Synthesis and characterization of closo-decahydrido-decaborate-phosphine mediated by palladium(0) complexes

Abstract

We report an efficient and selective preparation of an apical mono-functionalized closo-decahydrido-decaborate anion [B₁₀H₁₀]²⁻ with aromatic or aliphatic tertiary phosphines. Starting from [N(n-C₄H₉)₄]₂[1-B₁₀H₉I] and 0.25–0.5 equiv. of Pd(0)-PR₃, the corresponding [N(n-C₄H₉)₄][1-B₁₀H₉PR₃] compounds were obtained in high isolated yields under mild conditions. Additionally, a plausible mechanism was proposed based on experimental evidence.

---

Introduction

In the area of boron cluster chemistry,¹ the decaborate anion, [B₁₀H₁₀]²⁻, exhibits a 3D aromaticity, leading to unusual oxidative, hydrolytic and thermal stabilities.¹ It is also described as exhibiting a non-uniform electronic distribution within the cage and the charges of atoms in this cluster differ for different vertices (apical vs. equatorial).^2–4 These properties make such clusters susceptible to both nucleophilic and electrophilic substitutions.⁴ More precisely, the activation of the exo-B–H bond in the polyhedral closo-decahydrido-decaborate anion [B₁₀H₁₀]²⁻ has been a subject of intensive investigation in the last few decades.^5–9 Indeed, its derivatives have applications in medicine—notably in boron neutron capture therapy^10,11—materials sciences,¹² ion extraction systems,¹³ and aerogel nanomaterial preparation,¹⁴ as solid and liquid electrolytes¹⁵ and in tunable intermolecular charge transfer.¹⁶ However, selective heterofunctionalization of this cluster, notably with organophosphorus motifs, still remains challenging in terms of regioselectivity and efficiency. Earlier attempts to prepare mono-substituted phosphine derivatives have been largely restricted to [B₁₂H₁₂]^2−17–19 and carbaboranes.^20,21 In the case of the [B₁₀H₁₀]²⁻ anion, the scarce reports dealing with phosphine functionalization often exhibit poor regioselectivity and formation of a mixture of mono-, di- and/or polysubstituted products.²² In 1994, Todd and co-workers reported the synthesis of the 1,10-, 1,6- and 1,7(8) isomers of (PMe₂Ph)₂B₁₀H₈ in low yields using 2.5 equiv. of (PMe₂Ph)₂PdCl₂.²³ In 1999, Naoufal et al. reported that the combination of (PPh₃)₂PdCl₂ and CuI catalysts promoted the substitution of the diazo group in [1-N₂B₁₀H₉]⁻ by diphenylphosphine, leading to [1-(Ph₂PH)B₁₀H₉]⁻ species in a mixture with [1-Ph₂P(OH)B₁₀H₉]⁻.¹⁷ It is worth underlining that all these previous preparations required long and difficult synthetic procedures even if several products were obtained despite their tedious purification.

Based on a pioneering result of one of our research groups on the palladium promoted preparation of one example of closo-[1-B₁₂H₁₁PPh₃][Nn-Bu₄]₂ from closo-[1-IB₁₂H₁₁][Nn-Bu₄]₂,²⁴ we report herein a general and easy palladium-promoted synthesis of [N(n-C₄H₉)₄][1-B₁₀H₉PR₃] from [N(n-C₄H₉)₄]₂[1-B₁₀H₉I] using tertiary phosphines.

Results and discussion

Synthesis of closo-decahydrido-decaborate-phosphine clusters

The preparation of mono-anionic tertiary phosphine derivatives of the closo-decahydrido-decaborate cluster was first evaluated starting from a commercially available palladium precursor such as Pd(PPh₃)₄. Indeed, starting from 1 equiv. of iodinated boron cluster [N(n-C₄H₉)₄]₂[1-B₁₀H₉I] 1,²² in the presence of 10 mol% of Pd(PPh₃)₄ and 3 equiv. of Na₂CO₃ in dried and degassed THF at 50 °C for 2 h, the corresponding [N(n-C₄H₉)₄][1-B₁₀H₉(PPh₃)] 2a was obtained in 33% isolated yield (conv. = 40%). When increasing the amount of Pd(0) to 25 mol% (i.e., 1 equimolar amount of triphenylphosphine), full conversion was achieved and 2a was isolated in 80% yield. Noticeably, the reaction can be conducted with similar efficiency in different solvents, including CH₂Cl₂, 1,4-dioxane and acetonitrile. It should be noted that the reaction did not proceed at room temperature. Interestingly, in the absence of a base, the reaction exhibited the same efficiency.

The formation of the new compound 2a was easily followed in ³¹P NMR spectroscopy as a broad quartet was observed at 12.22 ppm (J_P–B = 186.1 Hz), which clearly demonstrated the formation of a B–P bond. Additionally, the ¹¹B NMR spectrum showed 4 signals at 14.62 ppm (d, J = 140 Hz, 1B), −9.98 ppm (d, J = 190 Hz, 1B), −22.62 ppm (d, J = 131 Hz, 4B) and −25.79 ppm (d, J = 131 Hz, 4B), which have the characteristic pattern of an apical-substituted decaborate cluster (Scheme 1).

Scheme 1 Preparation of [N(n-C₄H₉)₄][1-B₁₀H₉(PPh₃)] 2a. (a) ³¹P NMR spectrum for [B₁₀H₉PR₃][N(n-C₄H₉)₄] 2a and (b) ¹¹B NMR and ¹¹B{¹H} spectra for [B₁₀H₉PPh₃][N(n-C₄H₉)₄] 2a.

With these conditions in hand, the scope of the reaction was then evaluated (Scheme 2).

Scheme 2 Scope of the reaction.

Using 0.5 equiv. of Pd(PR₃)₂ precursors (PR₃ = P(o-tolyl)₃, P(m-Cl-C₆H₄)₃ and P(cyclohexyl)₃), the corresponding [N(n-C₄H₉)₄][1-B₁₀H₉PR₃] derivatives were obtained in high isolated yields after 0.5–3 h of reaction at 50 °C (2b, 87%; 2d, 76%; and 2e, 75%). Noticeably, 0.5 equiv. of palladium precursors Pd(PR₃)₂ was used in order to find the equimolar ratio between the phosphines and the [closo-B₁₀H₉-1-I]²⁻ derivative. To prepare compound 2c, 35 mol% of Pd(3,5-(CF₃)₂C₆H₃)₃P)₃ was used. After 4 h at 50 °C, the corresponding [N(n-C₄H₉)₄][B₁₀H₉(3,5-(CF₃)₂C₆H₃)₃P)] derivative 4c was isolated in 90% yield.

The availability of commercially available Pd(PR₃)_n is low, and a search for an alternative method starting from phosphine and palladium precursors was carried out. Thus, using 12.5 mol% of Pd₂(dba)₃ and 1 equiv. of PPh₃ in THF at 50 °C for 2 h led to the same results as starting with Pd(PPh₃)₄ as the corresponding [N(n-C₄H₉)₄][1-B₁₀H₉(PPh₃)] 2a was obtained in 80% isolated yield. Accordingly, the [N(n-C₄H₉)₄][1-B₁₀H₉(PBn₃)] compound 2f was prepared starting from 25 mol% of Pd₂dba₃ and 1.0 equiv. of tribenzylphosphine and isolated in 83% yield.

Characterization of compounds 2a–f

Mass spectrometry studies. The new compounds 2a–f were characterized by ESI-MS mass spectrometry. Molecular ions were consistent with the presence of the expected [N(n-C₄H₉)₄][1-B₁₀H₉(PPh₃)] complexes 2a–f (see the ESI†).

NMR studies. 2a–f exhibited NMR spectra consistent with the structures of the expected compounds (Scheme 1 and Table 1). Noticeably, in ³¹P NMR spectra, a typical quartet with a coupling constant ¹J_P–B ranging from 180 to 190 Hz was observed for all the derivatives.^22,23 The boron cluster was also classically identified by ¹¹B NMR. The spectra exhibited a classical pattern for apical-substituted closo-decaborate with four peaks: (i) two doublets for the boron atoms in equatorial positions (a shift in the range from −20.0 to −27.0 ppm) and one doublet for the boron at the unsubstituted apical position (a shift in the range from 11.9 to 20.3 ppm), with ¹J_B–H coupling values in the range of 128–136 Hz; and (ii) one doublet for the substituted boron at the apical position (shift in the range from −8.4 to −14.1 ppm) with a coupling constant ¹J_P–B ranging from 180 to 190 Hz.^22,23

Table 1 ³¹P and ¹¹B NMR data for 2a–f

	2a	2b	2c	2d	2e	2f
^31P NMR
δ (ppm)	12.22	18.03	16.84	13.48	12.52	3.37
^1 J P–B (Hz)	190	191	189	192	183	184
^11B NMR
δ (ppm)	14.62	14.14	20.34	16.51	11.98	13.43
^1 J B–H (Hz)	140	146	148	147	145	147
Integration	1B	1B	1B	1B	1B	1B
δ (ppm)	−9.98	−9.23	−14.08	−11.52	−12.12	−8.40
^1 J P–B (Hz)	190	191	189	192	185	188
Integration	1B	1B	1B	1B	1B	1B
δ (ppm)	−22.62	−22.00	−20.89	−22.06	−24.48	−23.31
^1 J B–H (Hz)	131	133	136	136	128	133
Integration	4B	4B	4B	4B	4B	4B
δ (ppm)	−25.79	−25.95	−24.85	−25.47	−26.68	−26.14
^1 J B–H (Hz)	131	129	134	134	135	133
Integration	4B	4B	4B	4B	4B	4B

---

IR studies. 2a–f exhibited a characteristic broad intense peak at 2470–2490 cm⁻¹ (ν, cm⁻¹: 2a, 2491; 2b, 2473; 2c, 2485; 2d, 2486; 2e, 2470; 2f, 2464).

X-ray crystal diffraction and molecular structures

Colorless monoclinic crystals of 2a, 2b, and 2f and triclinic crystals of 2c, 2d, and 2e were obtained from dichloromethane/diethyl ether solutions upon slow diffusion at 0 °C. Compounds 2a–f were then characterized by X-ray diffraction. Molecular crystal structures are shown in Fig. 1. Most salts in the series have a single ion pair in the asymmetric unit, except for 2d (two anions and two cations) and 2c (four anions and four cations). The large [Bu₄N]⁺ cation is disordered in most salts. For 2e, the crystal is disordered with some constraints and 2c shows disorder in the CF₃ group. The obtained crystal structure of 2d is slightly twinned. The crystallographic data (bond lengths, angles, and torsion angles) are summarized in Tables 2, 3, and Table S1.†

Fig. 1 Crystal structures (50% probability level for the displacement ellipsoids) of 2a–f. Tetrabutylammonium counterions and hydrogen atoms are omitted for clarity. In 2d and 2c, only one unique molecule is shown.

Table 2 Selected bond lengths (Å) for the series 2a–f

Selected bond lengths (Å)
	2a	2b	2c	2d	2e	2f
P–B	P1–B1 = 1.870(3)	P1–B1 = 1.898(3)	P1–B1 = 1.864(6)	P1–B1 = 1.858(6)	P1–B1 = 1.876(5)	P1–B1 = 1.871(5)
P–C	P1–C1 = 1.807(2)	P1–C8 = 1.821(3)	P1–C9 = 1.812(5)	P1–C13 = 1.805(5)	P1–C13B/A = 1.826(5)	P1–C15 = 1.823(4)
	P1–C13 = 1.809(2)	P1–C15 = 1.829(3)	P1–C1 = 1.816(5)	P1–C7 = 1.806(5)	P1–C7B/A = 1.833(5)	P1–C1 = 1.823(5)
	P1–C7 = 1.811(2)	P1–C1 = 1.829(3)	P1–C17 = 1.819(5)	P1–C1 = 1.835(5)	P1–C1A/B = 1.833(5)	P1–C8 = 1.827(5)
B–B	B1–B4 = 1.686(4)	B1–B4 = 1.688(4)	B1–B4 = 1.686(8)	B1–B4 = 1.686(8)	B1–B4 = 1.692(7)	B1–B4 = 1.676(7)
	B1–B5 = 1.686(4)	B1–B5 = 1.695(5)	B1–B5 = 1.685(8)	B1–B5 = 1.666(8)	B1–B5 = 1.698(7)	B1–B5 = 1.681(7)
	B1–B3 = 1.692(4)	B1–B3 = 1.698(4)	B1–B3 = 1.677(8)	B1–B3 = 1.701(8)	B1–B3 = 1.684(7)	B1–B3 = 1.675(7)
	B1–B2 = 1.692(4)	B1–B2 = 1.700(5)	B1–B2 = 1.693(8)	B1–B2 = 1.686(8)	B1–B2 = 1.693(8)	B1–B2 = 1.680(7)

---

Table 3 Selected angles (°) of the series 2a–f

Selected angles (°)
2a	2b	2c	2d	2e	2f
C1–P1–C13 = 107.95(12)	C8–P1–C15 = 105.27(13)	C9–P1–C1 = 105.8(2)	C13–P1–C7 = 105.6(2)	C13B/A–P1–C7B/A = 104.7(2)	C15–P1–C1 = 103.5(2)
C1–P1–C7 = 104.40(11)	C8–P1–C1 = 106.24(13)	C9–P1–C17 = 105.8(2)	C13–P1–C1 = 106.4(2)	C13A/B–P1–C1A/B = 110.9(2)	C15–P1–C8 = 103.9(2)
C13–P1–C7 = 107.03(11)	C15–P1–C1 = 107.78(14)	C1–P1–C17 = 103.5(2)	C7–P1–C1 = 106.5(2)	C7A/B–P1–C1A/B = 106.3(2)	C1–P1–C8 = 106.1(2)
C1–P1–B1 = 115.12(12)	C8–P1–B1 = 112.77(14)	C9–P1–B1 = 114.5(2)	C13–P1–B1 = 112.0(3)	C13B/A–P1–B1 = 111.2(2)	C15–P1–B1 = 113.7(2)
C13–P1–B1 = 109.26(12)	C1–P1–B1 = 114.09(14)	C1–P1–B1 = 114.1(2)	C7–P1–B1 = 114.2(2)	C7B/A–P1–B1 = 112.1(2)	C1–P1–B1 = 115.7(2)
C7–P1–B1 = 112.65(12)	C15–P1–B1 = 110.23(14)	C17–P1–B1 = 112.2(2)	C1–P1–B1 = 111.6(2)	C1A/B–P1–B1 = 111.5(2)	C8–P1–B1 = 112.7(2)

---

The B–P bond lengths in 2a–2f are in the range of 1.858–1.898 Å. These values are consistent with the ones described for similar substituted phosphine borane clusters such as 1.9055 and 1.91113 Å for 1,7-(PMe₂Ph)₂-closo-B₁₂H₁₀,¹⁷ 1.901 and 1.886 Å for 2,8-(PMe₂Ph)₂-closo-B₁₀H₈ ²³ or 1.928 Å for [B₁₂H₁₁PPh₃][N(n-C₄H₉)₄].¹⁷ Nevertheless, the B–P bond lengths in 2a–2f are slightly shorter than typical B–P single bonds (1.90–2.00 Å). Noticeably, in classical organophosphorus-borane derivatives, the B–P single-bond distance range is 1.90–2.00 Å and the double-bond distance range is 1.79–1.84 Å.²⁵ Such observations can be explained by potential π-overlap between the boron and phosphorus atoms.²⁶

P–C_α-carbons bond lengths in 2a–f (from 1.805 to 1.835 Å) aligned with expected values for phosphorus–carbon single bonds. The expected bond length for P–C in a phosphine is approximately 1.80–1.85 Å, with slight variations depending on the specific electronic and steric environments.^17,22,23

Within the boron cluster, the B–B bond lengths in 2a–f ranged from 1.66 to 1.70 Å, which is consistent with the expected range for boron hydride clusters (typically in the range of 1.6–1.8 Å), reflecting normal bonding interactions.²⁷ Measurements of distances between apical boron atoms and the equatorial plane revealed that the substitution at the B₁ position with the phosphorus group induced slight contraction of the boron cage (average 1.067 Å compared to 1.093 Å for the unsubstituted B₁₀H₁₀²⁻ motif). This contraction likely arises from electron-withdrawing effects or steric interactions associated with the phosphorus substituent. In contrast, B₁₀, which remains unsubstituted, retained a geometry similar to that of the unsubstituted cluster, showing no significant distortion.

The bond angles around the phosphorus centre deviated from the ideal tetrahedral geometry. As a representative example, in the crystal structure of 2d, the angles between the phenyl groups (C₁–P₁–C₁₃ = 106.4(2)°, C₁–P₁–C₇ = 106.5(2)°, and C₁₃–P₁–C₇ = 105.6(2)°) were compressed relative to the ideal 109.5°. In contrast, the bond angles of the cluster (C₁–P₁–B₁ = 111.6(2)°, C₇–P₁–B₁ = 114.2(2)°, and C₁₃–P₁–B₁ = 112.0(3)°) are larger than the ideal tetrahedral angle. This widening can be attributed to the steric bulk of the B₁₀H₉ cluster, which exerts significant repulsive force, pushing the phenyl groups away from the cluster, resulting in increased bond angles. This trend is similar for the rest of the 2a–2f series.

Mechanism studies

First of all, in order to confirm or exclude the involvement of radicals during the transformation, several radical scavengers such as 2,2,6,6-tetramethylpiperidinyloxyl (TEMPO), 2,6-di-tert-butyl-4-methyl-phenol (BHT) and galvinoxyl free radicals, have been used as additives. Thus, conducting the reaction of [NBu₄]₂[B₁₀H₉I] 1 (1 equiv.) with Pd(PPh₃)₄ (25 mol%) in the presence of a radical scavenger (1 equiv.) in THF at 50 °C for 2 h led to the corresponding compound 2a without loss of activity, which excluded a radical pathway.

To monitor the reaction progress, both ¹H and ³¹P NMR spectra were recorded. First of all, the formation of the [N(n-C₄H₉)₄][1-B₁₀H₉(PPh₃)] product 2a was monitored by ¹H-NMR depending on the temperature. At temperatures lower than 55 °C, no reaction occurred as confirmed by the presence of the characteristic signals δ = 7.25, δ = 7.19, and δ = 7.09 ppm of Pd(PPh₃)₄. After 15 min at 55 °C, Pd(PPh₃)₄ was fully transformed and product 2a was selectively obtained (aromatic signals at 7.98–7.88 and 7.60–7.45) (Fig. 2).

Fig. 2 ¹H NMR spectra for the reaction mixture between [N(n-C₄H₉)₄]₂[B₁₀H₉I] and Pd(PPh₃)₄ in d⁸-THF after 15 min at RT, 313 K (40 °C), 323 K (50 °C) and 328 K (55 °C).

Interestingly, when conducting the reaction with Pd(PPh₃)₄ and closo-[Bu₄N]₂[1-B₁₀H₉I] 1 in the presence of excess triphenylphosphine, no reaction occurred. This may suggest that the excess phosphine disfavoured the dissociation of the PPh₃ ligand from Pd(PPh₃)₄ in order to generate an active species able to promote an oxidative addition step. Additionally, during the substrate scope investigation, the electronic nature of substituents on the phosphine significantly affected the kinetics of the reaction. When an electron-donating substituent was used (e.g., –CH₃ of the tolyl group in 4b), the reaction time was reduced to 30 minutes. In contrast, using a phosphine with an electron-withdrawing group, such as CF₃ (complex 4c) or chlorine (complex 4d), led to a slower transformation with a reaction time extended to 4 and 3 hours, respectively. Indeed, phosphines bearing electron-withdrawing substituents should disfavour the oxidative addition, which thus seems an important step. With tricyclohexylphosphine palladium species, the reaction time was 3 h, showing that the hindrance of phosphine slowed down the reaction. Such observations may indicate that the oxidative step was involved and was the kinetic key step. It should be also noted that n-Bu₄NI crystals were isolated from the crude mixture obtained at the end of the reaction conducted under optimized conditions.

A plausible mechanism can be proposed when starting from the 18-electron complex Pd(PPh₃)₄; after non-reductive elimination of two phosphine ligands followed by the oxidative addition of the [B₁₀H₉I]⁻ anion, a palladium(0) [Bu₄N]₂[(1-B₁₀H₉)(I)Pd(PPh₃)₂] species was obtained. Due to steric hindrance caused by the presence of the decaborate entity and phosphine ligands, this latter complex should lead to an unusual elimination of [Bu₄N][(1-B₁₀H₉)(PPh₃)] and after coordination of PPh₃ gave the anionic species [Pd(I)(PPh₃)₂]⁻, which after elimination of n-Bu₄NI generated Pd(PPh₃)₂ that was able to perform a second oxidative addition of the closo-[Bu₄N]₂[1-B₁₀H₉I] cluster.²⁸ Nevertheless, it cannot be excluded that the tricoordinated anionic species [Pd(I)(PPh₃)₂]⁻ performed an oxidative addition with closo-[Bu₄N]₂[1-B₁₀H₉I], thus leading to a pentacoordinated intermediate despite steric hindrance of the cluster (Scheme 3).^29–32

Scheme 3 Proposed mechanism for the reaction of the phosphine complex with [N(n-C₄H₉)₄] [B₁₀H₉I].

Conclusions

We have developed a simple, convenient and efficient method for the preparation of mono-anionic phosphine derivatives of a closo-decaborate cluster starting from the closo-[Bu₄N]₂[1-B₁₀H₉I] cluster using an equimolar amount of the desired phosphine ligand associated with the Pd⁰ precursor. Both triaryl- and tri-alkyl phosphine clusters can be prepared with high selectivity and yields. Noticeably, the obtained phosphine clusters were characterized by NMR, HR-MS, IR and X-ray diffraction studies. Based on the experiments, a mechanistic pathway was proposed, suggesting notably the role of palladium(0) involved in the key step oxidative addition of the B–I bond.

Conflicts of interest

There are no conflicts to declare.

Data availability

Crystallographic data for compounds 2a–2f have been deposited at the CCDC under deposition numbers 2456501–2456506† and can be obtained from https://www.ccdc.cam.ac.uk. All other data including NMR spectra supporting this article have been included as part of the ESI.†

Acknowledgements

We wish to acknowledge the Université de Rennes (France), the Centre National de la Recherche Scientifique (CNRS, France), the Lebanese University (Lebanon), and CNRSL (Lebanon).

References

1. (a) N. S. Hosmane, Boron Science. New Technologies and Applications, Taylor & Francis Books/CRC Press, Boca Raton, 2016; (b) R. N. Grimes, Carboranes, Elsevier Science, Amsterdam, 2016; (c) N. S. Hosmane and R. D. Eagling, Handbook of Boron Science: With Applications in Organometallics, Catalysis, Materials and Medicine (In 4 Volumes), World Scientific, London, UK, 2018; (d) L. Mcconnell, Boron: Advances in Research and Applications, Nova Science Publishers, 2021; (e) Y. Zhu, Fundamentals and Applications of Boron Chemistry, Elsevier, 2022; (f) M. F. Hawthorne, Boranes and beyond: History and the man who created them, Springer Nature, 2023.
2. I. B. Sivaev, A. V. Prikaznov and D. Naoufal, Collect. Czech. Chem. Commun., 2010, 75, 1149–1199.
3. W. C. Ewing, 6- and 5-Halodecaboranes: Selective Syntheses from ClOSO-B10H10 (2-) and Use as Polyborane Building Blocks, 2010.
4. Z. Laila, F. Abi-Ghaida, S. Al Anwar, O. Yazbeck, R. Jahjah, R. Aoun, S. Tlais, A. Mehdi and D. Naoufal, Main Group Chem., 2015, 14, 301–312.
5. N. Mahfouz, F. Abi-Ghaida, F. A. Ghaida, Z. El Hajj, M. Diab, S. Floquet, A. Mehdi and D. Naoufal, ChemistrySelect, 2022, 7, e202200770.
6. V. V. Voinova, I. N. Klyukin, A. S. Novikov, A. Y. Kozmenkova, A. P. Zhdanov, K. Y. Zhizhin and N. T. Kuznetsov, Russ. J. Inorg. Chem., 2021, 66, 295–304.
7. V. M. Retivov, E. Y. Matveev, M. V. Lisovskiy, G. A. Razgonyaeva, L. I. Ochertyanova, K. Y. Zhizhin and N. T. Kuznetsova, Russ. Chem. Bull., 2010, 59, 550–555.
8. Z. Laila, O. Yazbeck, F. Abi Ghaida, M. Diab, S. El Anwar, M. Srour, A. Mehdi and D. Naoufal, J. Organomet. Chem., 2020, 910, 121132.
9. D. Naoufal, Z. Assi, E. Abdelhai, G. Ibrahim, O. Yazbeck, A. Hachem, H. Abdallah and M. El Masri, Inorg. Chim. Acta, 2012, 383, 33–37.
10. H. Hatanaka, J. Neurol., 1975, 209, 81–94.
11. C. Salt, A. J. Lennox, M. Takagaki, J. A. Maguire and N. S. Hosmane, Russ. Chem. Bull., 2004, 1871–1888.
12. Z. Yinghuai, Boron-Based Hybrid Nanostructures: Novel Applications of Modern Materials, in Hybrid Nanomaterials: Synthesis, Characterization, and Applications, eds. B. P. S. Chauhan, John Wiley & Sons, Inc., 1st edn, 2011, pp. 181–198.
13. D. Naoufal, B. Grüner, P. Selucký, B. Bonnetot and H. Mongeot, J. Radioanal. Nucl. Chem., 2005, 266, 145–148.
14. K. E. Yorov, A. P. Zhdanov, R. K. Kamilov, A. E. Baranchikov, G. P. Kopitsa, O. I. Pokrovskiy, A. L. Popov, O. S. Ivanova, L. Almásy, Y. G. Kolyagin, K. Y. Zhizhin and V. K. Ivanov, ACS Appl. Nano Mater., 2022, 5, 11529–11538.
15. A. Gigante, L. Duchêne, R. Moury, M. Pupier, A. Remhof and H. Hagemann, ChemSusChem, 2019, 12, 4832–4837.
16. L. Jacob, E. Rzeszotarska, M. Koyioni, R. Jakubowski, D. Pociecha, A. Pietrzak and P. Kaszyński, Chem. Mater., 2022, 34, 6476–6491.
17. R. Bernard, D. Cornu, D. Luneau, D. Naoufal, J.-P. Scharff and P. Miele, J. Organomet. Chem., 2005, 690, 2745–2749.
18. H. C. Miller, N. E. Miller and E. L. Muetterties, Inorg. Chem., 1964, 3, 1456–1463.
19. S. A. Jasper, J. Mattern, J. C. Huffman and L. J. Todd, Polyhedron, 2007, 26, 3793–3798.
20. J. F. Kleinsasser, E. D. Reinhart, J. Estrada, R. F. Jordan and V. Lavallo, Organometallics, 2018, 37, 4773–4783.
21. A. M. Spokoyny, C. D. Lewis, G. Teverovskiy and S. L. Buchwald, Organometallics, 2012, 31, 8478–8481.
22. D. Naoufal, B. Bonnetot, H. Mongeot and B. Grüner, Collect. Czech. Chem. Commun., 1999, 64, 856–864.
23. S. A. Jasper, R. B. Jones, J. Mattern, J. C. Huffman and L. J. Todd, Inorg. Chem., 1994, 33, 5620–5624.
24. E. Rzeszotarska, I. Novozhilova and P. Kaszyński, Inorg. Chem., 2017, 56, 14351–14356.
25. R. T. Paine and H. Noth, Chem. Rev., 1995, 95, 343–379.
26. D. C. Pestana and P. P. Power, J. Am. Chem. Soc., 1991, 113, 8426–8437.
27. Y.-F. Shen, C. Xu and L.-J. Cheng, RSC Adv., 2017, 7, 36755–36764.
28. S. Bouquillon, A. du Moulinet d’Hardemare, M.-T. Averbuch-Pouchot, F. Hénin and J. Muzart, Polyhedron, 1999, 18, 3511–3516.
29. C. Amatore, M. Azzabi and A. Jutand, J. Am. Chem. Soc., 1991, 113, 8375–8384.
30. C. Amatore, E. Carré, A. Jutand, H. Tanaka, Q. Ren and S. Tori, Chem. – Eur. J., 1996, 2, 957–966.
31. C. Amatore and A. Jutand, J. Organomet. Chem., 1999, 576, 254–278.
32. C. Amatore and A. Jutand, Acc. Chem. Res., 2000, 33, 314–321.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2456501–2456506. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5dt01302j

---