Proton transfer vs. oligophosphine formation by P–C/P–H σ-bond metathesis: decoding the competing Brønsted and Lewis type reactivities of imidazolio-phosphines

Abstract

Cationic imidazolio-phosphines show two-sided reactivity towards bases, undergoing either Brønsted-type proton transfer to imidazolio-phosphides or autocatalytic Lewis acid/base reaction cascades to yield P-free imidazolium ions and oligophosphines.

Reactions of Imidazolio‐Phosphides with Organotin Chlorides: Surprisingly Diverse

“…[15] However, the applied protocol entailing metalation of the PH-bond and subsequent salt metathesis with an organotin dichloride seems unamenable to imidazolio-derivatives A as the PH-bond is here effectively protected by the enhanced acidity of the CH-bonds at the imidazole ring, which redirects the attack of the metalating agent to the heterocycle. [16] Anticipating that the alkylation-deprotonation approach used to introduce alkyl or phosphinyl groups [3,[7][8][9] would allow to avoid this obstacle, we studied the reactions of imidazoliophosphides 1 a,b with Me 3 SnCl and Ph 3 SnCl in the presence of tertiary amines. To our surprise, we did not obtain any evidence of the expected substitution products, but observed the formation of mixtures containing substantial amounts of soluble imidazolium salts [2 a,b]Cl and intractable precipitates previously recognized as insoluble polyphosphorus species.…”

“…To our surprise, we did not obtain any evidence of the expected substitution products, but observed the formation of mixtures containing substantial amounts of soluble imidazolium salts [2 a,b]Cl and intractable precipitates previously recognized as insoluble polyphosphorus species. [7] 31 P NMR studies disclosed that tristannyl phosphine 3 [17] was formed as main [18] or even only detectable soluble phosphoruscontaining product with Ph 3 SnCl (Scheme 2) while both 4 [19] and tristannyl heptaphosphine 5 [20] were established as the products of the reaction of 1 a with Me 3 SnCl. Remarkably, the reactions proceeded in the same way, but even more cleanly, when the tertiary amine was omitted.…”

“…Having noticed that the successful stannylation of a metalated dihydroimidazolio-phosphide had in fact been carried out with an organotin dichloride, [15] we also explored the reactions of 1 a,b with equimolar quantities of Me 2 SnCl 2 and Ph 2 SnCl 2 in the presence of various proton scavengers. Anion bases (BuLi, KHMDS, LDA), which had been effectively employed in the alkylation of imidazolio-phosphides, [7] reacted in this case non-specifically, presumably because unwanted substitution at SnÀ Cl functionalities could not be suppressed. Isolation of any products was unfeasible, and these attempts were abandoned.…”

“…Nominal primary imidazolylidene-phosphinidenes A [1][2][3][4] have caught attention because of their multifaceted reactivity towards Brønsted acids and electrophiles. [5] Depending on the conditions, A may react via attachment of an electrophile to produce cationic phosphines B, [6][7] via PÀ H bond functionalization to afford substitution products C, [3,[7][8][9] or via condensation under cleavage of an imidazolium salt to furnish cyclic oligophosphines D, [7] respectively (Scheme 1). This varied reactivity can be rationalized in the framework of a valence bond description invoking canonical formulae A'-A''' as principal resonance structures.…”

“…The cations are then intermediates in the generation of C and D by alkylation/deprotonation or autocatalytic cascades of alkylation and intermolecular condensation steps. [7] Because their reactions reflect typical reactivity patterns of phosphides R 2 P À , we suggested addressing compounds A as zwitterionic imidazolio-phosphides. [7] This view is also in accord with the fact that a typical double bond reactivity of A is largely unknown, indicating that phosphaalkene structure A' (identified as the leading resonance structure in the description of parent A [10] ) appears to have little impact on the chemical behavior.…”