Substituent effects on o-carborane reactivity with cyclic (alkyl)(amino)carbenes

Abstract

The functionalization of carboranes, particularly at boron vertices, is a topic of extensive research. Here, we report the reactivity of ortho-carboranes (oCbs) with cyclic (alkyl)(amino)carbenes (CAACs) – a transformation that, surprisingly, has not been previously reported. We found that the substitution at the carbon atoms of an oCb strongly influences the reaction outcome. An unsubstituted oCb (H at carbon) reacts with the CAAC via carbene insertion into the C–H bond. However, when the carbon atoms of the oCb are substituted, the reaction is dictated by the electronic nature of these substituents: electron-withdrawing groups (EWGs) promote the formation of nido-type products via the nucleophilic attack by the CAAC at the B3 (or B6) position, whereas electron-donating groups (EDGs) favor insertion of the CAAC into the B–H bonds at these positions. The explanation of this reactivity pattern, along with the proposed mechanism of these processes, is supported experimentally and by density functional theory (DFT) computations.

---

Introduction

Carboranes have been known since the 1950s,¹ and have since found applications in various fields. For example, they enhance the heat resistance and hydrophobicity of polymers,² play a role in boron neutron capture therapy in medicine,^2,3 are used for electrochemical energy storage,^4,5 serve as ligand building blocks in catalyst design,^6–8 aid in metal recovery from radioactive waste,^9–11 and function as weakly coordinating anions.^12,13

ortho-Carboranes (oCbs) have gained significant attention in recent decades due to their unique features, such as non-classical carbon–boron and boron–boron bonds that are stabilized by σ-electron delocalization in the cluster. This electron delocalization is referred to as 3D aromaticity, in analogy to the 2D aromaticity of benzene rings, and similarly contributes to the stabilization of the cluster.^14,15

The two carbon atoms in oCbs alter the electronic properties of both the endo- and exoskeleton frameworks.^16,17 The functionalization of exoskeleton C–H and B–H bonds is key to the modification and use of this cluster in synthetic chemistry.^18,19 The C–H units in oCbs are acidic and susceptible to deprotonation by Brønsted bases such as ⁿBuLi, KHMDS, etc.^20–23 Once deprotonated, the oCbs can be further modified by installing various functional groups.^24–27,28 In contrast, the high s-character of the sp hybridized B atoms at the vertices increases their effective electronegativity, making the exoskeleton B–H bonds nearly nonpolar and therefore only weakly hydritic.^25,29,30 Consequently, functionalizing the B atoms in oCbs is more challenging.

Earlier work on oCb functionalization showed that the direct functionalization of cage B–H vertices in oCbs could be achieved via electrophilic substitution with various electrophiles under forced reaction conditions.^31–34 This approach was typically limited to substitution at the B(9,12)–H bonds, as these positions are the most electron-rich in oCbs.³⁵ The direct regioselective functionalization of other B–H vertices became possible with transition metal-catalyzed cross-coupling reactions, primarily employing noble metals such as Pd, Ir, Rh, and Ru, often facilitated by a directing group installed on oCbs.^36–38 Xie reported numerous examples of this chemistry over the years, introducing highly selective and efficient methods for B-functionalization in oCbs.^32,36–43 More recently, Xie has reported a series of impressive transition-metal-free strategies for direct B–H functionalization using Grignard reagents⁴⁴ and magnesium bisamides.⁴⁵

We became interested in the chemistry of oCbs with carbenes, as carbenes have been shown to react with both C–H and B–H vertices. In 1983, Jones Jr. reported the reaction between an oCb and carboethoxycarbene, an unstable singlet carbene formed in situ from the corresponding diazo compound,⁴⁶ which yielded a mixture of four isomers resulting from carbene insertion into B–H bonds.^47,48 Notably, Lee recently reported an impressive Rh(II)-catalyzed site- and enantioselective B–H insertion reaction using singlet carbenes.⁴⁹ With the development of stable N-heterocyclic carbenes (NHCs), their reactivity with oCbs was also studied. Smaller NHCs primarily acted as Brønsted bases, deprotonating the C–H bonds in oCbs,⁵⁰ while bulkier NHCs functioned as Lewis bases, attacking the most electrophilic boron centre (B3 or B6) and opening the cluster to a nido-type structure.^51,52

Surprisingly, the reactivity of cyclic (alkyl)(amino)carbenes (CAACs), which are both stronger σ-donors and better π-acceptors, with oCbs has not been reported.⁵³ We envisioned that the ambiphilicity of CAACs could lead to distinct reactivity patterns with oCbs, and decided to take on this task to fill this gap in the reactivity of carbenes and oCbs. In this paper, we present the reactivity of ^Me2CAAC ⁵⁴ with oCbs bearing different substituents at the carbon atoms (H, electron-withdrawing groups (EWGs), and electron-donating groups (EDGs)), demonstrating that these substituents play a crucial role in determining the reaction outcome. The mechanism, supported by both experiments and DFT calculations, explaining these distinct reactivities is also presented.

Results and discussion

First, the reaction between ^Me2CAAC and an equimolar amount of oCb was carried out in benzene at room temperature. The ¹³C-NMR spectrum recorded after 12 h showed the disappearance of the carbene carbon signal (δ = 312 ppm), indicating complete consumption of ^Me2CAAC. Analysis of the ¹H, ¹³C, and ¹¹B NMR spectra revealed that the reaction process was neither a mere deprotonation, a nucleophilic attack on the B atoms, nor an insertion into the B–H bond of the oCb, as previously reported for carbene reactions.^{46–48,50–52} Instead, the NMR data suggested a carbene insertion into the C–H bond, yielding product 1 (Fig. 1), a reactivity that, to the best of our knowledge, has not been previously reported. Evaporation of volatiles under vacuum afforded a yellowish precipitate, which was washed with pentane. Crystallization from benzene yielded transparent crystals, and single-crystal X-ray diffraction (SC-XRD) confirmed the molecular structure as insertion product 1 (Fig. 2).

Fig. 1 ^Me2CAAC insertion into the C–H bond of oCb.

Fig. 2 POV-ray depiction of 1. Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

We propose that the formation of 1 starts with the deprotonation of the oCb by ^Me2CAAC, which acts as a Brønsted base. This deprotonation generates an intermediate salt consisting of the cyclic iminium cation [^Me2CAAC-H]⁺ and the carboranyl anion oCb⁻ (Fig. 1). This assumption is supported by previous reports on NHC reactions with oCbs, where NHCs similarly acted as Brønsted bases.⁵⁰ However, unlike imidazolium cations, [NHC-H]⁺, which are poor electrophiles, the cyclic iminium [^Me2CAAC-H]⁺ is a significantly better electrophile. As a result, it undergoes nucleophilic attack by oCb⁻, leading to the formation of the insertion product 1 (Fig. 1).

As mentioned above, unstable singlet carbenes can react with oCbs to yield the B–H insertion products,^47,48 while NHCs react with oCbs, attacking the B3 (or B6) atom, leading to nido-type products.^51,52 This suggests that ^Me2CAAC might exhibit similar reactivity. However, achieving these transformations would require replacing the hydrogen atoms on the carbon atoms of oCbs with non-reactive “protecting” groups.

Thus, to force the reactivity with the B–H vertices, first dimethyl carborane (^Me2oCb) was synthesized⁵⁵ and reacted with ^Me2CAAC (1 : 1) in benzene at rt (Fig. 3). After 12 h, NMR spectra were recorded, showing the disappearance of the carbene carbon signal in the ¹³C NMR spectrum (δ = 312 ppm), and the appearance of new signals at δ = 3.07 ppm in the ¹H NMR spectrum and at δ = 63.17 and 62.43 ppm in the ¹³C NMR spectrum, which were attributed to the N–C–H fragment of the pyrrolidine type moiety that originated from the ^Me2CAAC, which was further confirmed by heteronuclear single quantum coherence (HSQC) NMR. Finally, the obtained product was crystallized from benzene, and its molecular structure was determined by SC-XRD (Fig. 4) as the product of the double insertion of ^Me2CAAC into the B–H bonds at the B3 and B6 positions (2^clscls) (Fig. 3).

Fig. 3 ^Me2CAAC insertion into the B–H bonds of ^Me2oCb.

Fig. 4 POV-ray depiction of 2^clscls. Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

It is important to note that even when an excess of ^Me2oCb (3 : 1) was used to prevent double insertion and favour the 1 : 1 reaction, the outcome remained the same, yielding product 2^clscls. Interestingly, unlike the previously reported reaction between oCb and bulky NHCs, where the carbene attacked the B-centre leading to a nido cluster,^{52 Me2}CAAC reacted with ^Me2oCb similarly to unstable carbenes, i.e., via B–H insertion,^47,48 but in a more selective manner.

We assumed that replacing the two methyl groups in ^Me2oCb with bulkier substituents at the carbon atoms of the oCb cluster could prevent double insertion and yield a 1 : 1 product. To test this hypothesis, ^Ph2oCb was prepared⁵⁶ and reacted with ^Me2CAAC, leading after 12 h to a 1 : 1 product resulting from ^Me2CAAC attack at the B3 atom of ^Ph2oCb, forming nido carborane 3^nido (Fig. 5) rather than the expected B–H insertion product. 3^nido was crystallized from CH₂Cl₂ as colorless crystals, and its structure was determined by SC-XRD (Fig. 6).

Fig. 5 Reaction of ^Me2CAAC and ^Ph2oCb – formation of the nido-type product 3^nido.

Fig. 6 POV-ray depiction of 3^nido. Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

Notably, the reaction of ^Ph2oCb with ^Me2CAAC, which leads to the nido product 3^nido (Fig. 5), resembles the previously reported reaction of bulky NHCs with oCbs.⁵¹ However, it differs significantly from the reaction of ^Me2CAAC with ^Me2oCb (Fig. 3). This contrast highlights the strong dependence of these reactions on the steric and electronic properties of the substituents at the carbon centres of the oCb. The Me groups are σ-donating, while the Ph groups are π-donating and σ-withdrawing.

We propose that the B–H insertion product 2^cls is formed via nido-type intermediate 2^nido (Fig. 7), structurally similar to 3^nido, which is formed as a result of an attack of ^Me2CAAC at the most electrophilic B3 (or B6) atom of the ^Me2oCb. A subsequent hydride migration from the B–H moiety in 2^nido to the cyclic iminium unit closes the cluster, yielding 2^cls. A similar reaction sequence with the second molecule of ^Me2CAAC ultimately leads to the formation of product 2^clscls (Fig. 7).

Fig. 7 Proposed pathway for B–H insertion and the influence of the substituents (EWG vs. EDG) on the reaction outcome.

It is noteworthy that while the attack of NHCs on the B-centres of oCbs, leading to nido carboranes, is well-documented, the subsequent hydride migration and cluster closure have not been reported and are actually rather rare phenomena in carboranes in general.^29,30,57,58 This difference likely stems from the distinct electronic properties of CAACs and NHCs. CAACs are more ambiphilic than NHCs, and after the formation of 2^nido, the iminium moiety becomes highly electrophilic, making it capable of accepting a hydride from the adjacent B–H unit (Fig. 7). Notably, an alternative mechanism similar to the one proposed for C–H insertion (Fig. 1) can be ruled out, as the hydrogen atoms at the B-centres of carborane are not acidic.

An obvious question arises: why does the reaction with ^Ph2oCb stop at the nido-type compound 3^nido and not proceed through hydride migration to the closo form? To explain this, we propose that hydride migration from the B-centre to the iminium moiety can be viewed as a nucleophilic substitution-like process. The more electron-rich the nido cage, the more favorable this nucleophilic substitution becomes, leading to B–H insertion products (Fig. 7). In contrast, if the nido cage is less electron-rich, hydride migration slows down or is entirely inhibited. In other words, when carborane is functionalized with an electron-donating group (EDG), such as methyl, the nido cage becomes more nucleophilic, facilitating hydride migration and cluster closure, ultimately yielding the B–H insertion products. However, if the substituents at the oCb are electron-withdrawing groups (EWGs), such as phenyl, the nido cage is less electron-rich and lacks sufficient nucleophilicity to promote hydride migration and cluster closure, thus halting the reaction at the nido stage (Fig. 7).

Additionally, steric factors may also influence the conversion from the nido to the closo form. If the structural reorganization required for hydride migration leads to significant steric repulsion between substituents at the carbon centres of the oCb, the closure of the cluster may be kinetically hindered or entirely suppressed.

To support this hypothesis, DFT computations of the hydride migration from the boron to a carbon atom, causing the transformation from the nido-type carboranes 2^nido and 3^nido to closo-type carboranes 2^cls and 3^cls (not formed in the reaction), respectively, were performed at the BP86-D3/def2-TZVP level of theory.^59–61 These computational results revealed that the hydride migration in 2^nido leading to 2^cls is a significantly exothermic (ΔH = −12.4 kcal mol⁻¹) and exergonic (ΔG = −9.7 kcal mol⁻¹) process, proceeding through a reasonable free energy barrier (ΔG^‡ = 21.8 kcal mol⁻¹) (Fig. 8, blue). In contrast, the same process for 3^nido leading to 3^cls, which was not observed experimentally, was calculated to be enthalpy and free energy neutral (ΔH = −2.5 and ΔG = −0.7 kcal mol⁻¹) with a free energy barrier of ΔG^‡ = 23.1 kcal mol⁻¹ (Fig. 8, red).

Fig. 8 DFT computed mechanism of the hydride migration from the boron to the carbon atom, producing from the nido-type compounds 2^nido and 3^nido closo-type carboranes 2^cls and 3^cls, respectively.

These computational results suggest that hydride migration in 3^nido leading to 3^cls should also be observed, and possibly was not observed due to the energy barrier. Consequently, we decided to heat 3^nido to force the formation of 3^cls. Thus, 3^nido dissolved in CDCl₃ was heated to 60 °C, and the reaction was monitored by NMR spectroscopy. As a result, after 12 h, the ¹H NMR spectrum of the reaction mixture showed the emergence of two new septet signals at δ = 3.12 and 4.05 ppm, while the intensity of the septet signal of 3^nido at δ = 2.84 ppm decreased. Probably, the reason for obtaining two new signals of septets (that correspond to the hydrogens of CH(CH₃)₂) instead of one septet signal as in 3^nido is due to the loss of symmetry in the molecule 3^cls; the hydride migrates to the empty orbital of ^Me2CAAC and the carbenic carbon becomes sp³ hybridized (see the SI for more details).

To provide additional experimental evidence for our hypothesis, we synthesized ^pCF3Ph2oCb with an electron-withdrawing substituent at the carbon atoms of the oCb, as well as a mono-substituted oCb bearing a bulky electron-withdrawing ortho-carboranyl substituent, ^oCboCb (biscarborane). Additionally, we prepared ^pTol2oCb, ^Anis2oCb, and ^Anil2oCb, which contain electron-donating substituents, along with ^SioCb, substituted with the bulky electron-donating Me₂^tBuSi-group. These compounds were then reacted with ^Me2CAAC (Fig. 9).

Fig. 9 Reactions of ^Me2CAAC with ^pTol2oCb, ^Anis2oCb, ^Anil2oCb, ^pCF3Ph₂oCb, ^SioCb, and ^oCboCb.

Thus, oCbs with electron-withdrawing groups (EWGs), ^pCF3Ph2oCb and ^oCboCb, reacted with ^Me2CAAC at r.t., leading to adduct formation and nido-type carboranes 4^nido and 5^nido (Fig. 9, path a). Interestingly, no deprotonation occurred despite the presence of protic C–H moieties in ^oCboCb. This suggests that either the ortho-carboranyl substituent provides sufficient steric protection to prevent deprotonation or its electronic effects favour nucleophilic attack at the B3 (or B6) centre. Notably, Peryshkov observed similar reactivity for the biscarboranylphosphine compound in the reaction with phosphines.⁶² Both 4^nido and 5^nido were crystallized, and their molecular structures were determined by SC-XRD (Fig. 10).

Fig. 10 POV-ray depiction of 4^nido (a) and 5^nido (b). Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

Importantly, heating 4^nido and 5^nido in CDCl₃ to 80 °C for 12 h did not lead to any observable changes in the NMR spectra, indicating that no hydride migration and transformation to closo carboranes occur in these cases (Fig. 9, path b). This supports our mechanistic suggestion that in oCbs with EWG substituents at the carbon atom(s), this transformation is precluded due to the low nucleophilicity of the nido-type open cluster. This is also supported by the DFT calculations (BP86-D3/def2-TZVP)^59–61 done for the hydride migration and transformation to closo carboranes for both 4^nido and 5^nido. For both compounds, 4^nido and 5^nido, this process was computed to be mildly endergonic (ΔG = 2.5 and 4.0 kcal mol⁻¹, respectively) (see Fig. 12, red and blue).

The oCbs with EDGs, ^pTol2oCb, and ^Anis2oCb, reacted with ^Me2CAAC at r.t., leading to nido-type carboranes 6^nido and 7^nido, respectively (Fig. 9, path c). The molecular structures of both compounds were determined by SC-XRD (Fig. 11). In contrast to 4^nido and 5^nido, heating of which did not lead to any observable change, heating 6^nido and 7^nido in CDCl₃ at 60 °C after 12 h led to clearly observable changes in the NMR spectra (see the SI for details), which were attributed to hydride migration and the closing of the cluster to closo forms 6^cls and 7^cls, respectively (Fig. 9, path d). In both cases, the reaction of ^pTol2oCb and ^Anis2oCb with ^Me2CAAC, the products of hydride migration and the closing of the nido carborane, 6^cls and 7^cls, respectively, were isolated by washing the crude solid with hexane, followed by crystallization from hexane. The molecular structures of 6^cls and 7^cls were determined by SC-XRD (Fig. 11). This again supports the suggested mechanism for this process, which is also supported by DFT calculations (BP86-D3/def2-TZVP)^59–61 that showed that in both cases the transformation from 6^nido and 7^nido to 6^cls and 7^cls, respectively, is exergonic (ΔG = −0.8 and −3.0 kcal mol⁻¹, respectively) (Fig. 12, brown and green).

Fig. 11 (a) POV-ray depiction of 6^nido (left) and 6^cls (right). Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity; (b) POV-ray depiction of 7^nido (left) and 7^cls (right). Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

Fig. 12 DFT computed mechanism of the hydride migration from the boron to the carbon atom in ^pCF3Ph2oCb, ^oCboCb, ^pTol2oCb, ^Anis2oCb, and ^SioCb.

The ^SioCb was reacted with ^Me2CAAC in benzene (1 : 1) at r.t. However, NMR spectra recorded after 12 h indicated that no reaction occurred. Therefore, the reaction was heated to 80 °C, which led to an observable change in the ¹H NMR spectra after 6 h, which revealed that ^SioCb and ^Me2CAAC had been fully consumed. Upon slow evaporation of benzene, crystals were formed, and their molecular structure was determined by SC-XRD as a 1 : 1 product of ^Me2CAAC insertion into the B3–H bond, product 8^cls (Fig. 9, path e; see Fig. 13a for the X-ray molecular structure of 8^cls). This result is also supported by DFT computations (BP86-D3/def2-TZVP),^59–61 which show that the conversion from the nido-type carborane 8^nido to 8^cls is strongly exergonic (ΔG = −7.7 kcal mol⁻¹) (Fig. 12, purple).

Fig. 13 POV-ray depiction of 8^cls (a) and 9^cls (b). Thermal ellipsoids at 30% probability, and non-relevant hydrogen atoms are omitted for clarity.

Notably, 8^nido was not observed in this reaction under these reaction conditions, which can be explained by the thermodynamics of the hydride migration, favouring the formation of 8^cls. Importantly, the obtained product 8^cls indicates that the ^tBuMe₂Si substituent was sufficiently bulky to prevent the reaction of ^Me2CAAC with the C–H bond of the ^SioCb. Furthermore, the obtained reaction was selective, and only one insertion occurred, which contrasts with the reaction of ^Me2oCb with ^Me2CAAC, which reacted 1 : 2, respectively, leading to 2^clscls (Fig. 3), which can be attributed to the steric factor in ^SioCb vs. ^Me2oCb, with the former being significantly bulkier.

Interestingly, the reaction of ^Anil2oCb with ^Me2CAAC in C₆D₆ did not produce any result at r.t.; however, upon prolonged heating at 60 °C for 10 days, it led to the clean formation of 9^cls (Fig. 9, path f). Product 9^cls was isolated by crystallization from hexane, and its X-ray molecular structure was determined using SC-XRD (Fig. 13b). Importantly, unlike in all other cases in which the insertion of the carbene into the B–H bond was observed at the most electrophilic B centre (B3), in this case, product 9^cls clearly indicates that the ^Me2CAAC insertion into the B–H bond in ^Anil2oCb occurred at the B4 position. This could be explained by the steric hindrance that the p-Et₂N-C₆H₄ substituents create at positions B3 and B6.

We examined possible mechanisms for the formation of 9^cls using DFT calculations (see the SI). Based on these computations, we propose that the reaction proceeds via initial attack of ^Me2CAAC at the B4 vertex of ^Anil2oCb, leading to a nido-type intermediate (9^nido), followed by hydride migration and closure to afford 9^cls.

An alternative pathway involving insertion of ^Me2CAAC into the B3–H bond, followed by vertex migration (B3 to B4)⁶³ is also possible. However, according to our calculations, this route is overall slightly less favorable thermodynamically and kinetically.

Finally, to understand the reason behind the formation of the 2 : 1 product in the reaction of ^Me2CAAC with ^Me2oCb leading to 2^clscls (see Fig. 3), we used DFT to compute the full reaction profile of this transformation at the B86-D3/def2SVP level of theory.^59–61 As a result, the reaction of ^Me2CAAC with ^Me2oCb leading to 2^nido is strongly exothermic and exergonic (ΔH = −30.8 and ΔG = −15.2 kcal mol⁻¹), proceeding via TS1 with the Gibbs free energy barrier of ΔG^‡ = 15.3 kcal mol⁻¹. The hydride migration in 2^nido leading to 2^cls is also exothermic and exergonic (ΔH = −14.9 and ΔG = −12.5 kcal mol⁻¹) with the Gibbs energy barrier (TS2) of ΔG^‡ = 19.4 kcal mol⁻¹, which is also the rate-determining step of this reaction. The attack of the second ^Me2CAAC on 2^cls leading to 2^clsnido is exothermic and exergonic (ΔH = −32.1 and ΔG = −17.3 kcal mol⁻¹) with an energy barrier of ΔG^‡ = 14.2 kcal mol⁻¹. The last step of the reaction is the hydride migration in 2^clsnido, leading to the final product 2^clscls, which is again highly exothermic and exergonic (ΔH = −14.9 and ΔG = −12.1 kcal mol⁻¹) with an energy barrier of ΔG^‡ = 18.8 kcal mol⁻¹ (Fig. 14). From this computed reaction coordinate, it becomes evident that as soon as 2^cls forms, it reacts directly with the second equivalent of ^Me2CAAC due to the lower Gibbs energy barriers of the following steps. This could be explained by the higher electrophilicity of the B6 centre in 2^cls in comparison with ^Me2oCb due to the substitution of the oCb cage by an electron-withdrawing cyclic amine. Importantly, these results can also explain why the 2 : 1 (^Me2CAAC : ^R2oCb) product was not obtained in all other cases. In addition to the steric factors that prevented the reaction of two molecules of ^Me2CAAC with different carboranes (^R2oCb), electronic factors affected these reactions in the following way: when the kinetic nido products were obtained as stable intermediates in 1 : 1 reactions, ^Me2CAAC was fully consumed, and therefore, 2 : 1 products were not obtained.

Fig. 14 DFT computed mechanism for the reaction between ^Me2oCb with ^Me2CAAC leading to the formation of 2^clscls.

Conclusions

To conclude, in this work, the reactivity of ^R2oCbs was studied with ^Me2CAAC. The reaction of parent oCb with ^Me2CAAC led to the insertion of the carbene into the C–H bond. When the carbons of the oCb were substituted, the reaction of ^Me2CAAC took a different course that was guided by the nature of the substituents. Electron-withdrawing groups (EWGs) at the carbon centre(s) of oCb led to nido-type carboranes as a result of the nucleophilic attack of ^Me2CAAC at the B3 (or B6) position. In contrast, electron-donating groups (EDGs) led to the insertion of the carbene into the B–H bond(s) at the B3 (or B6) position. Experimental and computational studies of the mechanism of these insertion processes revealed that this insertion proceeds via nido intermediates, by hydride migration from the B to C atom, and closing of the cluster. In the suggested mechanism, EWGs at the carbon centre stabilized the nido form and precluded the hydride migration and closing of the cluster; on the other hand, EDGs made the hydride transfer and closing of the cluster a favourable process. This study provided the first experimental evidence for the mechanism of nucleophilic substitution at B3 and/or B6 positions, as well as at the B4 (or B5) position in the case of large sterically encumbered substituents, and potentially will lead in the future to more controlled and selective substitution of oCbs, as well as paving the way towards metal-free substitutions at these positions.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data supporting this study are available in the supplementary information (SI). Supplementary information: NMR, MS, experimental and computational details (PDF). See DOI: https://doi.org/10.1039/d6qi00095a.

CCDC 2503802–2503813 and 2504008 contain the supplementary crystallographic data for this paper.^64a–m

Acknowledgements

This work was supported by the Israeli Science Foundation, Grant 195/22, and the Israel Ministry of Science Technology & Space, Grant 01032376.

References

1. R. N. Grimes, Introduction and History, in Carboranes, Academic Press, 2nd edn, 2011, pp. 1–6.
2. X. Zhang, L. M. Rendina and M. Müllner, Carborane-Containing Polymers: Synthesis, Properties, and Applications, ACS Polym. Au, 2023, 4, 7–33.
3. R. N. Grimes, Carboranes in medicine, in Carboranes, Academic Press, 2nd edn, 2011, pp. 1053–1082.
4. A. W. Tomich, J. Chen, V. Carta, J. Guo and V. Lavallo, Electrolyte Engineering with Carboranes for Next-Generation Mg Batteries, ACS Cent. Sci., 2024, 10, 264–271.
5. S. P. Fisher, A. W. Tomich, S. O. Lovera, J. F. Kleinsasser, J. Guo, M. J. Asay, H. M. Nelson and V. Lavallo, Nonclassical Applications of closo-Carborane Anions: From Main Group Chemistry and Catalysis to Energy Storage, Chem. Rev., 2019, 119, 8262–8290.
6. R. N. Grimes, Carboranes in catalysis, in Carboranes, Academic Press, 2nd edn, 2011, pp. 1037–1052.
7. D. Bawari, D. Toami, K. Jaiswal and R. Dobrovetsky, Hydrogen Splitting at a Single Phosphorus Centre and Its Use for Hydrogenation, Nat. Chem., 2024, 16, 1261–1266.
8. X. Tan, X. Wang, Z. H. Li and H. Wang, Borenium-Ion-Catalyzed C−H Borylation of Arenes, J. Am. Chem. Soc., 2022, 144, 23286–23291.
9. R. N. Grimes, Carboranes in other applications, in Carboranes, Academic Press, 2nd edn, 2011, pp. 1083–1105.
10. M. Keener, C. Hunt, T. G. Carroll, V. Kampel, R. Dobrovetsky, T. W. Hayton and G. Ménard, Redox-sitchable crboranes for uranium capture and release, Nature, 2020, 577, 625–655.
11. R. N. Grimes, Metallacarboranes of the transition and lanthanide elements, in Carboranes, Academic Press, 2nd edn, 2011, pp. 773–1014.
12. D. J. Liston, Y. J. Lee, W. R. Scheidt and C. A. Reed, Observations on silver salt metathesis reactions with very weakly coordinating anions, J. Am. Chem. Soc., 1989, 111, 6643–6648.
13. C. A. Reed, Carboranes: A New Class of Weakly Coordinating Anions for Strong Electrophiles, Oxidants, and Superacids, Acc. Chem. Res., 1998, 31, 133–139.
14. R. N. Grimes, Icosahedral carboranes: 1,7-C₂B₁₀H₁₂ and 1,12-C₂B₁₀H₁₂, in Carboranes, Academic Press, 2nd edn, 2011, pp. 541–674.
15. A. D. Ready, V. Tej Raviprolu, T. A. Kerr, J. W. Treacy, M. L. Matsumoto, P. E. Hammer, E. M. Sletten, K. N. Houk and A. M. Spokoyny, Carboranes without Cage Carbons: closo-Dodecaborate Mimics of Neutral closo-Carboranes, J. Am. Chem. Soc., 2025, 147, 25478–25488.
16. R. N. Grimes, Structure and Bonding, in Carboranes, Academic Press, 2nd edn, 2011, pp. 7–20.
17. N. N. Greenwood and A. Earnshaw, Boron, in Chemistry of the Elements, Butterworth-Heinemann, 2nd edn, 1997, pp. 139–215.
18. V. I. Bregadze, Dicarba-closo-dodecaboranes C₂B₁₀H₁₂ and their derivatives, Chem. Rev., 1992, 92, 209–223.
19. Y. Quan and Z. Xie, Controlled functionalization of o-carborane via transition metal catalyzed B−H activation, Chem. Soc. Rev., 2019, 48, 3660–3673.
20. M. F. Hawthorne, D. C. Young, P. M. Garrett, D. A. Owen, S. G. Schwerin, F. N. Tebbe and P. A. Wegner, reparation and characterization of the (3)-1,2- and (3)-1,7-dicarbadodecahydroundecaborate(-1) ions, J. Am. Chem. Soc., 1968, 90, 862–868.
21. L. I. Zakharkin and V. N. Kalinin, On the Reaction of Amines with Barenes, Tetrahedron Lett., 1965, 6, 407–409.
22. R. A. Wiesboeck and M. F. Hawthorne, Dicarbaundecaborane(13) and Derivatives, J. Am. Chem. Soc., 1964, 86, 1642–1643.
23. H. D. A. C. Jayaweera, M. M. Rahman, P. J. Pellechia, M. D. Smith and D. V. Peryshkov, Free Three-dimensional Carborane Carbanions, Chem. Sci., 2021, 12, 10441–10447.
24. K. P. Anderson, H. A. Mills, C. Mao, K. O. Kirlikovali, J. C. Axtell, A. L. Rheingold and A. M. Spokoyny, Improved synthesis of icosahedral carboranes containing exopolyhedral B−C and C−C bonds, Tetrahedron, 2019, 75, 187–191.
25. N. S. Hosmane, Boron Science: New Technologies and Applications, CRC Press, Boca Raton, 2011, pp. 3–20.
26. G.-X. Jin and Z. Xie, Carboranes Themed Issue, Dalton Trans., 2014, 43, 4924–5133.
27. V. I. Bregadze and Z. Xie, Boron Chemistry Themed Issue, Eur. J. Inorg. Chem., 2017, 4344.
28. T. L. Chan and Z. Xie, Synthesis, structure and aromaticity of carborane-fused carbo- and heterocycles, Chem. Sci., 2018, 9, 2284–2289.
29. S. O. Lovera, A. Gregory, K. Espinoza Morelos, P. Farias, V. Carta, C. B. Musgrave III and V. Lavallo, Noncatalyzed Intramolecular B−N and B−O Cross-Coupling of “Inert” Carboranes Lead to the Formation of an Unusual Oxoborane, via Reversible Cluster C−B Bond Scission, J. Am. Chem. Soc., 2025, 147, 17764–17771.
30. C. Douvris and J. Michl, Chemistry of the Carbacloso-dodecaborate(−) Anion, CB₁₁H₁₂⁻, Chem. Rev., 2013, 113, PR179–PR233.
31. Z. Zheng, W. Jiang, A. A. Zinn, C. B. Knobler and M. F. Hawthorne, Facile Electrophilic Iodination of Icosahedral Carboranes. Synthesis of Carborane Derivatives with Boron-Carbon Bonds via the Palladium-Catalyzed Reaction of Diiodocarboranes with Grignard Reagents, Inorg. Chem., 1995, 34, 2095–2100.
32. V. N. Lebedev, E. V. Balagurova, A. V. Polyakov, A. I. Yanovsky, Y. T. Struchkov and L. I. Zakharkin, Selective fluorination of o0 and m-carboranes. Synthesis of 9-monofluoro-, 9,12-difluoro-1,8,9,12-trifluoro-, and 8,9,10,12-tetrafluoro-o-carboranes and 9-monofluoro-, and 9,10-difluoro-m-carboranes. Molecular structure of 8,9,10,12-tetrafluoro-o-carborane, J. Organomet. Chem., 1990, 385, 307–318.
33. Y. Ma, H. Ren, Y. Wu, N. Li, F. Chen and X. Chen, B(9)-OH-o-Carboranes: Synthesis, Mechanism, and Property Exploration, J. Am. Chem. Soc., 2023, 145, 7331–7342.
34. F. Teixidor, G. Barberà, C. Viñas, R. Sillanpää and R. Kivekäs, Synthesis of Boron-Iodinated o-Carborane Derivatives. Water Stability of the Periodinated Monoprotic Salt, Inorg. Chem., 2006, 45, 3496–3498.
35. K. Wade, Electron Deficient Compounds, Springer, 1971.
36. R. Cheng, Z. Qiu and Z. Xie, Iridium-catalysed regioselective borylation of carboranes via direct B-H activation, Nat. Commun., 2017, 8, 14827.
37. Z. Qiu and Z. Xie, A Strategy for Selective Catalytic B–H Functionalization of o-Carboranes, Acc. Chem. Res., 2021, 54, 4065–4079.
38. M. Zhu, P. Wang, Z. Wu, Y. Zhong, L. Su, Y. Xin, A. M. Spokoyny, C. Zou and X. Mu, A Pd-catalyzed route to carborane-fused boron heterocycles, Chem. Sci., 2024, 15, 10392–10401.
39. Y. Quan and Z. Xie, Controlled functionalization of o-carborane via transition metal catalyzed B–H activation, Chem. Soc. Rev., 2019, 48, 3660–3673.
40. Z. Qiu, S. Ren and Z. Xie, Transition Metal-Carboryne Complexes: Synthesis, Bonding, and Reactivity, Acc. Chem. Res., 2011, 44, 299–309.
41. Y. Quan and Z. Xie, Palladium-Catalyzed Regioselective Diarylation of o-Carboranes By Direct Cage B−H Activation, Angew. Chem., Int. Ed., 2016, 55, 1295–1298.
42. R. Cheng, Z. Qiu and Z. Xie, Iridium–Catalyzed Regioselective B(3)–Alkenylation/B(3,6)–Dialkenylation of o–Carboranes by Direct B−H Activation, Chem. – Eur. J., 2020, 26, 7212–7218.
43. Y. Quan and Z. Xie, Iridium Catalyzed Regioselective Cage Boron Alkenylation of o-Carboranes via Direct Cage B–H Activation, J. Am. Chem. Soc., 2014, 136, 15513–15516.
44. C. Tang, J. Zhang, J. Zhang and Z. Xie, Regioselective Nucleophilic Alkylation/Arylation of B–H Bonds in o-Carboranes: An Alternative Method for Selective Cage Boron Functionalization, J. Am. Chem. Soc., 2018, 140, 16423–16427.
45. J. Zhang and Z. Xie, N-Ligand-Enabled, Aromatic Nucleophilic Amination of 1,2-Diaryl-o-Carboranes with (R(₂)N)(₂)Mg for Selective Synthesis of 4-R(₂)N-o-Carboranes and 2-R(₂)N-m-Carboranes, Angew. Chem., Int. Ed., 2022, 61, e202202675.
46. M. Jones, W. Ando, M. E. Hendrick, A. Kulczycki, P. M. Howley, K. F. Hummel and D. S. Malament, Irradiation of Methyl Diazomalonate in Solution. Reactions of Singlet and Triplet Carbenes with Carbon-Carbon Double Bonds, J. Am. Chem. Soc., 1972, 94, 7542–7549.
47. G.-X. Zheng and M. Jones, Reaction of (ethoxycarbonyl)carbene with o-carborane, J. Am. Chem. Soc., 1983, 105, 6487–6488.
48. K. Yuan and M. Jones, Carbenes Do React with p-Carborane, Tetrahedron Lett., 1992, 33, 7481–7484.
49. K. Lee, G. A. González-Montiel, H. Eom, T. H. Kim, H. C. Noh, A. O. Farah, H. R. Wise, D. Kim, P. H.-Y. Cheong and P. H. Lee, Site- and Enantioselective B–H Functionalization of Carboranes, Nat. Commun., 2025, 16, 4182.
50. N. Matsumi, Preparation of ionic liquids bearing o-carborane anion via N,N’-dialkylimidazol-2-ylidene carbene, J. Organomet. Chem., 2009, 694, 1612–1616.
51. C. E. Willans, C. A. Kilner and M. A. Fox, Deboronation and Deprotonation of ortho–Carborane with N–Heterocyclic Carbenes, Chem. – Eur. J., 2010, 16, 10644–10648.
52. F. Zheng and Z. Xie, Reaction of o-carboranes with sterically demanding N-heterocyclic carbene: synthesis and structural characterization of 1 : 1 adducts, Dalton Trans., 2012, 41, 12907–12914.
53. S. K. Kumar, A. Mishra, H. W. Roesky and K. C. Mondal, Recent Advances in the Domain of Cyclic (Alkyl)(Amino) Carbenes, Chem. – Asian J., 2022, 17, e202101301.
54. R. Jazzar, R. D. Dewhurst, J.-B. Bourg, B. Donnadieu, Y. Canac and G. Bertrand, Intramolecular “Hydroiminiumation” of Alkenes: Application to the Synthesis of Conjugate Acids of Cyclic Alkyl Amino Carbenes (CAACs), Angew. Chem., Int. Ed., 2007, 46, 2899–2902.
55. Y. Zhu, C. Li, M. Sudarmadji, H. M. Ng, B.-Y. Ao, J. A. Maguire and N. S. Hosmane, An efficient and recyclable catalytic system comprising nanopalladium(0) and a pyridinium salt of iron bis(dicarbollide) for oxidation of substituted benzyl alcohol and lignin, ChemistryOpen, 2012, 1, 67–70.
56. P. W. Causey, T. R. Besanger and J. F. Valliant, Synthesis and Screening of Mono- and Di-Aryl Technetium and Rhenium Metallocarboranes. A New Class of Probes for the Estrogen Receptor, J. Med. Chem., 2008, 51, 2833–2844.
57. M. A. Fox, A. K. Hughes and J. M. Malget, Cage-closing reactions of the nido-carborane anion 7,9-C₂B₉H₁₂⁻ and derivatives; formation of neutral 11-vertex carboranes by acidification, J. Chem. Soc., Dalton Trans., 2002, 3505–3517.
58. T. D. McGrath, A. Franken, J. A. Kautz and F. G. A. Stone, Synthesis of nickel–monocarbollide complexes by oxidative insertion, Inorg. Chem., 2005, 44, 8135–8144.
59. J. P. Perdew, Density-functional approximation for the correlation energy of the inhomogeneous electron gas, Phys. Rev. B: Condens. Matter Mater. Phys., 1986, 33, 8822–8824.
60. A. D. Becke, Density-functional exchange-energy approximation with correct asymptotic behavior, Phys. Rev. A, 1988, 38, 3098–3100.
61. F. Weigend and R. Ahlrichs, Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy, Phys. Chem. Chem. Phys., 2005, 7, 3279–3305.
62. Y. O. Wong, M. D. Smith and D. V. Peryshkov, Cage Opening of a Carborane Ligand by Metal Cluster Complexes, Chem. – Eur. J., 2016, 22, 6764–6767.
63. S.-h. Wu and M. Jones Jr., The mechanism of rearrangement of the icosahedral carboranes, J. Am. Chem. Soc., 1989, 111, 5373–5384.
64. (a) CCDC 2503802: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dt0; (b) CCDC 2503803: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dv1; (c) CCDC 2503804: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dw2; (d) CCDC 2503805: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dx3; (e) CCDC 2503806: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dy4; (f) CCDC 2503807: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1dz5; (g) CCDC 2503808: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f07; (h) CCDC 2503809: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f18; (i) CCDC 2503810: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f29; (j) CCDC 2503811: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f3b; (k) CCDC 2503812: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f4c; (l) CCDC 2503813: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1f5d; (m) CCDC 2504008: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2q1mgw.

---