Synthesis of Tetrahedranes Containing the Unique Bridging Hetero‐Dipnictogen Ligand EE′ (E ≠ E′=P, As, Sb, Bi)

Abstract

In order to improve and extend the rare class of tetrahedral mixed main group transition metal compounds, a new synthetic route for the complexes [{CpMo(CO)₂}₂(μ,η²:η²‐PE)] (E=As (1), Sb (2)) is described leading to higher yields and a decrease in reaction steps. Via this route, also the so far unknown heavier analogues containing AsSb (3 a), AsBi (4) and SbBi (5) ligands, respectively, are accessible. Single crystal X‐ray diffraction experiments and DFT calculations reveal that they represent very rare examples of compounds comprising covalent bonds between two different heavy pnictogen atoms, which show multiple bond character and are stabilised without any organic substituents. A simple one‐pot reaction of [CpMo(CO)₂]₂ with ME(SiMe₃)₂ (M=Li, K; E=P, As, Sb, Bi) and the subsequent addition of PCl₃, AsCl₃, SbCl₃ or BiCl₃, respectively, give the complexes 1–5. This synthesis is also transferable to the already known homo‐dipnictogen complexes [{CpMo(CO)₂}₂(μ,η²:η²‐E₂)] (E=P, As, Sb, Bi) resulting in higher yields comparable to those in the literature reported procedures and allows the introduction of the bulkier and better soluble Cp′ (Cp′=tert butylcyclopentadienyl) ligand.

New synthetic pathway towards organometallic tetrahedrane derivatives [{CpMo(CO)₂}₂(μ,η²:η²‐EE′)] (E≠E′=P, As, Sb, Bi) involving a hetero‐dipnictogen ligand is reported, leading to dramatic yield enhancements for already known compounds, the reduction of reaction steps and access to so far unknown AsSb, AsBi and SbBi ligand complexes, which feature unseen covalent bonds between two different heavy group 15 elements without organic substituents.

Introduction

Tetrahedral molecules such as the most simple organic representative, the parent tetrahedrane (tricyclo[1.1.0.0^2,4]butane), have always been of great scientific interest, not only due to their aesthetical attraction as a chemical equivalent of a platonic body, but also because of their unusual bonding situation, high ring strain and reactivity. ^[1] These properties point already to their mostly challenging syntheses and low stability. ^[1] The probably most prominent inorganic example of this class of compounds is white phosphorus (P₄), which can be prepared by sublimation of red phosphorus and is stable under exclusion of air. In contrast, its heavier counterpart, yellow arsenic (As₄), is highly unstable and undergoes rapid autocatalytic degradation under light exposure already. ^[2] Furthermore, the heavier homologue Sb₄ has been observed in the solid state only within thin antimony films received by the evaporation of Sb₄ molecules in ultra‐high vacuum, ^[3] while Bi₄ is only known in the gas phase. ^[4] Heteroatomic tetrahedranes built from p‐block elements are extremely scarce. Only three examples, namely AsP₃ (I) and the tetrahedrane derivatives P(C^tBu)₃ (II) and P₂(C^tBu)₂ (III), have been reported to date (Scheme 1b). ^[5] They are, however, highly sensitive to air. While I and III are pyrophoric, compound II degrades after 30 mins at room temperature.

Scheme 1.

a) Rare examples of compounds featuring covalent As−Bi and Sb−Bi bonds (left), as well as a covalent As−Sb bond, which is in equilibrium with its respective homo‐dipnictogen complexes (right); b) heteroatomic main‐group tetrahedranes. This work: tetrahedral [{CpMo(CO)₂}₂(μ,η²:η²‐EE′)] compounds containing scarce covalent hetero‐dipnictogen bonds.

By formal substitution of one or two atoms of the E₄ (E=P, As, Sb, Bi) tetrahedra with isolobal 15‐valence‐electron transition metal fragments, such as [CpMo(CO)₂], stable tetrahedra such as the air‐stable (in the solid state) polypnictogen ligand complexes [{CpMo(CO)₂}(η³‐E₃)] (“MoE₃”; E=P, As, Sb) ^[6] and [{CpMo(CO)₂}₂(μ,η²:η²‐E₂)] (“Mo₂E₂”; E=P, As, Sb, Bi)[ ^6a , ^6c , ⁷ ] are received (Scheme 1). They can easily be synthesised, characterised and handled and are therefore well‐suited starting materials for further investigations and have already proved to be excellent precursors for the formation of extended polypnictogen frameworks upon reduction, ^[8] coordination ^[9] and oxidation. ^[10] While for Mo₂E₂ complexes all representatives of E (E=P, As, Sb, Bi; except E=N) are known, the respective hetero‐dipnictogen complexes Mo₂EE′ have been far less investigated. Until now, only two compounds of this class, [{CpMo(CO)₂}₂(μ,η²:η²‐PE)] (E=As (1), Sb (2)), could be synthesised by Mays et al. in 1998 (Scheme 2). ^[11] Their approach based on the deprotonation of [{CpMo(CO)₂}₂(μ‐H)(μ‐PH₂)] (C), which yields the lithium salt of the anion [{CpMo(CO)₂}₂(μ‐PH₂)]⁻ (D). Further reaction with AsCl₃ or SbCl₃, respectively, gives access to 1 and 2 by elimination of one equivalent of LiCl and two equivalents of HCl. However, the latter can also protonate the anion D back to C, which leads to just moderate yields of 33 % (1) and 39 % (2), respectively. Another disadvantage of this synthetic route is the elaborate preparation of C, where first Mo₂P₂ (B) is generated by thermolysis of [CpMo(CO)₂]₂ (A) with white phosphorus to be further hydrolysed with NaOH in THF/H₂O. Following this route, C can only be obtained in a yield of just 15 %, which leads to an overall yield of 5.0 % for 1 and 5.9 % for 2, when starting from A. Moreover, a direct route affording C from A was developed by our group in 2003. ^[12] Additionally, only recently, we were able to show that D can be directly synthesised from B by a nucleophilic attack of OH⁻. ^[13] Mays et al. also tried to synthesise the respective PBi ligand complex by adding BiCl₃ to D, but failed, which is not overly surprising as complexes containing a P−Bi bond are very rare and only of a dative bonding nature and/or stabilised by bulky substituents. ^[14] The number of examples containing a covalent E−E′ bond dramatically decreases by moving on to the heavier pnictogen elements. ^[15] While three compounds for covalent As−Sb bonds were reported, ^[16] only two are known for As−Bi and Sb−Bi (Scheme 1a), ^[17] only half of which were characterised crystallographically. Between As and Sb, only single bonds were observed, whereas the other ones mentioned also form double bonds. However, all of them can only be stabilised by very bulky organic substituents or they have a distinct tendency to disproportionate and form homonuclear bonds (Scheme 1a; right). ^[16a]

Scheme 2.

Synthetic route for the tetrahedral molybdenum pnictogen complexes of the type {[Mo]₂(μ,η²:η²‐PE)} (E=P (B), As (1), Sb (2); [Mo]=CpMo(CO)₂) by Mays et al. in 1998. Yields for B, 1 and 2 are calculated for the overall reaction starting from A. This work: direct synthesis of D from A and avoidance of HCl elimination by replacing the protons with SiMe₃ groups.

These disadvantages and the perspective of making tetrahedral compounds with heavier hetero‐dipnictogen ligands accessible prompted us to develop a new synthetic route, which reduces the number of reaction steps and avoids the elimination of HCl to give much higher yields. In the following, we report a facile two‐step one‐pot synthesis for this type of compounds starting directly from A (Scheme 2), which is also transferable to the complexes Mo₂E₂, containing different Cp^R ligands such as the tert‐butyl substituted cyclopentadienyl ligand (Cp′=η⁵‐C₅H₄ tBu) and gives access to the so far unknown heavier hetero‐dipnictogen derivatives of 1 and 2 such as [{CpMo(CO)₂}₂(μ,η²:η²‐AsSb)] (3 a), [{CpMo(CO)₂}₂(μ,η²:η²‐AsBi)] (4) and [{CpMo(CO)₂}₂(μ,η²:η²‐SbBi)] (5) in higher yields. The latter represent the first examples of compounds containing covalent bonds between mixed heavier group 15 elements that do not bear any bulky organic substituents and are only stabilised by transition metal moieties.

Results and Discussion

When A is reacted with a solution of Li[P(SiMe₃)₂] (E1) in THF, an immediate colour change from orange‐brown to greenish red occurs suggesting the formation of the intermediate Li[{CpMo(CO)₂}₂{μ‐P(SiMe₃)₂}] (6 a) in solution (Scheme 3). Compound 6 a is a derivative of D in which the P‐bound hydrogen atoms are replaced by SiMe₃ groups. Further addition of AsCl₃ or SbCl₃, respectively, leads to the generation of the tetrahedral molybdenum pnictogen complexes [{CpMo(CO)₂}₂(μ,η²:η²‐PE′)] (E′=As (1), Sb (2)) in 69 % and 59 % isolated yields (Scheme 3), respectively, after chromatographic work‐up. In the last step, two equivalents of Me₃SiCl are released avoiding the disadvantageous elimination of HCl. This method not only avoids elaborate reaction steps, but also increases the yield of the compounds 1 and 2 dramatically, compared to the literature (5.0 % and 5.9 %; Table 1).

Scheme 3.

One‐pot synthesis of the tetrahedral hetero‐pnictogen complexes 1–5. Top: using NaPH₂ for the synthesis of 1 leading to HCl evolution in the second reaction step; bottom: using ME(SiMe₃)₂ (M=Li, K; E=P, As, Sb, Bi) for the synthesis of 1–5, which avoids HCl elimination resulting in higher yields.

Table 1.

All possible combinations of the intermediates 6 a–d with E′Cl₃ (E′=P, As, Sb, Bi) and the resulting products [{CpMo(CO)₂}₂(μ,η²:η²‐E E′)]. Yield improvements compared to the literature syntheses are given subjacent (referred to A).

	6 a (“P”)	6 b (“As”)	6 c (“Sb”)	6 d (“Bi”)
PCl3	P2 20 %→57 %	P As (1) 5 %→50 %	P Sb (2) 6 %→18 %	unsuccessful
AsCl3	P As (1) 5 %→69 %	As2 35 %→52 %	As Sb (3 a) 0 %→63 %	unsuccessful
SbCl3	P Sb (2) 6 %→59 %	As Sb (3 a) 0 %→51 %	Sb2 76 %→37 %	unsuccessful
BiCl3	unsuccessful	As Bi (4) 0 %→58 %	Sb Bi (5) 0 %→46 %	Bi2 14 %→25 %

The purity of 1 and 2 was proven by ³¹P and ¹H NMR spectroscopy as well as by elemental analysis and mass spectrometry. It has to be mentioned that compound 1 can also be synthesised by reacting A with NaPH₂ (yielding the anion D immediately in solution) and the subsequent addition of AsCl₃. But, here again, HCl evolves in the second reaction step leading to side reactions and a relatively low yield of 16 %, which, however, still exceeds three times the yield reported by Mays et al.

Compound Mo₂P₂ (B) can be obtained in the same manner by reacting the solution of the intermediate 6 a with PCl₃. Here, the yield can be increased from 20 % to 57 % (Table 1). However, the synthesis of an analogous complex with a PBi ligand by adding BiCl₃ to 6 a was not successful, as as it was also reported by themethod of Mays et al. Only small traces of Mo₂P₂ could be detected by NMR spectroscopy and mass spectrometry, which might suggest that the P−Bi bond is too unstable and the tetrahedral product is either not formed or decomposes rapidly.

Our synthetic route clearly shows that the number of reaction steps could be reduced and that the yield is considerably increased by an easy one‐pot reaction, especially for the hetero‐dipnictogen complexes 1 and 2. But the more challenging question arises whether the yet unknown hetero‐dipnictogen ligands, which only contain the heavier elements As, Sb and Bi, are also accessible. The reaction of A with either Li[As(SiMe₃)₂] (E2), K[Sb(SiMe₃)₂] (E3) or K[Bi(SiMe₃)₂] (E4) in THF leads to an immediate colour change from orange‐brown to greenish red (E2), greenish brown (E3) or bronze‐coloured (E4) suggesting the formation of the respective intermediates M[{CpMo(CO)₂}₂{μ‐E(SiMe₃)₂}] (M=Li, K; E=As (6 b), Sb (6 c), Bi (6 d)) in solution. The subsequent addition of AsCl₃, SbCl₃ or BiCl₃, respectively, and easy purification by column chromatography lead to the unprecedented complexes [{CpMo(CO)₂}₂(μ,η²:η²‐AsSb)] (3 a), [{CpMo(CO)₂}₂(μ,η²:η²‐AsBi)] (4) and [{CpMo(CO)₂}₂(μ,η²:η²‐SbBi)] (5) in remarkable isolated yields of 63 %, 58 % and 46 % (Scheme 3, Table 1).

Thus, all combinations of dipnictogen ligands can be synthesised by reacting the respective intermediate 6 a‐d with the appropriate pnictogen‐trihalide ECl₃ (except Mo₂PBi). While compound 3 a can be obtained in two ways, either by combining 6 b with SbCl₃ or 6 c with AsCl₃, 4 and 5 are only formed by the reaction of 6 b or 6 c, respectively, with BiCl₃, not by reacting 6 d with AsCl₃ or SbCl₃ (Table 1 or Scheme 3). This suggests that either the formation of 6 d is less favoured or that 6 d is less stable than its lighter counterparts.

The incorporated EE′ ligands feature very rare bonds between different heavier pnictogen atoms, especially within the compounds 3 a, 4 and 5. Compound 3 a represents the fourth, 4 and 5 only the third example of compounds with covalent As−Sb, ^[16] As‐Bi ^[17a] or Sb‐Bi[ ^17b , ^17c ] bonds, respectively. Moreover, they are the first examples of such covalent E−E′ bonds that are only stabilised by transition metals and do not bear any organic substituents. Therefore, these compounds can be regarded as complexes of the exotic diatomic As≡Sb, As≡Bi and Sb≡Bi molecules, respectively, which are the heaviest hetero‐pnictogen congeners of N₂.

Analogously to the synthesis of Mo₂P₂, the homo‐dipnictogen complexes Mo₂As₂, Mo₂Sb₂ and Mo₂Bi₂ can be synthesised via this method too, again resulting in a remarkable yield enhancement compared to their hitherto existing preparations (except for the antimony complex; Table 1). Additionally, this synthesis could be scaled‐up to a multigram scale, which enables the systematic investigation of the reactivity of this class of compounds. All the tetrahedral complexes are well soluble in dichloromethane and toluene and moderately soluble in ortho‐difluorobenzene and nonpolar solvents such as n‐hexane or n‐pentane. Overall, the heavier the pnictogen atoms are, the more the solubility decreases.

The products 1–5 can be crystallised from saturated CH₂Cl₂ solutions at −30 °C leading to dark red blocks suitable for single crystal X‐ray diffraction. Since 1 and 2 are already described in the literature, only the solid‐state structures of 3 a‐5 (Figure 1) will be discussed. They are isostructural with a distorted Mo₂EE′ tetrahedron as the central structural motif and crystallise in the monoclinic space groups P2/n (3 a), P2₁ (4) or I2/a (5), respectively, with two half molecules (3 a), one (4) or one half molecule (5) in the asymmetric unit. The AsSb ligand of 3 a exhibits a 50 : 50 disorder over the two sites with an average bond length of 2.515(1) Å, ^[18] which is in between those of the diarsenic (2.311(3) Å) ^[7a] and the diantimony complexes (2.678(1) Å). ^[19] Additionally, it is just slightly longer than the sum of the covalent radii for an As−Sb double bond (2.47 Å). ^[20] Therefore, it shows a structural similarity to all other existing complexes of the type [{CpMo(CO)₂}₂(μ,η²:η²‐EE′)]. It represents, however, the first example of an As−Sb bond with a multiple bond character. The same accounts for the compounds 4 and 5. The former bears an AsBi ligand with a bond length of 2.64(2) Å, ^[18] the latter an SbBi ligand with a bond length of 2.7916(2) Å. Both are again disordered over the two sites in a ratio of 50 : 50 and the bond distances are in between the sum of the respective single (As−Bi: 2.72 Å; Sb−Bi: 2.91 Å) and double bond (As=Bi: 2.55 Å; Sb=Bi: 2.77 Å) covalent radii, while 5 nears a double bond. ^[20] Additionally, the bond length of the SbBi ligand in 5 is again between those of the diantimony (2.678(1) Å) ^[19] and the dibismuth (2.838(1) Å) complexes. The same behaviour is observed for the Mo−Mo distances of 3 a‐5, which are slightly elongated in comparison to the respective lighter homo‐dipnictogen complex Mo₂E₂, but shortened in comparison to the heavier analogue. The Cp ligands in 3 a‐5 are arranged in an eclipsed manner to each other. Overall, the newly formed complexes 3 a‐5 can, on the one hand, be described as tetrahedra, isolobal to P₄ and As₄ and, on the other hand, as EE′ dumbbells stabilised by an Mo₂ unit, each featuring a formal triple bond. The rather short E−E′ distances are nicely reproduced by DFT calculations (c.f. Supporting Information). For example, the As−Sb distance in the optimized geometry of 3 a is with 2.515 Å in excellent agreement with the experimental value (vide supra). Moreover, the As−Sb multiple bond character is reflected in the Wiberg bond order with a Löwdin orthogonalized basis (see Supporting Information for details) of 1.40. The corresponding Wiberg bond orders of the As−Bi and Sb−Bi bonds in 4 and 5 are 1.35 and 1.36, respectively.

Figure 1.

Molecular structure of the unprecedented products 3 a (left), 4 (middle) and 5 (right). Anisotropic displacement is set to the 50 % probability level. The Cp ligands as well as the CO substituents are drawn translucent and the disorder is omitted for clarity. Selected bond lengths [Å]: 3 a: As−Sb 2.515(1), Mo1‐Mo1′ 3.0656(6); 4: As−Bi 2.64(2), Mo1‐Mo2 3.084(2); 5: Sb−Bi 2.7916(2), Mo1‐Mo1′ 3.1255(6).

The ¹H NMR spectra of 1–5 all feature one singlet in the characteristic region for Cp ligands. ^[21] Likewise, one singlet is observed in the ¹³C{¹H} NMR spectra as well as characteristic signals for the CO ligands. The phosphorus‐containing complexes 1 and 2 were characterised by ³¹P NMR spectroscopy, revealing the expected signals at δ=34.5 ppm (1 in C₆D₆) ^[22] and 98.8 ppm (2 in C₆D₆) ^[22] as reported in literature. ^[11] However, in some cases, both solutions show an additional small signal at −44.5 ppm, which can be attributed to trace impurity of Mo₂P₂. This indicates that a slight intermolecular exchange between the pnictogen atoms can take place in these syntheses and, therefore, small amounts of the homo‐dipnictogen complexes are generated which cannot be separated by column chromatography. The amount of the produced Mo₂P₂ is very little (ratio of 20 : 1 or higher in favour of 1 or 2, respectively) though. Small amounts of homo‐dipnictogen complexes can also be observed in the ¹H NMR spectra of 3 a–5 in some reactions. The identity of 1–5 has been unambiguously proven by mass spectrometry, where the molecular ion peak has been detected. In addition, in some cases, peaks of low intensity corresponding to the homo‐dipnictogen species have been also detected. The composition of 1–5 is further supported by the elemental analysis. All the synthesised tetrahedral compounds, including 3 b, 7 and 8 (vide infra), are stable in air, in contrast to other heteroatomic tetrahedranes.[ ^5a , ^5b , ^5c ]

It was also possible to isolate crystals of the intermediates 6 a–c (regardless of several attempts, no crystals for 6 d could be obtained), which were suitable for single‐crystal X‐ray diffraction by adding the respective crown ethers (12‐crown‐4 for lithium, 18‐crown‐6 for potassium) within the first reaction step and layering with n‐hexane at −30 °C (Figure 2). The anions of the intermediates 6 a–c are isostructural to each other as well as to D, which is a derivative of 6 a where the SiMe₃ groups are substituted by protons. They crystallise in the space groups P2/n (6 a), P2₁/n (6 b) and P ⁻1(6 c), respectively. In the newly formed anions, the pnictogenido [E(SiMe₃)₂]⁻ units bridge the Mo−Mo bond with E−Mo distances in the range between a single and a double bond (P−Mo: 2.4304(6) Å; As−Mo: 2.5170(3) Å; Sb−Mo: 2.6897(7) Å). The pnictogen atom features a distorted tetrahedral geometry. The Mo−Mo distances (6 a: 3.1890(5) Å; 6 b: 3.2445(3) Å; 6 c: 3.2598(8) Å) are dramatically increased compared to the starting material A (2.4477(12) Å) ^[23] reasoning that the original triple bond is completely degraded and only an elongated single bond between the molybdenum atoms is left. Additionally, the Mo−Mo distances increase from the lighter to the heavier pnictogen elements as expected. Compared to its derivative D, compound 6 a shows a similar Mo−Mo distance, but slightly elongated Mo−P bonds (D: 2.375(2)‐2.378(2) Å). This is caused by the sterically more demanding SiMe₃ groups in contrast to the P‐bound hydrogen atoms of D. A similar derivative of the arsenic compound 6 b is not known yet, but a related structural motif can be found in the neutral compound [{CpMo(CO)₂}₂(μ‐H)(μ‐AsH₂)] with a bridging [AsH₂]⁻ unit instead of [As(SiMe₃)₂]⁻. Interestingly, 6 c is the first crystallographically characterized compound with a stibenido unit bridging a dimolybdenum fragment. And, although the solid‐state structure of the bismuth derivative 6 d could not be determined, it probably reveals the same constitution, which is unprecedented for bismuth.

Figure 2.

Molecular structures of the intermediates 6 a (left), 6 b (middle) and 6 c (right). Anisotropic displacement is set to the 50 % probability level. Only the anionic part is shown. The cations and the crown ethers are omitted, and Cp as well as CO substituents are drawn translucent for clarity. Selected bond lengths [Å] and angles [°]: 6 a: P−Mo1/P−Mo1′ 2.4304(6), P−Si1/P−Si1′ 2.2574(8), Mo1‐Mo1′ 3.1889(5), Si1‐P−Si1′ 103.85(5); 6 b: As−Mo1 2.5213(3), As−Mo2 2.5128(3), As−Si1 2.3410(8), As−Si2 2.3540(8), Mo1‐Mo2 3.2445(3), Si1‐As−Si2 104.39(3); 6 c: Sb−Mo1 2.6887(8), Sb−Mo2 2.6907(7), Sb−Si1 2.561(2), Sb−Si2 2.559(2), Mo1‐Mo2 3.2597(8), Si1‐Sb−Si2 105.27(7).

The electronic structures of 6 a–d have been investigated by DFT calculations, which are in excellent agreement with the experimentally determined geometric parameters, especially the elongated Mo−Mo distances. Although the Mo−Mo distances in 6 a–d are longer than in the corresponding Mo₂EE’ derivatives 1–5, the Wiberg bond orders in Löwdin orthogonalized basis are only slightly lower (for example: Mo−Mo 3.199 Å, WBI 0.45 in 6 a vs. Mo−Mo 3.048 Å, WBI 0.53 in 1), indicating the presence of an elongated Mo−Mo single bond. This is also substantiated by the Intrinsic Bonding Orbitals, which show the presence of a Mo−Mo bond, although with additional orbital contribution from the CO groups (see Fig. S34 in the Supporting Information).

Besides making the novel hetero‐dipnictogen complexes 3 a–5 accessible, the synthetic strategy reported herein also allows to introduce substituted Cp ligands, such as tert‐butylcyclopentadienyl (Cp′), to vary the solubility and the electronic as well as the steric properties of the complexes. This might be crucial for further reactivity studies. The following complexes were synthesised by using [Cp′Mo(CO)₂]₂ as starting material exemplifying [{Cp′Mo(CO)₂}₂(μ,η²:η²‐As₂)] (7), [{Cp′Mo(CO)₂}₂(μ,η²:η²‐Sb₂)] (8) and [{Cp′Mo(CO)₂}₂(μ,η²:η²‐AsSb)] (3 b).

The complexes 7 and 8 had already been synthesised by our group in the past, ^[8b] even though with a remarkably low yield hampering the investigation of further reactivity. Nonetheless, they had already proved to be suitable starting materials for building different anionic pnictogen frameworks upon reduction. ^[8b] This shows the necessity of making these compounds available in a larger scale, which can be achieved via the new synthetic route. That way, the yields of the complexes 7 (5.4 %→55 %) and 8 (6.0 %→29 %) can be dramatically increased and also the unprecedented compound 3 b can be obtained in a remarkably good yield of 49 %.

Interestingly, the ¹H NMR spectrum of 3 b (in contrast to the spectra of 7 and 8) shows three multiplets in a ratio of 1 : 2 : 1 for the aromatic protons instead of the expected two triplets, which indicates that the protons in 2,5 and 3,4 position, respectively, are not chemical or magnetical equivalent in 3 b. This is confirmed by ¹³C{¹H} NMR spectroscopy, where four singlets instead of two for the proton bound C atoms of the Cp′ ring are observed.

Furthermore, orange red crystals suitable for X‐ray diffraction of the complexes 3 b, 7 and 8 could be obtained by cooling saturated CH₂Cl₂ solutions from room temperature to −30 °C. Their solid‐state structures (cf. Figure 3 for 3 b) show tetrahedral Mo₂ EE′ cores, which are almost identical to their Cp congeners.

Figure 3.

Molecular structure of the Cp′ derivative 3 b, exemplifying the isostructural compounds 3 b, 7 and 8. Anisotropic displacement is set to the 50 % probability level. The Cp′ ligands as well as the CO substituents are drawn translucent and the disorder is omitted for clarity.

Conclusions

We developed a new and easy one‐pot synthesis of air‐stable tetrahedral dimolybdenum dipnictogen complexes, which enables not only an impressive yield improvement (see Table 1) for the already known hetero‐dipnictogen complexes [{CpMo(CO)₂}₂(μ,η²:η²‐PE)] (E=As (1), Sb (2)) and the homo‐dipnictogen complexes Mo₂E₂ (E=P, As, (Sb), Bi), but, more importantly, also gives access to unprecedented complexes containing the heavier hetero‐dipnictogen ligands AsSb (3), AsBi (4) and SbBi (5). Now, these syntheses can also be carried out in a multigram scale. The complexes 3–5 extend the very rare class of E_n ligand complexes of the heavy pnictogen elements and the exotic tetrahedrane analogues, with all of them featuring very rare covalent bonds between two different heavy pnictogen atoms as well as representing the first ever examples in which these bonds can be stabilised without any organic substituents. Therefore, these compounds can be understood as the complexes of the exotic diatomic As≡Sb, As≡Bi and Sb≡Bi molecules with reduced E−E′ bond order, which are the heavy hetero‐pnictogen congeners of N₂. Within that, compound 3 contains the very first As−Sb bond with a multiple bond character.

The intermediates M[{CpMo(CO)₂}₂{μ‐E(SiMe₃)₂}] (M=Li, K; E=P (6 a), As (6 b), Sb (6 c)) could also be crystallographically characterized revealing anionic [E(SiMe₃)₂]⁻ units bridging a dimolybdenum fragment. The salt 6 c contains the first crystallographically characterized single stibenido unit of this kind.

Moreover, the new synthesis allowed to introduce the bulkier tert‐butylcyclopentadienyl (Cp′) ligand to vary the steric and electronic properties and enhance the solubility, which expands the possibilities of further reactivity studies. Overall, the newly synthesised E_n ligand complexes in large scale can be used as promising precursors for the synthesis of extended E_n ligand frameworks upon oxidation or reduction. Furthermore, due to the lack of Sb_n and Bi_n ligand complexes, the almost unexplored coordination behaviour of these compounds towards coinage and other transition metals can now be investigated.

Experimental Section

For experimental details see the Supporting Information.

Deposition numbers 2061901 (for 2), 2061902 (3 a), 2061903 (3 b), 2061904 (4), 2061905 (5), 2061906 (6 a), 2061907 (6 b), 2061908 (6 c), 2061909 (7), 2061910 (8), and 2072587 ([Cp′Mo(CO)₂(η³‐As₃)] contain(s) the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service.

Conflict of interest

The authors declare no conflicts of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supplementary

L. Dütsch, C. Riesinger, G. Balázs, M. Scheer, Chem. Eur. J. 2021, 27, 8804.