Absence of Evidence ≠ Evidence of Absence: Statistical Analysis of Inclusions in Multiferroic Thin Films

Assertions that a new material may offer particularly advantageous properties should always be subjected to careful critical evaluation, especially when those properties can be affected by the presence of inclusions at trace level. This is particularly important for claims relating to new multiferroic compounds, which can easily be confounded by unobserved second phase magnetic inclusions. We demonstrate an original methodology for the detection, localization and quantification of second phase inclusions in thin Aurivillius type films. Additionally, we develop a dedicated statistical model and demonstrate its application to the analysis of Bi(6)Ti(2.8)Fe(1.52)Mn(0.68)O(18) (B6TFMO) thin films, that makes it possible to put a high, defined confidence level (e.g. 99.5%) to the statement of ‘new single phase multiferroic materials’. While our methodology has been specifically developed for magnetic inclusions, it can easily be adapted to any other material system that can be affected by low level inclusions.