Modelling and fabrication of magnetic field sensors based on nanostructured material

Recently, there has been a growing interest towards the use of magnetic micro/nanoparticles in biomedicine for the labelling of biological species, in order to open new ways in drug delivery, biological analysis and cancer treatment via hyperthermia. These magnetic nanostructures are physically and chemically stable and permit potentially non-invasive detection and manipulation methods; furthermore, their dimensions can be tuned in the production phase, and their surface can be both coated with gold or polymers to increase biocompatibility, and functionalized with drugs that can be directly delivered in the required tissue. The use of magnetic micro/nanoparticles in in-vitro diagnostics requires the development of high-sensitivity magnetic field sensors that can be integrated in lab-on-chip systems. The best candidates for this objective are low magnetic field sensors based on nanostructured materials, like recent magnetoresistive nanodevices and miniaturized Hall effect sensors. The aim of this Thesis, which has been developed at the Istituto Nazionale di Ricerca Metrologica (INRIM), in Torino, is to study such kind of application, focusing on two types of nanostructured magnetic field sensors. The first type is represented by graphene Hall bars with micrometer size; graphene has been chosen due to its electronic properties and the possibility of reducing to nearly zero the space between sensor surface and the magnetic object to be detected, thus enabling the enhancement of measured signal. The devices here considered are made by using two different techniques for graphene production: chemical vapour deposition (CVD) on copper films and thermal decomposition of silicon carbide (SiC). Despite the different fabrication processes, these two materials show similar electronic properties. In the first part of this research activity, micrometre-sized Hall bars are produced using commercial CVD graphene. These devices have been experimentally characterized by measuring their electronic properties (carrier density and electron mobility). Furthermore, in order to investigate the reliability of graphene in sensing applications, the Hall bars noise spectra have been measured via cross-correlation method. Noise is one of the most limiting factors for the detection of low magnetic fields, but the measured flicker noise contribution is small enough to permit the realization of ultra-sensitive graphene Hall sensors. Then, Hall devices based on epitaxial graphene grown on SiC have been simulated using a comprehensive finite element model, in order to study their feasibility in the detection of a single 1 m size magnetic bead (commercial DynaBead). The model has been validated by comparison to experimental results obtained at National Physical Laboratory (NPL), Teddington (UK), within the Quantum Detection Group. The results have shown a high voltage response by using an ac-dc detection technique. Furthermore, the main micro-structural heterogeneities of epitaxial graphene (multi-layer islands and terraces) have been included in the model in order to study their influence on the sensor output. The numerical model has been then used to simulate graphene Hall sensors for the calibration of custom-made multilayered-Magnetic Force Microscopy (ML-MFM) probes, which are composed of two magnetic thin layers separated by a non-magnetic interlayer. The two layers can have quasi parallel (ferromagnetic state) or anti-parallel (anti-ferromagnetic state) magnetization. The magnetic moment of the ML-MFM probes is investigated via Scanning Gate Microscopy (SGM) imaging of graphene Hall bars, mapping the Hall voltage consequent to the application of the ML-MFM probe stray field. The SGM maps, measured at NPL, within the Quantum Detection Group, have been numerically reconstructed approximating the probe as a two-dipole system. A parametric study is performed to find the dipole parameters that guarantee the best fit with the experimental results. This methodology can lead to several possible sets of parameters, depending on the distance between the approximating dipoles and sensor surface. As an alternative approach, the starting values of the magnetic dipole parameters have been obtained by FFT processing of MFM images of a magnetic reference sample, measured with the ML-MFM probes at the Leibniz Institute for Solid State and Materials Research (IFW) in Dresden (Germany). Finally, a new type of magnetic particle sensor based on a permalloy antidot array thin film has been studied, by means of a micromagnetic numerical model previously developed at INRIM. The idea is to exploit the shifts in the frequencies of the Ferromagnetic Resonance (FMR) peaks of the antidot array to detect the presence of magnetic particles immobilized on the film surface. The sensing element is first biased with an external field and then excited with a Gaussian pulse orthogonal to it. The consequent magnetization time evolution is analyzed with FFT processing to obtain FMR spectra. The FMR peaks obtained with and without particles have been compared in order to calculate the frequency shifts caused by particle perturbation. The study includes the influence of the type of particle (commercial MagSignal and DynaBead), and the role of antidot array lattice features. Further analysis is going to be performed in order to obtain engineered antidot arrays with optimized sensing properties.