Restricted Lie algebras via monadic decomposition (Q1671755)

The paper proves that one can construct a functor \(\Lambda: \mathfrak{M}_2 \to \mathsf{Lie}_p\) such that \((P_2\widetilde{u}, \Lambda)\) defines an equivalence between \(\mathsf{Lie}_p\), the category of restricted Lie algebras over a field of characteristic \(p\) [\textit{N. Jacobson}, Trans. Am. Math. Soc. 42, 206--224 (1937; JFM 63.0873.03)], and \(\mathfrak{M}_2\), the Eilenberg-Moore category of the monad \(P_1 \widetilde{T}_1\) on \(_{P\widetilde{T}}\mathfrak{M}\), being \(\widetilde{T} : \mathfrak{M} \to Bialg(\mathfrak{M})\); \(\mathfrak{M}\) denotes the category of \(\mathbb{K}\)-vector spaces; \(P : Bialg(\mathfrak{M}) \to\mathsf{Lie}_p\) is the functor that computes the space of primitive elements of any bialgebra; \(P_1\) is the comparison functor associated to the monad \(P\widetilde{T}\) and \(\widetilde{T}_1\) is a left adjoint to \(P_1\). \(\widetilde{u} : \mathsf{Lie}_p \to Bialg(\mathfrak{M})\) is the functor computing the restricted universal enveloping algebra of a restricted Lie algebra, namely \(\widetilde{u}(L,[-,-], -^{[p]}) = \frac{\widetilde{U}(L,[-,-])}{I}\), where \(I\) is the ideal of the universal enveloping algebra \(\widetilde{U}(L,[-,-])\) generated by the elements of the form \(x^p - x^{[p]}\); \(P_2\) is a full and faithful right adjoint to the functor \(\widetilde{T}_2 : \mathfrak{M}_2 \to Bialg(\mathfrak{M})\) given in [\textit{A. Ardizzoni} et al., Appl. Categ. Struct. 23, No. 1, 93--105 (2015; Zbl 1310.18002)]. The main result establishes: \(\bullet\) The functor \(P\) is comparable. There is a functor \(\Lambda : \mathfrak{M}_2 \to \mathsf{Lie}_p\) such that \(\Lambda \circ P_2 = P\) and \(H_{\mathsf{Lie}_p} \circ \Lambda = U_{0,2}\). \(\bullet\) The adjunction \((\widetilde{T}_1, P_1)\) is idempotent. The functor \(P\) has a monadic decomposition of monadic length at most two. \(\bullet\) The pair \((P_2\widetilde{u}, \Lambda)\) is an adjoint equivalence of categories. An alternative approach to the proof of the main theorem by means of adjoint squares is provided in the final section of the paper.