Lie commutators in a free diassociative algebra (Q2220277)

A diassociative algebra is a vector space with two bilinear associative operations \(\dashv\) (left product) and \(\vdash\) (right product) satisfying the following identities \[a \dashv (b \dashv c) = a \dashv (b \vdash c), \ (a \dashv b) \vdash c = a \dashv (b \vdash c),\ (a \dashv b) \vdash c = (a \vdash b) \vdash c .\] There are two versions of Lie brackets for diassociative algebras: \begin{itemize} \item[\(\bullet\)] \([a, b] := a \dashv b- b \vdash a\) (Leibniz bracket or dicommutator) \item[\(\bullet\)]\(a \blacklozenge b := a \dashv b - a \vdash b\) \end{itemize} Let \(X\) be a set and \(DiAs(X)\) be a free diassociative algebra on a set \(X\). Let \(a_1, \ldots , a_n \in X\) and \(a = a_1 \ldots a_n\) be a monomial in \(DiAs(X)\) with some placement of parantheses and choice of operations. It is know, that each \(a = a_1 \ldots a_n\) can be re-written in a normal form \(a = (a1 \vdash \cdots \vdash a_{i-1}) \vdash a_i \dashv (a_{i+1} \dashv \cdots \dashv a_n).\) We write \(a_1 \ldots a_{i-1}\widehat {a_i}a_{i+1} \ldots a_n\) as a normal form of \(a\). In order to formulate the Leibniz criterion, the authors are introducing so-called Dynkin map \(D\) on \(DiAs(X).\) The map \(D : DiAs(X) \to DiAs(X)\) is the linear map, defined on the base elements as follows \begin{itemize} \item[\(\bullet\)] \(\widehat {a_1}a_2 \cdots a_n \to [[\cdots [a_1, a_2]\cdots ], a_n]\) \item[\(\bullet\)] \(a_1a_2 \cdots a_{i-1} \widehat{a_i}a_{i+1}\cdots a_n \to -[[\cdots [[a_i, [[\cdots [a_1, a_2] \cdots ], a_{i-1}]], a_{i+1}] \cdots ], a_n]\) if \(i > 1.\) \end{itemize} An element of \(DiAs(X)\) is called Leibniz element (or Lie di-element) of \(DiAs(X)\) if it can be expressed by elements of X in terms of Leibniz brackets. The main result of the present paper is the following theorem. \textbf{Theorem 1.} Let \(f \in DiAs (X)\) of degree \(n\). Then \(f\) is a Leibniz element of \(DiAs(X)\) if and only if \(D(f) = nf.\) Let \(D^{(\blacklozenge)}(X)\) be a subalgebra of \(DiAs(X)\) under the product \(\blacklozenge\) generated by set \(X.\) Below we have the criterion from the present paper for elements of \(DiAs(X)\) to be in subalgebra \(D^{(\blacklozenge)}(X).\) \textbf{Theorem 2.} Let \(f \in DiAs(X)\) of the form \(f = \sum\limits_{i=1}^n \lambda_i a_1\cdots \widehat { a_i} \cdots a_n.\) Then \(f \in D^{(\blacklozenge)} (X)\) if and only if \(\sum\limits_{i=1}^n \lambda_i = 0.\)