Strain intramolecular crowding

A possible explanation of these differences is that the reactions of the trivalent phosphorus compounds with carbonyl compounds may contain reversible and irreversible steps. The condensation of one mole of biacetyl with one of the phosphite probably contains several reversible steps leading eventually to the 1 1 oxyphosphorane. If this 1 1 adduct is of high energy as a result of ring strain and/or intramolecular crowding in the TBP, it might not be observable. The carbon-carbon condensation step in the formation of the 2 1 adducts is probably essentially irreversible under most conditions hence, once that step is achieved, there is an opportunity for isolation of the 2 1 phosphoranes. 

---