On the homology of completion and torsion (Q2015169)

Let \(A\) be a commutative ring and \(I\) a weakly proregular ideal (for instance any ideal if \(A\) is noetherian). Let \(\Lambda_I\) be the \(I\)-adically completion functor in the category of \(A\)-modules, and \(\Gamma_I\) be the \(I\)-torsion functor. In this paper their (left and right resp.) derived functors \textbf{L}\(\Lambda_I\) and \textbf{R}\( \Gamma_I\) are studied, proving among other results that they induce an equivalence of categories between the category of cohomologically \(I\)-torsion complexes (the full subcategory of the derived category of \(A\)-modules consisting in complexes \(C\) such that the canonical map \textbf{R}\( \Gamma_I C \to C\) is an isomorphism) and the category of cohomologically \(I\)-adically complete complexes (defined similarly with \textbf{L}\(\Lambda_I\)). While most of the results in this paper were proved previously (some of them under slightly different hypotheses) in papers by \textit{L. Alonso Tarrío, A. Jeremias} and \textit{J. Lipman} [Ann. Sci. Éc. Norm. Supér. (4) 30, No. 1, 1--39 (1997; Zbl 0894.14002)] (even in the context of quasi-compact separated schemes), \textit{W. G. Dwyer} and \textit{J. P. C. Greenlees} [Am. J. Math. 124, No. 1, 199--220 (2002; Zbl 1017.18008)], and \textit{P. Schenzel} [Math. Scand. 92, No. 2, 161--180 (2003; Zbl 1023.13011)], the authors of the paper under review believe their proofs are ``quite transparent''. The authors have also posted an erratum on \url{arXiv:1506.07765} correcting the proof of a theorem.