Noise and sensitivity of GMI sensing elements

International audienceIncreasing the intrinsic sensitivity of magnetic sensing elements requires theoretical models of their electromagnetic response as a function of relevant parameters, such as bias and excitation fields, material properties and geometry. These models are particularly useful when they yield simple expressions for the sensitivity (V/T) and noise spectral density (V/√Hz), two critical metrics directly related to the development of highly sensitive magnetometers. While we have developed such models for the noise [1] and the sensitivity [2] of giant magnetoimpedance (GMI) microwires and started to test their experimental validity [3], further work is needed in order to fully explore the interrelations between the design parameters and to extract and validate some useful design rules.Here, we review and refine our initial models of noise and sensitivity and systematically explore the consequences of these models on the design of GMI sensing elements. Specifically, we review the effect of excitation fields, material properties and geometric parameters and highlight the trade-off involved on various performance metrics of the sensors. Special emphasis is placed on the difficulties related to the experimental determination of the relevant physical properties of soft magnetic microwires. We propose an experimental procedure for addressing some of these difficulties. Finally we analyze different avenues of improving the performance of GMI sensors, distinguishing between the causes related to ideally perfect wires and those related to defects and imperfections of real wires. Our most recent set of data lead us to believe that there is still significant room for improving the equivalent magnetic noise density and that reducing the material imperfections is currently one of the main challenges for achieving this goal