PROFESSIONAL, SCIENTIFIC INTERESTS
My current scientific interests are connected with the collective properties of complex systems, studied from the point of view of their magnetic properties. The non-linear behavior of hysteretical systems is a clear sign of the complex phenomena occurring at a lower level. I worked along the years on the following topics:

Magnetic Thin Films
We produce magnetic thin films and multilayers DC/RF sputtering. The films obtained can be subjected to successive steps of photolitography.
One of the main characterization phases is the magnetooptical investigation, which can be quantitative (magnetooptical hysteresis measurement) or qualitative (domain observation by high resolution Kerr effect / Faraday effect)
Garnet films were investigated with the Faraday optical bench. The interest of this study lies in the investigation of a very complex domain structure: from the stripe domain structure, exhibiting labirint-like patterns, to the bubble domain structure. Avalanches of all sizes are observed in the stripe domain state, triggered by small variations of the external field.
References: [Par96,Mag00,Ver00,Ver01,Ver03]

Non transparent thin films must instead be studied with the high resolution Kerr effect, which allows an important characterization of the magnetization process and its correlation to the film structure.
References: [Pas05-1,Pas05-2,Pas06,Cel07-1,Cel07-2,Coi08,Pas08,Chi08,Cel08, Pas09-1, Pas09-2, Pas10-1, Mag10-1, Boa09]

The use of fast CCD cameras allowed to investigate the properties of domain wall displacement in disordered media, correlating it to the Barkhausen effect.
References: [Zap03-1,San04,San06,Dur08-1, Mag09]

By using a very fast, strobed camera, we are able to reach a time interval of 50ps between frames, investigating the magnetization behaviour at very high excitation frequencies.
References: [Mag10-2,Mag11-1]







Barkhausen Effect in ferromagnetic materials. The Barkhausen Effect is one of the most evident marks of complex low-level behavior in the magnetic systems. It appears as a sequence of stochastic jumps of the magnetization value along the hysteresis loop. These jumps arise from the irreversible domain wall motion, triggered by local instabilities. It has been studied both from the experimental point of view, and with computer simulations based on a stochastic model.
References: [Ber94,Dur95-1,Dur95-2,Dur95-3,Dur96]





The 1-degree of freedom domain wall motion (ABBM) model. This remarkably simple model of hysteresis starts from the definition of a random, one-dimensional energy landscape over which a domain wall moves under the action of an applied field. Despite its simple description, the results that one can obtain - using different, possibly stochastic, definitions of the energy - are very interesting.
Starting from this model, the functional integration approach to hysteresis is a very general approach that allows to describe hysteresis as a property arising from the features of the energy landscape over which the domain wall moves. Its connections with the ABBM and the Preisach models hasve been clearly identified.
References: [Mag96,Bea97,Mag99-2,Ber01-2,Ber99-1,Ber99-2]













The investigation of Landau-Lifschitz-Gilbert dynamics under circularly polarized magnetic fileds can give remarkably simple results when the magnetic system possesses a symmetrical shape and anisotropy.
Further studies it was observed how the methods employed in the discussed investigations could be applied in the study of devices employing spin polarized currents, such as spin valves
References: [Ber01-1,Ber01-3,Ber01-4,Mag01,Mag02,Mag03,Ber05-3,Ber05-1,Ber05-4,Ber05-5]



















The Random Field Ising Model (RFIM). The Hamiltonian of this very simple model is the two dimensional Ising system hamiltonian, but with an additional term, representing a "noise" term (random field) at each spin site. As a consequence, we do not observe the typical square hysteresis loops of the Ising systems, but loops which range from the square loop to the paramagnetic loop, according to the intensity of the random field. According to the evolution rules of the system, as the external field changes, we observe along the loop sudden magnetization changes, when cluster of neighbour spins flip to the opposite magnetization value, exactly as we can observe Barkhausen jumps in real materials.
It has been investigated the structure of the set of metastable states of the model, and its connection to the property of reachability by applied field histories.
The study of RFIM lead to the connection between garnet films hysteresis loops / domain structure, and Ising models, namely the Dipolar Random Field Ising model, or DRFIM.
In this model a dipolar field is added to the RFIM Hamiltonian: the result is the possibility to obtain stripe domain structures and garnet hysteresis loops - a remarkable result, never before obtained.
References: [Mag99-1,Mag00,Dan02,Zap02,Zap03-2,Col04,Bas04,Ala05,Mag05,Bas06,Bor07-1,Ber07, Bor08-2, Bor10-1]





Other topics covered in these years: