Cellularity of diagram algebras as twisted semigroup algebras. (Q875073)

The main result of the paper under review is the following statement: Theorem. Let \(R\) be a commutative ring with unit and \(S\) be a finite semigroup with an antiinvolution \(*\). Assume that every \(\mathcal D\)-class of \(S\) contains an idempotent, fixed by \(*\), and that the group algebra (over \(R\)) of the maximal subgroup, corresponding to this idempotent, is cellular. Then for every associative \(R\)-twisting of the multiplication of \(S\), compatible with \(*\) and the cellularity of the maximal subgroups (the letter is a lengthy technical condition), the twisted semigroup algebra of \(S\) over \(R\) is cellular. Examples include, in particular, partition, Brauer and Temperley-Lieb algebras and their cyclotomic analogues.