3,2,1 and stop! An innovative, straightforward and clean route for the flash synthesis of metallacarboranes

Abstract

[Co(C₂B₉H₁₁)₂]⁻, analogs to metallocene, can be readily synthesised in high yield by a fast and clean solvent-free reaction of [HNMe₃][C₂B₉H₁₂] with CoCl₂·xH₂O. This innovative approach, applied to both ortho and meta isomers, yields the desired structures by simply heating the solid compounds to high temperature for a very short time.

---

Metallacarboranes are 3D boron hydride clusters that incorporate metal atoms or units in their polyhedral skeletons.¹ They find many applications in materials, sensors/biosensors and medicine, among others.² Metallacarboranes [M(C₂B₉H₁₁)₂]⁻ (M = Fe³⁺, Co³⁺), which were reported in 1965 ³ are the most studied. [M(C₂B₉H₁₁)₂]⁻ were synthesised from the nido [C₂B₉H₁₂]⁻ cluster in two steps: (i) first, deprotonation of [C₂B₉H₁₂]⁻ produces nido [C₂B₉H₁₁]²⁻ which then (ii) reacts with metal-containing reagents yielding stable anionic metallacarboranes.³ The insertion of metal into the nido [C₂B₉H₁₂]⁻ ion is extremely versatile for two reasons: (i) the nido [C₂B₉H₁₂]⁻ ion withstands harsh reaction conditions (e.g. strong bases or reflux temperatures) and (ii) completion of the icosahedron is, in general, strongly thermodynamic and kinetic favoured. The synthesis of [M(C₂B₉H₁₁)₂]⁻ has been reported in both non-aqueous and aqueous media by using NaH in dry THF or a freshly prepared hot 40% NaOH aqueous solution, respectively, to deprotonate the H_bridge of the nido-cluster.⁴ The complexation reaction with MCl₂ required the reaction mixture to be stirred at reflux temperature under nitrogen for 3 hours. In both methods the work-up to isolate the pure [M(C₂B₉H₁₁)₂]⁻ is tedious.⁴ A one-pot reaction method in anhydrous DME that uses K[t-BuO] as a base and anhydrous MCl₂ was reported to produce [M(C₂B₉H₁₁)₂]⁻ derivatives in high yield requiring much less work-up (Scheme 1a).⁵ This route is, however, time consuming and implies dealing with DME, a solvent with a high boiling point and hence, difficult to remove from the reaction flask.

Scheme 1

Solvent-free transformations are an essential approach to the sustainability of organic and organometallic synthesis⁶ through the 21st century due to its efficiency in minimizing waste.^7,8 In this work, the new expeditive syntheses of [M(C₂B₉H₁₁)₂]⁻ (M = Co³⁺) and their C_c-substituted derivative compounds have been established by exploring the reactivity of nido [HNMe₃][C₂B₉H₁₂] with CoCl₂·xH₂O under solvent-free conditions in the absence of bases at high temperature during a very short time (minutes). The main objective was to design an easy, one-pot route to synthesise the pristine [Co(C₂B₉H₁₁)₂]⁻ and their C_c-substituted derivatives in a high yield in a rapid and efficient reaction.

The solvent-free reaction between nido [HNMe₃][7,8-C₂B₉H₁₂] and CoCl₂·xH₂O (x = 0, 6) was performed by heating the two solid reagents at 350 °C (also at 250 °C) into a Pyrex tube. As displayed in Table 1, several parameters such as temperature, reaction time, nature of the CoCl₂ reagent (anhydrous or hydrated) as well as the CoCl₂/[NHMe₃][C₂B₉H₁₂] molar ratio were studied.

Table 1 Representative parameters for the complexation reaction

Cluster	Entry	Conditions		Molar ratio CoCl2/[HNMe3][nido-C2B9H12]	Yield %
		Time (min)	T (°C)
a Yield related to the cobaltabis(dicarbollide) mixture.
	1	3	350	1.5 CoCl2 anh.	83
	2	5	350	1.5 CoCl2 anh.	85
	3	7	350	1.5 CoCl2 anh.	88
	4	7	350	1.5 CoCl2·6H2O	90
	5	8	350	1.5 CoCl2·6H2O	88
	6	2	350	1.5 CoCl2 anh.	42
	7	3	350	1.5 CoCl2 anh.	46
	8	3	470	1.5 CoCl2 anh.	68
	9	5	350	1.5 CoCl2 anh.	75
	10	10	350	1.5 CoCl2 anh.	90
	11	2 + 6	350 + 470	1.5 CoCl2·6H2O	87
	12	2 + 6	350 + 470	1.5 CoCl2·6H2O	88
	13	2 + 6	350 + 470	2.5 CoCl2·6H2O	94
	14	2 + 6	350 + 470	2.5 CoCl2 anh.	60
	15	10	350	5.5 CoCl2 anh.	83
	16	2	350	2.5 CoCl2 anh..	82^a
	17	8	250	2.5 CoCl2 anh.	68^a
	18	8	250	2.5 CoCl2·6H2O	78^a
	19	2	350	2.5 CoCl2 anh.	68^a
	20	8	250	2.5 CoCl2 anh.	62^a
	21	8	250	2.5 CoCl2·6H2O	70^a
	22	8	470	2.5 CoCl2 anh.	50
	23	5	350	2.5 CoCl2 anh.	40
	24	10	350	2.5 CoCl2 anh.	47
	25	15	350	2 CoCl2 anh.	52
	26	15	350	2.5 CoCl2 anh.	74

---

Upon completion of the reaction, water was added to the crude product before extracting with ethyl ether. Surprisingly, the formation of cobalt metal was not observed. This was typical in the wet methods as Co(II) dismutates to Co(III) and Co.⁴ The orange organic phase was separated from the pink aqueous phase, the solvent removed in vacuum, and the residue taken up in water. This solution was treated with an aqueous solution of Cl[NMe₄] to give the less soluble [NMe₄][3,3′-Co(1,2-C₂B₉H₁₁)₂] complex, which was confirmed by ¹¹B- and ¹H NMR spectra. The yield provided in Table 1 is related to the [NMe₄]⁺ cation in all experiments.

As demonstrated in entries 3 and 4, no significant effect on final yield was observed when using anhydrous CoCl₂ with respect to the CoCl₂·6H₂O.

Once the synthesis of the pristine [3,3′-Co(1,2-C₂B₉H₁₁)₂]⁻ complex was achieved, we tested if the synthesis of C_cluster-substituted cobaltabis(dicarbollide) complexes could be accomplished. Reaction was run by using the same method and conditions but starting with the monosubstituted [HNMe₃][8-R-7,8-C₂B₉H₁₀] (R = Me, Ph) ligands.

This new reaction led to the formation of [NMe₄][3,3′-Co(1-Me-1,2-C₂B₉H₁₀)₂] and [NMe₄][3,3′-Co(1-Ph-1,2-C₂B₉H₁₀)₂] sandwich with yields equal to 90% after 10 min and 94% after 8 min, respectively (entries 10 and 13). It was noticed that the presence of one methyl or aryl group linked to the C_cluster decreases the reaction's yield; however, very good yields were still obtained in only 10 minutes of heating in comparison with the traditional method that takes around 1440 minutes to give rise to the same compound, but with a yield difference of approx. 5%. Time of reaction also has a positive impact on the final yield. As illustrated in Table 1, entries 7 and 9, the yield rises significantly with increase in reaction time, namely from 3 to 5 minutes. Furthermore, if temperature is increased by 120 °C (entries 7 and 8), a consequently boost in yield of about 22% is observed. Reactions with CoCl₂·xH₂O (x = 0, 6) have also been conducted at 250 °C just extending the reaction time 6 extra minutes with comparable yields to these in the table.

Based on these utmost positive results, the complexation reactions of disubstituted nido units ([HNMe₃][7-R-8-R′-7,8-C₂B₉H₁₀] (R = R′ = Et; R = Me, R′ = Et and R = Ph, R′ = Et)) were investigated. Our approach was shown to work with both [HNMe₃][7-Me-8-Et-7,8-C₂B₉H₁₀] and [HNMe₃][7,8-Et₂-7,8-C₂B₉H₁₀], leading to [Co(C₂B₉H₁₁)₂]⁻ derivatives with yields above 70%. However, the ethyl units were lost in part (vide infra). On the other hand, the linkage of the aryl substituent to the C₂B₃ face in the presence of the ethyl group hinders the formation of the complex. Therefore, our next step was to study the effect of the lone pair of electrons on the complexation reactions. Starting from closo 1-R-2-R′-1,2-C₂B₁₀H₁₀ (R = R′ = SEt; R = Me, R′ = SEt and R = Ph, R′ = SEt), via a partial deboronation reaction, nido ligands [HNMe₃][7-R-8-R′-7,8-C₂B₉H₁₀] (R = R′ = SEt; R = Me, R′ = SEt and R = Ph, R′ = SEt)⁹ were synthesised (see ESI†).

Out of these three compounds, only [HNMe₃][7-Me-8-SEt-7,8-C₂B₉H₁₀], yielded 50% of the target complex, which confirms the inhibition of the solid state reaction when hindrance and/or electronic alterations exist (entry 22). Nevertheless, the yield of the [HNMe₃][3,3′-Co(1-Me-2-SEt-C₂B₉H₉)₂] was low compared to the previously obtained complexes. The yield for this particular [HNMe₃][7-Me-8-SEt-7,8-C₂B₉H₁₀] ligand using traditional methods was of 79% with K[t-BuO] and 35% when NaH was employed.⁵ Having obtained such good results and satisfactory yields by using this new solid state complexation reaction on ortho–nido species, the behaviour of the meta–nido isomer, [HNMe₃][7,9-C₂B₉H₁₂], was our target.

The [NMe₄][2,2′-Co(1,7-C₂B₉H₁₁)₂] complex was obtained in 74% yield after heating at 350 °C for 15 min. The meta-isomer needed more time to produce a comparable good yield to the ortho-isomer. Table 1 summarises the different guidelines used to find the optimal parameters for all obtained sandwich complexes. The direct isolation of the complexes having cobalt in the formal +3 oxidation state was not accompanied by the formation of cobalt metal. This observation suggests that the initially Co²⁺ is rapidly oxidized to form the more stable d⁶ metallabis(dicarbollide) complex 18 e⁻ count avoiding reduction from Co²⁺ to Co⁰. This is proof that the free-solvent synthesis of [3,3′-Co(1,2-C₂B₉H₁₁)₂]⁻ takes place via another mechanism than the previously reported synthesis in solution.

To discern on the reaction pathway, several studies were run. The solvent-free reaction's progress was controlled by the colour change of the mixture as well as liberation of gas. To determine the identity of the evolved gas, a water solution monitored by a pH meter was prepared. As the pH of this aqueous solution did not change, the generated gas was not NMe₃, but probably hydrogen. The next study was to measure the pH of the aqueous phase of the extraction process, which provided a value of 6.33 suggesting that an acid had been generated. Both experiments along with the observation that no Co⁰ was produced and the ¹¹B NMR evidence that [3,3′-Co(1,2-C₂B₉H₁₁)₂]⁻ was obtained, agree with the balanced equation displayed at Scheme 1(b). The solvent-free synthetic reaction takes place by an oxidation–reduction reaction that involves the two redox couples Co³⁺/Co²⁺ and H₂O/OH⁻. Moreover, X-ray diffraction of [HNMe₃][2,2′-Co(1,7-C₂B₉H₁₁)₂] fully supports that the cation is [HNMe₃]⁺ (Fig. 1).¹⁰

Fig. 1 Perspective view of the [HNMe₃][2,2′-Co(1,7-C₂B₉H₁₁)₂] unit with 50% ellipsoids. Selected bond lengths (Å): C1–B3=C7–B3 = 1.727(7) Å, C1–B6=C7–B11 = 1.683(7) Å, B11–B6 = 1.706(7) Å.

The MS spectra of [NMe₄][Co(1-Me-2-Et-1,2-C₂B₉H₉)₂] and [NMe₄][Co(1,2-Et₂-1,2-C₂B₉H₉)₂] are clear in showing that some of the ethyl groups are lost/won during the formation of the cobaltabis(dicarbollide) cluster from the nido components and CoCl₂(xH₂O), x = 0, 6. We do not know at what stage of the complex building the transfer of ethyl units occurs, but it is clear that at 250 °C some of the ethyl units are lost or won. Our interpretation is that at some stage of the building process β-hydride elimination occurs, (ESI†) that is consistent with NMR spectra indicating co-existence of some [Co(C₂B₉H₁₁)₂]⁻ derivatives.

Conclusions

A new, fast and environmentally-friendly solid state reaction for the syntheses of cobaltabis(dicarbollides) is presented. Our approach is a significant improvement on the traditional syntheses in solution in both speed of reaction and generated yield. We demonstrate that the [Co(C₂B₉H₁₁)₂]⁻ building reaction works well with starting plain, single or double C_cluster-substituted nido clusters. However, care has to be taken when β-hydride elimination may take place because in this case the resulting [Co(C₂B₉H₁₁)₂]⁻ derivative may be different to the expected one based on the precursor nido cluster due to the β-hydride elimination, as commonly occurs in organometallic chemistry. The presence of substituents bearing a free pair of electrons also influences the complex formation. Meta-isomers also give rise to the corresponding cobaltabis(dicarbollide), but in this case, the reaction time needs to be increased. A suggested mechanism of the complexation reaction is proposed, based on identifying the chemical nature of the evolved gas, the pH of the mixture, the crystal structure of the target complex and the absence of Co²⁺ dismutation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work has been supported by MINECO (CTQ2016-75150-R), Generalitat de Catalunya (2017 SGR 1720).

Notes and references

1. R. N. Grimes, Carboranes, Elsevier Inc., New York, 3^rd edn, 2016; C. E. Housecroft, Cluster Molecules of the p-Block Elements, Oxford University Press Inc., New York, 1994.
2. N. S. Hosmane and R. Eagling, Handbook of Boron Science with applications in Organometallics, Catalysis and Medicine (4 Vols.), World Scientific Publishing Europe Ltd., London, 2018; R. Ziólkowski, A. B. Olejniczak, L. Górski, J. Janusik, Z. J. Leśnikowski and E. Malinowska, Bioelectrochemistry, 2012, 87, 78; T. García-Mendiola, V. Bayon-Pizarro, A. Zaulet, I. Fuentes, F. Pariente, F. Teixidor, C. Viñas and E. Lorenzo, Chem. Sci., 2016, 7, 5786; E. Hey-Hawkins and C. Viñas, Boron-Based Compounds: Potential and Emerging Applications in Medicine, John Wiley & Sons Ltd., 2018.
3. M. F. Hawthorne, D. C. Young and P. A. Wegner, J. Am. Chem. Soc., 1965, 87, 1818; M. F. Hawthorne and T. D. Andrews, J. Chem. Soc., Chem. Commun., 1965, 443.
4. M. F. Hawthorne, D. C. Young, T. D. Andrews, D. V. Howe, R. L. Pilling, A. D. Pitts, M. Reintjes, L. F. Warren Jr. and P. A. Wegner, J. Am. Chem. Soc., 1968, 90, 879.
5. C. Viñas, J. Pedrajas, J. Bertran, F. Teixidor, R. Kivekäs and R. Sillanpää, Inorg. Chem., 1997, 36, 2482.
6. G. Kaupp, Top. Curr. Chem., 2005, 254, 95; K. Tanaka and F. Toda, Chem. Rev., 2000, 100, 1025; G. W. V. Cave, C. L. Raston and J. L. Scott, Chem. Commun., 2001, 2159; G. Rothenberg, A. P. Downie, C. L. Raston and J. L. Scott, J. Am. Chem. Soc., 2001, 123, 8701; V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, Chem. Commun., 2002, 1606; V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, J. Am. Chem. Soc., 2002, 124, 6244.
7. R. Noyori, Chem. Commun., 2005, 1807; M. Avalos, R. Babiano, P. Cintas, J. L. Jiménez and J. C. Palacios, Angew. Chem., Int. Ed., 2006, 45, 3904; M. Jansen and J. C. Schön, Angew. Chem., Int. Ed., 2006, 45, 3406.
8. R. D. Rogers and K. R. Seddon, Science, 2003, 302, 792.
9. M. F. Hawthorne, D. C. Young, P. M. Garrett, D. A. Owen, S. G. Schwerin, F. N. Tebbe and P. A. Wegner, J. Am. Chem. Soc., 1968, 90, 862; C. Viñas, J. Pedrajas, J. Bertran, F. Teixidor, R. Kivekäs and R. Sillanpää, Inorg. Chem., 1997, 36, 2482; C. Viñas, J. Pedrajas, F. Teixidor, R. Kivekäs, R. Sillanpää and A. J. Welch, Inorg. Chem., 1997, 36, 2988.
10. Crystal data for C₇H₃₂B₁₈CoN: monoclinic, space group C2/m, a = 9.130(11) Å, b = 10.446(10) Å, c = 11.101(9) Å, V = 1014.44 ų, Z = 2, T = 100(2) K, measured reflections 24 871, independent reflections 856, R_int = 0.0325, R₁ = 0.0677, wR₂ = 0.1743. CCDC 1882757.†.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis characterization of the cobaltabis(dicarbollide) complexes and X-ray studies of [HNMe₃][2,2′-Co(1,7-C₂B₉H₁₂)₂]. CCDC 1882757. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c8gc03943g

---