Selective synthesis of the missing tiara-like Ni₁₀, Ni₅ and Ni₆ thiolates by the C–S bond cleavage of bis(thioether) molecules with a DFT study†

Suparno Debnath , Sandip Giri and Ganesan Mani *
Department of Chemistry, Indian Institute of Technology Kharagpur, Kharagpur, 721302 India. E-mail: gmani@chem.iitkgp.ac.in; Fax: +91 3222 282252; Tel: +91 3222 282320

First published on 14th August 2024

---

---

Introduction

The X-ray structure of the first cyclic nickel thiolate [Ni₆(SC₂H₅)₁₂] was reported in 1965¹ by Woodward and co-workers. Since then, several other macrocyclic toroidal systems having the general formula [Ni(SR)₂]_n, where R = alkyl and aryl group and n = 4–6, 8–12, have been reported.^2–17 These atomically precise molecular multinuclear homoleptic complexes are called nanoclusters and have attracted attention owing to their unusual optical, electronic and catalytic properties which are different from those of nanoparticles.^18,19 For example, toroidal nickel thiolates have been demonstrated as catalysts for several important reactions such as the reduction of dinitrogen to ammonia²⁰ and dihydrogen^21,22 and oxygen^23–25 evolution, oxidation²⁶ and reduction^17,27 of organic compounds. In addition, they have fascinating cyclic structures having precise compositions with structural features including three-layers of S/Ni/S atoms with a prismatic polyhedral arrangement of the bridging nonbonding coplanar sulfur atoms in a staggered conformation, resembling a tiara or toroid with a specific cavity size depending on the number of nickel atoms. These toroidal nickel complexes are air stable and handled conveniently. In general, their synthesis is relatively easy, and conventionally involves reactions between nickel salts and either neutral thiol or thiolate species which, however, often result in the formation of a mixture of cyclic nickel thiolates.^17,28,29 In some cases, two different thiolates^30,31 and disulfides^32,33 have been used in synthesis, giving a mixture of products^31,32 with poor yields.³³ In addition, the treatment of NaSPh with nickel salt has been reported to only give the homoleptic cyclic [Ni(SR)₂]_n system with n = 9 and 11.¹⁶ Recently, the metathesis reaction of the hydrated nickel(II) chloride with the in situ generated sodium arylthiolate has been reported to give a mixture of new cyclic systems [Ni(S-tol-p)₂]_n (n = 9–12).¹⁷ These reactions have prompted us to find an alternative synthetic route that is easy and will give selectively one cyclic nickel thiolate, which is of paramount importance for follow-up studies such as those on reactivity, catalysis and biological applications.³⁴ However, the targeted synthesis of any missing member of a toroidal system [Ni(SR)₂]_n having different n values and alkyl or aryl groups remains a challenge and open for research studies. Besides, successful synthesis of larger toroidal systems (n ≥ 9) is probably governed by the statement of Dance et al. that a guest species inside the cavity or a re-entrant ligand is required for stabilization.¹⁵

Transition metal (TM) mediated and TM-free C–S bond cleavage methods have extensively been used in organic synthesis for forming new C–C and C–heteroatom bonds.^35,36 In addition, the C–S bond cleavage reactions have been used to construct metal-thiolate complexes which are relevant to metalloenzymes, metal–organic frameworks,³⁷ and other thiolate complexes^38,39 and remain to be a process in development for hydrodesulfurization of fossil fuels.⁴⁰ Wu and co-workers have reported two novel high-nuclearity nickel thiolates having ring and cage structures by the C–S bond cleavage.⁴¹ However, toroidal nickel thiolates have not been reported via C–S bond cleavage of organosulfur compounds to date.

As a continuation of our research on pincer metal complexes and their catalytic properties,^42,43 we set out to synthesize a few pincer frameworks containing sulfur atoms instead of phosphorus owing to the demonstrated hemilability of thioether molecules.^44–47 Herein, we report the synthesis of a new class of sulfur analogues of the reported pincer bis(phosphine) ligands^42,43 and their unexpected C–S bond cleavages in the presence of nickel salts leading to selective and easy synthetic routes for the hitherto unknown toroidal Ni(II) thiolates: [Ni₁₀(SPh)₂₀] containing encapsulated CH₂Cl₂ and [Ni₅(SCH₂Ph)₁₀] as well as for the known cyclic system [Ni₆(SCH₂Ph)₁₂] with single crystal X-ray structures, isolation of by-products and DFT calculations.

Results and discussion

Synthesis of bis(thioether)

To expand the scope of the reaction between primary diamines, secondary phosphine and paraformaldehyde giving the procarbene bis(phosphine) ligand,^42,43 the reaction between diamines, thiols and paraformaldehyde in a 1:2:3 molar ratio under heating conditions was carried out to give compounds L1–L5 in excellent yields up to 99% (Scheme 1). More discussion is given in the ESI.† The UV-vis spectra of L1–L5 are given in Fig. 1. In contrast to L1–L4, the presence of aromatic π-electrons in L5 gives two absorption maxima close to the visible region at 337 and 347 nm in CH₃CN, which can be due to n to π* transitions, accounting for its light pink color.

Having synthesized a series of bis(thioether) compounds, the 1:1 molar ratio reaction between L1 and NiCl₂·6H₂O in methanol was carried out at room temperature or under refluxing conditions (Scheme 2). To our surprise, a dark brown precipitate was formed immediately. The precipitate was only sparingly soluble in polar solvents such as DCM and THF; crystallization from these solvents failed to give crystals. Gratifyingly, when a solution of compound L1 in DCM was layered with an ethanolic solution of NiCl₂·6H₂O, dark brown block shaped crystals began to form as the solutions diffused slowly. The single crystal X-ray diffraction study revealed that it is not a pincer NHC bound nickel(II) complex as expected based on the phosphine analogue,^42,43 but is the new toroidal Ni(II) thiolate molecule, [Ni₁₀(SPh)₂₀] 1, obtained as crystals in 53% yield. All bands in the IR spectrum of these crystals match with those of the precipitate obtained from the same reaction at room temperature or 75 °C, confirming that the same product 1 formed in both cases (see ESI, Fig. S33†). This is also supported by the CHN analysis of both crystals and the precipitate. In addition, crystals of 1 were also obtained from the reaction of L1 with other nickel salts such as Ni(OAc)₂·4H₂O, Ni(ClO₄)₂·6H₂O, or Ni(NO₃)₂·6H₂O, suggesting that the formation of Ni₁₀ thiolate does not depend on the counter anion. Furthermore, L3 having the phenylthio group reacts in a similar way to NiCl₂·6H₂O or Ni(OAc)₂·4H₂O to result in the formation of complex 1. The slow diffusion of the ethanolic solution of NiCl₂·6H₂O or Ni(OAc)₂·4H₂O into the CH₂Cl₂ solution of compound L3 at room temperature afforded single crystals of 1.

To gain more insights into the reaction, the reaction filtrate after collecting the precipitate of 1 was subjected to column chromatographic separation and analyzed by the HRMS method. Diphenyl disulfide 2 was isolated in 10–17% yield from both the reactions of L1 and L3. The HRMS(ESI⁺) spectra of the reaction mixture display the [M + H]⁺ ion peaks for the formation of compounds 3, 3a and 4 as minor by-products (see ESI, Fig. S38 and S40†). These observations suggest that L1 and L3 undergo facile C–S bond cleavages induced by the nickel(II) ions in methanol to give complex 1 and 3 or 4 containing a methoxy group which can be from the methanol solvent. It is noted that Wu and co-workers have reported the high-nuclearity nickel thiolates by the C–S bond cleavage without isolating by-products.⁴¹

A plausible mechanism for the formation of toroidal nickel thiolates is given in the ESI (Fig. S1†) and is based on the isolated and detected by-products formed in their synthesis (see ESI, Fig. S38–S40 and S50–S52† for HRMS). The coordination of the thioether sulfur to the nickel atom can lead to the formation of the pyrimidinium ion linked nickel hydride complex Bvia the abstraction of hydride from the “NCH₂N” moiety (Chart 1). This type of abstraction has been proposed and supported by DFT calculations for the formation of the bis(phosphine)-NHC coordinated nickel complex.⁴⁸ Complex B undergoes C–S bond cleavage owing to the strong Ni–S bond. Subsequently, the hydride transfer and nucleophilic attack by the solvent methanol with the elimination of HCl take place. Eventually, a bis(thiolate) nickel complex is formed, which then oligomerizes to give the cyclic nickel thiolates depending upon the size of the alkyl/aryl group attached to the sulfur atom.

The isolation of the new cyclic Ni₁₀ thiolate encouraged us to carry out analogous reactions using other compounds L2, L4 and L5. As shown in Scheme 2, again to our surprise, a new cyclic Ni₅ thiolate complex 5 was isolated as a dark brown solid in 52% yield from the reaction of L2 with NiCl₂·6H₂O or Ni(OAc)₂·4H₂O in methanol under refluxing conditions. Conversely, the cyclic Ni₆ thiolate complex 8 was isolated as a dark brown precipitate in 56% yield from the reaction of L4 with Ni(acac)₂ in CH₃CN at room temperature (Scheme 2). Column separations of the reaction mixtures of L2 and L4 gave the corresponding disulfide by-product 6 in about 20% yield. In addition to compound 3 detected by the HRMS(ESI⁺) method, a new compound 7 resulting from one C–S bond cleavage was isolated in 23% yield from the reaction of L2. The formation of these by-products in the reactions of L2 and L4 suggests a similar mechanism of the C–S bond cleavage leading to the Ni₅ and Ni₆ thiolates and that the nuclearity of the complex is controlled by the steric bulk of the substituent attached to the sulfur atom. The presence of benzyl groups in 5 and 8 makes them more soluble in DCM and THF compared to 1, and hence, 5 and 8 were characterized by the NMR method as well. Dark brown needle shaped single crystals of complexes 5 and 8 were grown by the slow evaporation of their solutions in DCM/methanol.

Interestingly, unlike L1–L4, no precipitation or an abrupt color change was observed when L5 was treated with NiCl₂·6H₂O or Ni(OAc)₂·4H₂O in a 1:1 molar ratio in methanol at room temperature or 75 °C, suggesting no reaction or no C–S bond cleavage (Scheme 2). L5 was not consumed as monitored by TLC even after refluxing for 48 h. Even this reaction in the presence of 4 equiv. of KOBu^t failed to yield the expected carbene coordinated nickel complex (see the ESI for HRMS spectra, Fig. S58†). The main difference between these ligands is the presence of the fused aromatic naphthalene moiety, which makes L5 more planar and rigid. Owing to geometric constraints, nickel is unable to activate the C–H bond of the “NCH₂N” moiety like B in Chart 1 and hence subsequent C–S bond cleavage does not take place. Conversely, L1–L4 have saturated flexible backbones and the C–H bond activation readily takes place.

Structure description

The molecular structure of 1 is given in Fig. 2. It is unique and the missing member of the reported series [Ni(SPh)₂]_n with n = 9 and 11 synthesized using sodium arylthiolate. Complex 1 contains a guest species inside its cavity and exhibits a shape that is different from that of the recently reported guest-free system [Ni(S-tol-p)₂]₁₀.¹⁷ It crystallizes in the triclinic P space group and the asymmetric unit constitutes a half molecule containing five Ni and ten thiolates along with 1.5 molecules of CH₂Cl₂ as the solvent of crystallization. The centre of inversion in the middle of the molecule generates the full molecule possessing a pseudo-D_2h symmetry formulated as 1·3CH₂Cl₂. Its top view (Fig. 2A) shows ‘two opposite vertices truncated rhombus shape’ having mixed convex/concave vertices. The encapsulated CH₂Cl₂ at the special position, disordered over two positions as generated by the inversion centre, is tilted with respect to the coplanar arrangement of 10 nickel atoms and possesses short contacts with the sulfur atoms with distances ranging from 3.266 to 3.499 Å, which are shorter than the sum of the van der Waals radii of S and Cl atoms (3.50 to 3.70 Å),⁴⁹ indicating weak non-covalent interactions. These interactions can be the driving force for its formation, as suggested by Dance and coworkers.¹⁵ The side view of the skeleton shows three layers of S/Ni/S atoms; the middle nickel layer is sandwiched between two parallel layers (up and down) of coplanar sulfur atoms each lying between two nickel atoms (staggered), giving a tiara-like view (Fig. 2B).

Important metric parameters of 1 are summarized in Tables S3 and S4 (ESI†) and their average values are given Table 1 along with reported values of other cyclic systems. This molecule has a highly deformed or distorted convex/concave-like geometry owing to the large difference in the three inner-ring dihedral angles (δ = 199, 118 and 114°) between two adjacent NiS₄ square planes. This difference is less in the reported [Ni(S-tol-p)₂]₁₀ (131.6 and 207.8°).¹⁷ The average interior angle (θ₁ = 82.0°) is smaller than 83° found in [Ni(SPh)₂]_n (n = 9 and 11) and the average exterior angle (θ₂ = 97.8°) is close to those of the other reported toroidal nickel thiolates. The other interior angle (Ni–S–Ni, φ) is of two types owing to the shape of the molecule. The average φ angle of 80.5° at the convex region is smaller than 94.7° at the concave region, both of which lie between the values found in the reported SPh coordinated toroidal systems with n = 9 and 11. The angle (φ) of 80.5° in 1 is smaller than 86.1° in [Ni(S-tol-p)₂]₁₀ owing to its different shape (Table 1).

---

The average Ni–S bond length is 2.21 Å which is close to those found in the reported toroidal systems [Ni(SPh)₂]_n (n = 4, 5, 6, 8, 9, 10 and 12) (Table 1). The average nonbonding Ni⋯Ni distance at the convex regions (2.856(7) Å) is shorter than that (3.235(6) Å) at the concave regions. The overall average nonbonding Ni⋯Ni distance of 3.01 Å in 1 is expected to be greater than that in [Ni(SPh)₂]₉ (3.05 Å) owing to the expansion of the ring and, however, this closer value is attributed to the presence of a 1:1.5 ratio of concave:convex regions in 1 as compared to the 1:2 ratio in [Ni(SPh)₂]₉. Nevertheless, 3.01 Å is slightly shorter than 3.08 Å found in the larger size molecule [Ni(SPh)₂]₁₁. The thiolate phenyl groups adopt three different conformations about the nickel square planes in 1: alternate axial and equatorial, two axial and two equatorial, and three axial and one equatorial. The average deviation of the nickel atom from its mean square plane is 0.08 Å. The distances between Ni atoms which are colinear across the ring range from 4.702 to 11.864 Å.

The X-ray structure of the pentanuclear complex, [Ni₅(SCH₂Ph)₁₀], 5, is given in Fig. 3. It is a new derivative of the Ni₅ toroidal molecule containing benzyl mercaptide groups crystallized in the monoclinic P2₁/c space group. The molecule contains five contiguous nickel square planes linked to each other through their opposite edges containing the bridging nonbonded S⋯S atoms. The top view shows two distorted pentagon shapes; the inner and outer portions are staggered (Fig. 3A). The inner pentagon is formed by five nickel atoms with a Ni⋯Ni distance of 2.8196(4) Å and the outer one is formed by the bridging thiolate vertices. Each pentagon has an approximate D_5h point group. The benzyl mercaptide groups are oriented alternately in the axial and equatorial positions about the square planes of nickel atoms Ni1 and Ni3–Ni5 except Ni2 in which two are in axial and two in equatorial positions owing to an odd number of nickel atoms. The average Ni–S and Ni–Ni distances of 2.20 Å and 2.82 Å, respectively, are close to those in other reported pentagons (Table 1). Similarly, the average interior S–Ni–S (θ₁ = 81.5°), exterior S–Ni–S (θ₂ = 98.8°) and another interior Ni–S–Ni (φ = 79.8°) angles are consistent with those of the previously reported Ni₅ toroidal molecule (Table 1).

The X-ray structure of the toroidal hexanuclear complex [Ni₆(SCH₂Ph)₁₂] 8 is given in Fig. 4. The molecule crystallizes in the tetragonal P2₁c space group. The asymmetric unit constitutes half of the molecule and the whole molecule was generated by the C₂ axis passing through Ni1 and Ni4 atoms. The molecule adopts a regular convex hexagon shape formed by six nickel atoms bridged by 12 benzyl mercaptides without any encapsulated solvent of crystallization. Although the structure of 8 is similar to that of the reported toroidal Ni₆, complex 8 was synthesized by the direct reaction between L4 and nickel salt at room temperature with quantification of yield, unlike the synthetic method for the reported Ni₆ that involves nickel salt, tetraoctylammonium bromide, NaBH₄, PhCH₂SH and a temperature around 10 °C.⁵⁰ In addition, the by-product dibenzyl disulfide was isolated. In accordance with the results reported by Dance et al., as the ring size increases, the angle at the bridging sulfur atom (Ni–S–Ni, φ) and the Ni⋯Ni distance tend to increase with a decrease in the pyramidality around the sulfur atom. As given in Table 1, the average φ value increases from 79.8° in Ni₅ (5) to 83.1° in Ni₆ (8) and the Ni⋯Ni distance also increases from 2.82 to 2.91 Å. The average of the sum of the angles around each sulfur atom decreases from 305.0° in 5 to 304.5° in 8. As observed in other structures, the Ni–S, θ₁ and θ₂ values are not much affected as the ring size increases.

The UV-vis absorption spectra of complexes 1, 5 and 8 in THF are given in Fig. 5. Complex 5 shows intense absorption peaks at 348 and 415 nm and a broad peak at 526 nm, which are similar to those in the UV-vis spectrum of [Ni₅(μ-(ⁱPr)₂NCH₂CH₂S)₁₀].⁶ Complex 8 shows a similar spectrum but its peaks are shifted and appear at 352, 410, 521 and 579 nm with more intensity, as the number of Ni atoms increases (n = 5 → 6) (Fig. 5A). Similar absorption spectra have been reported for [Ni₆(PET)₁₂] (PET = phenylethanethiolate).^23,25,28,29 The UV-vis spectrum of 1 exhibits peaks at 339, 468 and 603 nm (Fig. 5B), appearing similar to those shown by [Ni(S-tol-p)₂]₁₀.¹⁷

DFT calculations

DFT calculations were performed to understand the electronic structure of all three nickel toroidal molecules. Geometry optimization was performed using the X-ray structure coordinates. The calculated metric parameters are close to those found in their X-ray structures. The Kohn–Sham orbital energy levels in each complex are given in Fig. 6 and the diagrams of selected HOMOs and LUMOs are given in the ESI (Fig. S4–S6†). The HOMO–LUMO gaps in 5, 8 and 1 were calculated to be 3.012, 3.169, and 3.212 eV, respectively (see ESI, Table S9†). This energy gap increases as the number of nickel atoms in the structure increases, indicating their increasing relative stability.

Excited state analysis followed by calculations of the percentages of holes and electrons responsible for specific transitions were performed to generate UV-vis spectra of 5, 8 and 1 shown in Fig. 7. Two transitions with significant oscillator strength were found for 5, that is, the ground state S0 to the excited state S74 (322.47 nm, 3.8448 eV), and S0 to S29 (415.76 nm, 2.9821 eV) (see ESI, Table S10†). These peaks are closer to the experimentally observed absorptions at 348 and 415 nm for 5 (Fig. 5A). Similarly, for 8, the energy of transitions corresponding to S0 to S71 (343.39 nm, 3.6106 eV) and S0 to S35 (411.38 nm, 3.0139 eV) match with the experimental values. For the Ni₁₀ system 1, to get the calculated spectrum, the geometry optimization was performed after replacing all phenyl groups with methyl groups owing to the rapid increase in computational cost with the number of states. The optimized geometry of [Ni₁₀(SMe)₂₀] appears similar to that of 1 and is given in the ESI (Fig. S7 and S8†). For the hypothetic [Ni₁₀(SMe)₂₀], three transitions were found: two with significant oscillator strengths, S0 to S179 (349.98 nm, 3.5426 eV) and S0 to S60 (449.51 nm, 2.7582 eV), and one with lower oscillator strength, S0 to S34 (609.26 nm, 2.0350 eV), which are closer to the experimentally observed absorption peaks found for 1 (Fig. 5B).

To understand the nature of interactions with which one dichloromethane is present inside the cavity of Ni₁₀ thiolate 1, a non-covalent interaction (NCI)⁵¹ analysis was performed. Topological parameters including electron density ρ(r), Laplacian value Δρ(r), electrostatic potential V(r), and interaction energy (E_int) at the bond critical point (3, −1) between sulfur and chlorine atoms are summarized in Table S18.† The relatively low electron density and positive value of Laplacian indicate the formation of nonbonding weak interactions. The negative value of interaction energy indicates an attractive interaction. Furthermore, the quantitative negative value of sign(λ₂)ρ shows the formation of non-covalent bonding. In addition, the plot of the reduced density gradient (RDG) and the sign(λ₂)ρ provides a visual overview of different interactions, represented in various colors, where λ₂ is the second largest eigen value of the electron density Hessian matrix (Fig. S9†) in which the blue and green isosurface colors indicate weak interactions, while red indicates repulsive interactions. The molecular graphs of 1 obtained from QTAIM and non-covalent interaction methods are given in Fig. 8. The green isosurface and orange critical points between sulfur and chlorine atoms indicate the non-covalent bonding in 1 (also see Fig. S10 in the ESI†).

Conclusions

In conclusion, the contrasting reaction of a series of new bis(thioether) molecules with nickel salts led to the isolation of the hitherto unknown cyclic Ni(II) thiolates (1 and 5) formed by the C–S bond cleavages. This method of synthesis is direct and selective, yielding one cyclic system. It is also convenient as both the bis(thioether) compounds and cyclic nickel complexes were synthesized by one pot methods. The nuclearity of the ring system depends not only on the nature of the substituents attached to the sulfur atom, but also on the backbone of the bis(thioether). Their X-ray structures show an eye-catching tiara-like view with a toroidal shape owing to the prismatic staggered arrangement of sulfur atoms. Their electronic structures, absorption bands, and the non-covalent bonding interactions by the guest species inside the cavity were studied by DFT calculations. Synthesis of cyclic thiolates of other metals with varying alkyl/aryl groups attached to the sulfur atoms in L1–L5 and catalytic applications are under progress.

Experimental section

Synthesis of 1,3-bis((phenylthio)methyl)hexahydropyrimidine, L1

A suspension of 1,3-diaminopropane (4.20 mL, 3.696 g, 49.86 mmol), paraformaldehyde (4.954 g, 163.44 mmol) and thiophenol (10.70 mL, 11.556 g, 104.89 mmol) was heated to 100 °C using an oil bath with stirring for 1 h. The resulting colourless viscous oily compound was cooled to 0 °C for 4 h to give L1 as a colourless crystalline solid (16.278 g, 49.251 mmol, 99%). ¹H NMR (400 MHz, CDCl₃): δ 1.68 (m, 2H, NCH₂CH₂CH₂N), 2.70 (t, J(H,H) = 4.0, 4H, NCH₂CH₂CH₂N), 3.52 (s, 2H, NCH₂N), 4.42 (s, 4H, NCH₂S), 7.20–7.47 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 23.6, 49.7, 64.5, 71.3, 126.7, 129.1, 131.6, 137.5. UV-vis NIR (10⁻³ M, CH₃CN): λ_max(nm) = 204 (0.913), 252 (0.533). HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₁₈H₂₃N₂S₂: 331.1297, found: 331.1275.

Synthesis of 1,3-bis((benzylthio)methyl)hexahydropyrimidine, L2

A suspension of 1,3-diaminopropane (4.20 mL, 3.696 g, 49.86 mmol), paraformaldehyde (4.954 g, 163.44 mmol) and benzylmercaptan (12.30 mL, 13.01 g, 104.75 mmol) was heated to 100 °C using an oil bath with stirring for 1 h to give L2 as an oily viscous liquid (17.820 g, 49.70 mmol, 99%). ¹H NMR (500 MHz, CDCl₃): δ 1.70 (m, 2H, NCH₂CH₂CH₂N), 2.69 (t, J(H,H) = 5.0, 4H, NCH₂CH₂CH₂N), 3.44 (s, 2H, NCH₂N), 3.76 (d, J(H,H) = 5.0, 4H, SCH₂C), 3.79 (d, J(H,H) = 5.0, 4H, NCH₂S), 7.27–7.34 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 23.3, 36.1, 50.4, 58.2, 71.8, 127.1, 128.6, 129.2, 138.8. UV-vis NIR (10⁻³ M, CH₃CN): λ_max(nm) = 207 (0.880). HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₂₀H₂₇N₂S₂: 359.1610, found: 359.1626.

Synthesis of 4-methyl-1,3-bis((phenylthio)methyl)imidazolidine, L3

A suspension of 1,2-diaminopropane (4.26 mL, 3.706 g, 50.67 mmol), paraformaldehyde (4.954 g, 163.44 mmol) and thiophenol (10.70 mL, 11.556 g, 104.89 mmol) was heated to 100 °C using an oil bath with stirring for 1 h to give L3 as an oily viscous compound (16.310 g, 49.350 mmol, 97%). ¹H NMR (400 MHz, CDCl₃): δ 0.89 (d, J(H,H) = 8.0, 3H, CH₃), 2.51 (t, J(H,H) = 6.0, 1H, imidazolidine CH₂), 2.95 (dd, J(H,H) = 6.0, 6.0, 1H, imidazolidine CH₂), 3.14 (m, 1H, imidazolidine CH), 3.67 (dd, J(H,H) = 24.0, 4.0, 2H, NCH₂N), 4.34 (s, 2H, NCH₂S), 4.39 (d, J(H,H) = 4.0, 2H, NCH₂S), 7.05–7.35 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 17.2, 54.7, 57.1, 59.2, 61.2, 70.6, 126.6, 126.7, 129.1, 131.3, 131.8, 137.4, 137.5. UV-vis NIR (10⁻³ M, CH₃CN): λ_max(nm) = 205 (1.100), 254 (0.746). HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₁₈H₂₃N₂S₂: 331.1297, found: 331.1279.

Synthesis of 1,3-bis((benzylthio)methyl)-4-methylimidazolidine, L4

A suspension of 1,2-diaminopropane (4.26 mL, 3.706 g, 50.67 mmol), paraformaldehyde (4.954 g, 163.44 mmol) and benzylmercaptan (12.30 mL, 13.01 g, 104.75 mmol) was heated to 100 °C using an oil bath with stirring for 1 h to give L4 as a colourless viscous oily compound (17.785 g, 49.60 mmol, 98%). ¹H NMR (500 MHz, CDCl₃): δ 1.10 (br m, 3H, imidazoline CH₃), 2.54 (br s, 1H, imidazolidine CH), 3.07 (br s, 2H, imidazolidine CH₂), 3.58 (t, J(H,H) = 2.5, 1H, NCH₂N), 3.78 (m, 1H, NCH₂N), 3.83 (br m, 6H, NCH₂S and NCH₂C), 3.86 (m, 1H, NCH₂S), 3.97–4.00 (m, 1H, NCH₂S), 7.27–7.36 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 18.2, 35.8, 36.0, 54.6, 56.0, 56.2, 58.3, 72.6, 127.0, 127.1, 128.6, 129.1, 129.2, 138.9, 138.9. UV-vis NIR (10⁻³ M, CH₃CN): λ_max(nm) = 208 (0.903). HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₂₀H₂₇N₂S₂: 359.1610, found: 359.1650.

Synthesis of 1,3-bis((phenylthio)methyl)-2,3-dihydro-1H-perimidine, L5

A suspension of 1,8-diaminonaphthalene (7.91 g, 50.0 mmol), paraformaldehyde (4.954 g, 163.44 mmol) and thiophenol (10.70 mL, 11.556 g, 104.89 mmol) was heated to 100 °C using an oil bath with stirring for 1 h to give L5 as a light pink coloured solid (20.46 g, 49.35 mmol, 99%). ¹H NMR (500 MHz, CDCl₃): δ 4.27 (s, 2H, NCH₂N), 4.99 (s, 4H, NCH₂S), 6.69 (dd, J(H,H) = 7.5, 2.5, 2H, naphthalene ring), 7.31–7.48 (m, 14H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 58.5, 65.8, 106.2, 116.4, 119.2, 126.6, 127.5, 129.2, 132.9, 135.1, 136.00, 140.3. UV-vis NIR (10⁻³ M, CH₃CN): λ_max(nm) = 208 (1.040), 237 (0.814), 337 (0.273), 347 (0.287). HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₂₅H₂₃N₂S₂: 415.1297, found: 415.1256.

Synthesis of [Ni₁₀(SPh)₂₀], 1

To a solution of NiCl₂·6H₂O (0.396 g, 1.666 mmol) in methanol (10 mL) was added dropwise a solution of ligand L1 (0.500 g, 1.513 mmol) in methanol (10 mL) with stirring. A dark brown coloured precipitate was formed immediately. The resulting dark brown reaction mixture was refluxed at 75 °C for 16 h, and then allowed to cool down to room temperature and filtered. The dark brown solid was thoroughly washed with methanol (3 × 5 mL) and dried under vacuum to give 1 (0.276 g, 0.091 mmol, 55%). The complex 1 is insoluble in most of the organic solvents and only sparingly soluble in DCM and THF.

A dark brown precipitate of complex 1 (0.276 g, 0.091 mmol, 55%) was also obtained by refluxing a solution of NiCl₂·6H₂O (0.396 g, 1.666 mmol) or Ni(OAc)₂·4H₂O (0.414 g, 1.664 mmol) in methanol (20 mL) and ligand L3 (0.500 g, 1.513 mmol) at 75 °C for 16 h.

Isolation of by-products from the above reaction

After collecting the brown precipitate of complex 1 from the reaction mixture using L1 and NiCl₂·6H₂O, the light green coloured filtrate solution was concentrated using a rotary evaporator and loaded onto a silica gel column. Elution using ethyl acetate/hexane (v/v = 1:5) gave diphenyl disulfide 2 as a colourless solid (0.033 g, 0.151 mmol, 10% based on ligand L1). Similarly, the reaction filtrate obtained using L3 and NiCl₂·6H₂O was concentrated using the rotary evaporator and loaded onto a silica gel column. Elution using ethyl acetate/hexane (v/v = 1:5) gave diphenyl disulfide 2 as a colorless solid (0.056 g, 0.256 mmol, 17% based on ligand L3).

Crystallization of 1

In a test tube, to a solution of ligand L1 (0.200 g, 0.605 mmol) in DCM (10 mL) was added slowly ethanol (2 mL) followed by a solution of NiCl₂·6H₂O (0.158 g, 0.665 mmol), Ni(OAc)₂·4H₂O (0.166 g, 0.667 mmol), Ni(ClO₄)₂·6H₂O (0.243 g, 0.664 mmol) or Ni(NO₃)₂·6H₂O (0.194 g, 0.667 mmol) in ethanol (10 mL) so that two layers were formed. The slow diffusion of these two layers of solutions gave 1 as dark brown crystals formed over a period of 10 days at room temperature.

Similarly, single crystals of complex 1 (0.103 g, 0.372 mmol) were also obtained by the slow diffusion of a solution of NiCl₂·6H₂O (0.158 g, 0.665 mmol) or Ni(OAc)₂·4H₂O (0.166 g, 0.667 mmol) in ethanol (10 mL) into a solution of ligand L3 (0.200 g, 0.605 mmol) in DCM (10 mL) at room temperature. In both cases, crystals were separated, washed with methanol (3 × 5 mL) followed by n-pentane (3 × 5 mL), and dried under vacuum to give complex 1 (0.107 g, 0.035 mmol, 53%).

For 1: ¹H NMR (500 MHz, CDCl₃): δ 7.02–7.77 (br m, 100 H, ArH). UV-vis NIR (1.67 × 10⁻⁵ M, THF): λ_max(nm) = 339 (0.434), 468 (0.373), 603 (0.073). Anal. calcd for 1·CH₂Cl₂ C₁₂₁H₁₀₂Cl₂Ni₁₀S₂₀: C, 50.90; H, 3.60; S, 22.46, found: C, 50.80; H, 3.36; S, 22.17.

For 2: ¹H NMR (500 MHz, CDCl₃): δ 7.21–7.51 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 127.3, 127.7, 129.2, 137.2.

The other minor by-products 1-(methoxymethyl)-3-methylhexahydropyrimidne 3, 1-methyl-3-((phenylthio)methyl)hexahydropyrimidine 3a and 1-(methoxymethyl)-3,4-dimethylimidazolidine 4 could not be isolated, but were detected by the HRMS method. For 3: HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₇H₁₇N₂O: 145.1335, found: 145.1332. For 3a: HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₁₂H₁₉N₂S: 223.1263, found: 223.1273. For 4: HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₇H₁₇N₂O: 145.1335, found: 145.1332.

Synthesis of [Ni₅(SCH₂Ph)₁₀], 5

To a solution of NiCl₂·6H₂O (0.364 g, 1.531 mmol) or Ni(OAc)₂·4H₂O (0.382 g, 1.535 mmol) in methanol (10 mL) was added slowly a solution of ligand L2 (0.500 g, 1.394 mmol) in methanol (10 mL) with stirring. The resulting dark brown reaction mixture was refluxed at 75 °C for 16 h, and then allowed to cool down to room temperature and filtered. The dark brown solid was washed with methanol (3 × 5 mL) and dried under vacuum to give 5 (0.243 g, 0.159 mmol, 52%). This complex is relatively well soluble in DCM and THF.

Isolation of by-products from the above reaction

After collecting the brown precipitate of complex 5, the filtrate solution was concentrated using the rotary evaporator and loaded onto a silica gel column. Elution using ethyl acetate/hexane (v/v = 1:5) gave the first fraction from which the solvents were removed under vacuum to give PhCH₂SSCH₂Ph 6 as a colorless powdery solid (0.069 g, 0.280 mmol, 20%). Further elution using chloroform/methanol (v/v = 90/10) gave the second fraction from which the solvents were removed under vacuum to give 1-((benzylthio)methyl)-3-methylhexahydropyrimidine 7 as a colorless solid (0.076 g, 0.321 mmol, 23%).

Crystallization: The whole solid of 5 was first dissolved in DCM (20 mL) and then methanol (20 mL) was added. The resulting clear solution was kept for slow evaporation at room temperature to give dark brown needle shaped crystals of 5 for 3 days. The crystals were separated, washed with methanol (3 × 5 mL) and dried under vacuum (0.234 g, 0.153 mmol, 50%). ¹H NMR (500 MHz, CDCl₃): δ = 3.00 (br s, 20H, SCH₂Ph), 6.91–7.62 (br m, 50H, ArH). UV-vis NIR (0.33 × 10⁻⁶ M, THF): λ_max(nm) = 348 (0.663), 415 (0.196), 526 (0.092). Anal. calcd for C₇₀H₇₀Ni₅S₁₀: C, 55.12; H, 4.63; S, 21.02, found: C, 55.55; H, 4.39; S, 20.62.

For 6: ¹H NMR (500 MHz, CDCl₃): δ 3.60 (s, 4H, SCH₂C), 7.22–7.32 (m, 10H, ArH). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 43.5, 127.6, 128.6, 129.5, 137.5.

For 7: ¹H NMR (500 MHz, CDCl₃): δ 1.63 (m, 2H, NCH₂CH₂CH₂N), 2.17 (s, 3H, NCH₃), 2.37 (s, 2H, NCH₂CH₂CH₂NCH₃), 2.53 (t, J(H,H) = 5.0, 2H, NCH₂CH₂CH₂NCH₃), 3.11 (s, 2H, NCH₂N), 3.68 (s, 2H, SCH₂C), 3.69 (s, 2H, NCH₂S), 7.20–7.24 (m, 5H, Ph). ¹³C{¹H} NMR (125.75 MHz, CDCl₃): δ 23.6, 36.0, 42.8, 49.9, 54.2, 58.4, 75.5, 127.0, 128.5, 129.1, 138.8. HRMS (ESI⁺): calcd m/z for [M + H]⁺ C₁₃H₂₁N₂S: 237.1420, found: 237.1437.

Synthesis of [Ni₆(SCH₂Ph)₁₂], 8

To a solution of [Ni(acac)₂] (0.394 g, 1.534 mmol) in acetonitrile (10 mL) was added slowly ligand L4 (0.500 g, 1.394 mmol) in acetonitrile (10 mL) with stirring at room temperature. A dark brown colored precipitate was formed immediately. The solution was stirred for 16 h and then filtered. The dark brown solid was washed with methanol (3 × 5 mL) and dried under vacuum to give complex 8 (0.262 g, 0.143 mmol, 56%).

Isolation of by-products from the above reaction

After collecting brown precipitate of complex 8, the filtrate solution was concentrated using the rotary evaporator and loaded onto a silica gel column. Elution using ethyl acetate/hexane (v/v = 1:5) gave dibenzyl disulfide 6 as a colourless powdery solid (0.065 g, 0.264 mmol, 19% based on ligand L4).

Crystallization: The whole solid of 8 was dissolved in DCM (20 mL) and then methanol (20 mL) was added. The resulting clear solution was kept for slow evaporation at room temperature to give 8 as dark brown needle shaped crystals after 3 days. The crystals were separated, washed with methanol (3 × 5 mL), and dried under vacuum (0.253 g, 0.138 mmol, 54%). ¹H NMR (500 MHz, CDCl₃): δ 2.90 (s, 12H, SCH₂C), 3.19 (s, 12H, SCH₂C), 6.89–7.63 (m, 60H, ArH). UV-vis NIR (0.33 × 10⁻⁶ M, THF): λ_max(nm) = 352(0.828), 410 (0.367), 521 (0.111), 579 (0.085).

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. G. M. conceived the whole work and supervised, and directed S. D. to carry out all experiments. S. D. collected all experimental data and S. G. performed all DFT calculations.

Data availability

The data for this article such as the general Experimental section, crystallographic refinement data, mercury structures, and NMR, HRMS, ATR-FTIR, and DFT calculation methods and data are given as the ESI,†. Crystallographic data files (cif) have been deposited at the Cambridge Crystal Structure Database with CCDC numbers: 2370249–2370251.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We are thankful to STARS (MoE-STARS/STARS-2/2023-0130), Ministry of Education, Indian Institute of Science, Bangalore, India for financial support. We thank Dr. Vasudevan Subramaniyan for completing the crystal structure disorder.

References

1. P. Woodward, L. F. Dahl, E. W. Abel and B. C. Crosse, J. Am. Chem. Soc., 1965, 87, 5251 CrossRef CAS .
2. T. Krüger, B. Krebs and G. Henkel, Angew. Chem., Int. Ed. Engl., 1989, 28, 61 CrossRef .
3. M. Kriege and G. Henkel, Z. Naturforsch., B: J. Chem. Sci., 1987, 42, 1121 CrossRef .
4. W. Gaete, J. Ros, X. Solans, M. Font-Altaba and J. L. Brianso, Inorg. Chem., 1984, 23, 39 CrossRef .
5. B.-K. Koo, E. Block, H. Kang, S. Liu and J. Zubieta, Polyhedron, 1988, 7, 1397 CrossRef .
6. A. H. Mahmoudkhani and V. Langer, Inorg. Chim. Acta, 1999, 294, 83 CrossRef .
7. T. A. Wark and D. W. Stephan, Organometallics, 1989, 8, 2836 CrossRef CAS .
8. K. Schulbert and R. Mattes, Z. Naturforsch., B: J. Chem. Sci., 1994, 49, 770 CrossRef CAS .
9. H. Feld, A. Leute, D. Rading, A. Benninghoven, G. Henkel, T. Krüger and B. Krebs, Z. Naturforsch., B: J. Chem. Sci., 1992, 47, 929 CrossRef CAS .
10. R. O. Gould and M. M. Harding, J. Chem. Soc. A, 1970, 875 RSC .
11. A. H. Mahmoudkhani and V. Langer, Polyhedron, 1999, 18, 3407 CrossRef CAS .
12. M. Capdevila, P. Ginzales-Duarte, J. Sola, C. Foces-Foces, F. H. Cano and M. Martinez-Ripoll, Polyhedron, 1989, 8, 1253 CrossRef CAS .
13. H. Barrera, J. C. Bayon, J. Suades, C. Germain and J. P. Declerq, Polyhedron, 1984, 3, 969 CrossRef CAS .
14. C. Jin-Hua and K. Bei-Sheng, Chin. J. Struct. Chem., 1993, 12, 397 Search PubMed .
15. I. G. Dance, M. I. Scudder and R. Secomb, Inorg. Chem., 1985, 24, 1201 CrossRef .
16. S. A. Ivanov, M. A. Kozee, W. A. Merril, S. Agarwal and L. F. Dahl, J. Chem. Soc., Dalton Trans., 2002, 4105 RSC .
17. H. Xiang, R. Cheng, C. Ruan, C. Meng, Y. Gan, W. Cheng, Y. Zhao, C. Xu, J. Li and C. Yao, Nanoscale, 2024, 16, 4563 RSC .
18. A. Mathew and T. Pradeep, Part. Part. Syst. Charact., 2014, 31, 1017–1053 CrossRef .
19. F. Zheng, W. Zhang, Q. Guo, B. Yu, D. Wang and W. Chen, Coord. Chem. Rev., 2024, 502, 215603 CrossRef CAS .
20. M. P. Maman, T. Gurusamy, A. K. Pal, R. Jana, K. Ramanujam, A. Datta and S. Mandal, Angew. Chem., Int. Ed., 2023, 62, e202305462 CrossRef CAS .
21. R. Angamuthu and E. Bouwman, Phys. Chem. Chem. Phys., 2009, 11, 5578 RSC .
22. H. N. Kagalwala, E. Gottlieb, G. Li, T. Li, R. Jin and S. Bernhard, Inorg. Chem., 2013, 52, 9094 CrossRef CAS PubMed .
23. D. R. Kauffman, D. Alfonso, D. N. Tafen, J. Lekse, C. Wang, X. Deng, J. Lee, H. Jang, J.-s. Lee, S. Kumar and C. Matranga, ACS Catal., 2016, 6, 1225 CrossRef CAS .
24. S. Srinivasan, Z. Liu, S. House and R. Jin, Inorg. Chem., 2023, 62, 1875 CrossRef CAS PubMed .
25. K. S. Joya, L. Sinatra, L. G. AbdulHalim, C. P. Joshi, M. N. Hedhili, O. M. Bakr and I. Hussain, Nanoscale, 2016, 8, 9695 RSC .
26. A. M. S. Pembere, C. Cui, R. Amumula, H. Wu, P. An, T. Liang and Z. Luo, Phys. Chem. Chem. Phys., 2019, 21, 17933 RSC .
27. M. Zhu, S. Zhou, C. Yao, L. Liao and Z. Wu, Nanoscale, 2014, 6, 14195 RSC .
28. Y. Pan, J. Chen, S. Gong and Z. Wang, Dalton Trans., 2018, 47, 11097 RSC .
29. S. Funaki, T. Kawawaki, T. Okada, K. Takemae, S. Hossain, Y. Niihori, T. Naito, M. Takagi, T. Shimazaki, S. Kikkawa, S. Yamazoe, M. Tachikawa and Y. Negishi, Nanoscale, 2023, 15, 5201 RSC .
30. C. Zhang, S. Takada, M. Kölzer, T. Matsumoto and K. Tatsumi, Angew. Chem., Int. Ed., 2006, 45, 3768 CrossRef CAS PubMed .
31. C. Zhang, T. Matsumoto, M. Samoc, S. Petrie, S. Meng, T. C. Corkery, R. Stranger, J. Zhang, M. G. Humphrey and K. Tatsumi, Angew. Chem., Int. Ed., 2010, 49, 4209 CrossRef CAS .
32. C. Tan, M. Jin, H. Zhang, S. Hu, T. Sheng and X. Wu, CrystEngComm, 2015, 17, 5110 RSC .
33. C. Tan, M. Jin, X. Ma, Q. Zhu, Y. Huang, Y. Wang, S. Hu, T. Sheng and X. Wu, Dalton Trans., 2012, 41, 8472 RSC .
34. P. Bose, K. K. Ramankutty, P. Chakraborty, E. Khatun and T. Pradeep, Nanoscale, 2024, 16, 1446 RSC .
35. S. Huang, M. Wang and X. Jiang, Chem. Soc. Rev., 2022, 51, 8351 RSC .
36. K. Yang, Q. Li, Z. Li and Z. Sun, Chem. Commun., 2023, 59, 5343 RSC .
37. H.-B. Zhu and S.-H. Gou, Coord. Chem. Rev., 2011, 255, 318 CrossRef CAS .
38. T. Ganguly, A. B. Chakraborty and A. Majumdar, ACS Org. Inorg. Au, 2023, 3, 332 CrossRef CAS .
39. K. Shen, M. Gennari, C. Philouze, A. Velic, S. Demeshko, F. Meyer and C. Duboc, Inorg. Chem., 2024, 63, 9119 CrossRef CAS .
40. T. Ganguly, A. Das, M. Jana and A. Majumdar, Inorg. Chem., 2018, 57, 11306 CrossRef CAS PubMed .
41. H. Jiang, T. Sheng, S. Bai, S. Hu, X. Wang, R. Fu, P. Yu and X. Wu, Inorg. Chem., 2013, 52, 12305 CrossRef CAS PubMed .
42. V. Subramaniyan, B. Dutta, A. Govindaraj and G. Mani, Dalton Trans., 2019, 48, 7203 RSC .
43. A. Kumar, R. Gupta and G. Mani, Organometallics, 2023, 42, 732 CrossRef CAS .
44. C. Fliedel, G. Schnee and P. Braunstein, Dalton Trans., 2009, 2474 RSC .
45. H. V. Huynh, C. H. Yeo and Y. X. Chew, Organometallics, 2010, 29, 1479 CrossRef CAS .
46. J. Vila, M. Solà, T. Achard, S. Bellemin-Laponnaz, A. Pla-Quintana and A. Roglans, ACS Catal., 2023, 13, 3201 CrossRef CAS PubMed .
47. K. N. Gavrilov, I. V. Chuchelkin, V. K. Gavrilov, S. V. Zheglov, I. D. Firsin, V. M. Trunina, N. E. Borisova, Y. P. Bityak, O. A. Maloshitskaya, V. A. Tafeenko, V. S. Zimarev and N. S. Goulioukina, Organometallics, 2023, 42, 1985 CrossRef CAS .
48. F. He, C. Gourlaouen, H. Pang and P. Braunstein, Chem. – Eur. J., 2022, 28, e202200507 CrossRef CAS .
49. A. J. Bondi, J. Phys. Chem., 1964, 68, 441 CrossRef CAS .
50. F. Tian, J. Chen, F. Chen, Y. Liu, Y. Xu and R. Chen, Appl. Catal., B, 2021, 292, 120158 CrossRef CAS .
51. E. R. Johnson, S. Keinan, P. Mori-Sánchez, J. Contreras-García, A. J. Cohen and W. Yang, J. Am. Chem. Soc., 2010, 132, 6498 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available: NMR, IR, HRMS, UV-Vis, crystallographic data, crystal structures, and DFT calculations. CCDC 2370249–2370251. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4dt02047b

This journal is © The Royal Society of Chemistry 2024