Partial trace ideals and Berger's conjecture (Q2667900)

Let \(k\) be a perfect field and \(R\) be a reduced local \(k\)-algebra of dimension one. There is a long standing conjecture of R.W. Berger that \(R\) is regular if and only if the universally finite differential module \(\Omega_{R/k}\) is torsion-free. The paper under review proves Berger's conjecture when \(h(\Omega_{R/k})\) is sufficiently small, where \(h\) of the \(R\)-module \(M\) is defined by \[h(M)=\min\{\lambda (R/J)\mid \exists \text{ homomorphism }f:M\to R\text{ such that } J=f(M)\},\] and \(\lambda(.)\) represents the length of an \(R\)-module. Any ideal \(J=f(M)\) with \(\lambda(R/J)=h(M)\) is called a \textit{partial trace ideal} of \(M\). The inspiration for the notion of \(h(M)\) came from Scheja's proof of the Berger conjecture in the graded case. (Recall that the trace ideal of \(M\) is \(\sum \alpha(M)\) as \(\alpha\) roams over all of \(\operatorname{Hom}_R(M,R)\). This explains the name partial trace ideal.) It is shown that if \(R\) is a Gorenstein domain, then \(h(M)\le \lambda (R/\mathfrak C)\), for all \(M\), where \(\mathfrak C\) is the conductor ideal \(R:\bar R\) of \(R\) in the integral closure \(\bar R\). Instructions are given for computing \(h(\Omega_{R/k})\) by computer and the results of a few such computations are included.