Chemistry of early and late transition metallaboranes: synthesis and structural characterization of periodinated dimolybdaborane [(Cp*Mo)₂B₄H₃I₅]

Synthesis of tantalaboranes 2 and 3

While studying the reactivity of several metallaborane clusters of group 5–9 [29–37] with various organic and inorganic compounds, use of [Fe₂(CO)₉] and [Ru₃(CO)₁₂] turned out to be the most profitable procedures [34–40]. As a result, we performed the reaction of [(Cp*Ta)₂(BH₃)₂Cl₂], 1 with [Ru₃(CO)₁₂] that led to the isolation of [Cp*TaCl(μ-Cl)B₂H₄Ru₃(CO)₈], 2 and [(Cp*Ta)₂(B₂H₆)(B₂H₄Cl₂)], 3 in moderate yields (Scheme 1). Compound 2 is reasonably stable in air, thus, can be separated by thin-layer chromatography, allowing the characterization of pure materials. The ¹¹B NMR spectrum of 2 displays two distinct resonances at δ = 112.9 and 57.9 ppm. The ¹H NMR shows a single resonance for Cp* protons at δ = 1.90 ppm along with three upfield chemical shifts at δ = –7.63, –7.79, –16.61 ppm. The broad resonances at δ = –7.63 and –7.79 ppm have been assigned for Ru-H-B and Ta-H-B, respectively. The sharp peak at δ = –16.61 ppm stands for Ru-H-Ru. The IR spectrum shows intense bands at 2056, 2029 and 1953 cm^–1, characteristic of terminal carbonyl groups and a band at 2417 cm^–1 for the terminal B–H stretch.

Scheme 1

Synthesis of tantalaboranes 2 and 3.

The molecular structure of 2, shown in Fig. 1, is fully consistent with solution spectroscopic data. The structure can be defined as capped square pyramid where the square is composed Ta1-Ru1-Ru2-B2 and B1 is at the pyramidal position. Further Ru3 is capped on the B1-Ru1-Ru2 face. Another interesting feature of 2 is the bridged chlorine atom between one Ta and one Ru atom. The square pyramidal geometry of compound, 2 can be compared with tantalaborane [Cp*TaCl₂B₄H₈] [32]. The two clusters display similar B–B distances and the metal–apical boron distances whereas, the metal–basal boron distance (2.269 Å) is significantly shorter in 2 as compared to [Cp*TaCl₂B₄H₈] (2.413 Å). Similarly, the average Ru-B and Ru-Ru distance of 2.178 Å and 2.835 Å are well matched to those of analogues pileo-[2,3-(Cp*Ru)₂(μ-H)B₄H₇] [8]. Compound 2, to the best of our knowledge is the first structually known metallaborane having both early and late tranistion metal atoms in the capped squrare pyramidal arrangement [Chart 2 (I-V)] [41–45]. The bond length of bridged chloride ligand between Ta-Ru can be compared with reported Cl-bridged and unbridged metal clusters, [Ru₃(CO)₁₀(μ-Cl)(μ-AuPPh₃)], [(Cp*Rh)Ru₄H₂(μ-Cl)(CO)₁₂B] [45] and [Ru₃(μ-Cl)(μ-PPh₂)₃(CO)₇] [46]. Interestingly, all the Ru atoms in 2 have 18 electrons around the metal center, however, the Ta-centre possesses 16 electrons (considering two electrons from the bridged chlorine atom).

Fig. 1

Molecular structure of [Cp*TaCl(μ-Cl)B₂H₄Ru₃(CO)₈], 2 (carbonyl ligands on Ru atoms were omitted for clarity). Selected bond lengths (Å) and angles (°): Ta1-B1 2.270(7), Ta1-B2 2.268(7), Ta1-Cl1 2.4752(17), Ta1-Cl2 2.4047(17), B1-B2 1.714(10), Ru1-B1 2.112(7), Ru2-B1 2.234(6), Ru2-B2 2.327(7), Ru3-B1 2.190(7), Ru1-Ru2 2.9091(7), Ru1-Ta1 2.8151(6); Ta1-B1-B2 67.8(3), Ta1-B1-Ru1 79.9(2), Ru2-B1-Ta1 95.5(2), Ru3-B1-Ta1 159.4(4), Ru1-B1-Ru2 84.0(2).

Chart 2

pileo-square pyramidal metallaborane clusters.

Considering cluster 2 as fused cluster, the Mingos’s fusion formalism gave 66e [Ru₂Ta₁B₂ square pyramid (54) + Ru₃B tetrahedron (50) – Ru₂B triangle (38) = 66e]. On the other hand, the total electron count of cluster can generally be determined by adding the valence electrons contributed by the metal atoms and the electrons donated to the metals by the ligands [47, 48]. Thus, the total number of cluster valence electrons (cve) available for 2 is 64 [Cp*TaCl (11e) + μ-Cl (3e) + 2B (6e) + 4H (4e) + 3 Ru (24e) + 8CO (16e) = 64e], two electrons short to the capped square pyramidal geometry. It might be due to the fact that Cp*TaCl is not a “conical” fragment rather a ML₄ fragment with one frontier orbital less. So the cluster 2 is electronically unsaturated cluster and having both early and late transition metal atoms in the cluster core.

Cluster 3 was obtained as orange red solid in 20 % yield and characterized by spectroscopic and mass spectrometric data. In the mass spectrum, the highest envelope was observed at m/z 757, consistent with [(Cp*Ta)₂(B₂H₆)(B₂H₄Cl₂)]. The ¹¹B NMR displays two resonances at δ = 12.5 and –5.7 ppm in 1:1 ratio indicating a higher symmetric arrangements. The ¹H NMR spectrum shows resonances for the Cp* protons at δ = 2.42 ppm and broad peaks at δ = –9.02 and –10.41 ppm for Ta-H-B hydrogens along with B–H terminal signal at 3.98 ppm. Further in an attempt to find the position of hydrogen atoms in 3, 2D ¹¹B{¹H}/¹H{¹¹B} HSQC experiment was undertaken. The study shows that the boron at δ = –5.7 ppm coupled with both the protons at δ = –10.41 and 3.98 ppm; whereas the boron at δ = 12.5 ppm coupled with proton at –9.02 ppm. This suggests that the boron resonating at 12.5 ppm devoid of any terminal hydrogen. Further, to observe the fluxionality of 3, the variable-temperature ¹H{¹¹B} and ¹¹B{¹H} NMR was recorded. The result signifies the absence of any possibility of fluxionality in 3. The IR spectrum shows a band at 2403 cm^–1 due to the terminal B–H stretch.

Synthesis of periodinated [(Cp*Mo)₂B₄I₅H₃], 4

The high yield synthesis of dimolybdaboranes allowed us to synthesize their stable derivatives [49–53]. Earlier, we have described the synthesis of B–H substituted dimolybdaboranes [50] from the reaction of [Cp*MoCl₄] and [LiBH₄.thf], followed by the addition of [BHCl₂.SMe₂]. Further, extension of this chemistry towards dichalcogenide ligands [49], RE-ER [R = Ph, CH₂Ph, 2,6-(^tBu)₂-C₆H₂OH, (CH₃)₃C); E = S, Se] yielded variety of hybrid clusters. Recently we have reported the B-iodination of [(Cp*Mo)₂B₄H₄S₂] and [(Cp*Mo)₂B₅H₉] using NaI, which led to the isolation of iodinated dimolybdathiaborane derivatives [51]. However, all our attempts to synthesis periodinated [(Cp*Mo)₂(BI)₅H₄] or [(Cp*Mo)₂B₄I₄S₂] failed. Therefore, in a challenge to synthesize periodinated molybdaborane, we have renewed this chemistry in different reaction conditions and iodine sources. As a result, thermolysis of an in situ generated intermediate, produced from [Cp*MoCl₄] and [LiBH₄.thf], in presence of MeI afforded periodinated dimolybdatetraborane [(Cp*Mo)₂B₄H₃I₅], 4 (Scheme 2).

Scheme 2

Synthesis of [(Cp*Mo)₂B₄H₃I₅] 4.

The solid state structure of the compound 4, shown in Figure 2, was ascertained by X-ray structure analysis. The molecule has near perfect C_s symmetry and can be viewed as bicapped tetrahedron geometry. The Mo1-Mo_1-B2-B3 atoms are arranged in tetrahedra whereas the other two B-I moieties [B1(I1) and B4(I4)] are in capping position (Figure 2). The total cluster valence electrons for compound 4 is 44 which is isoelectronic to rhenaborane cluster [(Cp*ReH)₂B₄H₈] [54] and [(Cp*Cr)2(CO)2B4H6] [55]. The Mo-Mo distance of 2.729 Å in 4 is considerably longer than the analogues [(Cp*Mo)₂B₂S₂H₂(μ-η¹-S)] (2.635 Å) and [(Cp*Mo)₂B₂H₅(BSePh)₂(μ-η¹-SePh)] (2.694 Å) [49]. Although the average Mo-B distances in 4 and [(Cp*Mo)₂B₂H₅(BSePh)₂(μ-η¹-SePh)] are similar (2.274 Å, 2.275 Å respectively), the B-B distances of former is 0.5 Å longer than the later.

Fig. 2

Molecular structure of [(Cp*Mo)₂B₄H3I₅], 4. Selected bond lengths (Å) and angles (°): B1-B2 1.570(3), B2-B3 1.679(16), B3-B4 1.840(3), B1-I1 2.199(12), B3-I3 2.178(12), B1-Mo1 2.297(9), B2-Mo1 2.272(8), I5-Mo1 2.815(9); B2-B1-I1 132.5(12), B2-B1-Mo1 69.0(5), I1-B1-Mo1 140.7(4), B1-B2-B3 128.0(11), Mo1-Mo1-I5 61.00(11).

The ¹¹B NMR spectrum of 4 displays four distinct resonances at δ = 53.3, 36.1, 31.3, 13.9 ppm. The ¹H NMR shows two signals for Cp* protons at δ = 1.98 and 1.90 ppm along with three upfield chemical shifts at δ = –4.75, –6.01 and –8.39 ppm. Although, the bridging hydrogens could not be located in the solid state structure, but their presence has been ascertained by ¹H NMR spectroscopy.

Synthesis of brominated rhodaborane clusters

Although the metallaboranes offer reaction potentials similar to those of polyhedral boranes and transition metal clusters [56–60], the additional feature is the competition between the metal and boron sites. For example, borane displacement vs. metal fragment displacement, ligand substitution at boron vs. metal sites, removal of M-H-B vs B–H–B protons [61–63]. Nevertheless, the isolation of chlorinated tantalaborane 3 and the existence of perbrominated cobaltaborane [28], led us to extend this chemistry to Rh-system. As a result, we have performed the reaction of [(Cp*Rh)₃(B₄H₄)], 5 with PtBr₂, that yielded partial brominated rhodaboranes [(Cp*Rh)₃(BH)(BBr)₃], 6 and [(Cp*Rh)₃(BH)₃(BBr)], 7 (Scheme 3). Prolonged reaction time or drastic reaction conditions were unsuccessful for perbromination in 5.

Scheme 3

A) Synthesis of closo-[(Cp*Co)₂B₄H₂Br₄]; B) Synthesis of 6 and 7.

The mass spectrometric data of 6 suggests a molecular formula of C₃₀H₄₆B₄Br₃Rh₃. The ¹¹B{¹H} NMR reveals the presence of three resonances in 1 : 2 : 1 ratio whereas the coupled ¹¹B NMR shows different pattern. The resonances at δ = 89.3 and 62.6 ppm showed no coupling in ¹¹B NMR suggests the B–Br environment. The ¹H NMR shows one broad peak at 8.62 ppm for BH_t proton; two peaks for Cp* protons at 2 : 1 ratio. The IR spectrum shows a broad band at 2423 cm^–1 for the terminal B–H stretches. The molecular structure of compound 6 confirms the structural inferences made on the basis of spectroscopic results. The solid state structure of 6 is shown in Fig. 3, where this compound is seen to be a cluster analogue of 5 and [(Cp*Co)₃B₄H₄] [64]. The substitution at the terminal B–H changes the internal distances. The average Rh-Rh distance lengthens from 2.667 Å (in 5) to 2.691 Å (in 6) whereas; the distance of unique borylene atom (B4) to the triangular metal plane shortens form 1.388 Å (in 5) to 1.354 Å (in 6). The average B-Br bond distances of 1.97 Å in 6 are very close [(Cp*Co)₂(BBr)₄H₂] [28]. The geometry of cluster 7 is established by comparison of its spectroscopic data with those for capped octahedral cluster 5 and other related species [65, 66].

Fig. 3

Molecular structure of [(Cp*Rh)₃(BH)(BBr)₃], 6 (the Cp* ligands on the Rh are excluded for clarity). Selected bond lengths (Å) and angles (°): Rh1-Rh2 2.6914(11), Rh2-Rh3 2.697(11), Rh1-B2 2.092(14), B2-B3 1.71(2), B1-B2 1.69(2), Rh1-B4 2.049(13), B2-Br2 1.979(13), B1-Br3 1.963(14); Rh1-Rh2-Rh3 60.00(3), Rh1-B2-Rh2 79.9(5), B1-B2-B3 59.2(8).

General procedures and instrumentation

All the syntheses were carried out under an argon atmosphere with standard Schlenk and glovebox techniques. Solvents were dried by common methods and distilled under N₂ before use. Compound 1 [29], [(Cp*Rh)₃(BH)₄] [8] and [Cp*MoCl₄] [67] were prepared according to literature methods, while other chemicals [Cp*TaCl₄], [LiBH₄·thf], [Ru₃(CO)₁₂], [PtBr₂] were obtained commercially and used as received. The external reference for the ¹¹B NMR, [Bu₄N(B₃H₈)] was synthesized as per literature methods [68]. Thin layer chromatography was carried on 250 mm dia aluminum supported silica gel TLC plates (Merck TLC Plates). NMR spectra were recorded on a 400 and 500 MHz Bruker FT-NMR spectrometer. Residual solvent protons were used as reference (δ, ppm, d₆-benzene, 7.16) while a sealed tube containing [Bu₄N(B₃H₈)] in d₆-benzene (δ_B, ppm, –30.07) was used as an external reference for the ¹¹B NMR. The infrared spectra were recorded on a Nicolet iS10 spectrometer. MALDI-TOF mass-spectra of the compounds were obtained on a Bruker Ultraflextreme using 2,5-dihydroxybenzoic acid as a matrix and a ground steel target plate. Microanalyses for C and H were performed on Perkin Elmer Instruments series II model 2400.