem. A, 2019, 123, 1030–1043 CrossRef CAS PubMed.
  22. (a) H. Duddeck, Prog. Nucl. Magn. Reson. Spectrosc., 1995, 27, 1–323 CrossRef CAS; (b) B. A. Demko and R. E. Wasylishen, Prog. Nucl. Magn. Reson. Spectrosc., 2009, 54, 208–238 CrossRef CAS; (c) S. Sen, D. C. Kaseman, I. Hung and Z. Gan, J. Phys. Chem. B, 2015, 119, 5747–5753 CrossRef CAS PubMed.
  23. (a) D. Wei, M. Han and Y. Yu, Nat. Sci. Rep., 2017, 6376, 1–8 Search PubMed; (b) M. N. Garaga, U. Werner-Zwanziger and J. W. Zwanziger, Inorg. Chem., 2018, 57, 892–898 CrossRef CAS.
  24. K. A. Scott and J. T. Njardarson, Top. Curr. Chem., 2018, 376(5), 1–34 CAS.
  25. (a) A. Sasaki, L. B. Ibarrab and S. Wimperis, Phys. Chem. Chem. Phys., 2017, 19, 24082–24089 RSC; (b) L. A. O’Dell and I. L. Moudrakovski, J. Magn. Reson., 2010, 207, 345–347 CrossRef PubMed; (c) A. Sutrisno, V. V. Terskikh and Y. Huang, Chem. Commun., 2009, 186–188 RSC.
  26. (a) A. F. Cozzolino, P. J. W. Elder and I. Vargas-Baca, Coord. Chem. Rev., 2011, 255, 1426–1438 CrossRef CAS; (b) A. Kremer, A. Fermi, N. Biot, J. Wouters and D. Bonifazi, Chem. – Eur. J., 2016, 22, 5665–5675 CrossRef CAS.
  27. (a) A. M. S. Riel, H.-T. Huynh, O. Jeannin, O. Berryman and M. Fourmigué, Cryst. Growth Des., 2019, 19, 1418–1425 CrossRef CAS; (b) H.-T. Huynh, O. Jeannin and M. Fourmigué, Chem. Commun., 2017, 53, 8467–8469 RSC; (c) O. Jeannin, H.-T. Huynh, A. M. S. Riel and M. Fourmigué, New J. Chem., 2018, 42, 10502–10509 RSC.
  28. H. K. Spencer, V. Lakshmikantham and M. P. Cava, J. Am. Chem. Soc., 1977, 99, 1470–1473 CrossRef CAS.
  29. P. Scilabra, J. S. Murray, G. Terraneo and G. Resnati, Cryst. Growth Des., 2019, 19, 1149–1154 CrossRef CAS.
  30. (a) J. Viger-Gravel, J. E. Meyer, I. Korobkov and D. L. Bryce, CrystEngComm, 2014, 16, 7285–7297 RSC; (b) J. Viger-Gravel, I. Korobkov and D. L. Bryce, Cryst. Growth Des., 2011, 11, 4984–4995 CrossRef CAS.
  31. (a) M. J. Collins, J. A. Ripmeester and J. F. Sawyer, J. Am. Chem. Soc., 1987, 109, 4113–4115 CrossRef CAS; (b) M. J. Collins, J. A. Ripmeester and J. F. Sawyer, J. Am. Chem. Soc., 1988, 110, 8583–8590 CrossRef CAS.
  32. (a) K. Grenader, M. Kind, L. Silies, A. Peters, J. W. Bats, M. Bolte and A. Terfort, J. Mol. Struct., 2013, 1039, 61–70 CrossRef CAS; (b) F. Cataldo, Polyhedron, 2000, 19, 681–688 CrossRef CAS; (c) F. Cataldo, J. Inorg. Organomet. Polym., 2003, 13, 87–98 CrossRef CAS.
  33. V. Vasylyeva, L. Catalano, C. Nervi, R. Gobetto, P. Metrangolo and G. Resnati, CrystEngComm, 2016, 18, 2247–2250 RSC.
  34. J. Herzfeld and A. E. Berger, J. Chem. Phys., 1980, 73, 6021–6030 CrossRef CAS.
  35. D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B. Alonso, J.-O. Durand, B. Bujoli, Z. Gan and G. Hoatson, Magn. Reson. Chem., 2002, 40, 70–76 CrossRef CAS.
  36. K. Eichele, WSolids NMR Simulation Package, Universität Tübingen, 2014.

Footnote

Electronic supplementary information (ESI) available: X-ray crystallographic, methods, additional NMR data, additional computational results. CCDC 1956409–1956411. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c9cp06267j

This journal is © the Owner Societies 2020