Phosphacycles as Building Blocks for Main Group Cages

Abstract

As part of an ongoing interest in phosphorus/carbon compounds [1] we have focused on the reactivity and synthetic utility of possible sources of electrophilic PX 2 + and AsX 2 + fragments with a view to accessing novel main group architectures. It has been suggested [2] that there is an analogy between these reactive pnictenium species and carbenes. For example, Krossing and co-workers reported [3] that PX 3 (X = Br, I) reacts in CH 2 Cl 2 at À78 8C with Ag[Al(OR F ) 4 ] (R F = C(CF 3 ) 3 ) and P 4 to give the novel phosphorus rich molecules [P 5 X 2 ][Al(OR F ) 4 ] through a process thought to involve the insertion of the electrophilic carbenoid species [PX 2 ][Al-(OR F ) 4 ] into a PÀP bond of P 4 . Allied to this interesting observation is the report by Weigand and co-workers [4] that P 4 reacts, possibly by a similar reaction pathway, in a melt of Ph 2 PCl and GaCl 3 (1:1 at 60 8C) to form a structurally related molecule [P 5 Ph 2 ][GaCl 4 ]. Similarly, Burford and co-workers have used related methodology to prepare an extensive series of catena-phosphorus cations. [5]In our initial studies we focused on the reaction of 2,4,6tri-tert-butyl-1,3,5-triphosphabenzene (1) with latent sources of PX 2 + and AsX 2 + cations. It is interesting to note that 1 reacts with the Arduengo-type carbene 1,3,4,5-tetramethylimidazol-2-ylidene to give a ring-contracted 1,2,4-triphosphole, [6] whereas the silicon analogues, stable bis-(amine)silylenes, afford (4+1) cycloaddition products. [7] It has also been found experimentally that in the case of the reaction of 1 with sources of H + , CH 3 [9] h 1 -P-bonded cationic adducts are formed.As a first step we used density functional theory (see Supporting Information) to compare the relative thermodynamic stabilities of possible products which might result from the reaction of 1 with EX 3 (E = P, As ; X = Cl, Br) and the Lewis acid GaCl 3 . These calculations indicated that under thermodynamic control, a PX 2 + electrophile could react with 1 to give a h 1 -P-bonded adduct. For the phosphenium reagents, alternative products arising from electrophilic attack at carbon, ring contraction (analogous to the product from the reaction with an Arduengo carbene [6] ) or 1,2-addition are less favorable. In contrast, for the arsenium systems, although h 1 -As coordination is preferred, the energy difference between the alternative products is small. We can in fact envisage a possible reaction pathway to such a h 1 -P/Asbonded cation where the reaction of ECl 3 (E = P or As) with GaCl 3 leads to the direct formation of the ion-separated electrophilic species [ECl 2 ] + [GaCl 4 ] À , which can then be captured by 1 (Scheme 1 a). However, there is an alternative and perhaps more plausible pathway not involving a free ECl 2 + cation and that is through a synchronous S N 2 (P/As) push-pull (GaCl 3 ) process as illustrated in Scheme 1 b. [10] This alternative route can be related to the proposal by Wild, Radom, and co-workers that phosphenium cations can be readily transferred between unsaturate...

A Novel N,P,C Cage Complex Formed by Rearrangement of a Tricyclic Phosphirane Complex: On the Importance of Non‐covalent Interactions

“…Early milestones in P,C cage chemistry are represented by tetraphosphacubane I 3 and hexaphosphapentaprismane II 4 (Scheme ), both obtained by phosphaalkyne cyclooligomerization 5. Other synthetic strategies for P,C cage structures include cycloaddition reactions of phosphaalkynes with cyclobutadienes6 ( III ), thermolysis of 1,2,3‐tris(pentamethylcyclopentadienyl)cyclotriphosphene7 ( IV ) and reactions of triphosphabenzene with phosphenium salts8 ( V ). P,C cage structures containing other heteroatoms directly bound to phosphorus,9 such as oxygen10 and nitrogen,11 are also known.…”