research papers
Reactions of Rh2(CH3COO)4 with and thiolates: a structural study
aDepartment of Chemistry, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada T2N 1N4
*Correspondence e-mail: faridehj@ucalgary.ca
The structural differences between the aerobic reaction products of Rh2(AcO)4 (1; AcO− = CH3COO−) with and thiolates in non-aqueous media are probed by X-ray absorption spectroscopy. For this study, ethanethiol, dihydrolipoic acid (DHLA; a dithiol) and their sodium thiolate salts were used. Coordination of simple to the axial positions of Rh2(AcO)4 with Rh—SH bonds of 2.5–2.6 Å keeps the RhII—RhII bond intact (2.41 ± 0.02 Å) but leads to a colour change from emerald green to burgundy. Time-dependent density functional theory (TD-DFT) calculations were performed to explain the observed shifts in the electronic (UV–vis) absorption spectra. The corresponding sodium thiolates, however, break up the Rh2(AcO)4 framework in the presence of O2 to form an oligomeric chain of triply S-bridged Rh(III) ions, each with six Rh—S (2.36 ± 0.02 Å) bonds. The RhIII⋯RhIII distance, 3.18 ± 0.02 Å, in the chain is similar to that previously found for the aerobic reaction product from aqueous solutions of Rh2(AcO)4 and glutathione (H3A), {Na2[Rh2III(HA)4]·7H2O}n, in which each Rh(III) ion is surrounded by about four Rh—S (2.33 ± 0.02 Å) and about two Rh—O (2.08 ± 0.02 Å). The reaction products obtained in this study can be used to predict how dirhodium(II) tetracarboxylates would react with cysteine-rich proteins and such as metallothioneins.
1. Introduction
Dirhodium(II) tetracarboxylates [Rh2(RCOO)4] are compounds with specific applications in biological and chemical systems due to the stability of their Rh24+ core, resulting from the single Rh—Rh bond supported by four carboxylate bridges in a `lantern' structure. Their anti-tumor activity was initially reported by the Bear group (Erck et al., 1974; Bear et al., 1975). They also function as catalysts, especially with chiral ligands R that promote and in carbene insertion reactions into C—H bonds, such as cyclopropanation and C—H functionalization (Hansen & Davies, 2008), or insertion into polar X—H (X = S, N, O, etc.) bonds (Miller & Moody, 1995; García et al., 1996; Zhang et al., 2003). Reactions involving this family of compounds generally start with the formation of an adduct by a nucleophilic ligand L (Lewis base) coordinating in a monodentate manner to the axial positions, [Rh2(RCOO)4L1–2] (see 1 in Fig. 1) (Chifotides & Dunbar, 2005). For example, [Rh24+]-carbenoids are expected to form in the initial step of the above catalytic insertion reactions (García et al., 1996; Zhang et al., 2003). While the axial ligands are labile and easily exchanged, the bridging carboxylate groups are kinetically inert (Bear et al., 1979), and their substitution is proposed to occur stepwise. After an initial axial coordination of, for example, (N/O)-donor ligands with chelating (e.g. bipyridine) and/or bridging ability (e.g. guanine), the probable next step is formation of axial–equatorial (ax–eq) and finally eq–eq bonds (Yoshimura et al., 2003; Chifotides et al., 2004; Dunbar et al., 1994; Deubel & Chifotides, 2007). It has been suggested that the biological activity of dirhodium(II) carboxylates requires easily accessible axial positions (Aguirre et al., 2007).
Dirhodium(II) tetracarboxylates form adducts with nearly all common donor atoms, including sulfur (Chifotides & Dunbar, 2005). While the thiol (—SH) is important both from biological and chemical points of view, structural information for reaction products of and thiolates with dirhodium(II) carboxylates is limited. Such information would be useful, for example, in elucidating reactions of this family of anti-tumour active compounds with thiol-containing biomolecules, and have prompted several studies (Erck et al., 1976; Howard et al., 1976, 1977; Pneumatikakis & Psaroulis, 1980; Jakimowicz et al., 2000; Sorasaenee et al., 2002, 2003; Wong & Stillman, 2016; Wong et al., 2017; Jalilehvand et al., 2017; Enriquez Garcia & Jalilehvand, 2018).
The only reported crystal structures for such compounds are those of Rh2(AcO)4 (1, AcO− = CH3COO−) with axially coordinated benzylthiol or thiophenol groups (Christoph & Tolbert, 1980; Felthouse, 1982). Using combined data from different techniques such as electrospray ionization (ESI-MS) and extended X-ray absorption fine-structure (EXAFS) spectroscopy, we recently reported that the aerobic reactions of Rh2(AcO)4 with cysteine (H2Cys) and glutathione (GSH; denoted H3A in its triprotonated form) in aqueous media at physiological pH result in the formation of oligomeric products {Na2[RhIII2(S,N-Cys)4]·5H2O}m and {Na2[RhIII2(S-HA)4]·7H2O}n (i.e. RhIII–GSH; 2 in Fig. 1), with two and three thiolate groups, respectively, bridging between the two Rh(III) ions (Jalilehvand et al., 2017; Enriquez Garcia & Jalilehvand, 2018). Also, at early stages of the reaction between Rh2(AcO)4 and penicillamine (3,3′-dimethylcysteine) at the pH of mixing (∼3.0), we detected ESI-MS peaks corresponding to [RhII2(AcO)4(HPen)]−, [RhII2(AcO)4(Pen)]2- and [RhIII2(AcO)4(HPen)2]2- ions, and also made similar observations for the reaction with N-acetylcysteine (H2NAC). Therefore, we proposed that two deprotonated thiol groups are needed to oxidize the Rh(II) ions to Rh(III) (Jalilehvand et al., 2017).
Wong et al. in another recent study, based on ESI-MS and density functional theory (DFT) calculations, suggested that, when Rh2(AcO)4 reacts with excess glutathione in aqueous solution under anaerobic conditions, deprotonated glutathione (GS−) binds to the axial positions of Rh2(AcO)4 to form [RhII2(AcO)4(GS)(H2O)]− and [RhII2(AcO)4(GS)2]2− complexes (Wong et al., 2017). They also proposed that Rh2(AcO)4 releases all carboxylate ligands when reacting with the β-domain of human metallothionein (β-MT) 1a under anaerobic conditions. According to the authors the metallated RhII2-β-MT retains the Rh—Rh bond `with each Rh replacing four of the cysteine thiol protons', creating a square pyramidal coordination geometry around each Rh(II) ion in the dirhodium core (Wong & Stillman, 2016). However, Sorasaenee et al. showed that the anaerobic reaction of [RhII2(AcO)2(bpy)2(CH3CN)2](BF4)2 (3) (bpy = 2,2′-bipyridine) with excess of the monodentate ligand sodium benzylthiolate, NaC6H5S, led to formation of a dirhodium(II) complex [RhII(μ-S-C6H5S)(η1-S-C6H5S)(bpy)]2·CH3OH (4), with two bridging thiolates and another two thiolates opposite the Rh—Rh bond (see Fig. 1). The authors described this reaction as `an important first step in understanding the metabolism of dirhodium anticancer compounds with thiol-containing and proteins' (Sorasaenee et al., 2003).
Sorasaenee et al. also reported that the aerobic reaction of a similar complex [RhII2(AcO)2(phen)2(CH3CN)2](PF6)2 (phen = 1,10-phenanthroline) with excess 2-aminothiophenol (C6H7NS, a ligand with chelating ability) resulted in a mononuclear complex, [RhIII(η2-C6H7NS)(η1-C6H7NS)2(phen)] (5), while the aerobic reaction of complex 3 with four-fold C6H7NS yielded both mono- and bi-nuclear Rh(III) complexes, [RhIII(μ-S,N-C6H6NS)(η1-S-C6H6NS)(bpy)]22+ (6a) and [RhIII(S,N-C6H6NS)2(bpy)]+ (6b), co-crystallized in the same (Sorasaenee et al., 2002). The presence of the (N—N) ligands bpy and phen have probably promoted the formation of the monomeric Rh(III) species 5 and 6b (Jalilehvand et al., 2017). No has been reported for reaction product(s) of dirhodium(II) tetracarboxylate with thiolates.
Our goal in the current study was to gain better understanding of the structural changes that occur around the rhodium ions when dirhodium(II) tetracarboxylates [Rh2(RCOO)4] react with and thiolates under aerobic conditions. For this purpose, we used X-ray absorption spectroscopy to investigate reactions of Rh2(AcO)4 with two thiol-containing ligands ethanethiol (C2H5SH) and dihydrolipoic acid (DHLA) and also with their sodium thiolate salts. We chose the dithiol molecule DHLA (Fig. 2) to explore whether a molecule that can form an S,S-chelate with metal ions (Ioannou & Tsivgoulis, 2014) would facilitate the breakdown of the Rh2(AcO)4 structure under aerobic conditions. We have previously compared such an effect for cysteine, which behaves as an (S,N)- or (S,N,O)-donor ligand in its reaction product with Rh2(AcO)4, with N-acetylcysteine that prefers to bind as a monodentate thiolate ligand rather than forming a S,O-chelate (Jalilehvand et al., 2017). In the current study, we also performed time-dependent density functional theory (TD-DFT) calculations to explain the observed shifts in the UV–vis absorption spectra of Rh2(AcO)4 when dissolved in ethanethiol.
2. Experimental section
Sodium hydroxide, ethanethiol (EtSH), sodium ethanethiolate (NaEtS), (±)-α-lipoic acid (LA) were used without further purification. Dirhodium(II) tetraacetate was used as supplied by Pressure Company Co. Extra dry oxygen (99.6%) was supplied by Praxair. Dihydrolipoic acid (DHLA) and its sodium dithiolate salt Na3(DTLA) were prepared according to literature procedures (Spuches et al., 2005; Kubicki et al., 1995). Methanol (MeOH) was dried by refluxing twice over magnesium turnings under argon atmosphere, whereas dry tetrahydrofuran (THF) was obtained by refluxing over sodium/benzophenone and distilling immediately before use. Ethanol (EtOH) was degassed by passing argon through a boiled solution until cooled down to room temperature. Rhodium analysis was performed by the Canadian Microanalytical Services Ltd.
2.1. Sample preparation
2.1.1. Synthesis of Na3(DTLA)
Sodium metal (24.0 mmol) was added to a transparent solution of DHLA (4.80 mmol) in dry THF (20.0 ml) under argon atmosphere. The mixture was refluxed (70°C) under a stream of argon for 24 h. A white precipitate was formed, which was filtered under inert atmosphere and washed with THF (Kubicki et al., 1995). The solid was stored in a to avoid decomposition.
2.1.2. Solutions of Rh2(AcO)4 in ethanethiol, Rh2(AcO)4(EtSH)2 (7)
Dissolution of Rh2(AcO)4 (0.125 mmol) in 2.50 ml ethanethiol (33.8 mmol, d = 0.839 g ml−1) resulted in a clear burgundy solution after 30 min stirring under a stream of argon gas to prevent thiol (—SH) oxidation to disulfide (—S—S—); this solution was used for the data collection (C1 = 50 mM). For UV–vis measurement, a (C1 = 5.0 mM) was prepared by dissolving Rh2(AcO)4 (0.0025 mmol) in 0.50 ml neat ethanethiol. Another UV–vis spectrum was collected for a 1.0 mM solution of Rh2(AcO)4 (0.050 mmol) dissolved in 50.0 mM ethanethiol in chloroform (total volume = 50.00 ml). The solutions were kept under aerobic conditions prior to, or during, the physical measurements.
2.1.3. [Rh2(AcO)4(DHLA)]n solid (8)
To a DHLA solution (0.904 mmol) in 5.0 ml EtOH, a suspension of Rh2(AcO)4 (0.226 mmol) in 15.0 ml EtOH was added dropwise, resulting in a blue–green solution. The reaction was carried out under a stream of argon in O2-free solvent to avoid oxidation of thiol groups. After 24 h of stirring at room temperature a violet precipitate formed that was filtered in air, washed with EtOH and dried in a desiccator under vacuum. The solid was insoluble in common solvents. Calcd. for [Rh2(AcO)4(DHLA)]n (Rh2C16H28O10S2)n: C, 29.55; H, 4.34. Found: C, 29.81; H, 4.34; N, 0.02.
2.1.4. Aerobic reaction of Rh2(AcO)4 with sodium ethanethiolate (NaEtS) and Na3(DTLA), the sodium dithiolate salt of DHLA (9 and 10)
A solution of the sodium thiolate salt (2.26 mmol) in 10.0 ml of dry MeOH was added dropwise to a completely dissolved solution of Rh2(AcO)4 (0.226 mmol) in 180.0 ml dry MeOH under a stream of argon. Dry MeOH was used to avoid protonation of the sodium salts. The solution immediately turned dark orange–red and a precipitate was observed after 2 h. After stirring for 24 h, dry O2 was vigorously bubbled through the reaction mixture for 15 min, and stirring continued under O2 atmosphere for another 24 h. The reaction mixture was concentrated to 5 ml and the precipitate filtered off and then re-suspended in a 10.0 ml water–methanol (1:1) mixture to wash away any sodium acetate residue. The suspension was filtered to collect the orange precipitate, which was washed with distilled water and methanol and then dried under vacuum. The solid product was diamagnetic and insoluble in water and most organic solvents. Found for the Rh(III) ethanethiolate product (9): C, 24.02; H, 4.93; S, 23.80; Rh, 31.49. Elemental analysis calculations were carried out for several combinations of Rh(III), CH3CH2S−, Na+ and CH3COO−, and the composition closest to the experimental data was [Na3Rh7(C2H5S)17(CH3COO)7(H2O)4]n (Na3Rh7C48H114O18S17)n: C, 24.92; H, 4.97; S, 23.56; Rh, 31.13. Compound 10 is also an oligomer (or polymer) with similar structure as 9 (see below).
2.2. Methods
measurements on the solid samples were performed using a Johnson Matthey balance.
2.2.1. Electronic spectroscopy
A Cary 300 UV–vis double-beam spectrophotometer was used to measure the UV–vis spectra of Rh2(AcO)4 in water (C1 = 1.0 mM), in neat ethanethiol (C1 = 5.0 mM), or in 50.0 mM ethanethiol in chloroform (C1 = 1.0 mM), using the corresponding solvents as blank, with 1 mm path length quartz cells as sample holder, and 1.0 nm energy resolution.
2.2.2. Vibrational spectroscopy
The IR spectrum of a KBr disc of the [Rh2(AcO)4(DHLA)]n solid (7) was measured using a Thermo Nicolet NEXUS 470 FT-IR ESP spectrometer (4 cm−1 resolution). The IR spectrum of pure DHLA was obtained by placing a drop of DHLA (oil) between two BaF2 disks (window cut-off 870 cm−1).
2.2.3. X-ray absorption spectroscopy: data collection
Rh K-edge X-ray absorption spectra were measured at room temperature at BL7-3 (500 mA) at the Stanford Synchrotron Radiation Lightsource (SSRL) operating under 3 GeV. Higher-order harmonics were rejected by detuning a Si(220) (ϕ = 0°) double-crystal monochromator to 50% of maximum intensity (I0) at the end of the Rh K-edge scan range. The ion chambers (I0, I1 and I2) were filled with nitrogen (N2) and the Lytle detector with argon (Ar). The X-ray energy was internally calibrated by assigning the first inflection point in the of a Rh foil (placed between the I1 and I2 ion chambers) to 23219.80 eV. The monochromator energy step in the X-ray absorption near-edge structure (XANES) region was 0.3 eV. The solution of [Rh2(AcO)4(EtSH)2] (7) was held inside a 2 mm pinhole sample holder with Kapton windows, collecting the data at 120 K to suppress the strong odour of EtSH (boiling point = 308 K; freezing point = 125 K). The solid samples 1, 8, 9 and 10 were mixed with boron nitride (sample: BN = 70:30 w/w), finely ground and pressed in a 1 mm Al frame with Mylar tape as window material. Five to seven scans were collected for each sample in transmission mode (except for 1, which was measured in fluorescence mode). All individual scans were compared prior to averaging to ensure that no radiation damage occurred during measurement.
2.2.4. X-ray absorption spectroscopy: data analysis
WinXAS3.1 program (Ressler, 1998), by removing a first-order polynomial background in the pre-edge region, followed by edge step normalization. The (E0), defined as the first inflection point in the varied over a narrow range: 23226.0 eV (for 1), 23225.8 eV (7), 23225.6 eV (8), 23225.4 (9), 23225.3 (10). The E0 value was used for converting the energy unit to photoelectron wavevector, k (Å−1), where k = [(8π2me/h2)(E − E0)]1/2. The oscillation was then extracted by removing the atomic background absorption above the edge, using a seven-segment cubic spline.
oscillations were extracted using theFor modelling theoretical χ(k), the FEFF7.0 program (Zabinsky et al., 1995; Ankudinov & Rehr, 1997) was used to obtain ab initio calculated amplitude feff(k)i, phase shift ϕij(k), and λ(k) functions [equation (1)] for coordination models obtained from the crystal structures of [Rh2(AcO)4(H2O)2] (Cotton et al., 1971), {Rh2(AcO)4[(PhCH2)2S]2} (Cambridge Structural Database (CSD) code: DEHBIB) (Clark et al., 1985) and ΔΛ-[Rh{Ir(aet)3}2]Br3 (CSD: AQUVOX; Haet = 2-aminoethanethiol) (Mahboob et al., 2004), as well as the optimized geometry for [Rh4(EtS)15]3− (see the supporting information),
oscillations,The input file for the FEFF program was created by the ATOMS program (Ravel, 2001), using the atomic coordinates of the above crystal structures. For the ΔΛ-[Rh{Ir(aet)3}2]Br3 structure the terminal Ir atoms were replaced by Rh in the ATOMS input file, because of their similar ionic radii, Rh(III) 0.665 Å and Ir(III) 0.68 Å (Shannon, 1976).
The least-squares curve-fitting of the k3-weighted model function χ(k) to the experimental unfiltered oscillation was performed over the k-range 2.8–18 Å−1. For each backscattering path around the central Rh atom, the distance (R), the Debye–Waller factor parameter (σ2) and in some cases the (N) were refined, while ΔE0 was refined as a common value for all scattering paths in the model. The amplitude reduction factor (S0 2) was refined to 0.92 for 1 and fixed at this value for all other fittings. The accuracy of bond distances, σ2 and refined N values is within ±0.02 Å, ±0.001 Å2 and ±10–15%, respectively.
2.2.5. XANES calculations
The Rh K-edge XANES simulations were performed by means of the FEFF8.10 program (Ankudinov et al., 1998), using the atomic coordinates of the DFT optimized geometries of [Rh2(AcO)4(EtSH)2] and [Rh4(SEt)15]3− and muffin-tin potentials with 15% overlap. The best results were obtained when the Hedin–Lundqvist exchange-correlation potential was used for the The input files for both XANES calculations are presented in the supporting information.
2.2.6. Computational calculations
The electronic structure calculations were performed using the Gaussian09 package (Frisch et al., 2016). Initial atomic coordinates were imported from the of [Rh2(AcO)4(H2O)2] (Cotton et al., 1971), and were used as the starting point for the geometry optimizations. The Becke–Half-and-Half–LYP density functional was employed with the 6-31G(d) basis set for all non-metal atoms and the LANL2DZ pseudopotential for the Rh atoms. All calculated vibrational frequencies for the optimized geometries were positive. The optimized geometry of [Rh2(AcO)4(H2O)2] in its ground state was compared with the reported (Cotton et al., 1971) to test the combination of functional and basis sets. All Rh—Rh and Rh—Oeq distances in the optimized geometry were within ±0.02 Å of those in the crystal structure.
TD-DFT calculations were performed for the optimized geometries of [Rh2(AcO)4(H2O)2], [Rh2(AcO)4(H2S)2] and [Rh2(AcO)4(EtSH)2] placed inside a solvent cavity, considering the first 100 lowest vertical excitations. The solvent effects were described by means of the Polarizable Continuum Model (PCM) (Frisch et al., 2016), using water for all three systems, and also chloroform for [Rh2(AcO)4(EtSH)2].
The optimization and TD-DFT calculations were repeated using the same functional, but this time with the 6-311++G(2d,p) basis set, including the LANL2TZ(f) pseudopotential for Rh, to study whether this more comprehensive and flexible combination of basis sets would significantly affect the results. Both models resulted in similar optimization parameters and transition energies, and identity of the excitations, with comparable TD-DFT-reported state compositions in terms of fractional contributions to the excitations. These results suggest that the original combination of basis sets describes the system adequately.
3. Results and discussion
3.1. Electronic absorption spectroscopy
The UV–vis spectrum of the emerald green aqueous solution of Rh2(AcO)4 (1) showed the characteristic broad band of [Rh2(AcO)4(H2O)2] with λmax = 584 nm (Boyar & Robinson, 1983), which previously has been attributed to a π*(Rh24+) → σ*(Rh24+) transition (band I); see Fig. 3 (Wilson & Taube, 1975; Norman et al., 1979; Norman & Kolari, 1978). Schematic (MO) diagrams for the dirhodium(II) core, with calculated energy levels of Rh2(AcO)4 and [Rh2(AcO)4(H2O)2] are shown in Figs. S1a, S1b and S2a, respectively, of the supporting information. The π*(Rh24+) → σ*(Rh24+) transition is sensitive to the nature of the donor atom in the axially coordinated ligands (L) in the [Rh2(AcO)4L2] complexes, shifting to shorter wavelength in the following order: O < S < Nsp3 < Nsp2 < S = O (Kitchens & Bear, 1969). Burgundy solutions with λmax = 555 nm (blue shift ∼29 nm) are obtained when dissolving Rh2(AcO)4 in neat ethanethiol (C1 = 5.0 mM), or in 50.0 mM ethanethiol in chloroform (C1 = 1.0 mM); see Fig. 3.
This blue shift for band I is smaller than those observed for the bis(thioether) adducts of Rh2(AcO)4 with methionine (λmax = 537 nm) (Chen & Kostic, 1988; Jakimowicz et al., 2000), diethylsulfide (541 nm) (Kitchens & Bear, 1969) and dibenzylsulfide (546 nm) (Clark et al., 1985). The largest shift reported is for the S-coordinated dimethylsulfoxide (DMSO) adduct (λmax = 497 nm) (Kitchens & Bear, 1969; Johnson et al., 1963). Dubicki & Martin proposed based on a self-consistent charge and configuration (SCCC-MO) calculation that such blue shifts relative to [Rh2(AcO)4(H2O)2] are probably caused by an increasing axial interaction between the ligands, i.e. the soft S atom of the axial thioether/thiol groups and the σ*(Rh24+) orbital of the soft Rh(II) ions, which has a high 4dz2 contribution and is directed toward the ligands L along the linear L—Rh—Rh—L axis. The stronger the of the axial ligand L, the higher the energy of the antibonding σ*(Rh24+) orbital (Dubicki & Martin, 1970). The observed colour change should therefore be due to the influence of the axial ethanethiol ligands on the π*(Rh24+) → σ*(Rh24+) transition in its Rh2(AcO)4 adduct (also see Section 3.2) (Felthouse, 1982; Dubicki & Martin, 1970). The retention of the Rh—Rh bond in the ethanethiol adduct of Rh2(AcO)4 is further confirmed by spectroscopy (see below).
Because of the high solvent absorption in neat ethanethiol below ∼300 nm, we also measured the UV–vis spectrum of Rh2(AcO)4 dissolved in 50.0 mM ethanethiol in chloroform (C1 = 1.0 mM), which enabled us to observe an intense band at 295 nm. A similar absorption has been observed at 308 nm in the UV–vis spectra of the bis(thioether) adducts of 1: [Rh2(AcO)4{S(CH2Ph)2}2] (Clark et al., 1985) and [Rh2(AcO)4{methionine}2] (Chen & Kostic, 1988; Enriquez Garcia et al., 2018).
3.2. Computational studies
To interpret the observed UV–vis absorption bands, DFT calculations were performed followed by TD-DFT calculations. As a starting point for the optimizations, the molecular structure of the aqua complex was adapted from the 2(AcO)4(H2O)2] (Cotton et al., 1971); see Section 2.2.6. Then the aqua oxygen atoms were replaced with sulfur to obtain [Rh2(AcO)4(H2S)2] in order to investigate the effect of the axial donor atom. This was followed by replacing one hydrogen atom of each H2S ligand with an ethyl group to obtain [Rh2(AcO)4(EtSH)2]. The calculated energy diagrams of the MOs for [Rh2(AcO)4L2] (L = H2O, H2S and EtSH) are shown in Fig. 4 and the corresponding graphic representations of each are displayed in Figs. S2a–S2c.
of [RhThe energies of the calculated MOs for [Rh2(AcO)4(H2O)2] are in agreement with those calculated by Wong et al., using the CAM-B3LYP functional (Wong et al., 2017). Fig. 4 shows that by changing H2O to H2S as the axial ligand the energy of the σ(Rh24+) MO increases. Note that this orbital has some contribution from the σ*(Rh—X) orbital, originating from the axially coordinated donor atom X (i.e. O or S), beside the Rh—Rh σ interaction (Sowa et al., 1983). Therefore, the increase in σ bonding interaction between the soft sulfur and Rh(II) atoms will also increase the σ*(Rh—S) contribution and destabilizes the σ(Rh24+) MO orbital in [Rh2(AcO)4(H2S)2]. Replacing one hydrogen atom in H2S with an ethyl group in [Rh2(AcO)4(EtSH)2] increases the σ*(Rh—S) contribution and raises the energy of the σ(Rh24+) orbital further.
The TD-DFT calculations were performed for [Rh2(AcO)4(EtSH)2] in a cavity surrounded by a continuum corresponding to either water or chloroform, to simulate the effect of the experimental conditions (see Section 2.2.6). The resulting calculated transition energies for [Rh2(AcO)4(EtSH)2] were very similar (see Fig. S3) and only the results for water will be discussed hereafter.
The TD-DFT calculations show two excitations with weak oscillator strength (f ≃ 0.003–0.004) at 514 nm for [Rh2(AcO)4(EtSH)2] corresponding to the π*(Rh24+) → σ*(Rh24+) transition (with ∼76% contribution) originating from the two π*(Rh24+) orbitals (see Fig. S4). This transition is blue-shifted 27 nm compared with the aqua complex (541 nm). The extent of this blue shift is in agreement with the 29 nm blue shift of band I observed experimentally for the solutions of 1 dissolved in water and ethanethiol (see Fig. 3). This blue shift was previously described as due to destabilization of the σ*(Rh24+) MO when the of the axial ligand increased, moving the π*(Rh24+) → σ*(Rh24+) transition toward higher energies (Dubicki & Martin, 1970). However, our DFT calculations show the σ*(Rh24+) MO energy to be slightly lower in the ethanethiol adduct than in the aqua complex (Fig. 4), suggesting that this π*(Rh24+) → σ*(Rh24+) transition, although predominant, is not the only contributing factor to the blue shift of band I. In the UV region, our computational results for [Rh2(AcO)4(EtSH)2] show two vertical excitations with high oscillator strength (f ≥ 0.3) in the 280–290 nm region (Fig. S4). The MO transition that contributes significantly to both excitations is σ(Rh24+) → σ*(Rh24+), depicted in Fig. 4. Similar calculations for the aqua complex [Rh2(AcO)4(H2O)2] imply a high contribution (70%) of the same transition to the band at 239 nm (Fig. S2a and Table S1). The UV band shift to longer wavelengths for the thiol-coordinated adduct [Rh2(AcO)4(EtSH)2] relative to that of the aqua complex corresponds to the destabilization of the σ(Rh2) orbital. The increase in its energy level leads to a decrease in the energy of the σ(Rh24+) → σ*(Rh24+) excitation and absorption at longer wavelengths. A second transition with significant contribution to the oscillator strength (38%) is from one of the π(Rh24+) orbitals to the σ*(Rh24+), as shown in Figs. 4 and S4.
Thus, our TD-DFT calculations for [Rh2(AcO)4(EtSH)2] show that the electronic transitions π(Rh24+) → σ*(Rh24+) and σ(Rh24+) → σ*(Rh24+) closely match the experimentally observed band at 295 nm (see Fig. 3). Note that for the aqua complex [Rh2(AcO)4(H2O)2] the σ(Rh24+) → σ*(Rh24+) transition occurs at 247 nm (see Fig. 3) (Norman & Kolari, 1978).
3.3. Rh K-edge X-ray absorption spectroscopy
Fig. 5 compares the k3-weighted Rh K-edge spectra and the corresponding Fourier transforms (FTs) of solid Rh2(AcO)4 (1) and its solution in ethanethiol, [Rh2(AcO)4(EtSH)2] (7), [Rh2(AcO)4(DHLA)]n solid (8), as well as the reaction products of Rh2(AcO)4 with excess amount of the sodium thiolate salts NaEtS (9) and Na3(DTLA) (10).
The Fourier transforms of the 2(AcO)4 in ethanethiol (7), and for the [Rh2(AcO)4(DHLA)]n solid (8) show a distinct peak for the Rh—Rh bond in a similar position as the corresponding FT peak for Rh2(AcO)4 in Fig. 5. For the aerobic reaction products of Rh2(AcO)4 with large excess (mole ratio 1:10) of sodium ethanethiolate (9), or the sodium dithiolate salt of DHLA (10), the oscillations and the corresponding FTs differ considerably from those of Rh2(AcO)4 as shown in Fig. 5. The spectra of products 9 and 10 almost superimpose (Fig. S5). For comparison with the RhIII–GSH product (2), see Section 3.3.3.
spectra obtained for a solution of RhChanges in the XANES spectra of these compounds are also informative. Fig. 6 (top) displays similar features for solid Rh2(AcO)4 (1) and its solution in ethanethiol [Rh2(AcO)4(EtSH)2] (7). However, the XANES spectrum of the reaction product of Rh2(AcO)4 with excess amount of sodium ethanethiolate (9) differs from its solution in ethanethiol (7). Thus, a drastic change in the electronic environment around the Rh ions occurs when Rh2(AcO)4 reacts with sodium ethanethiolate under aerobic conditions.
To some extent the first inflection point (E0) of the absorption edges is also affected: for the Rh(II) compounds 1 and 7, the E0 values are 23226.0 and 23225.8, respectively, while for the Rh(III) compound 9 the first inflection point occurs at slightly lower energy (23225.4 eV). For a higher higher energy is generally expected to excite a core electron. However, in this case, the Rh(III) ions in 9 are surrounded by six electron-rich thiolate groups, which provide electron density to the Rh(III) ions. The calculated Mulliken charge of +0.79 for the Rh(III) ions in Rh2S3 (Tan & Harris, 1998), with a similar Rh surrounding as in 9 (Section 3.3.3), is consistent with the lower energy required for exciting core electrons of Rh(III) ions in 9.
3.3.1. Ethanethiol solution of Rh2(AcO)4
The Rh K-edge spectrum of the ethanethiol solution of Rh2(AcO)4 fits well to a model consisting of four Rh—O (2.03 ± 0.02 Å), one Rh—Rh (2.41 ± 0.02 Å) and one Rh—S (2.52 ± 0.02 Å) distances around each Rh ion; see the spectrum of 7 in Fig. 5 and Table 1. These bond lengths are comparable with the Rh—Rh bond lengths 2.4020 (3) Å and 2.4024 (7) Å, and corresponding Rh—Saxial distances 2.551 (2) Å and 2.548 (1) Å in the crystal structures of [Rh2(AcO)4(HSCH2Ph)2] and [Rh2(AcO)4(HSPh)2], respectively (Christoph & Tolbert, 1980; Felthouse, 1982). This result is consistent with a dominating [Rh2(AcO)4(EtSH)2] (7) complex in the ethanethiol solution, with EtSH molecules occupying the axial sites of the Rh2(AcO)4 core (Fig. 7).
‡Additional refined paths: 4Rh⋯C (R = 2.96 Å, σ2 = 0.0031 Å2), 4Rh⋯Oeq′ (R = 3.09 Å, σ2 = 0.0023 Å2). §Additional refined paths: 4Rh⋯C (R = 2.88 Å, σ2 = 0.0030 Å2), 4Rh⋯Oeq′ (R = 3.08 Å, σ2 = 0.0061 Å2). ¶Additional refined path {using FEFF files from the optimized geometry of [Rh4(EtS)15]3−}: 6.2Rh—S—Rh—S (nleg = 4; along the linear S—Rh—S entities) (R = 4.73 Å, σ2 = 0.0087 Å2), correlating its to Rh—S. ††Additional refined path: 6.3Rh—S—Rh—S (nleg = 4) (R = 4.72 Å, σ2 = 0.0097 Å2), correlating its to Rh—S. ‡‡Concentrated solution of RhIII–GSH reaction product with three bridging thiolato groups between the Rh(III) ions (Enriquez Garcia & Jalilehvand, 2018). |
The elongation of the Rh—Rh bond distance from 2.38 ± 0.02 Å in solid Rh2(AcO)4 (1) to 2.41 ± 0.02 Å in [Rh2(AcO)4(EtSH)2] (7) can be attributed to the trans influence of the axially coordinated ethanethiol ligands, and the destabilization of the σ(Rh24+) (see Fig. 4, and Section 3.2). For a series of [Rh2(AcO)4L2] complexes [L = pyridine, NH(Et)2, CO, PF3, P(OPh)3, P(Ph)3, P(OCH3)3], it has been shown that axial ligands L, both with strong σ-donating (basicity) or with strong π-accepting ability, can weaken and lengthen the Rh—Rh bond. For example, the elongation of the Rh—Rh distance to 2.430 (3) Å in the [Rh2(AcO)4(PF3)2] complex has been explained as a result of the strong π-acceptance ability of PF3 to the empty PF σ*-orbital, despite its weak σ-donation (due to the electronegative F atoms) (Christoph & Koh, 1979; Orpen & Connelly, 1990). So ethanethiol, which is a fairly weak σ-donor but a good π-acceptor (to a σ*-orbital) (Kraatz et al., 1993), should have a similar effect on the Rh—Rh bond. Christoph & Koh proposed that π-back donation of electrons to the π-acceptor ligands in [Rh2(AcO)4L2] complexes would occur from filled high-energy MO orbitals with π rather than π* symmetry (since removing electron density from π* orbitals would strengthen the Rh—Rh bond) (Christoph & Koh, 1979).
3.3.2. Reaction of Rh2(AcO)4 with dihydrolipoic acid
We have previously shown that the reaction of Rh2(AcO)4 with cysteine at mole ratio of 1:4 leads to formation of a {Rh2(HCys)2(Cys)2·4H2O}n precipitate at the pH of mixing (pH = 3.2), and to oligomeric {Na2[Rh2(Cys)4]·5H2O}n species at pH = 7.4 in which the cysteinate ligands act as (S,N)- or (S,N,O)-chelates forming dithiolate bridges between two Rh(III) ions (Jalilehvand et al., 2017). Dihydrolipoic acid (DHLA) could act as a dithiol ligand and form an (S,S)-chelate upon deprotonation. Therefore, we investigated the reaction of Rh2(AcO)4 with DHLA in ethanol as solvent, since DHLA is insoluble in water. The product was a violet precipitate, [Rh2(AcO)4(DHLA)]n (8), which was not soluble in common solvents (e.g. water, acetonitrile, etc.). Least-squares curve-fitting of its Rh K-edge spectrum showed a similar Rh—Rh distance (2.41 ± 0.02 Å) to that in [Rh2(AcO)4(EtSH)2] (7); see Table 1. Also, a Rh—S scattering path could be fitted at 2.61 ± 0.02 Å, which is slightly longer than that of 7 (2.52 ± 0.02 Å). Fig. 8 displays the separate contributions from different scattering paths used in the model fitting of the [Rh2(AcO)4(DHLA)]n solid (8).
In the FT-IR spectrum of the solid 8, the S—H stretching vibration of DHLA appears at 2530 cm−1, shifted to lower frequencies relative to that of pure DHLA at 2561 cm−1 (Fig. S6; literature values 2547–2557 cm−1) (Nikolić et al., 2014; Zhang et al., 2013), suggesting that the S—H bond becomes weaker upon coordination of DHLA to the axial positions of Rh2(AcO)4. Also the S—H stretching frequency for benzylthiol coordinated in [Rh2(AcO)4(HSCH2Ph)2] (2550 cm−1) shows a similar shift relative to that of pure benzylthiol (2565 cm−1) (Christoph & Tolbert, 1980; Rajalingam et al., 2010). It is unlikely that the DHLA carboxyl group is involved in any bonding interaction in this compound, e.g. replacing an acetate group in the Rh2(AcO)4 cage of 8, since its 1:1 Rh2(AcO)4:DHLA composition from elemental analysis confirms that the number of AcO− ligands remains unchanged. Moreover, the C=O stretching vibrational frequency of the COOH group in 8 (1708 cm−1) is very similar to that in pure DHLA (1707 cm−1), and the νasym(AcO−) vibrational band shifts slightly to a higher frequency (1587 cm−1), relative to that of Rh2(AcO)4 [1579 cm−1; reported at 1586 cm−1 at 80 K (Clark & Hempleman, 1988)]; see Figs. S6 and S7a.
The above results show that DHLA in [Rh2(AcO)4(DHLA)]n (8) behaves like ethanethiol in [Rh2(AcO)4(EtSH)2] (7). Each thiol group of DHLA coordinates via its S atom to the axial position of a Rh2(AcO)4 unit, thus forming a bridge rather than an (S,S)-chelate; i.e. it does not behave like the (S,N)-donor ligand cysteine, which replaces the acetate groups. The proposed oligomeric structure in Fig. 7 (middle) is deduced from the 1:1 ratio of Rh2(AcO)4:DHLA in 8, which is insoluble in all common solvents.
3.3.3. Reactions of Rh2(AcO)4 with sodium ethanethiolate (NaEtS) and Na3(DTLA), the sodium dithiolate salt of DHLA
Ethanethiol and DHLA are both weak acids in water (pKa = 10.6 and 10.7, respectively) (Bhattacharyya & Rohrer, 2012; Fuchs, 1997); however, since their reactions with Rh2(AcO)4 were not carried out in aqueous media, formation of their conjugate bases was not promoted. Therefore, we reacted their sodium thiolate salts NaEtS and Na3(DTLA) with Rh2(AcO)4 in mole ratio 10:1 under aerobic conditions, which produced the precipitates 9 and 10, respectively. These products were found to be diamagnetic based on measurements and insoluble in common solvents, and therefore considered to be oligomeric (or polymeric) in nature. The [Na3Rh7(CH3CH2S)17(CH3COO)70.4H2O]n is closest to the elemental analysis of 9 (see Section 2.1.4). The presence of bridging acetate groups in compound 9 was confirmed by assigning a band at 1541 cm−1 in its IR spectrum to COO− stretching (see Fig. S7c).
The 9 and 10 were quite different from those of the thiol (EtSH and DHLA) reaction products [Rh2(AcO)4(EtSH)2] (7) and [Rh2(AcO)4(DHLA)]n (8); see Fig. 5. The XANES spectra of 7 and 9 also display a drastic change in the electronic environment around the Rh ions in these compounds (see Fig. 6). Analyses of the data using the FEFF files from the optimized geometry for [Rh4(EtS)15]3−, which allowed fitting the FT feature at (R − α) ≃ 4 Å, revealed six Rh—S and about two RhIII⋯RhIII interactions with mean distances of 2.37 ± 0.02 Å and 3.18 ± 0.02 Å, respectively, around each Rh ion in 9 and 10 (see Table 1); the separate contributions from each scattering path to the spectrum of the solid 10 are shown in Fig. 9. These distances are close to the average Rh—S and Rh⋯Rh distances, Rh—Save = 2.365 Å and RhIII⋯RhIII = 3.208 Å, in the of rhodium(III) sulfide (Rh2S3), which comprises of distorted [ReS6] octahedra and [SRh4] tetrahedra (Parthé et al., 1967; Tan & Harris, 1998). Also, in the [Cp*Rh(μ-SPh)3Rh(μ-SPh)3RhCp*]Cl crystal (CSD code: RUYROT), the average distances around the central Rh(III) ion surrounded by six bridging thiolate S atoms are Rh—Save = 2.371 Å and RhIII⋯RhIII = 3.218 Å (Boudreau et al., 2010). Oxidation of the Rh(II) ions in Rh2(AcO)4 (1) to Rh(III) ions in the rhodium-thiolate products 9 and 10 probably occurs in a similar way as the reaction between 1 and glutathione, cysteine and its derivatives, where O2 is reduced to (O22−) (Jalilehvand et al., 2017; Enriquez Garcia & Jalilehvand, 2018).
oscillations and the corresponding FTs forThe 9 was also fitted, using the FEFF files from the ΔΛ-[Rh{Ir(aet)3}2]Br3 structure, to simulated oscillations that also included an Rh—O scattering path because of the presence of acetate groups shown by its FT-IR spectrum (Models II–IV in Table S2). Assuming an RhS5O coordination (Model II) resulted in reasonable distances and σ2 values; however, the fitting residual increased relative to that of RhS6 coordination (Model I); see Fig. S8. of coordination numbers for all scattering paths (Rh—O, Rh—S and Rh⋯Rh) led to rejection of the Rh—O path. Fig. 7 (right) displays our proposed structure for compound 9, while coordination of CH3COO− to Rh(III) ions cannot be ruled out.
spectrum ofThe XANES features of the oligomeric RhIII–GSH product, {Na2[RhIII2(HA)4]·7H2O}n (2), resemble those of the reaction product of 1 with excess sodium ethanethiolate (NaEtS) (9); see Fig. 6. The FT of the k3-weighted spectrum of 2 shows a small peak at ∼2.8 Å (not corrected for phase shift), as also for compounds 9 and 10; see Fig. 5 (right). Thus, the model compounds 9 and 10 support our previously proposed structure for the RhIII–GSH product (2), in which three glutathione thiolato groups bridge two Rh(III) ions in each dimeric unit, with the Rh⋯Rh distance refined to 3.11 ± 0.02 Å (see Table 1) (Enriquez Garcia & Jalilehvand, 2018). The main difference is that each Rh(III) ion is surrounded by four thiolates and two O atoms (from —COO− or H2O) in 2, but six thiolates in 9 and 10.
3.4. XANES simulation
Fig. 10 shows experimental XANES spectra for [Rh2(AcO)4(EtSH)2] (7, left) and for the reaction product of Rh2(AcO)4 with excess NaEtS (9, right), compared with XANES spectra simulated for DFT optimized geometries of [Rh2(AcO)4(EtSH)2] and [Rh4(EtS)15]3−, respectively, by means of the FEFF8.10 program (Ankudinov et al., 1998). The absorption features in the experimental XANES spectra are well reproduced in the simulated XANES spectra, indicating that [Rh4(EtS)15]3−, which is similar to the proposed structure in Fig. 7 (right), is a good structural model for 9. However, both simulated spectra show an energy shift of about +20 eV relative to the K-edge Attempts to reduce the shift using the so-called correction card were unsuccessful. Similar energy shifts of simulated versus experimental XANES spectra have been previously reported (Bosman & Thieme, 2009), and explained based on the inaccuracy in the potentials and the photoelectron self-energy (Rehr & Albers, 2000; Gilbert et al., 2003).
4. Conclusions
The current study shows that monothiol or dithiol ligands with —SH as their only 2(AcO)4, keeping the Rh24+ cage intact (RhII—RhII = 2.41 ± 0.02 Å). The energy levels of the σ(Rh24+) and π(Rh24+) orbitals increase, while σ*(Rh24+) decreases relative to the corresponding orbitals of the aqua complex [Rh2(AcO)4(H2O)2], shifting the corresponding absorption bands due to σ(Rh24+), π(Rh24+) → σ*(Rh24+) and π*(Rh24+) → σ*(Rh24+) transitions in the UV and visible regions.
coordinate to the axial positions of RhIn contrast, ligands with thiolate as their only coordination site facilitate the oxidation of Rh2(AcO)4 under aerobic conditions and break up its cage structure, forming stable oligomeric Rh(III) d6 low-spin species with three bridging thiolate groups between the Rh(III) ions, each with a RhS6 coordination environment (RhIII⋯RhIII ≃ 3.18 ± 0.02 Å).
Even though the current study was carried out in non-aqueous media, the present results may serve as general schemes for `aerobic' reactions of dirhodium(II) tetracarboxylates with thiol and thiolate ligands. These structural models are consistent with our proposed structure for the aerobic reaction product of Rh2(AcO)4 with glutathione at physiological pH, {Na2[RhIII2(HA)4]·7H2O}n, with three bridging thiolates between the Rh(III) ions in dimeric units (Enriquez Garcia & Jalilehvand, 2018). The present results are also applicable for interactions of dirhodium(II) carboxylates with metallothioneins (MTs), the thiol-rich proteins. We propose that under aerobic and physiological conditions the reaction product of Rh2(AcO)4 with β-MT has a similar structure as the RhIII–GSH compound (2 in Fig. 1), or a dimeric unit of the model compound 9 (Fig. 7, right), with three β-MT cysteinyl thiolates bridging between the two Rh(III) ions.
Supporting information
Simulated UV-vis spectra for [Rh2(AcO)4(EtSH)2] in water and in chloroform; main excitations in the simulated UV-vis spectra of [Rh2(AcO)4(H2O)2] and [Rh2(AcO)4(EtSH)2] in water in the 200-350 nm and 400-700 nm regions; https://doi.org/10.1107/S160057751900033X/hf5373sup1.pdf
diagrams and MO energies for [Rh2(AcO)4], [Rh2(AcO)4(H2O)2], [Rh2(AcO)4(H2S)2] and [Rh2(AcO)4(EtSH)2]; simulated electronic transitions in the 200-800 nm region for [Rh2(AcO)4(H2O)2] and expected absorption wavelengths; superimposed Rh K-edge spectra of compounds 9 and 10; FT-IR spectra of pure DHLA, and solids Rh2(AcO)4, Na(AcO), 8 and 9; FEFF 8.10 input files for simulating the XANES spectra of [Rh2(AcO)4(EtSH)2] and [Rh4(EtS)15]^3-. DOI:Acknowledgements
We are grateful to Professor Arvi Rauk for helpful discussions regarding the theoretical calculations results. AEG acknowledges University of Calgary Eyes High, and Faculty of Science Dean's Open Competitions Doctoral Scholarships. This work was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Canadian Foundation for Innovation (CFI), the Province of Alberta (Department of Innovation and Science) and the University of Calgary (URGC SEED Grant). Theoretical calculations were performed using the computer resources provided by WestGrid (https://www.westgrid.ca) and Compute/Calcul. Canada (https://www.computecanada.ca). X-ray absorption data collection was carried out at the Stanford Synchrotron Radiation Lightsource (SSRL; Proposal No. 3637). Use of the SSRL, SLAC National Accelerator Laboratory, is supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. The SSRL Structural Molecular Biology Program is supported by the DOE Office of Biological and Environmental Research, and by the National Institutes of Health, National Institute of General Medical Sciences (including P41GM103393). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of NSERC, NIGMS or NIH.
Funding information
The following funding is acknowledged: Canada Foundation for Innovation (grant No. 9479 to FJ); Department of Innovation and Science of Province of Alberta (grant to FJ); National Institutes of Health (NIH) – National Institute of General Medical Sciences (grant No. P41 GM103393 to Keith O. Hodgson); Natural Sciences and Engineering Research Council of Canada (grant No. RGPIN-2008-250406 to FJ; grant No. RGPIN-2016-04546 to FJ); University Research Grant Committee, University of Calgary (grant No. SEED Grant 1031019479 to FJ).
References
Aguirre, J. D., Lutterman, D. A., Angeles-Boza, A. M., Dunbar, K. R. & Turro, C. (2007). Inorg. Chem. 46, 7494–7502. CrossRef CAS Google Scholar
Ankudinov, A. L., Ravel, B., Rehr, J. J. & Conradson, S. D. (1998). Phys. Rev. B, 58, 7565–7576. Web of Science CrossRef CAS Google Scholar
Ankudinov, A. L. & Rehr, J. J. (1997). Phys. Rev. B, 56, R1712–R1716. CrossRef CAS Google Scholar
Bear, J. L., Gray, H. B. Jr, Rainen, L., Chang, I. M., Howard, R., Serio, G. & Kimball, A. P. (1975). Cancer Chemother. Rep. 59, 611–620. Google Scholar
Bear, J. L., Howard, R. A. & Korn, J. E. (1979). Inorg. Chim. Acta, 32, 123–126. CAS Google Scholar
Bhattacharyya, L. & Rohrer, J. S. (2012). Editors. Applications of Ion Chromatography for Pharmaceutical and Biological Products (Appendix 1), pp. 451–453. New York: John Wiley & Sons, Inc. Google Scholar
Bosman, E. & Thieme, J. (2009). J. Phys. Conf. Ser. 186, 012004. CrossRef Google Scholar
Boudreau, J., Grenier-Desbiens, J. & Fontaine, F.-G. (2010). Eur. J. Inorg. Chem. 2010, 2158–2164. CrossRef Google Scholar
Boyar, E. B. & Robinson, S. D. (1983). Coord. Chem. Rev. 50, 109–208. CrossRef CAS Web of Science Google Scholar
Chen, J. & Kostic, N. M. (1988). Inorg. Chem. 27, 2682–2687. CrossRef CAS Google Scholar
Chifotides, H. T. & Dunbar, K. R. (2005). Rhodium Compounds, Multiple Bonds Between Metal Atoms, edited by F. A. Cotton, C. A. Murillo & R. A. Walton. New York: Springer Science and Business Media Inc. Google Scholar
Chifotides, H. T., Koomen, J. M., Kang, M. J., Tichy, S. E., Dunbar, K. R. & Russell, D. H. (2004). Inorg. Chem. 43, 6177–6187. CrossRef CAS Google Scholar
Christoph, G. G. & Koh, Y. B. (1979). J. Am. Chem. Soc. 101, 1422–1434. CrossRef CAS Web of Science Google Scholar
Christoph, G. G. & Tolbert, M. (1980). Am. Crystallogr. Assoc. Ser. 2, 7, 39. Google Scholar
Clark, R. J. H. & Hempleman, A. J. (1988). Croat. Chem. Acta, 61, 313–329. CAS Google Scholar
Clark, R. J. H., Hempleman, A. J., Dawes, H. M., Hursthouse, M. B. & Flint, C. D. (1985). J. Chem. Soc. Dalton Trans. pp. 1775–1780. CrossRef Google Scholar
Cotton, F. A., DeBoer, B. G., LaPrade, M. D., Pipal, J. R. & Ucko, D. A. (1971). Acta Cryst. B27, 1664–1671. CrossRef CAS IUCr Journals Web of Science Google Scholar
Deubel, D. V. & Chifotides, H. T. (2007). Chem. Commun. pp. 3438–3440. CrossRef Google Scholar
Dubicki, L. & Martin, R. L. (1970). Inorg. Chem. 9, 673–675. CrossRef CAS Google Scholar
Dunbar, K. R., Matonic, J. H., Saharan, V. P., Crawford, C. A. & Christou, G. (1994). J. Am. Chem. Soc. 116, 2201–2202. CrossRef CAS Google Scholar
Enriquez Garcia, A. & Jalilehvand, F. (2018). J. Biol. Inorg. Chem. 23, 231–239. CrossRef CAS Google Scholar
Enriquez Garcia, A., Jalilehvand, F., Niksirat, P. & Gelfand, B. S. (2018). Inorg. Chem. 57, 12787–12799. CrossRef CAS Google Scholar
Erck, A., Rainen, L., Whileyman, J., Chang, I.-M., Kimball, A. P. & Bear, J. (1974). Proc. Soc. Exp. Biol. Med. 145, 1278–1283. CrossRef CAS Google Scholar
Erck, A., Sherwood, E., Bear, J. L. & Kimball, A. P. (1976). Cancer Res. 36, 2204–2209. CAS Google Scholar
Felthouse, T. R. (1982). Prog. Inorg. Chem., 29, 73–166. CAS Google Scholar
Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Scalmani, G., Barone, V., Petersson, G. A., Nakatsuji, H., Li, X., Caricato, M., Marenich, A., Bloino, J., Janesko, B. G., Gomperts, R., Mennucci, B., Hratchian, H. P., Ortiz, J. V., Izmaylov, A. F., Sonnenberg, J. L., Williams-Young, D., Ding, F., Lipparini, F., Egidi, F., Goings, J., Peng, B., Petrone, A., Henderson, T., Ranasinghe, D., Zakrzewski, V. G., Gao, J., Rega, N., Zheng, G., Liang, W., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Throssell, K., Montgomery, J. A., Peralta, J. J. E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin, K. N., Staroverov, V. N., Keith, T., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J. C., Iyengar, S. S., Tomasi, J., Cossi, M., Millam, J. M., Klene, M., Adamo, C., Cammi, R., Ochterski, J. W., Martin, R. L., Morokuma, K., Farkas, O., Foresman, J. B. & Fox, D. J. (2016). Gaussian 16, Revision B.01. Gaussian, Inc., Wallingford, CT, USA. Google Scholar
Fuchs, J. (1997). Lipoic Acid in Health and Disease. New York: CRC Press. Google Scholar
García, C. F., McKervey, M. A. & Ye, T. (1996). Chem. Commun. pp. 1465–1466. Google Scholar
Gilbert, B., Frazer, B. H., Belz, A., Conrad, P. G., Nealson, K. H., Haskel, D., Lang, J. C., Srajer, G. & De Stasio, G. (2003). J. Phys. Chem. A, 107, 2839–2847. Web of Science CrossRef CAS Google Scholar
Hansen, J. & Davies, H. M. L. (2008). Coord. Chem. Rev. 252, 545–555. CrossRef CAS Google Scholar
Howard, R. A., Spring, T. G. & Bear, J. L. (1976). Cancer Res. 36, 4402–4405. CAS Google Scholar
Howard, R. A., Spring, T. G. & Bear, J. L. (1977). J. Clin. Hematol. Oncol. 7, 391–400. CAS Google Scholar
Ioannou, P. V. & Tsivgoulis, G. M. (2014). Monatsh. Chem. 145, 897–909. CrossRef CAS Google Scholar
Jakimowicz, P., Ostropolska, L. & Pruchnik, F. P. (2000). Met.-Based Drugs, 7, 201–209. CrossRef CAS Google Scholar
Jalilehvand, F., Enriquez Garcia, A. & Niksirat, P. (2017). Am. Chem. Soc. Omega, 2, 6174–6186. CAS Google Scholar
Johnson, S. A., Hunt, H. R. & Neumann, H. M. (1963). Inorg. Chem. 2, 960–962. CrossRef CAS Google Scholar
Kitchens, J. & Bear, J. L. (1969). J. Inorg. Nucl. Chem. 31, 2415–2421. CrossRef CAS Google Scholar
Kraatz, H. B., Jacobsen, H., Ziegler, T. & Boorman, P. M. (1993). Organometallics, 12, 76–80. CrossRef CAS Google Scholar
Kubicki, M. M., Oudet, P., Martin, C. & Barré, C. (1995). J. Chem. Soc. Dalton Trans. pp. 3699–3704. CrossRef Google Scholar
Mahboob, N., Miyashita, Y., Yamada, Y., Fujisawa, K. & Okamoto, K. (2004). Inorg. Chim. Acta, 357, 75–82. Web of Science CrossRef CAS Google Scholar
Miller, D. J. & Moody, C. J. (1995). Tetrahedron, 51, 10811–10843. CrossRef CAS Google Scholar
Nikolić, R. S., Krstić, N. S., Nikolić, G. M., Kocić, G. M., Cakić, M. D. & Anđelković, D. H. (2014). Polyhedron, 80, 223–227. Google Scholar
Norman, J. G. & Kolari, H. J. (1978). J. Am. Chem. Soc. 100, 791–799. CrossRef CAS Google Scholar
Norman, J. G., Renzoni, G. E. & Case, D. A. (1979). J. Am. Chem. Soc. 101, 5256–5267. CrossRef CAS Google Scholar
Orpen, A. G. & Connelly, N. G. (1990). Organometallics, 9, 1206–1210. CrossRef CAS Web of Science Google Scholar
Parthé, E., Hohnke, E. & Hulliger, F. (1967). Acta Cryst. 23, 832–840. CrossRef IUCr Journals Google Scholar
Pneumatikakis, G. & Psaroulis, P. (1980). Inorg. Chim. Acta, 46, 97–100. CrossRef CAS Google Scholar
Rajalingam, K., Hallmann, L., Strunskus, T., Bashir, A., Wöll, C. & Tuczek, F. (2010). Phys. Chem. Chem. Phys. 12, 4390–4399. CrossRef CAS Google Scholar
Ravel, B. (2001). J. Synchrotron Rad. 8, 314–316. Web of Science CrossRef CAS IUCr Journals Google Scholar
Rehr, J. J. & Albers, R. C. (2000). Rev. Mod. Phys. 72, 621–654. Web of Science CrossRef CAS Google Scholar
Ressler, T. (1998). J. Synchrotron Rad. 5, 118–122. Web of Science CrossRef CAS IUCr Journals Google Scholar
Shannon, R. D. (1976). Acta Cryst. A32, 751–767. CrossRef CAS IUCr Journals Web of Science Google Scholar
Sorasaenee, K., Galán-Mascarós, J. R. & Dunbar, K. R. (2002). Inorg. Chem. 41, 433–436. CrossRef CAS Google Scholar
Sorasaenee, K., Galán-Mascarós, J. R. & Dunbar, K. R. (2003). Inorg. Chem. 42, 661–663. CrossRef CAS Google Scholar
Sowa, T., Kawamura, T., Shida, T. & Yonezawa, T. (1983). Inorg. Chem. 22, 56–61. CrossRef CAS Google Scholar
Spuches, A. M., Kruszyna, H. G., Rich, A. M. & Wilcox, D. E. (2005). Inorg. Chem. 44, 2964–2972. CrossRef CAS Google Scholar
Tan, A. & Harris, S. (1998). Inorg. Chem. 37, 2215–2222. CrossRef CAS Google Scholar
Wilson, C. R. & Taube, H. (1975). Inorg. Chem. 14, 2276–2279. CrossRef CAS Google Scholar
Wong, D. L. & Stillman, M. J. (2016). Chem. Commun. 52, 5698–5701. CrossRef CAS Google Scholar
Wong, D. L., Zhang, A., Faponle, A. S., de Visser, S. P. & Stillman, M. J. (2017). Metallomics, 9, 501–516. CrossRef CAS Google Scholar
Yoshimura, T., Umakoshi, K. & Sasaki, Y. (2003). Inorg. Chem. 42, 7106–7115. CrossRef CAS Google Scholar
Zabinsky, S. I., Rehr, J. J., Ankudinov, A., Albers, R. C. & Eller, M. J. (1995). Phys. Rev. B, 52, 2995–3009. CrossRef CAS Web of Science Google Scholar
Zhang, C., Sun, X., Li, J. & Liu, Y.-N. (2013). Nanoscale, 5, 6261–6264. CrossRef CAS Google Scholar
Zhang, X., Ma, M. & Wang, J. (2003). ARKIVOC, (ii), 84–91. Google Scholar
© International Union of Crystallography. Prior permission is not required to reproduce short quotations, tables and figures from this article, provided the original authors and source are cited. For more information, click here.