Oxidative addition chemistry of bis(4,5-dimethoxybenzo)-1,2,5,6-tetrathiocin with cyclopentadienyl metal carbonyl complexes and the mechanochemical transformation of CpCo(dmobdt) to [CpCo(dmobdt)]₂

Abstract

The oxidative addition of tetramethoxy-1,2,5,6-dibenzotetrathiocin (1) to CpCo(CO)₂, [CpFe(CO)₂]₂, [CpMo(CO)₃]₂ and CpV(CO)₄ in toluene under microwave irradiation (150 °C, 30 min) afforded the complexes CpCo(dmobdt) (2), [Cp₂Fe₂(dmobt)(CO)₂] (3), [Cp₂Mo₂(dmobdt)₂] (4) and [(Cp)₂V₂(dmobdt)₂] (5) [dmobdt = 4,5-dimethoxybenzene-1,2-dithiolate dianion, (MeO)₂C₆H₂S₂²⁻]. Vacuum sublimation of 2 afforded dark blue crystals of the monomeric 16e⁻ complex, CpCo(dmobdt) (2a), whereas recrystallisation from CH₂Cl₂ afforded dark blue crystals of the dimeric 18e⁻ polymorph, [CpCo(dmodt)]₂ (2b). The structures of 2a, 2b, 3, 4 and 5 were determined by X-ray diffraction. DSC studies on 2a and 2b indicated distinct melting points (206 and 216 °C respectively) and VT-PXRD revealed no thermally-induced phase change between 2a and 2b. Conversely, mechanochemical grinding of 2a revealed an irreversible phase transition to 2b.

---

Introduction

The term “dithiolene” was introduced to describe the redox-active “dithiolene” (dt) ligand system,¹ in which the ligand can variously adopt one of the following redox states: the dithiolate dianion, dt²⁻, the semi-1,2-dithione dt˙⁻ and neutral dithione, dt (Scheme 1). Common dithiolene ligands include mnt²⁻, tdas²⁻, tfd²⁻ and dmit²⁻ in their dithiolate forms (Scheme 2) and their metal complexes have been explored for materials properties.² Applications of such complexes range from non-linear optics,^2,3 sensitisers in dye sensitised solar cells (DSSCs),⁴ conducting^5,6 and magnetic materials^7,8 through to catalysis⁹ and models for metalloenzymes.¹⁰

Scheme 1 Redox states of the dithiolene ligand.

Scheme 2 Select dithiolate ligands.

The ability to tune the redox properties of the dithiolene ligand is important for such applications and dithiolate ligands with both electron-withdrawing cyano groups, such as dcbdt²⁻ and cbdt²⁻ as well as π-electron donating groups such as dmobdt²⁻ are known (Scheme 2).^11,12 Traditionally metal dithiolates are prepared via three synthetic routes: (i) salt metathesis, whereby s-block metal dithiolates and d-block metal salts undergo ligand exchange to form d-block ditholates; (ii) condensation of dithiols with d-metal oxo, alkoxo and amido precursors and (iii) the oxidative addition of dithietes to low valent transition metals.¹³ Benzene dithiolate (bdt²⁻) and its derivatives are attractive ligands due to the ability to tune both the steric and electronic properties of the dithiolene ligand through judicious choice of electron-donating, electron-withdrawing and/or sterically demanding groups. Yet benzene and toluene dithiolate ligands (bdt²⁻ and tdt²⁻, Scheme 2) dominate this family with over 53% of benzene dithiolate derivatives in the CSD (release 5.46, Nov 2024) comprising bdt²⁻ or tdt²⁻. This is most likely due to the commercial availability of benzene and toluene dithiols.^14,15 Synthetic route (iii) has been limited because of the small number of dithiete precursors, R₂C₂S₂ (R = CF₃, CO₂Me), with benzodithiete reportedly unstable above 180 K.^13,15 Previously, our group has described the oxidative addition of the 8-membered 1,2,5,6-tetrathiocin (1) to zero valent group 10 transition metal complexes such as Ni(COD)₂, M₂dba₃ (M = Pd, Pt) or M(PR₃)_n as an efficient route to group 10 metal dithiolate complexes (Scheme 3).^16–19 This approach was recently extended to oxidative addition to the cobalt(I) complex, CpCo(CO)₂, generating CpCo(L) where L is the dithiolate derivative of benzo-15-crown-5 or benzo-18-crown-6.¹² To explore the generality of this reaction, we have now explored the microwave-assisted oxidative addition reaction of tetra-methoxy-dibenzotetrathiocin (1) to a range of early/mid/late 3d/4d transition metal ions of composition CpM(CO)_n (M = V, Fe, Co and Mo), affording [CpM(dmobdt)]_n (2an = 1 M = Co; 2bn = 2, M = Co; 4, n = 2, M = V; 5n = 2, M = Mo), and Cp₂Fe₂(CO)₂(dmobdt) (3) (Scheme 3). Within this series of complexes, the dithiolate adopts a chelating coordination mode but either one or both S atoms show the ability to also adopt a μ₂ bridging mode to satisfy the stereo-electronic demands of the metal centre. In particular, the solid-state conversion between 2a and 2b is explored and contrasted to previous studies on the closely related CpCo(bdt)/[CpCo(bdt)]₂ system.^20,21

Scheme 3 Microwave-assisted oxidative addition reactions of tetrathiocin 1 with low valent metals afford dithiolate complexes.

Results and discussion

Synthetic procedures

Bis(4,5-dimethoxybenzo)-1,2,5,6-tetrathiocin, 1, was prepared according to the previously described literature procedure.^22,23 The metal-dithiolate complexes, 2–5, were all prepared by oxidative addition of the tetrathiocin 1 to the M(I) complexes CpM(CO)_n in sealed containers using microwave irradiation.

Reaction of tetrathiocin 1 with cyclopentadienylcobalt dicarbonyl in toluene under microwave irradiation (150 °C, 30 min) formed an intense deep blue suspension. Extraction with chloroform and drying in vacuo afforded a dark blue solid from which dark blue crystals of the monomeric complex CpCo(dmobdt) (2a) were grown by vacuum sublimation onto a cold-finger (+175 to −3 °C). On the other hand, recrystallisation from dichloromethane selectively affords dark blue crystals of the dimeric complex, [CpCo(dmodt)]₂ (2b).

The reaction of tetrathiocin 1 with an excess of cyclopentadienyl iron dicarbonyl dimer, [CpFe(CO)₂]₂, under microwave irradiation (150 °C, 30 min) formed a brown/black suspension. Recrystallisation of the solid from chloroform afforded orange crystals of the dimeric iron dithiolate complex, [Cp₂Fe₂(CO)₂(dmobdt)] (3). In an analogous fashion, reaction of 1 with an excess of [CpMo(CO)₃]₂ or CpV(CO)₄ afforded the dimeric bis-dithiolate metal complexes [CpMo(dmobdt)]₂ (4) and [CpV(dmobdt)]₂ (5).

Crystallographic details

Dark blue crystals of 2a were formed via vacuum sublimation and were found to crystallize in the orthorhombic space group Pbca with one molecule per asymmetric unit. The structure of 2a (Fig. 1) is a two-legged piano stool in which the Co atom is located just 0.044 Å from the C₆S₂Co mean plane, with a fold angle between CoS₂ and C₂S₂ planes of just 3.23°. The Co–S bond lengths fall in the range of 2.111(2)–2.128(2) Å. The C–S bond lengths are in the range of 1.726(6)–1.733(6) Å, typical of a dithiolate dianion. The Cp ring plane is essentially perpendicular to the C₂S₂ plane, at 85.03°. The Cp-centroid to Co distances of 1.657 Å is comparable with those of other 16e⁻ cyclopentadienyl cobalt(III) benzenedithiolate structures (1.640–1.656 Å).^21,24–27

Fig. 1 Crystal structure of (top) 2a and (bottom) 2b. Colour code: Co^III, blue; S, yellow; C, black; O, red. Hydrogen atoms omitted for clarity.

Crystals of 2b were grown by slow diffusion of hexane into a saturated CH₂Cl₂ solution of 2. It crystallizes in the monoclinic space group P2₁/c with half a molecule in the asymmetric unit. The structure of the 18e⁻ dimer adopts a three-legged piano stool geometry at the Co(III) centre with an η⁵-coordinate Cp ring and three sulfur atoms, comprising S1 and S2 from a chelating dithiolate ligand and additional coordination from a S atom of a symmetry related CpCo(dmobdt) moiety (S1′ at −x, 1 − y, 1 − z). The Co centre is displaced from the C₆S₂Co plane of the chelating dithiolate by 0.598 Å such that the angle between the C₂S₂ and CoS₂ planes (Fig. 2) is 22.98° (c.f. 3.23° for 2a). The C–S bonds in the dithiolate are 1.779(2) and 1.760(3) Å for C11–S1 and C12–S2 respectively, a little longer than in monomeric 2a. These are intermediate between a standard C–S single bond (1.83 Å) and a thioketone (1.70 Å) and typical of other cyclopentadienyl cobalt benzenedithiolate complexes (1.75(2) Å). The Co–S bond associated with the μ₂-bridging S atoms, Co1-S2′, has a bond length of 2.2672(7) Å, similar to the monodentate Co1–S1 and Co1–S2 bond lengths (2.2309(8) and 2.2499(7) Å respectively). These Co–S bond lengths are ca. 0.1 Å longer than in 2a. Isolation of both 16e⁻ monomer and 18e⁻ dimer forms has been observed in CpCo(bdt) and [CpCo(bdt)]₂.²¹

Fig. 2 Crystal Structure of 3. Colour code: Fe^II, orange; S, yellow; C, black; O, red. Hydrogen atoms omitted for clarity, Fe–Fe bond represented with grey line.

Complex 3, Cp₂Fe₂(CO)₂(dmobdt), was crystalized from a saturated CH₂Cl₂ solution layered with hexane and adopts the triclinic space group P1̄ with one molecule in the asymmetric unit (Fig. 2). Complex 3 comprises a CpFe(dmobdt) unit linked to a CpFe(CO)₂ unit with the two Fe centres bridged by a μ₂-CO ligand and one of the S atoms (S2) of the dmobdt²⁻ which additionally adopts a μ₂-bridging mode. Both Fe1 and Fe2 adopt a three-legged piano stool geometry. The Fe–S bond lengths range from 2.1875(4)–2.2702(4) Å, similar to other Fe dithiolate dimers reported in the CSD (Table S4, SI), and comparable to the Co–S bond lengths in 2a and 2b. Comparison of the C–O distances within the two carbonyl groups reveals a longer distance of 1.178(2) Å for the bridging carbonyl compared to 1.143(2) Å for the terminal carbonyl. This is consistent with a weakened C–O bond due to greater back-donation of electron density into the π* orbital of the CO bond from the two Fe centres. The two Cp rings are positioned in a ‘cis’ conformation with a Cp_centroid–Fe–Fe–Cp_centroid torsion angle of 2.34°. The Fe–Fe distance between the two crystallographically independent Fe centres is 2.567(3) Å. Notably, the only benzo-1,2-dithiolate complex on the CSD, structurally analogous to 3 is the parent complex Cp₂Fe₂(CO)₂(bdt), described by Fan and coworkers in their studies of proton reduction.²⁸

Complex 4, [CpMo(dmobdt)]₂, crystallizes in three separate crystal forms dependent upon crystallisation conditions: slow evaporation of 4 from dichloromethane affords the solvate-free structure 4 (Fig. 3 (top)), whereas slow evaporation from chloroform affords 4·CHCl₃ (see Fig. S1, SI). Notably, attempts to prepare 4 by slow diffusion of hexanes into a saturated dichloromethane solution of 4 afforded an additional solvate, 4·C₆H₁₄ (see Fig. S1, SI)! There is negligible difference in the geometries of 4 across these three structures and the structure of the non-solvated form is discussed here. Complex 4 crystallizes in the monoclinic space group P2₁/n with half a molecule in the asymmetric unit, with the two Mo centres related via inversion. The dimer comprises two Mo centres, each comprising four-legged piano stool geometries, bonded to one η⁵-Cp ligand and four μ₂-S bridging atoms associated with the two bridging dmobdt²⁻ dithiolate ligands. This coordination geometry is notably different to the cobalt complex 2b which nominally has the same composition, [CpCo(dmobdt)]₂, but only has two of the four S atoms adopting μ₂-bridging modes. The structure is analogous to that reported by Miller et al. for the related tdt²⁻ complex, [CpMo(tdt)]₂.²⁹ The Mo–S bond lengths are 2.4597(5) and 2.675(4) Å, typical for Mo-dithiolate compounds (Table S5, SI) and a little longer than those associated with the related 3d metal complexes 2 and 3. The C–S distances (1.791(1) and 1.793(1) Å) are similar to those in the previously reported tdt²⁻ analogue as is the Mo–Mo distance of 2.5909(4) Å. The ring planes of the two crystallographically equivalent Cp groups are co planar, as are the ring planes of the two C₆S₂ groups. Conversely, the Cp and C₆S₂ rings are inclined at an angle of 11.03°.

Fig. 3 Crystal structures of 4 (top) and 5 (bottom). Colour code: Mo, light blue; V, grey; S, yellow; C, black; O, red. Hydrogen atoms omitted for clarity. Close M–M contacts represented with grey line.

Crystals of 5 were isolated by slow evaporation of a dichloromethane solution. Complex 5 crystalizes in the monoclinic space group P2₁/c with half a molecule in the asymmetric unit. The dimeric structure is analogous to 4 (Fig. 4) with the two V centres adopting a four-legged piano stool geometry with one η⁵-Cp ligand and four μ₂-S bridging atoms bonded to each metal centre. The V–S bond lengths are 2.437(1) and 2.4439(9) Å, a little longer than 2 and 3 and closer to those observed in 4 (2.4597(5) and 2.675(4) Å). A CSD search revealed six vanadium dithiolate complexes with a similar sandwich like topology,^30–33 but only one is formed with a benzenedithiolate derivative.³⁴ As with 4, the C–S bond lengths in 5 (1.778(3) and 1.775(3) Å) correspond well to those in the previously published complex, as does the V–V distance of 2.5185(6) Å. Also similar to 4, 5 comprises two sets of co-planar ring planes with respect to the Cp groups and the C₆S₂ groups, with the angle between the two sets of planes being 7.09°.

Fig. 4 PXRD patterns illustrating the effect of mechanochemical grinding on 2a: solid black line, phase pure 2a at room temperature; dotted black line, simulation of 2a from SC-XRD; solid green line lightly ground sample of 2a (resulting in a mix of 2a and 2b); solid red line, 2a after sustained grinding (affording 2b); dotted red line, simulation of 2b from SC-XRD.

Bonding. The monomer 2a comprises a formal Lewis acidic 16e⁻ Co(III) ion and the dithiolate S atoms are Lewis basic with each possessing 2 lone pairs. On forming the dimer, 2b, one S atom of each dithiolate adopts a μ₂-bridging mode in which the bridging S atom donates an additional lone pair to Co such that the Co(III) ions in 2b are each formally 18e⁻ species. Notably, the stronger donating effect of the Cp* ligand, weakens the Lewis acidity of the Co(III) centre, suppressing dimerization and all reported structures of Cp*Co(dt) complexes are monomeric in the solid state.^35–37 Previous ¹H solution NMR studies on CpCo(tdt) and CpCo(S₂C₆Cl₂H₂) reveal a singlet for the Cp ring, indicating the presence of a single species present (either monomer or dimer), although CpCo(S₂C₆Cl₃H) exhibited concentration dependent ¹H NMR spectra which were consistent with trace amounts of dimer present at high concentration.^26,36 In the current case, 2 exhibits a singlet for the Cp ring plus singlets for the aromatic and methoxy protons. The latter two resonances are diagnostic of the monomer being present in solution, since the two S atoms (and hence the corresponding phenylene and methoxy protons) are chemically equivalent in 2a but not in 2b. In this context the dimeric complex [Pt(PPh₃)(dmobdt)]₂ shows inequivalent aromatic and methoxy peaks, indicative of the dimer being present in solution.¹⁸ Notably the ASAP MS data (a direct insertion technique) was able to discriminate the two complexes with 2a exhibiting an [M + H]⁺ peak (m/z = 324.9767), whereas 2b exhibited a [M + H]⁺ peak at m/z = 648.9449.

For the Cp₂Fe₂(CO)₂(dmobdt) complex 3, both Fe centres can be considered to fulfil the 18e⁻ rule provided there is an Fe–Fe bond. This is supported by the ¹H NMR spectrum which exhibits sharp resonances which are neither paramagnetically broadened nor paramagnetically shifted. The ¹H NMR of 3 exhibit two chemically distinct ¹H resonances (4.87 and 4.81 ppm) reflecting two chemically distinct Cp rings, similar to those in Cp₂Fe₂(bdt)(CO)₂ and Cp₂Fe₂ (3,6-Cl₂bdt)(CO)₂ recorded in d₃-MeCN.²⁸ The ¹H NMR resonances of the dmobdt²⁻ ligand are consistent with a dithiolate ligand in which the two dithiolate S atoms (and hence aromatic and alkoxy protons) adopt chemically distinct environments, with two aromatic C–H and two methoxy ¹H resonances. These spectroscopic data for 3 are consistent with retention of the same molecular structure in both the solid state and solution. In addition, IR spectra reveal two distinct CO stretching modes at 1962 and 1749 cm⁻¹, corresponding to the terminal and μ₂-bridging CO groups. In the related complexes Cp₂Fe₂(bdt)(CO)₂ and Cp₂Fe₂(3,6-Cl₂bdt)(CO)₂, the terminal CO stretching vibration occurs notably higher in energy (1987 and 1992 cm⁻¹ respectively), reflecting the stronger donor nature of the dmobdt²⁻ ligand. Similarly, the bridging μ-CO in both Cp₂Fe₂(bdt)(CO)₂ and Cp₂Fe₂(3,6-Cl₂bdt)(CO)₂ also occur higher in frequency (1804 and 1811 cm⁻¹ respectively) than 3. Notably complexes 2, 4 and 5 all reveal complete elimination of CO during oxidative addition of 1 to the CpM(CO)_n precursor but 3 does not. Elimination of two equivalents of CO from 3 would afford Cp₂Fe₂(dmobdt). Structures of this type are formally 16e⁻ compounds and have been reported for several complexes with strong donor cyclopentadienyl groups including Cp*₂Fe₂(bdt) and Cp*₂Fe₂(dmobdt).^38,39 In the case of 3 there was no evidence for formation of Cp₂Fe₂(dmobdt) within the reaction mixture. In this regard, photolysis of [CpFe(CO)₂]₂ with one equivalent of bdtH₂ forms Cp₂Fe₂(CO)₂(bdt) (analogous to 3), whereas photolysis of [CpFe(CO)₂]₂ with two equivalents of bdtH₂ formed the dimer [CpFe(bdt)]₂, structurally similar to 2b.²⁸ In this work attempts to form [CpFe(dmobdt)]₂, compositionally analogous to 2b, 4 and 5via stoichiometric control were unsuccessful: reaction of [CpFe(CO)₂]₂ with both 0.5 or 1.0 equivalent of 1 afforded only the mono-dithiolate complex 3.

For complex 4, the compound is formally 17e⁻, or 18e⁻ if a Mo–Mo bond is present. This complex is structurally similar to the dimers [CpMo(bdt)]₂ and [CpMo(tdt)]₂.²⁹ The sharpness of the ¹H NMR spectrum and the observed chemical shift range are indicative of a diamagnetic complex, supporting the hypothesis of a formal Mo–Mo bond. Conversely for complex 5, the vanadium centre is formally 16e⁻, or 18e⁻ for a V=V double bond. Previous work has described the synthesis of three other [V₂-dithiolate] compounds: [V₂(edt)₄] is described as containing either a V–V single or double bond based upon crystallographic data, the diamagnetic nature of the compound and MO calculations.^31,32 On the other hand, [Cp₂V₂(S₂)(S₂C₄F₆)] is weakly paramagnetic.³³ Work by Stephan described the synthesis of the analogous dimer, [CpV(bdt)]₂ but did not report any NMR data.³⁴

In this work NMR data was similarly difficult to obtain with standard ¹H NMR procedures not affording an interpretable spectrum, suggesting 5 retains its paramagnetic nature as described by Stephan, leading to significant line broadening, peak shifting and reduced signal intensities.³⁴

Phase behaviour of 2. Derivatives of CpCo(bdt) have been found to adopt either monomeric 16e⁻ or 18e⁻ dimeric structures (eqn (1)):⁴⁰

[CpCo(bdt)]₂ ⇌ 2 CpCo(bdt) (1)

Formation of dimers is enthalpically favoured through formation of two S → Co dative covalent bonds, whereas the monomeric 16e⁻ complexes are entropically favoured. Nomura et al. showed from dilute solution UV/vis studies that the entropically favoured monomer phase dominates in solution.²⁶ Nevertheless, at high concentration, ¹H NMR studies reflected the presence of the dimer form as the minor component (monomer : dimer ∼ 40: 1 for CpCo(S₂C₆Cl₄)). The related selenium analogue, [CpCo(Se₂C₆H₄)]₂, appears more stable in the dimer form and reveal ΔH_dim = −60 kJ mol⁻¹ in d⁶-benzene and ΔS_dim = −120 J K⁻¹ mol⁻¹. Dimer formation is favored by (i) enhancing the Lewis basicity of the dithiolate S atoms and (ii) increasing the Lewis acidity of the cobalt(III) centre. In this regard, the strongly donor Cp* ligand reduces the Lewis acidity of Co(III) and crystal structures of Cp*Co(bdt) derivatives are all monomeric in the solid state. Conversely, crystal structures of CpCo(bdt) derivatives are reported to be variously monomers or dimers in the solid state, indicating a subtle balance between monomer and dimer forms. Among these derivatives, CpCo(bdt) (6) has previously appeared unique in forming both monomer (6a) and dimer (6b) phases. Slow crystallization was reported to favour 6b (enthalpic product) whereas sublimation or rapid crystallization favors 6a as the entropic product. For 6, the two phases were reported to interconvert in the solid state with the dimerization process 6a(s) → 6b(s) occurring slowly at room temperature and the reverse process, 6b(s) → 6a(s) occurring at 150 °C. DSC studies revealed ΔH_rxn = +18.9 kJ mol⁻¹ for the process 6b(s) → 6a(s) at 150–160 °C.

In the current study, we similarly found that monomeric 2a is isolated by sublimation at elevated temperatures, whereas crystal growth from solution at room temperature forms dimeric 2b. Room temperature powder X-ray diffraction (PXRD) studies on unground samples confirmed the phase purity of monomeric 2a and dimeric 2b phases, by comparison with simulated PXRD profiles based on low temperature single crystal structure determinations (Fig. S2 and S3, SI). Although the two PXRD profiles share certain similarities, 2a and 2b could be readily distinguished on the basis of two reflections: the (0 1 1) reflection present in 2b at 2θ ∼ 12° and the (2 0 0) reflection at 2θ ∼ 20° associated with 2a (Fig. 4). Variable temperature PXRD measurements on 2b (Fig. S4, SI) show no change in the PXRD profile between room temperature and 150 °C (the upper limit of our measurements).

DSC studies reveal that 2a melts at 206 °C and 2b melts at 216 °C (Fig. S7, SI). The enthalpies of fusion are Δ_fusH = +21.9 kJ mol⁻¹ for 2a and Δ_fusH = +51.9 kJ mol⁻¹ for 2b. This provides an approximate Δ_dimH [2 × 2a(s) → 2b(s)] of −8.1 kJ mol⁻¹. This is comparable with that observed for the process 2 × 6a(s) → 6b(s) at −18.9 kJ mol⁻¹. After cooling 2b from the melt (−10 K min⁻¹) and stabilizing at 25 °C for 5 minutes, a re-heat cycle (+10 K min⁻¹) revealed the emergence of a feature at 206 °C diagnostic of 2a in addition to the expected melting of 2b at 216 °C (Fig. S8, SI). This indicated that melting affords a mixture of 2a and 2b.

Although there was no evidence for a solid-state phase transition from 2a(s) to 2b(s) upon standing, nor conversion from 2b(s) to 2a(s) upon heating, we were intrigued by initial PXRD measurements: our PXRD samples are typically lightly ground prior to measurement to minimize preferred orientation effects and a sample of 2a subjected to brief grinding afforded a PXRD pattern reflecting the presence of both 2a and 2b. Optical examination of the remaining pristine sample suggested that 2a was homogeneous, suggesting a mechanochemical transformation from 2a to 2b (Fig. 4). Upon prolonged grinding, complete conversion of 2a into 2b was achieved (Fig. 4). Attempts to convert dimer 2b into monomeric 2a upon grinding were unsuccessful revealing dimerization is an irreversible process at ambient temperature. This led us to contemplate the different solid state behaviours of 2 and 6.

As pointed out by West, solid-state reactions can only occur if reactive centres are in the right orientation and sufficiently close together.⁴¹ This suggests that molecular displacements in solid state reactions are typically short. Miller's study on 6 revealed that monomer 6a has two molecules in the asymmetric unit. Each crystallographically independent molecule forms a pair of close, intermolecular Co⋯S contacts (Co1⋯S1 at 4.703 Å and Co2⋯S3 at 5.060 Å) linking molecules related via inversion symmetry (Fig. 5a). In both cases, shortening these contacts generates a centrosymmetric dimer with Co–S bonds of 2.272 Å (Fig. 5b). In the case of 2a, there is a single molecule in the asymmetric unit. This molecule exhibits a close intermolecular Co⋯S contact (4.719 Å, Fig. 5c) similar to that observed in 6a but this pair of molecules is not related via a crystallographic inversion centre. Instead, these close Co⋯S contacts in 2a are related via a glide plane and are therefore not of the correct orientation to undergo centrosymmetric dimerization. The space group Pbca observed for complex 2a does exhibit a crystallographic inversion centre, but molecules related by inversion exhibit long Co⋯S separations at 12.747 and 13.735 Å (Fig. 5d). These are well beyond the short molecular displacements typically associated with solid state transformations and are consistent with the lack of thermal interconversion between 2a and 2b in the solid state. The mechanochemically-driven transition from 2a to 2b, presumably displaces monomers with respect to the inversion. A comparison of the unit cells of 2a and 2b (Table S1, SI) indicate that the two structures are related via a doubling of the crystallographic c-axis. The packing of 2a and 2b in the bc plane (Fig. 6) highlights the similarity in the two structures. While the transformation from 2a to 2b is thermodynamically favoured based on (i) the density rule,⁴² and (ii) the estimated difference in Δ_fusH, there is likely a large activation barrier to this structural transformation. This is not achieved thermally up to the melting point but is seemingly driven mechanochemically at ambient temperature. Recent studies have highlighted the importance of mechanical stress reducing the reaction energy barrier for shear-driven transformations.⁴³ Similar investigations on dimeric complexes 4 and 5 revealed no evidence for similar phase transitions, likely due to the instability of the 12e⁻ CpV(dmobdt) and 13e⁻ CpMo(dmobdt) monomers in relation to 16e⁻ CpCo(dmobdt).

Fig. 5 (top) Comparison of the molecular displacements with respect to inversion in (a) the monomer 6a, (b) the dimer 6b; (bottom): (c) the closest Co⋯S contact between monomers in 2a and (d) monomers related via inversion symmetry in 2a.

Fig. 6 (top) Molecular packing of 2a viewed along the crystallographic a-axis, highlighting inversion centres (•); (bottom) molecular packing of 2b, also viewed along the crystallographic a-axis. [H atoms omitted for clarity].

Conclusions

In summary, the current study expands the synthetic methodology of microwave-assisted oxidative addition reactions of tetrathiocins, such as 1, as a route to transition metal dithiolate complexes for a variety of early, mid and late transition metals from the 3d and 4d series. In the current study oxidative addition of 1 to half-sandwich cyclopentadienyl metal carbonyl complexes occurs with partial or complete elimination of CO. In particular, the cobalt complexes 2a and 2b form distinct monomeric and dimeric structures with the entropically favored monomer formed by sublimation at high temperature and the dimer form afforded by recrystallization at room temperature. Heating the dimer into the melt phase generates some monomer based on DSC measurements, whereas mechanochemical grinding of 2a affords 2bvia an irreversible mechanochemical phase transition.

Conflicts of interest

There are no conflicts to declare.

Data availability

A summary of analytical data and instrumentation used for elemental analysis, IR, MS, NMR; plots of PXRD and DSC data are freely available and summarized in the experimental section of the SI. See DOI: https://doi.org/10.1039/d5dt01691f.

CCDC 2470686–2470691 and 2471388 (2a, 2b, 3, 4, 4·CHCl₃, 4.C₆H₁₂ and 5) contain the supplementary crystallographic data for this paper.^44(a–g) The crystallographic data supporting this article have also been included as part of the SI.

Acknowledgements

The work was supported through NSERC DG (J. M. R. DG 2020-04627; M. P. DG-2018-0425/2024-04043; C. L. B. M. DG 2024-04186), NSERC RTI (J. M. R. RTI-2022-00005; M. P. RTI-2020-00310) and CFI/ORF award (LOF-212442). DJC thanks the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant (#101151188).

References

1. (a) J. A. McCleverty and T. J. Meyer, Comprehensive Coordination Chemistry II: From Biology to Nanotechnology, Newnes, 2003; (b) R. Eisenberg and H. B. Gray, Inorg. Chem., 2011, 50, 9741–9751.
2. S. D. Cummings, L. T. Cheng and R. Eisenberg, Chem. Mater., 1997, 9, 440–450.
3. L. Pilia, D. Marinotto, M. Pizzotti, F. Tessore and N. Robertson, J. Phys. Chem. C, 2016, 120, 19286–19294.
4. (a) G. Li, M. F. Mark, H. Lv, D. W. McCamant and R. Eisenberg, J. Am. Chem. Soc., 2018, 140, 2575–2586; (b) A. Singh, P. Singh, G. Kociok-Köhn, M. Trivedi, A. Kumar, R. Chauhan, S. B. Rane, C. Terashima, S. W. Gosavi and A. Fujishima, New J. Chem., 2018, 42, 9306–9316.
5. M. Yamashita, Y. Sato, Y. Kasahara, S. Kasahara, T. Shibauchi and Y. Matsuda, Sci. Rep., 2022, 12, 9187.
6. K. Takada, M. Ito, N. Fukui and H. Nishihara, Commun. Chem., 2024, 7, 186.
7. M. Tran, Y. Wang, B. Dzikovski, M. E. Lahm, Y. Xie, P. Wei, V. V. Klepov, H. F. Schaefer and G. H. Robinson, J. Am. Chem. Soc., 2024, 146, 16340–16347.
8. F. Santanni, M. Briganti, G. Serrano, E. Salvadori, A. Veneri, C. Batistoni, S. F. Russi, S. Menichetti, M. Mannini, M. Chiesa, L. Sorace and R. Sessoli, JACS Au, 2023, 3, 1250–1262.
9. (a) K. Koshiba, K. Yamauchi and K. Sakai, Angew. Chem., Int. Ed., 2017, 56, 4247–4251; (b) Z. Liu, C. Xu, J. del Pozo, S. Torker and A. H. Hoveyda, J. Am. Chem. Soc., 2019, 141, 7137–7146.
10. J. A. Denny and M. Y. Darensbourg, Chem. Rev., 2015, 115, 5248–5273.
11. H. Alves, D. Simão, I. C. Santos, V. Gama, R. T. Henriques, H. Novais and M. Almeida, Eur. J. Inorg. Chem., 2004, 2004, 1318–1329.
12. L. K. Watanabe, Z. S. Ahmed, J. J. Hayward, E. Heyer, C. L. B. Macdonald and J. M. Rawson, Organometallics, 2022, 41, 226–234.
13. L. K. Watanabe, J. D. Wrixon and J. M. Rawson, Dalton Trans., 2021, 50, 13620–13633.
14. Y. Nakamura, T. Matsumoto, Y. Sakazume, J. Murata and H. C. Chang, Chem. – Eur. J., 2018, 24, 7398–7409.
15. E. J. Wharton and J. A. McCleverty, J. Chem. Soc. A, 1969, 2258–2266.
16. J. D. Wrixon, J. J. Hayward and J. M. Rawson, Inorg. Chem., 2015, 54, 9384–9386.
17. L. K. Watanabe, J. D. Wrixon, Z. S. Ahmed, J. J. Hayward, P. Abbasi, M. Pilkington, C. L. B. MacDonald and J. M. Rawson, Dalton Trans., 2020, 49, 9086–9093.
18. J. D. Wrixon, Z. S. Ahmed, M. U. Anwar, Y. Beldjoudi, N. Hamidouche, J. J. Hayward and J. M. Rawson, Polyhedron, 2016, 108, 115–121.
19. J. D. Wrixon, J. J. Hayward, O. Raza and J. M. Rawson, Dalton Trans., 2013, 43, 2134–2139.
20. G. B. Jameson, H. R. Oswald and H. R. Beer, J. Am. Chem. Soc., 1984, 106, 1669–1675.
21. E. J. Miller, T. B. Brill, A. L. Rheingold and W. C. Fultz, J. Am. Chem. Soc., 1983, 105, 7580–7584.
22. K. W. Stender, N. Wolki and G. Klar, Phosphorus, Sulfur Silicon Relat. Elem., 1989, 42, 111–114.
23. A. Alberola, D. Eisler, R. J. Less, E. Navarro-Moratalla and J. M. Rawson, Chem. Commun., 2010, 46, 6114–6116.
24. S. Tsukada, M. Kondo, H. Sato and T. Gunji, Polyhedron, 2016, 117, 265–272.
25. M. Nomura, E. Tsukano, C. Fujita-Takayama, T. Sugiyama and M. Kajitani, J. Organomet. Chem., 2009, 694, 3116–3124.
26. M. Nomura, T. Sasao, T. Hashimoto, T. Sugiyama and M. Kajitani, Inorg. Chim. Acta, 2010, 363, 3647–3653.
27. M. Nomura and M. Fourmigué, J. Organomet. Chem., 2007, 692, 2491–2499.
28. W. L. J. Loke and W. Y. Fan, Int. J. Hydrogen Energy, 2020, 45, 31976–31984.
29. W. K. Miller, R. C. Haltiwanger, M. C. Vanderveer and M. R. Dubois, Inorg. Chem., 1983, 22, 2973–2979.
30. O. A. Rajan, M. McKenna, J. Noordik, R. C. Haltiwanger and M. R. DuBois, Organometallics, 1984, 3, 831–840.
31. D. Szeymies, B. Krebs and G. Henkel, Angew. Chem., Int. Ed. Engl., 1983, 22, 885–886.
32. J. R. Dorfman and R. H. Holm, Inorg. Chem., 1983, 22, 3179–3181.
33. C. M. Bolinger, T. B. Rauchfuss and A. L. Rheingold, J. Am. Chem. Soc., 1983, 105, 6321–6323.
34. D. W. Stephan, Inorg. Chem., 1992, 31, 4218–4223.
35. M. Nomura, E. Tsukano, C. Fujita-Takayama, T. Sugiyama and M. Kajitani, J. Organomet. Chem., 2009, 694, 3116–3124.
36. S. Tsukada, M. Kondo, H. Sato and T. Gunji, Polyhedron, 2016, 117, 265–272.
37. M. Nomura and M. Fourmigué, Inorg. Chem., 2008, 47, 1301–1312.
38. D. Yang, Y. Li, B. Wang, X. Zhao, L. Su, S. Chen, P. Tong, Y. Luo and J. Qu, Inorg. Chem., 2015, 54, 10243–10249.
39. T. Sun, S. Xu, D. Yang, L. Su, B. Wang and J. Qu, Eur. J. Inorg. Chem., 2020, 2020, 4263–4269.
40. M. Nomura, Dalton Trans., 2011, 40, 2112–2140.
41. A. R. West, Solid State Chemistry and its Applications, J. Wiley and Sons, Chichester, UK, 1984.
42. A. Burger and R. Ramberger, Microchim. Acta, 1979, 72, 273–316.
43. F. H. Bhuiyan, Y.-S. Li, S. H. Kim and A. Martini, Sci. Rep., 2024, 14, 2992.
44. (a) CCDC 2470686: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxyk3; (b) CCDC 2470687: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxyl4; (c) CCDC 2470688: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxym5; (d) CCDC 2470689: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxyn6; (e) CCDC 2470690: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxyp7; (f) CCDC 2470691: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nxyq8; (g) CCDC 2471388: Experimental Crystal Structure Determination,  DOI:10.5517/ccdc.csd.cc2nyp6j.

---