Competition between C–H⋯S and S–H⋯Cl H-bonds in a CHCl₃–H₂S complex: a combined matrix isolation IR spectroscopic and quantum chemical investigation

Abstract

H-bonded complexes between CHCl₃ and H₂S have been studied in a cold and inert argon matrix using IR spectroscopy. Both molecules were found to act as both a H-bond donor and acceptor, resulting in two different conformers. The more stable one (binding energy 3.25 kcal mol⁻¹) was bound by a C–H⋯S H-bond, while the less stable one (1.90 kcal mol⁻¹) by a S–H⋯Cl H-bond. The H-bonded complex formation has been confirmed by monitoring the spectral changes in ν_C–H and ν_S–H fundamental vibrations. The ν_C–H mode exhibited red shifts by 22.7 cm⁻¹ upon C–H⋯S H-bond formation, while the formation of a S–H⋯Cl H-bond resulted in 24.2 and 25.4 cm⁻¹ red shifts in the ν_S–H modes. The barrier for conversion of the less stable conformer to the more stable one was found to be 0.6 kcal mol⁻¹. The rigid matrix environment prevented any detectable population transfer that required significant relative movement of the monomeric moieties. Dispersion interaction was found to significantly contribute to the overall stabilization of both conformers, more so for the S–H⋯Cl H-bonded one.

---

1. Introduction

Hydrogen bonds (H-bonds) involving one or more sulfur atom(s), commonly known as sulfur-centered H-bonds (SCHBs), are known to play significant roles in various biological and chemical processes.^1–7 They are known to influence protein structures and their activities.^2,7–9 Also, they facilitate the catalytic activities of sulfur-containing enzymes by assisting in substrate recognition.^9–11 They have been found to influence crystal packing and biomolecular assembly along with molecular recognition.^12–17

SCHBs are considered weak compared to the classical H-bond because of S atoms' lower electronegativity (2.58 on the Pauling scale), leading to a weaker polarity of the S–H bonds compared to O–H and N–H bonds (electronegativity of O and N are 3.44 and 3.04, respectively). Dispersion interaction has been shown to play a vital role in stabilizing SCHBs, while electrostatics does the same in classical H-bonds.^18,19 However, in some cases, SCHBs have been predicted to exhibit strength comparable to classical H-bonds.^20,21

The sulfur atom has been found to act both as an H-bond donor (HBD)^22–24 and acceptor (HBA)^21,24–35 in SCHBs, both experimentally^{21,24,27–30,32,35} and theoretically,^{21,27,28,31,33,34} by various research groups. Recently, Biswal et al. reviewed the potential of the S atom to act as an HBA in binary H-bonded complexes.³⁶ In most of these cases, the SCHBs have an electronegative element as the HBD, while only a few studies on C–H⋯S H-bonds exist.^26,29–31 Presence of C–H⋯S H-bonds has been confirmed in various biomolecules as well as in organic and inorganic crystals.^15,16,32,37 Ghosh et al. have shown that the C–H⋯S H-bonds possess all the characteristics of H-bonding interaction.²⁶ Later, Fargher et al. discussed in detail its importance in small and large biomolecules.²⁵ Numerous instances of SCHBs with S as the acceptor^{25–29,31,38} and with a S–H group as the donor^23,39–45 are available in the literature. Zhou et al. reported that, between methionine and cysteine, the S–H group in the latter acts as a comparatively better HBD than S atoms in both amino acids as HBA.² The sulfhydryl group has also been shown to participate in S–H⋯π type H-bond interactions, apart from the σ type SCHBs.^17,38,46,47

H₂S is the smallest neutral molecule that can form SCHBs as both HBD and HBA, similar to H₂O in classical H-bonds. Recently, it has been shown that a H₂O–H₂S mixed dimer can have two almost isoenergetic conformers, separated only by ∼0.23 kcal mol⁻¹, bound by S–H⋯O and O–H⋯S H-bonds.⁴⁸ Furthermore, the interactions were found to greatly influence the structures of larger H₂O–H₂S clusters; O–H⋯S H-bonded conformers preferred cyclic structure, whereas S–H⋯O H-bonded conformers preferred caged structures.⁴⁸ Therefore, experimental investigation of the competition between two weak SCHBs, with the S atom acting as HBD and HBA, is important to understand how that competition dictates structures of molecular assemblies and, consequently, their properties.

A number of recent experiments are available in the literature that investigate how competition between two intermolecular H-bonds impacts the energetic and structure of binary complexes. Zhao et al. studied the competition between O–H⋯π and O–H⋯Cl interaction between 2,2,2-trifluoromethanol and 3-chloro-2-methyl-1-propene, where the O–H⋯π bound structure was favored by 1 kJ mol⁻¹ over its O–H⋯Cl counterpart.⁴⁹ Similarly, Mukhopadhyay et al.⁵⁰ showed that the O–H⋯O bound linear complex between p-fluorophenol and 2,5-dihydrofuran prefers a folded structure with an additional C–H⋯π interaction.⁵⁰ On the other hand, the intramolecular O–H⋯O H-bond in 1,2-cyclohexanedione is weakened upon the formation of an intermolecular C–H⋯O H-bond with chloroform.⁵¹ However, no such studies exist on the competition between two SCHBs, one with S atom as the HBA and the other with a S–H group as the HBD.

Very recently, we have carried out an extensive spectral assignment of the H₂S monomer and its clusters up to tetramer in an argon matrix,⁵² which prompted us to investigate a possible competition between the molecule's capacity to form very weak H-bonds, both as HBD and HBA. We have chosen CHCl₃ as the second molecule due to the following two reasons: firstly, H₂S binding with the C–H donor of CHCl₃ results in one of the weakest known SCHBs with S as the HBA. Secondly, S–H as the HBD and CHCl₃ as the HBA would result in an S–H⋯Cl H-bond, an interaction that has not yet been studied experimentally, to the best of our knowledge. In the present work, we have studied the H-bonded complex formation between H₂S, CHCl₃, and CDCl₃ in an argon matrix. Spectral modulations in the ν_S–H and ν_C–H/D modes of the two molecules have been monitored to assess how interplay between two types of intermolecular H-bonds dictates the structure, energetic and relative population of the different possible intermolecular complexes. Quantum chemical calculations have been carried out to aid in the interpretation of the experimental findings.

2. Methodology

2.1. Experimental

H₂S, CHCl₃, and CDCl₃ were procured from Sigma Aldrich and were used as such without any further purification. The purity of H₂S was 99.5%, and that of CHCl₃ was 99.8%, whereas the deuterium content in CDCl₃ was 99.8%. Ultra-high purity (N6.0 grade) Argon (Ar) was procured from Messer and was used as the carrier gas to form an inert matrix. All the experiments were carried out with a sample to carrier gas mixing ratio ranging from 1:1000 to 1:10 000, while the mixing ratios between the two samples were varied from 1:1 to 1:10.

The CHCl₃–H₂S–Ar gas mixtures in the desired mixing ratios were prepared within a custom-made stainless-steel gas mixing chamber. The gas mixing chamber is connected to vacuum pumps to maintain a base pressure of 1 × 10⁻⁵ mbar. A digital pressure gauge (Pfeiffer CMR361) was used to monitor the pressure during the preparation of gas mixtures. Subsequently, these gas mixtures were deposited through a nozzle (0.6 mm inner diameter) onto a KBr window mounted on a cold head, maintained at a temperature of 7 K. A closed-cycle helium cryostat (Janis CCS-100/204N) was used to maintain the cold head temperature. The cold head region was evacuated using a turbomolecular pump (Pfeiffer TC400) backed by a rotary vane pump (Pfeiffer Duo 6). The pressure (∼5 × 10⁻⁶ mbar) was monitored using a Pirani-Penning gauge (Pfeiffer MPT200). The gas mixtures were deposited for 60 minutes at a controlled flow rate within 3.8 mmol h⁻¹, maintained by a gas dosing valve (Pfeiffer EVN116).

The initially deposited gas mixtures were annealed by gradually increasing the temperature to 35 K to allow diffusion of samples and formation of the intermolecular complexes within the matrix. Subsequently, the matrix was cooled back to 7 K after annealing. All the spectra reported here were recorded using a Bruker Tensor II FTIR spectrophotometer, equipped with a DLaTGS detector and at 1 cm⁻¹ instrument resolution and summed over 500 scans. Further details of the experimental conditions maintained are discussed in our previous work.^53,54

2.2. Quantum chemical

Geometry optimizations and frequency calculations of all the studied monomers and association complexes (1:1 and 1:2) have been performed at the ωB97X-D/aug-cc-pV(D+d)Z level. (aug-cc-pV(X+d)Z basis set referred to as aV(X+d)Z hereafter). This particular level of calculation has been shown to provide good geometric and energetic results for intermolecular complexes bound by SCHBs, including C–H⋯S H-bonds.^26,48,55 Several other functionals in conjunction with higher basis sets, along with the MP2 method, were also tested. However, none of them predicted usable spectral shifts (Table S1, ESI†) and the ωB97X-D/aV(D+d)Z level was found to provide a significantly better estimation of spectral shifts, in both the ν_C–H and ν_S–H regions. The normal mode frequencies calculated at the ωB97X-D/aV(D+d)Z level have been scaled using a scaling factor of 0.963 to match with experimental results. This scaling factor has recently been shown to provide a good match with the experimental frequencies of H₂S dimers and oligomers.⁵² Furthermore, single point energy calculation of all the studied species has been performed at the CCSD(T)-F12/cc-pVTZ-F12 level for geometries optimized at the ωB97X-D/aV(D+d)Z level. We have shown in our earlier works,^48,55,56 for multiple systems bound by SCHBs, that the CCSD(T)-F12/cc-pVTZ-F12//ωB97X-D/aV(D+d)Z level provides reliable geometric and energetic values that closely agree with experimental and high level theoretical results.

Besides, localized molecular orbital energy decomposition analysis (LMO-EDA)⁵⁷ has been performed to separate the total interaction energies into various contributing energy components. LMO-EDA calculations have been carried out at the CCSD(T)/aV(Q+d)Z level. Binding energies calculated at this level were found to be appreciably close to both CCSD(T)-F12 values reported herein, CCSD(T)/CBS values for H₂S clusters¹⁸ and also the experimental binding energy of the H₂S dimer.¹⁸ According to this methodology, the total interaction energy (E_INT) can be written as follows:

E_INT = E_ES + E_EX + E_REP + E_POL + E_DISP (1)

where E_ES is the electrostatic interaction energy term, E_EX and E_REP are the exchange and repulsion energy components, respectively. E_POL and E_DISP are the polarization and dispersion energy components, respectively.

Furthermore, the percentage contribution of all the attractive energy components to the total attractive energy has been calculated using the following equation.

where X = ES, EX, POL, and DISP.

Furthermore, to confirm the formation of H-bonds, atoms in molecules (AIM) and natural bond orbital (NBO) analyses have also been performed. Geometry optimization, frequency calculation, and NBO analysis⁵⁸ have been performed using the G16 suite of programs.⁵⁹ AIM analyses⁶⁰ have been performed using the Multiwfn program package.⁶¹ CCSD(T)-F12 calculations have been performed using the ORCA suite of programs.⁶² LMO-EDA calculations were carried out using the GAMESS program.⁶³

3. Results & discussion

3.1 Experimental results

3.1.1 IR Spectra of CHCl₃, CDCl₃, and H₂S.
3.1.1.1 CHCl₃. ν _C–H transition of monomeric CHCl₃ in an Ar matrix is observed at 3053.7 cm⁻¹, in accordance with earlier reported results (Fig. 1A).^64,65 Two more bands were observed at 3034.7 and 3038.6 cm⁻¹, red shifted by −19.0 and −13.1 cm⁻¹, respectively, from the monomeric ν_C–H band. These bands have been attributed in earlier reports to the matrix site, and 1:1 CHCl₃–H₂O complex, respectively.^66,67 Another band at 3045.2 cm⁻¹ was observed in close proximity to a band, reported by Sruthi et al.⁶⁵ at 3045.7 cm⁻¹ and Pal et al.³⁰ at 3045 cm⁻¹. The band has been assigned to the ν_C–H band of the CHCl₃ dimer, as per the earlier reports.

Fig. 1 IR spectra of (A) CHCl₃, (B) CDCl₃, and (C) H₂S in the ν_C–H, ν_C–D, and ν_S–H regions at sample to Ar mixing ratios of 1:1000, 1:2000, and 1:5000 from top to bottom. Preannealed and annealed spectra are shown in black and red, respectively.

It is evident from Fig. 1A that only a very marginal amount of (CHCl₃)₂ was detected at a 1:5000 mixing ratio, while it was more pronounced at 1:2000, especially after annealing. On the other hand, dimer signature is very prominent even in the preannealed spectrum 1:1000 and becomes even stronger after annealing. Although experiments have been carried out at higher CHCl₃:Ar mixing proportion, only 1:5000 and 1:2000 CHCl₃:Ar mixing ratios have been considered to analyze 1:1 complex formation between CHCl₃ and H₂S, as any higher order complex formation would be less in these cases.

3.1.1.2 CDCl₃. Similarly, ν_C–D transition of monomeric CDCl₃, known to appear at 2276.8 cm⁻¹ in an Ar matrix,⁶⁵ has been identified accordingly (Fig. 1B). The intensity of the CDCl₃ν_C–D band in the Ar matrix is lower than that of the CHCl₃ν_C–H band, similar to what was observed in nonpolar solutions.⁶⁸ The CDCl₃:Ar mixing ratio was varied from 1:5000 to 1:1000. Similar to CHCl₃, a total of three bands at 2284.9, 2271.0, and 2266.4 cm⁻¹, corresponding to the site, homodimer, and CDCl₃–H₂O complex, respectively, were observed. CDCl₃:Ar mixing ratios of 1:5000 and 1:2000 have been considered to analyze the 1:1 complex between CDCl₃ and H₂S for the same reason as that of CHCl₃.
3.1.1.3 H₂S. There has been a long-standing ambiguity over assignment of ν_S–H bands of H₂S monomer (ν₁ and ν₃ modes) in an Ar matrix,^69,70 which has been recently addressed.⁵² Based on that, 2634.4 and 2648.0 cm⁻¹ bands have been assigned to monomer ν₁ and ν₃ modes, respectively. All the remaining bands appearing within 2550–2650 cm⁻¹ have been assigned to H₂S clusters, from dimer to tetramer. Among the most prominent bands, 2581.5 and 2568.4 cm⁻¹ bands were assigned to dimers and trimers, respectively (Fig. 1C).

H₂S has a high tendency to form oligomers in a cold, inert matrix.^52,71 It has been observed that at no H₂S:Ar mixing ratio chosen in this study, the dimer and oligomer bands become less intense than the monomer bands. Similar to CHCl₃ and CDCl₃, the H₂S:Ar mixing ratio has also been fixed at 1:5000 and 1:2000 while studying the 1:1 complex formation.

3.1.2 Effect of H₂S on ν_C–H mode of CHCl₃. The spectral changes brought about by H₂S in the ν_C–H mode of CHCl₃ are shown in Fig. 2A and Fig. S1 (ESI†). A new band appeared at 3031.0 cm⁻¹, red shifted by 22.7 cm⁻¹ from the monomeric ν_C–H transition upon the addition of H₂S. The CHCl₃ homodimer band at 3045.7 cm⁻¹, CHCl₃–H₂O complex band at 3038.6 cm⁻¹, and matrix sites at 3034.7 cm⁻¹ were found to decrease upon the addition of H₂S and also with increasing H₂S concentration. Therefore, the 3031.0 cm⁻¹ band is assigned to the C–H⋯S H-bonded ν_C–H mode of the CHCl₃–H₂S complex (Table 1). A bar diagram of the intensity modulation for the 3031.0 cm⁻¹ band has been provided in Fig. S2 (ESI†) for quantitative estimation of the spectral changes.

Fig. 2 Changes in the ν_C–H region of CHCl₃ (A) and ν_C–D region of CDCl₃, (B) upon addition of H₂S. The CHCl₃(CDCl₃):H₂S:Ar mixing ratios in A and B are 1:2:5000; 1:1:5000, and 1:0:5000, both from top to bottom. Preannealed and annealed spectra are shown in black and red, respectively.

Table 1 Experimentally observed ν_C–H/D, ν_S–H, Δν_C–H/D, and Δν_S–H (in cm⁻¹) in the complexes of CHCl₃, CDCl₃ and H₂S

	ν C–H/D	ΔνC–H/D	ν S–H	ΔνS–H
CHCl3	3053.7	—	—	—
H2S	—	—	2648.0	—
			2634.4
CHCl3–H2S	3031.0	−22.7	2622.6	−25.4
			2610.2	−24.2
CHCl3–(H2S)2	—	—	2558.4	−89.6
(CHCl3)2H2S	3026.5	−27.2	2599.6	−48.4
	3020.9	−32.8
CDCl3	2276.8	—	—	—
CDCl3–H2S	2260.9	−15.9	2622.6	−25.4
			2610.2	−24.2
CDCl3–(H2S)2	—	—	2558.4	−89.6
(CDCl3)2H2S	2258.1	−18.7	2599.6	−48.4
	2254.1	−22.7

---

Finally, it is to be noted that the development of a weak and broad spectral feature was also observed in the presence of H₂S on the lower wavenumber side of the 3031.0 cm⁻¹ band, with maxima at ∼3021 cm⁻¹. However, this feature was visible only after annealing. We understand that this feature could have resulted from the formation of mixed clusters larger than a dimer, but it could not be confidently assigned to any specific species. Further experiments were carried out using CDCl₃ to ascertain the H-bonded complex formation and support the above spectral assignment.

3.1.3 Effect of H₂S on the ν_C–D mode of CDCl₃. Similar to the ν_C–H mode of CHCl₃, a new band red shifted from the ν_C–D band of the CDCl₃ monomer appeared at 2260.9 cm⁻¹ upon addition of H₂S (Fig. 2B and Fig. S1, ESI†). The appearance of this new band was accompanied by a concordant decrement in the intensity of the bands due to the homodimer (2271.0 cm⁻¹), water complex (2266.4 cm⁻¹) and sites (2284.9 cm⁻¹), besides the monomer band (2276.8 cm⁻¹). The newly appearing band was also found to intensify with increasing H₂S concentration as well as upon annealing. Most importantly, the extent of red shift (15.9 cm⁻¹) is ∼1/√2 times the same in ν_C–H of CHCl₃ (22.7 cm⁻¹). Therefore, the 2260.9 cm⁻¹ band has been assigned to the ν_C–D band of the 1:1 complex between CDCl₃ and H₂S (Table 1). Similar to what was observed for CHCl₃ (Fig. 2A), a weak and broad spectral feature developed upon the addition of H₂S on the lower wavenumber side of the 2260.9 cm⁻¹ band, with maxima at ∼2254 cm⁻¹, visible only after annealing.

The above discussion clearly indicates the association complex formation of CHCl₃ (and CDCl₃) with H₂S. Spectral shift in ν_C–H/D indicates C–H/D to act as HBD and H₂S acts as a HBA through the S atom to form a C–H⋯S bound H-bonded complex between CHCl₃ (CDCl₃) and H₂S.

In addition, spectral modulation in the ν_S–H region of H₂S has also been studied due to the complex formation between CHCl₃ and CDCl₃ with H₂S. This will further aid in the characterization of CHCl₃ (CDCl₃) complexes with H₂S.

3.1.4 Effect of CHCl₃ (and CDCl₃) on ν_S–H mode of H₂S. Spectral changes in the ν_S–H mode of H₂S upon addition of CHCl₃ are shown in Fig. 3A and Fig. S1 (ESI†). The same for the addition of CDCl₃ is shown in Fig. 3B for the same set of H₂S:Ar mixing ratios. As evident from these figures, spectral changes brought about by both CHCl₃ and CDCl₃ were identical in nature. Three new bands at 2558.4, 2610.2, and 2622.6 cm⁻¹, all of which red shifted with respect to the ν₁ and ν₃ modes of the H₂S monomer (2634.4 and 2648.0 cm⁻¹, respectively) were observed. All three bands showed intensity enhancement upon annealing. A fourth band also appeared upon annealing at 2599.6 cm⁻¹. A bar diagram of the intensity modulation for the 2558.4, 2599.6, 2610.2, and 2622.6 cm⁻¹ bands has been provided in Fig. S2 (ESI†) for quantitative estimation of the spectral changes. The intensities of the 2610.2 and 2622.6 cm⁻¹ bands were found to increase with increase in concentration of CHCl₃ (and CDCl₃), while that of the 2558.4 cm⁻¹ bands was found to decrease. On the other hand, the 2599.6 cm⁻¹ band showed an even more intriguing behavior. It showed intensity enhancement with increment in H₂S concentration but opposite behavior with CHCl₃ (and CDCl₃) concentration. All these changes were found to be accompanied by a decrease of the H₂S monomer bands, along with the homodimer band at 2581.5 cm⁻¹, the trimer band at 2568.4 cm⁻¹, and other homomeric cluster bands within the 2550–2580 cm⁻¹ range.

Fig. 3 Changes in the ν_S–H region of H₂S upon the addition of CHCl₃ (A) and CDCl₃ (B). The CHCl₃(CDCl₃):H₂S:Ar mixing ratios in A and B are 2:1:5000; 1:1:5000, and 0:1:5000, both from top to bottom. Preannealed and annealed spectra are shown in black and red, respectively.

From the above discussion, the 2610.2 and 2622.6 cm⁻¹ bands have been assigned to the S–H⋯Cl H-bonded ν_S–H modes of CHCl₃–H₂S and CDCl₃–H₂S complexes (Table 1). It is worth mentioning that the spacing between these two bands (12.4 cm⁻¹) is very close to that between the ν₁ and ν₃ modes of the H₂S monomer (13.6 cm⁻¹), supporting the above assignment. The 2558.4 cm⁻¹ is assigned to CHCl₃(H₂S)₂ and CDCl₃(H₂S)₂ complexes, as its intensity increases with H₂S concentration, but not with CHCl₃ (and CDCl₃) concentration. Consequently, the 2599.6 cm⁻¹ band has been assigned to (CHCl₃)₂–H₂S and (CDCl₃)₂–H₂S complexes as its intensity increases with increasing CHCl₃ (and CDCl₃) concentration, but not with H₂S concentration.

Multiple bands of H₂S dimer and oligomers are known to appear within the 2550–2650 cm⁻¹ range of the ν_S–H region.⁵² In particular, one band at 2623.0 cm⁻¹, which is known to belong to the cyclic H₂S tetramer, is very close to the 2622.6 cm⁻¹ band belonging to the CHCl₃–H₂S complex.⁵² Therefore, experiments were conducted at a lower H₂S:Ar mixing ratio of 1:10 000 to minimize the probability of H₂S cluster formation that would help identify and analyse the complex bands. The CHCl₃:H₂S mixing ratios of 5:1 and 10:1 were used in these experiments to facilitate complex formation (Fig. 4A).

Fig. 4 (A) Changes in the ν_S–H region of H₂S upon addition of CHCl₃. The CHCl₃:H₂S:Ar mixing ratios are 0:1:10 000 (bottom trace; black); 5:1:10 000 (middle trace; red), and 10:1:10 000 (top trace; blue). (B) Intensity modulations of ν_S–H bands upon varying CHCl₃ concentration. The CHCl₃:H₂S:Ar mixing ratios are 5:1:10 000 (red), and 10:1:10 000 (blue).

No discernible bands are visible in the spectrum of only H₂S (bottom trace of Fig. 4A) near any of the three bands that appear upon the addition of CHCl₃ at 2599.6, 2610.2, and 2622.6 cm⁻¹ (middle and top traces). Furthermore, it is evident from the plot in Fig. 4B that the intensity of the 2610.2 and 2622.6 cm⁻¹ bands is nearly doubled as the concentration of CHCl₃ is doubled. However, the 2599.6 cm⁻¹ band is nearly quadrupled at the same time. This behavior signifies that the 2610.2 and 2622.6 cm⁻¹ bands are due to CHCl₃–H₂S complex formation, whereas the 2599.6 cm⁻¹ band is due to (CHCl₃)₂–H₂S complex formation. This band can undoubtedly be assigned to a 2:1 complex between CHCl₃ and H₂S.

Formation of both C–H⋯S and S–H⋯Cl H-bonds is evident from the discussions in the preceding subsections. It is also apparent that both red shifted ν_C–H and ν_S–H bands exhibit enhancement in intensities during annealing for every CHCl₃(CDCl₃):H₂S:Ar mixing ratio. This observation indicates enhancement in the population of both C–H⋯S and S–H⋯Cl H-bonded conformers, i.e., conversion of the less stable conformer to a more stable one is not significant upon annealing. Furthermore, the spectra of both the ν_C–H (ν_C–D) and ν_S–H regions have been shown together in Fig. S3–S5 (ESI†) for clear visibility of the relative intensities of the two spectral regions.

3.2 Computational results

Detailed quantum chemical calculations have been performed to support the various experimental observations and understand the spectral behaviors, which are discussed below.

3.2.1 CHCl₃–H₂S complex. Calculations at the ωB97X-D/aVDZ+d level predicted two conformers for the 1:1 CHCl₃–H₂S complex (Fig. 5). The more stable one (CS-1) is bound by a single C–H⋯S H-bond, while the less stable complex (CS-2) possesses three S–H⋯Cl H-bonds, formed by both S–H groups of H₂S. The C–H⋯S bond length was found to be 2.760 Å, and the binding energy of CS-1 was −3.25 kcal mol⁻¹ at the CCSD(T)-F12/cc-pVTZ-F12 level. On the other hand, the S–H⋯Cl bond lengths were within 2.981–3.284 Å, and the binding energy of CS-2 was only −1.9 kcal mol⁻¹. This larger binding energy of CS-1 over CS-2 would result in a higher population of the former and that was evident from the significantly more intense C–H⋯S H-bonded ν_C–H mode of CHCl₃ (and ν_C–D mode of CDCl₃) compared to the S–H⋯Cl H-bonded ν_S–H mode of H₂S.

Fig. 5 Molecular geometries of all conformers of CHCl₃–H₂S, CHCl₃(H₂S)₂ and (CHCl₃)₂H₂S optimized at the ωB97X-D/aV(D+d)Z level, along with their binding energetics, and ZPE corrected binding energies (in parenthesis) (in kcal mol⁻¹) calculated at the CCSD(T)-F12/cc-pVTZ-F12 level.

Normal mode frequency calculations predicted 21.2 cm⁻¹ red shift for the ν_C–H mode of CHCl₃ of CS-1, while a 4.4 cm⁻¹ blue shift for CS-2 (Table 2). On the other hand, ν₁ and ν₃ modes of H₂S were predicted to undergo 10.3 and 12.2 cm⁻¹ red shifts, respectively, in CS-1, whereas 16.3 and 17.2 cm⁻¹ red shift in CS-2, in the same order (Table 2). When compared against the experimentally observed spectral shifts, it is evident that the 3031.0 cm⁻¹ band, red shifted by 22.7 cm⁻¹ from the ν_C–H CHCl₃ monomer belongs to CS-1. In contrast, the 2610.2 and 2622.6 cm⁻¹ bands, red shifted by 24.2 and 25.4 cm⁻¹, respectively, from the ν_S–H bands of H₂S monomers belong to CS-2.

Table 2 Scaled normal mode frequencies of ν_C–H, ν^asym_S–H and ν^sym_S–H modes along with Δν_C–H, Δν^asym_S–H and Δν^sym_S–H modes (both in cm⁻¹) for all conformers of CHCl₃–H₂S, CHCl₃(H₂S)₂ and (CHCl₃)₂H₂S calculated at the ωB97X-D/aV(D+d)Z level

	ν C–H	ΔνC–H	ν ^asymS–H	Δν^asymS–H	ν ^symS–H	Δν^symS–H
Monomer	3065.4		2659.3		2644.0
CS-1	3045.0	−20.4	2649.4	−9.9	2632.3	−11.7
CS-2	3069.6	+4.2	2643.6	−15.7	2627.4	−16.6
CSS-1	3017.3	−48.1	2644.6	−14.7	2622.2	−21.8
			2641.4	−17.9	2556.8	−87.2
CSS-2	3044.7	−20.7	2646.5	−12.8	2630.7	−13.3
			2646.1	−13.2	2630.6	−13.4
CSS-3	3067.2	+1.8	2653.5	−5.8	2637.8	−6.2
			2644.8	−14.5	2588.5	−55.5
CCS-1	3074.2	+8.8	2637.4	−21.9	2623.0	−21.0
	3050.4	−15.0
CCS-2	3070.5	+5.1	2650.8	−8.5	2634.1	−9.9
	3060.8	−4.6
CCS-3	3081.4	+16.0	2648.5	−10.8	2634.0	−10.0
	3042.4	−22.9
CCS-4	3067.2	+1.8	2648.2	−11.1	2631.7	−12.3
	3056.8	−8.5
CCS-5	3061.1	−4.3	2644.3	−15.0	2628.9	−15.1
	3067.8	+2.4
CCS-6	3076.2	+10.8	2646.5	−12.8	2630.2	−13.7
	3045.6	−19.8
CCS-7	3090.5	+25.1	2644.3	−15.0	2628.8	−15.2
	3046.2	−19.2
CCS-8	3070.8	+5.4	2649.8	−9.5	2635.2	−8.8
	3072.4	+7.0
CCS-9	3075.6	+10.2	2647.9	−11.4	2631.7	−12.3
	3071.7	+6.4
CCS-10	3073.9	+8.6	2647.3	−12.0	2631.7	−12.3
	3069.8	+4.4
CCS-11	3070.3	+4.9	2648.2	−11.1	2630.9	−13.1
	3052.0	−13.4

---

Furthermore, different aspects of the C–H⋯S H-bond in the CHCl₃–H₂S complex are compared against earlier works. Fargher et al.²⁵ extensively examined C–H⋯S interaction in a comprehensive dataset encompassing over 423 000 C–H⋯S contacts within more than 86 000 structures. Their findings revealed that C–H⋯S contacts exhibit nonlinearity, with the most prevalent H⋯S contact occurring within 3.12–3.25 Å.

The C–H⋯S angle in the CS-1 complex was found to be 137.9° and S⋯H length 2.760 Å, which was notably shorter compared to the range reported by Fargher et al. A prior investigation by Fargher et al. has elucidated that the energies of C–H⋯S interactions typically range between 1–3 kcal mol⁻¹.²⁵ In this context, the −3.25 kcal mol⁻¹ binding energy of the C–H⋯S H-bonded CHCl₃–H₂S complex aligns well. Given the absence of documentation on S–H⋯Cl H-bonds in the existing literature to the best of our knowledge, our examination is limited to the comparison of only C–H⋯S interaction.

As discussed in the previous section, the intensities of the spectral bands belonging to both CS-1 and CS-2 increase upon annealing. Therefore, conversion of the less stable CS-2 to more stable CS-1 conformer must have been insignificant, if it happened at all. This could be due to kinetically unfavorable interconversion as a result of large barrier height insurmountable at 35 K temperature. A relaxed potential energy scan along ∠Cl–C–S was performed to understand whether the energy barrier was responsible for hindered interconversion. The calculated barrier was ∼0.6 kcal mol⁻¹ (Fig. 6), which was significantly higher than the thermal energy available at the annealing temperature.

Fig. 6 Potential energy landscape for the CHCl₃–H₂S complex along ∠Cl–C–S.

It is worth mentioning that annealing-induced interconversion between two conformers has been reported for a C–H⋯O H-bonded camphor–CHCl₃ complex separated by a very small barrier height of only 0.1–0.2 kcal mol⁻¹, while no such phenomenon could occur for O–H⋯O H-bonded camphor–methanol and camphor–phenol complexes where the barriers were 1.0 and 1.3 kcal mol⁻¹, respectively.⁷² The barrier in this case is somewhat intermediate of the abovementioned two barriers. Besides, interconversion between the two conformers of H-bonded complexes with camphor required the H-bond donors (i.e., chloroform, methanol, and phenol) to move from one side of the carbonyl oxygen to the other side. On the other hand, here in the CHCl₃–H₂S complex, H₂S needs to rotate by ∼150° and move ∼7.6 Å around CHCl₃ for interconversion between CS-1 and CS-2 to happen. A full rotational maneuver by such a larger distance within the rigid, inert matrix is deemed highly improbable. Consequently, annealing experiments resulted in the formation of both CS-1 and CS-2 conformers, with very little or no interconversion between the two.

As discussed in the preceding subsections, intensity modulations of a number of bands in both the ν_C–H/D and ν_S–H regions upon concentration variations indicate the presence of higher order complexes. This possibility finds strong support from the self-association of both CHCl₃ and H₂S, even at the lowest mixing ratios that were evident from experimental spectra in Fig. 1. In order to check that possibility, quantum chemical calculations have also been carried out for CHCl₃(H₂S)₂ and (CHCl₃)₂H₂S.

3.2.2 CHCl₃(H₂S)₂ complex. A total of three conformers were identified for CHCl₃(H₂S)₂ (Fig. 5). The most (CSS-1) and least stable (CSS-3) conformers have both H₂S molecules mutually connected via S–H⋯S H-bonds, while the remaining conformer (CSS-2) has CHCl₃ in the middle flanked by two H₂S molecules. As the binding energy of CSS-1 (−6.67 kcal mol⁻¹) is significantly higher than the remaining two (−5.20 and −4.91 kcal mol⁻¹ for CSS-2 and CSS-3, respectively), the population of CSS-2 and CSS-3 is expected to be negligibly small in the matrix. CSS-1 was found to be held together by one S–H⋯S and C–H⋯S H-bond each along with two S–H⋯Cl H-bonds. None of CSS-2 and CSS-3 have all three types of H-bonds present in CSS-1. While CSS-2 has one C–H⋯S H-bond and three S–H⋯Cl H-bonds, CSS-3 has one S–H⋯S H-bond and four S–H⋯Cl H-bonds. The C–H⋯S H-bond lengths were found to be within 2.626–2.728 Å, whereas the S–H⋯Cl H-bond lengths were within 2.894–3.226 Å. When compared against the same H-bonds in the two binary complexes (CS-1 and CS-2), both C–H⋯S and S–H⋯Cl H-bonds are found to be shorter. Furthermore, the S–H⋯S bond length was found to be 2.717 and 2.757 Å, in CSS-1 and CSS-3, respectively. Both these values and especially the one in CSS-1, are shorter compared to 2.772 Å in the H₂S dimer.⁵⁶ Therefore, it can be inferred that these weak SCHBs show mutual cooperative stabilization.

Normal mode frequency calculations predicted the ν_S–H mode of the S–H⋯S H-bonded S–H group to undergo 90.6 cm⁻¹ red shifts. When compared against experiments, it is evident that the 2558.4 cm⁻¹ band, red shifted by 89.6 cm⁻¹ from the monomeric ν₃ band of H₂S belongs to the S–H⋯S H-bonded S–H group of CSS-1. It is worth noting here that the ν_S–H transition of the S–H⋯S H-bonded S–H group in the H₂S dimer appears at 2581.5 cm⁻¹. A comparison of the two transition frequencies indicates that the ν_S–H transition of the S–H⋯S H-bonded S–H group shows 23.1 cm⁻¹ larger red shift in CHCl₃(H₂S)₂ compared to that in (H₂S)₂. This observation further supports the energetic and geometrical features of cooperative strengthening of H-bonds in the complex. On the other hand, the ν_C–H mode of CHCl₃ was predicted to show 50.0 cm⁻¹ red shift in CSS-1. As discussed in the previous subsection, experimental spectra showed the appearance of a broad feature near 3020.9 cm⁻¹. This broad feature could be due to the formation of a CHCl₃(H₂S)₂ complex. However, a clear assignment of the ν_C–H spectral band, corresponding to the CSS-1 conformer could not be achieved.

3.2.3 (CHCl₃)₂–H₂S complex. Similar to the CHCl₃(H₂S)₂ complex, calculations were also carried out for the (CHCl₃)₂H₂S complex. A total of eleven conformers were identified for (CHCl₃)₂H₂S (Fig. 5). The conformers were found to possess multiple NCIs, namely, C–H⋯S, S–H⋯Cl, and C–H⋯Cl H-bonds along with Cl⋯Cl, and S⋯Cl interactions, as is evident from both AIM and NBO analysis (Tables S2, S3 and Fig. S6, ESI†). The most stable conformer, CCS-1, was found to possess two S–H⋯Cl, C–H⋯S, and C–H⋯Cl H-bonds each. Only the least stable conformer (CCS-11) does not have any C–H⋯Cl H-bonds. Similarly, CCS-8, CCS-9, and CCS-10 do not possess any C–H⋯S H-bonds, while the S–H⋯Cl H-bond was absent in the CCS-6, CCS-7, and CCS-11 conformers. In all these cases, the C–H⋯S H-bond lengths were within 2.700–3.296 Å, S–H⋯Cl H-bond lengths were within 2.929–3.320 Å, and C–H⋯Cl H-bond lengths were found to vary within 2.762–3.790 Å. In some cases, the C–H⋯S and S–H⋯Cl length was found to be shorter than that observed in the two binary complexes (CS-1 and CS-2), suggesting a cooperative stabilization in them. The binding energies of the conformers were found to vary within −8.20 to −5.54 kcal mol⁻¹. As the binding energy of CCS-1 (−8.20 kcal mol⁻¹) is significantly higher than the remaining ten, the population of the rest would be negligibly small in the matrix. Also, the same is reflected in the experimental spectra, where only one identifiable band of (CHCl₃)₂–H₂S is observed.

Normal mode frequency calculations predicted the spectral shift in the ν_C–H mode of CHCl₃ to be within −23.9 cm⁻¹ to +26.1 cm⁻¹, whereas the shift in ν₁ and ν₃ modes of H₂S to be within −8.9 to −22.8 cm⁻¹ (Table 2). However, only one experimental band at 2599.6 cm⁻¹ in the ν_S–H region could be assigned to the (CHCl₃)₂H₂S complex. Calculation predicted a red shift of −22.8 and −21.8 cm⁻¹ in the ν₁ and ν₃ modes of H₂S in the CCS-1 conformer, where both S–H bonds are H-bonded with the Cl atoms of CHCl₃. These calculated spectral shifts are ∼5–6 cm⁻¹ greater than the same in S–H⋯Cl H-bonded binary complex CS-2. The 2599.6 cm⁻¹ band is 10.6 cm⁻¹ more red shifted compared to the experimental band assigned to the C-2 conformer.

Similar to CHCl₃(H₂S)₂, the large difference in binding energies between the two most stable (CHCl₃)₂H₂S conformers indicates that only CCS-1 would have a detectable population in the matrix. The remaining conformers would have negligible population to induce any noticeable spectral sign. Also, the spectral shift in ν_S–H of CCS-1 is close to that observed experimentally by the 2599.6 cm⁻¹ band. Therefore, the 2599.6 cm⁻¹ band can be assigned to the CCS-1 conformer. However, a clear assignment of the ν_C–H band corresponding to a particular conformer could not be achieved.

Based on the above discussions it can be stated that certain theoretical results don’t match very closely with the experimental findings, especially the predicted magnitude of spectral shifts due to complex formation with the observed values. This could be due to the fact that the calculation did not consider the influence of matrix environment and there is no robust theoretical model available yet that explicitly considers matrix effects on binding energies and spectral shifts.

3.2.4 Energy decomposition analysis. Dispersion interaction is known to play an important role in stabilizing SCHBs, and its contribution is generally greater than that in classical H-bonds.^25,26 Therefore, energy decomposition analysis has been performed for both binary complexes, i.e., CS-1 and CS-2, to obtain individual energy components of the total interaction energy to understand the nature and magnitude of various forces contributing to the total binding energies of these complexes (Table 3). For comparison purposes, the values for the S–H⋯S H-bonded H₂S dimer are also provided in Table 3.

Table 3 EDA for the CS-1 and CS-2 conformer compared with (H₂S)₂ calculated at the CCSD(T)/aV(Q+d)Z level. The energy values are given in kcal mol⁻¹

Complex	E ES	%EES	E EX	%EEX	E POL	%EPOL	E DISP	%EDISP	E REP	E TOT
a Ref. 18.
CS-1	−3.96	25.3	−7.28	46.6	−1.15	7.4	−3.24	20.7	12.24	−3.39
CS-2	−1.23	13.6	4.35	48.2	−0.40	4.4	−3.04	33.7	7.05	−1.97
(H2S)2^a	−2.51	23.9	−5.27	50.2	−0.93	8.9	−1.78	16.4	8.77	−1.72

---

The values in Table 3 show that E_EX exhibits the highest contribution among the attractive interactions. Upon comparing the percentage contributions of different attractive interactions in CS-1 and CS-2 with those of (H₂S)₂, it becomes evident that CS-1 demonstrates a marginally higher contribution from E_ES (25.3%) to that of (H₂S)₂ (23.9%), while CS-2 exhibits a significantly lower contribution (13.6%), almost half of the above values. The trend is exactly opposite for E_DISP; its percentage contribution in CS-2 (33.7%) is more than double that in (H₂S)₂ (16.4%), while the same in CS-1 (20.7%) is a little higher than in (H₂S)₂. The contribution of E_POL to overall stabilization is found to be lower, always below 10%, for all three binary complexes. Nevertheless, here also the value for CS-2 (4.4%) is significantly different compared to CS-1 (7.4%) and (H₂S)₂ (8.9%).

Therefore, in summary, it can be inferred that in terms of the percentage contributions of various interactions the C–H⋯S H-bond in CS-1 is very similar to the S–H⋯S H-bond in (H₂S)₂. On the other hand, electrostatics plays an appreciably minor role in stabilizing the S–H⋯Cl H-bonded CS-2, which is found to have a significantly higher contribution from dispersion interaction compared to both C–H⋯S and S–H⋯S H-bonds. In order to have a clearer idea about the contribution of dispersion interaction in stabilizing the CHCl₃–H₂S complex, its value against the total stabilization energies has been compared with earlier investigations of binary complexes featuring SCHBs (Table 4).

Table 4 A comparison table for the dispersion contribution to the overall stabilization energy. All values are in kcal mol⁻¹

Complex	SCHB	E DISP	E Total		Method	Ref.
H2S–C6H6	S–H⋯π	−3.56	−2.85	125	RVS-SCF	4
H2S–indole	S–H⋯π	−2.95	−3.34	88	HF/aug-cc-pVDZ	20
Indole–Me2S	N–H⋯S	−6.57	−5.75	114
H2S-3-methylindole	S–H⋯π	−3.02	−3.26	93
3-Methylindole-Me2S	N–H⋯S	−6.69	−5.66	118
H2S–indole	S–H⋯π	−7.45	−5.22	143	RVS-SCF	38
p-Cresol-diethylsulfide (TT)	O–H⋯S	−4.00	−7.30	55	HF/aug-cc-pVDZ	28
p-Cresol–H2S	O–H⋯S	−2.58	−3.68	70
H2S–MeOH	S–H⋯O	−2.13	−1.97	108	MP2/aug-cc-pVDZ	73
MeOH–H2S	O–H⋯S	−1.70	−1.61	106
p-Fluorophenol–H2S	O–H⋯S	−2.24	−4.12	54		74
H2S–H2S	S–H⋯S	−0.78	−0.87	89	MP2/6-311++G**	23
H2S–MeOH	S–H⋯O	−2.61	−1.60	61
H2S-diethylether (coplanar)	S–H⋯O	−2.84	−2.47	87
H2S-diethylether (perpendicular)	S–H⋯O	−2.67	−3.00	113
H2S-dibutylether (coplanar)	S–H⋯O	−3.12	−2.79	89
H2S-dibutylether (perpendicular)	S–H⋯O	−2.97	−3.54	119
H2S-1,4-dioxane (coplanar)	S–H⋯O	−2.63	−2.12	81
H2S-1,4-dioxane (perpendicular)	S–H⋯O	−2.35	−2.39	102
1,2,4,5-Tetracyanobenzene-H2S	C–H⋯S	−2.62	−4.09	64	ωB97XD/cc-pVDZ	26
1,2,4,5-Tetracyanobenzene-MeSH	C–H⋯S	−3.20	−5.15	62
1,2,4,5-Tetracyanobenzene-EtSH	C–H⋯S	−5.95	−5.84	102
1,2,4,5-Tetracyanobenzene-Me2S	C–H⋯S	−5.65	−6.02	94
1,2,4,5-Tetracyanobenzene-tetrahydrothiophene	C–H⋯S	−7.03	−6.93	101
p-Aminophenol-Me2S	O–H⋯S	−4.73	−6.20	75	MP2-cp/aug-cc-pVDZ	75
p-Cresol-Me2S	O–H⋯S	−4.64	−6.20	74
Phenol-Me2S	O–H⋯S	−4.56	−6.24	72
p-Fluorophenol-Me2S	O–H⋯S	−4.51	−6.44	70
cis-β-Naphthol-Me2S	O–H⋯S	−5.04	−6.62	75
p-Chlorophenol-Me2S	O–H⋯S	−4.55	−6.56	69
p-Cyanophenol-Me2S	O–H⋯S	−4.39	−6.94	64
HCONH2-diethyldisulfide	N–H⋯S	−7.15	−7.27	98	SAPT2+(3)δMP2/aug-cc-pVTZ	76
HCONH2–Me2S	N–H⋯S	−6.09	−7.79	78
CHCl3-dimethylsulfide	C–H⋯S	−6.97	−4.65	149	SAPT2/aug-cc-pVDZ	30
CHCl3–H2S (CS-1)	C–H⋯S	−3.24	−3.25	100	CCSD(T)/aV(Q+d)Z	This work
CHCl3–H2S (CS-2)	S–H⋯Cl	−3.04	−1.90	160

---

As evident from Table 4, the values of dispersion interaction are either lesser than or similar to the corresponding overall stabilization energies (i.e., entries in column 5 are ≤100%) for most of the studies involving C–H⋯S interactions, except for the CHCl₃-dimethylsulfide complex (149%).³⁰ Here also, the value for C–H⋯S H-bonded CS-1 (100%) matched closely with the earlier reports. Notably, there is no such scope of comparison for the S–H⋯Cl H-bonded CS-2 as no prior studies were available in the literature on S–H⋯Cl H-bonded complexes. Nevertheless, the value for CS-2 (160%) is the highest percentage dispersion contribution documented thus far for any binary complexes held by SCHBs. No previous work reports a value greater than 150%. These values provide a quantitative estimation of the contribution of dispersion interaction in total binding energy of the binary complexes bound by SCHBs. The very large value for CS-2 shows that dispersion interaction makes a major contribution towards overall stabilization of the S–H⋯Cl H-bond. When compared against earlier works, the newly reported S–H⋯Cl H-bond is found to have one of the largest dispersion contributions among all the SCHBs reported to date.

To further substantiate the observed H-bond formations in the complex, NBO analysis has been conducted and is discussed below.

3.2.4 NBO analysis. NBO analysis has been performed to find the hyperconjugative charge transfer (CT) interaction taking place in the complexes (Table S2, ESI†). The NBO analysis suggests charge transfer taking place from the lone pair of the HBAs (S and Cl atoms) to the antibonding orbital of the HBDs (C–H and S–H) (Table S2, ESI†). The CT interaction energy in the case of conformer CS-1 was found to be 4.49 kcal mol⁻¹, whereas the same in the conformer CS-2 was found to be 1.66 kcal mol⁻¹. Therefore, CT energy shows greater interaction strength of the C–H⋯S H-bond over the S–H⋯Cl H-bond. Similarly, in the case of the CSS-1 conformer of the CHCl₃(H₂S)₂ complex, the CT energy was found to be 5.28 and 1.22 kcal mol⁻¹ for n(S) → σ*(S–H), and n(Cl) → σ*(S–H) CT interaction, respectively. In the case of conformer CSS-2, the n(S) → σ*(S–H) CT energy accounts for 5.35 kcal mol⁻¹, whereas the same for n(Cl) → σ*(S–H) CT interaction accounts for 1.93 kcal mol⁻¹. Similarly, for the CSS-3 conformer, the n(S) → σ*(S–H) CT energy was found to be 5.39 kcal mol⁻¹, whereas the same for n(Cl) → σ*(S–H) CT interaction was found to be 2.23 kcal mol⁻¹. The CT interaction for all (CHCl₃)₂–H₂S conformers has also been found. The most stable CCS-1 conformer has n(S) → σ*(C–H) CT energy 5.95 and 0.4 kcal mol⁻¹, whereas n(Cl) → σ*(C–H), and n(Cl) → σ*(S–H) CT energy of 5.08, and 2.92 kcal mol⁻¹, respectively. The CT interactions are indicative of the formation of H-bond formation in these complexes.

Furthermore, AIM analyses have been performed, which shows (3,−1) BCPs at the bond path between molecules (Fig. S6, ESI†). The ρ_CP and ∇²ρ_CP values corresponding to the BCP have been provided in Table S3 (ESI†). This further supports the evidence for the formation of both C–H⋯S and S–H⋯Cl H-bonds. Also AIM analysis was found to corroborate the NBO results. The NCI plot clearly shows the attractive regions between the two molecules, which is indicated by the minima observed in the reduced density gradient plot (Fig. S7, ESI†). The reduced density gradient approaches zero, whenever the sign λ₂ × ρ approaches zero. These analyses further support the formation of a H-bonded complex in the system.

4. Summary

A H-bonded CHCl₃–H₂S complex has been investigated in a cold argon matrix using FTIR spectroscopy. A new band at 3031.0 cm⁻¹, red shifted by 22.7 cm⁻¹ from the monomeric ν_C–H band of CHCl₃ was observed due to the formation of a C–H⋯S H-bonded CHCl₃–H₂S binary complex. Similar spectral features were observed for CDCl₃, where the new band at 2260.9 cm⁻¹ was 15.9 cm⁻¹ red shifted from the monomeric ν_C–D band. On the other hand, two new bands at 2610.2 and 2622.6 cm⁻¹, red shifted 24.2 and 25.4 cm⁻¹ from the monomeric ν₁ and ν₃ modes of H₂S appeared due to a S–H⋯Cl H-bonded conformer of CHCl₃–H₂S binary complex. Calculations predicted the C–H⋯S H-bonded conformer of the CHCl₃–H₂S binary complex (CS-1) to be 1.35 kcal mol⁻¹ more stable than its S–H⋯Cl H-bonded counterpart (CS-2), duly supporting the much higher intensity of the ν_C–H band compared to that of ν_S–H bands of the binary complexes.

No sign of interconversion between the two conformers could be identified experimentally. Subsequently, a relaxed PES scan revealed the energy barrier of converting CS-2 to CS-1 (∼0.6 kcal mol⁻¹) to be too high to overcome at annealing temperature (35 K) and the large relative motions of the two monomers to be improbable in the rigid matrix environment.

Intensity patterns of two more bands at 2558.4 and 2599.6 cm⁻¹ in the ν_S–H region indicated the presence of larger complexes: the first one belonging to S–H⋯S H-bonded ν_S–H mode of CHCl₃-(H₂S)₂ and the second one the S–H⋯Cl H-bonded ν_S–H mode of (CHCl₃)₂–H₂S. Both H-bonds exhibited cooperative strengthening as evident from the 23.1 cm⁻¹ larger spectral shift for the S–H⋯S H-bond in H₂S dimer and 10.6 cm⁻¹ higher spectral shift for the S–H⋯Cl H-bond in the CHCl₃–H₂S binary complex. Both the above bands were assigned to the most stable conformers of CHCl₃–(H₂S)₂ and (CHCl₃)₂–H₂S complexes based on calculated normal mode frequencies.

Energy decomposition analyses reveal dispersion interaction to play an important role in stabilizing SCHBs, which accounted for over one fifth of all the attractive interactions for the C–H⋯S H-bonded complex and over one third for S–H⋯Cl H-bonded one. In fact, the magnitude of dispersion contribution against overall interaction energy of the binary complex has been found to be the greatest for the latter among all the SCHBs reported to date. Additionally, AIM and NBO calculations performed have been found to support the above findings and they were found to corroborate with each other. This study represents the first experimental evidence of an S–H⋯Cl H-bond interaction. Additionally, it offers a comprehensive understanding of the coexistence of two weak SCHBs, namely, C–H⋯S and S–H⋯Cl H-bonds.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

BKO, Monu, and AK acknowledge MNIT Jaipur, CSIR, and CSIR-UGC, respectively, for research fellowships. BB acknowledges SERB-DST, Govt. of India for financial support through sanctioned project (CRG/2020/001601).

References

1. E. N. Baker and R. E. Hubbard, Prog. Biophys. Mol. Biol., 1984, 44, 97–179.
2. P. Zhou, F. Tian, F. Lv and Z. Shang, Proteins: Struct., Funct., Bioinf., 2009, 76, 151–163.
3. L.-H. Xu, Q. Liu, R. Suenram, F. J. Lovas, A. H. Walker, J. Jensen and A. Samuels, J. Mol. Spectrosc., 2004, 228, 243–250.
4. D. L. Crittenden, J. Phys. Chem. A, 2009, 113, 1663–1669.
5. T.-a Okamura, Y. Ushijima, Y. Omi and K. Onitsuka, Inorg. Chem., 2013, 52, 381–394.
6. T.-A. Okamura, T. Kaga, S. Yamashita, R. Furuya and K. Onitsuka, J. Org. Chem., 2017, 82, 2187–2192.
7. B. W. Beck, Q. Xie and T. Ichiye, Biophys. J., 2001, 81, 601–613.
8. K. Mazmanian, K. Sargsyan, C. D. Grauffel, T. Dudev and C. Lim, J. Phys. Chem. B, 2016, 120, 10288–10296.
9. L. M. Gregoret, S. D. Rader, R. J. Fletterick and F. E. Cohen, Proteins: Struct., Funct., Bioinf., 1991, 9, 99–107.
10. A. D. Pagar, M. D. Patil, D. T. Flood, T. H. Yoo, P. E. Dawson and H. Yun, Chem. Rev., 2021, 121, 6173–6245.
11. R. Reddi, K. K. Singarapu, D. Pal and A. Addlagatta, Mol. BioSyst., 2016, 12, 2408–2416.
12. V. S. Minkov and E. V. Boldyreva, J. Phys. Chem. B, 2013, 117, 14247–14260.
13. S. Pathania, R. K. Narang and R. K. Rawal, Eur. J. Med. Chem., 2019, 180, 486–508.
14. A. Nangia and G. R. Desiraju, J. Mol. Struct., 1999, 474, 65–79.
15. H.-P. Zhou, Y.-M. Zhu, J.-J. Chen, Z.-J. Hu, J.-Y. Wu, Y. Xie, M.-H. Jiang, X.-T. Tao and Y.-P. Tian, Inorg. Chem. Commun., 2006, 9, 90–92.
16. J. J. Novoa, M. C. Rovira, C. Rovira, J. Veciana and J. Tarrés, Adv. Mater., 1995, 7, 233.
17. R. T. Saragi, M. Juanes, R. Pinacho, J. E. Rubio, J. A. Fernández and A. Lesarri, Symmetry, 2021, 13, 2022.
18. S. Sarkar and B. Bandyopadhyay, Phys. Chem. Chem. Phys., 2019, 21, 25439–25448.
19. S. Ghosh, S. Bhattacharyya and S. Wategaonkar, J. Phys. Chem. A, 2015, 119, 10863–10870.
20. H. S. Biswal and S. Wategaonkar, J. Phys. Chem. A, 2009, 113, 12763–12773.
21. C. L. Andersen, C. S. Jensen, K. Mackeprang, L. Du, S. Jørgensen and H. G. Kjaergaard, J. Phys. Chem. A, 2014, 118, 11074–11082.
22. A. Paul and R. Thomas, J. Mol. Liq., 2022, 348, 118078.
23. A. Bhattacherjee, Y. Matsuda, A. Fujii and S. Wategaonkar, J. Phys. Chem. A, 2015, 119, 1117–1126.
24. V. R. Mundlapati, S. Ghosh, A. Bhattacherjee, P. Tiwari and H. S. Biswal, J. Phys. Chem. Lett., 2015, 6, 1385–1389.
25. H. A. Fargher, T. J. Sherbow, M. M. Haley, D. W. Johnson and M. D. Pluth, Chem. Soc. Rev., 2022, 51, 1454–1469.
26. S. Ghosh, P. Chopra and S. Wategaonkar, Phys. Chem. Chem. Phys., 2020, 22, 17482–17493.
27. H. S. Biswal, E. Gloaguen, Y. Loquais, B. Tardivel and M. Mons, J. Phys. Chem. Lett., 2012, 3, 755–759.
28. H. S. Biswal and S. Wategaonkar, J. Phys. Chem. A, 2010, 114, 5947–5957.
29. E. J. Cocinero, R. Sánchez, S. Blanco, A. Lesarri, J. C. López and J. L. Alonso, Chem. Phys. Lett., 2005, 402, 4–10.
30. D. Pal, H. Charaya and S. Chakraborty, ChemPhysChem, 2023, 24, e202300124.
31. M. Domagała and S. J. Grabowski, Chem. Phys., 2010, 367, 1–6.
32. L. K. Macreadie, A. J. Edwards, A. S. Chesman and D. R. Turner, Aust. J. Chem., 2014, 67, 1829–1839.
33. S. François, M.-M. Rohmer, M. Bénard, A. C. Moreland and T. B. Rauchfuss, J. Am. Chem. Soc., 2000, 122, 12743–12750.
34. A. Nowroozi, H. Roohi, H. Hajiabadi, H. Raissi, E. Khalilinia and M. N. Birgan, Comput. Theor. Chem., 2011, 963, 517–524.
35. E.-M. Sung and M. D. Harmony, J. Am. Chem. Soc., 1977, 99, 5603–5608.
36. H. S. Biswal, S. Bhattacharyya, A. Bhattacherjee and S. Wategaonkar, Int. Rev. Phys. Chem., 2015, 34, 99–160.
37. C. Ma, Q. Zhang, R. Zhang and D. Wang, Chem. – Eur. J., 2006, 12, 420–428.
38. H. S. Biswal and S. Wategaonkar, J. Phys. Chem. A, 2009, 113, 12774–12782.
39. A. Chand, D. K. Sahoo, A. Rana, S. Jena and H. S. Biswal, Acc. Chem. Res., 2020, 53, 1580–1592.
40. G. Duan, V. H. Smith Jr and D. F. Weaver, Mol. Phys., 2001, 99, 1689–1699.
41. K. Grzechnik, K. Rutkowski and Z. Mielke, J. Mol. Struct., 2012, 1009, 96–102.
42. I. A. Lobo, P. A. Robertson, L. Villani, D. J. Wilson and E. G. Robertson, J. Phys. Chem. A, 2018, 122, 7171–7180.
43. M. H. Graneri, D. A. Wild and A. J. McKinley, J. Mol. Spectrosc., 2021, 378, 111440.
44. K. K. Mishra, K. Borish, G. Singh, P. Panwaria, S. Metya, M. Madhusudhan and A. Das, J. Phys. Chem. Lett., 2021, 12, 1228–1235.
45. M. Solimannejad, M. Gharabaghi and S. Scheiner, J. Chem. Phys., 2011, 134, 024312.
46. Y. Jin, W. Li, R. T. Saragi, M. Juanes, C. Pérez, A. Lesarri and G. Feng, Phys. Chem. Chem. Phys., 2023, 25, 12174–12181.
47. E. M. Cabaleiro-Lago, J. Rodríguez-Otero and Á. Peña-Gallego, J. Chem. Phys., 2011, 135, 134310.
48. Monu, B. K. Oram and B. Bandyopadhyay, Comput. Theor. Chem., 2022, 1213, 113740.
49. H. Zhao, S. Tang, Q. Zhang and L. Du, RSC Adv., 2017, 7, 22485–22491.
50. D. P. Mukhopadhyay, S. Biswas, A. Chattopadhyay and T. Chakraborty, J. Phys. Chem. A, 2018, 122, 3787–3797.
51. A. K. Samanta, P. Banerjee, B. Bandyopadhyay, P. Pandey and T. Chakraborty, J. Phys. Chem. A, 2017, 121, 6012–6020.
52. Monu, B. K. Oram and B. Bandyopadhyay, J. Phys. Chem. A, 2024, 128, 6703–6713.
53. B. K. Oram, Monu and B. Bandyopadhyay, Mol. Phys., 2022, 120, e2061387.
54. B. K. Oram, Monu and B. Bandyopadhyay, Spectrochim. Acta, Part A, 2020, 230, 118070.
55. Monu, B. K. Oram and B. Bandyopadhyay, Comput. Theor. Chem., 2023, 1225, 114133.
56. Monu, B. K. Oram and B. Bandyopadhyay, Phys. Chem. Chem. Phys., 2021, 23, 18044–18057.
57. P. Su and H. Li, J. Chem. Phys., 2009, 131, 014102.
58. A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926.
59. M. E. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, G. Scalmani, V. Barone, G. Petersson and H. Nakatsuji, Gaussian, Inc., Wallingford CT, 2016.
60. R. F. Bader, Acc. Chem. Res., 1985, 18, 9–15.
61. T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580–592.
62. F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2, 73–78.
63. M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen and S. Su, J. Comput. Chem., 1993, 14, 1347–1363.
64. S. Oswald and S. Coussan, Low Temp. Phys., 2019, 45, 639–648.
65. P. Sruthi, S. Chandra, N. Ramanathan and K. Sundararajan, J. Chem. Phys., 2020, 153, 174305.
66. S. Paulson and A. Barnes, J. Mol. Struct., 1982, 80, 151–158.
67. F. Ito, J. Chem. Phys., 2012, 137, 014505.
68. B. K. Oram, Monu and B. Bandyopadhyay, J. Mol. Struct., 2023, 136749.
69. E. Isoniemi, M. Petterson and L. L. Khriachtchev, J. Phys. Chem. A, 1999, 103, 679–685.
70. A. Barnes and J. Howells, J. Chem. Soc., Faraday Trans. 2, 1972, 68, 729–736.
71. S. Y. Tang and C. W. Brown, J. Raman Spectrosc., 1974, 2, 209–215.
72. P. Banerjee, P. Pandey and B. Bandyopadhyay, Spectrochim. Acta, Part A, 2019, 209, 186–195.
73. A. Bhattacherjee, Y. Matsuda, A. Fujii and S. Wategaonkar, ChemPhysChem, 2013, 14, 905–914.
74. S. Bhattacharyya and S. Wategaonkar, J. Phys. Chem. A, 2014, 118, 9386–9396.
75. S. Bhattacharyya, S. Ghosh and S. Wategaonkar, Phys. Chem. Chem. Phys., 2021, 23, 5718–5739.
76. T. Lu, J. Zhang, Y. Xu, Z. Wang, G. Feng and Z. Zeng, Spectrochim. Acta, Part A, 2023, 288, 122199.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cp03509g

---