Cationic congeners captured

Main-group atoms often come in small clusters whose nuclearity and charge are governed by electron-counting rules. P₄ is arguably the most well-known of these clusters, but several other tetrahedra such as P₃As and [P₃Ga]²⁻ are isolable. Given the diversity of neutral and anionic 20-valence-electron tetrahedra, there is a striking paucity of cationic derivatives, only a handful of which have been observed, and even then only in the gas phase. A team led by Ingo Krossing has now addressed this by describing in Chemical Science the solution-phase generation and NMR characterization of three cationic congeners [P₃E]⁺ (E = S, Se, Te).

Credit: David Schilter/Springer Nature Limited

An early glimpse of heteroatomic tetrahedral cations came almost half a century ago with the observation of [Pn₃E]⁺ (Pn = P, As; E = S, Se) as fragment ions in mass spectra of Pn₄E₃. These ions, which can be viewed as group 15 tetrahedra in which one atom is replaced with a group 16 cation, were rediscovered by Krossing’s team in mass spectra of Ag⁺–P₄E₃ mixtures. This prompted the team to find a scalable and rational synthesis of [P₃E]⁺ (E = S, Se, Te), which necessitated the E^iv cations [Cl₃E]⁺ as precursors. These strong electrophiles are only compatible with weakly coordinating anions such as the aluminates the team had previously developed. The aluminates are suitably ‘inert’ and their salts are soluble in CH₂Cl₂, a good solvent for nucleophiles such as [Nb(OAr)₃(P₃)]⁻ — the source of P₃³⁻ that had previously been combined with Cl₃As to prepare P₃As. “The [Nb(OAr)₃(P₃)]⁻ reagent got chlorinated by [Cl₃E]⁺ to give PCl₃, but when we instead treated [Cl₃E]⁺ with P₄ we got a substitution reaction that delivered the [P₃E]⁺ target and Cl₃P as major products.”

Having generated their [P₃E]⁺ tetrahedra, Krossing and colleagues found that their ³¹P NMR resonances follow the trend P₄ < [P₃S]⁺ < [P₃Se]⁺ < [P₃Te]⁺, with the latter two ions giving rise to unusually high ⁷⁷Se and ¹²⁵Te chemical shifts for E^iv species. The team rationalized these shifts in terms of resonance structures in which chalcogen cations E⁴⁺ are magnetically shielded by aromatic P₃³⁻ rings. This model is consistent with the large and negative nuclear-independent chemical shifts calculated for the tetrahedra — values that are diagnostic of aromaticity at the centre of a cluster. In this regard, the new cations are similar to P₄, which does not have six 2-centre 2-electron bonds and four lone pairs but instead has a more delocalized electronic structure, as evidenced by P₄ undergoing protonation at its edges rather than its vertices. The four tetrahedra have similar frontier molecular orbitals, although those calculated for [P₃E]⁺ lie at lower energies than those for P₄ on account of charge, and there is some lifting of degeneracy on shifting from T_d to C_3v symmetry. The differences between P₄ and [P₃E]⁺ are more marked when it comes to reactivity — the cations are potent electrophiles with large calculated F⁻ affinities, so it is not surprising that they had eluded synthesis for so long.

when we instead treated [Cl₃E]⁺ with P₄ we got a substitution reaction that delivered the [P₃E]⁺ target

Despite these encouraging results, the story of [P₃E]⁺ continues to be written. Indeed, Krossing’s team got [P₃E]⁺ and Cl₃P but could not isolate the former as a salt because of a competing reaction that affords [Cl₂P₅]⁺ and Cl₂E₂. The team is on the lookout for more selective synthetic routes, including the replacement of [Cl₃E]⁺ with [F₃E]⁺, even stronger electrophiles that would afford [P₃E]⁺ along with volatile and easily removable F₃P. As is so often the case with reactive cations, the challenge here would be to find suitable anions. For example, the anions for which [F₃E]⁺ salts are known are incompatible with the [P₃E]⁺ target. “We do have ideas to continue and are working on this, but this is difficult chemistry and will take some time,” anticipates Krossing.