MESA+ University of Twente
Research Business & Innovation About MESA+ Education

Mohand Talanana (promotion date: 12 July 2006)

Spin transport from first-principles: metallic multilayers and a model spin-valve transistor


Promotion Date: 12 July 2006

Mohand Talanana

Principally in this work I focused on the development of an efficient and flexible method which uses no adjustable parameters to study the electronic transport in complex and inhomogeneous nanostructures, e.g. hybrid systems combining magnetic, nonmagnetic and/or semiconductor materials. We are not only interested in calculating the quantities (conductances or resistances) that define the transport of electrical current, but we also want to know how this is happening, particularly analysing the details of “scattering” of electrons and how does this affects these transport properties.

What was your thesis about?

Principally in this work I focused on the development of an efficient and flexible method which uses no adjustable parameters to study the electronic transport in complex and inhomogeneous nanostructures, e.g. hybrid systems combining magnetic, nonmagnetic and/or semiconductor materials. We are not only interested in calculating the quantities (conductances or resistances) that define the transport of electrical current, but we also want to know how this is happening, particularly analysing the details of “scattering” of electrons and how does this affects these transport properties. Over the last two decades the size of the engineered devices is continuously shrinking down to nanoscale dimensions. This has given rise to the discovery of new physical effects, such as giant magnetoresistance (GMR), and exotic properties. This lead to a huge technological

impact. These properties were exploited for storage devices (hard disk drives in computers), magnetic memories, magnetic field sensors, ... The characteristic length scale of these devices requires that the correct description of electronic, magnetic and transport properties is done within a full quantum-mechanical treatment. In this approach, the motion of the electron is governed by the Schrödinger equation. The electrons are the carriers of the electrical current. The quantum theory is a framework which describes a particle as being a wave and a particle at the same time with a mass and a charge. Besides these two properties, the electron possesses a spin (an intrinsic angular momentum) having two projections, it can be up or down. The origin of magnetism is due to the spin of the electron. The populations of electrons with spin up and spin down projections in iron (Fe) are different, hence Fe is naturally magnetic and so are Cobalt (Co) and Nickel (Ni). Transport properties observed in metallic multilayers (sandwiches of thin films alternating magnetic and nonmagnetic metals) displaying the GMR effect are closely related to the magnetism of the above mentioned materials and their alloys. The

manipulation of this property of the electron and the right choice of the materials for making nanostructures will add substantially more capability and performance to electronic devices. Based on the electronic structure and nature of materials, magnetic and transport quantities are calculated and predictions are made.

What was your specific angle ?

You need efficient tools and flexible methods which can easily be adapted to various scenarios and materials to understand what is “happening” as far as the motion of the electron is concerned. How do the electrons behave in this specific materials and conditions? This behaviour gives rise to the various properties of materials mentioned earlier. I was involved in a substantial development of a transport code, written exclusively at the CMS group, to calculate transmission and reflection matrices which are the central quantities to study electronic transport. This code is flexible and efficient such that realistic structures, similar to experiments, can be easily handled. If, for instance, you stick materials together there are always imperfections at the contact during the process of growing one material on top of the other. Experimentally little is known though about the morphology at the contact of grown materials. The presence of these defects can have a large influence on the electrical and magnetic response of the devices. A rigorous description of the properties of such devices should take into account these imperfections.

In our research we try as best as we can to link the experimental world with our calculations.With our results and predictions a better insight is brought to help experimentalist as well as theorists to have a better understanding of the the physical phenomena. Moreover, a very quantitative agreement between theory and experiment is usually obtained. Sometimes with our calculations we make predictions of properties that have not been observed yet, particularly we have obtained a large anisotropy in the electrical conductance where it is the least expected if not at all, Al/Ag & Al/Au. Our results have been the subject of numerous oral and poster presentations locally as well as around the world.

Where have you been?

We have been invited to America and Japan and all over Europe to present the research that has been carried out at the CMS group. We have various collaborations, Chinese academy of science, Michigan State university, Delft, ... and Twente. I myself gave oral presentations or hang up poster presentations in various European countries and the US.

Can you mention one phenomenon that you have predicted and not observed yet in an experiment?

As mentioned earlier we obtained a large anisotropy in the electrical conductance in Al/Ag & Al/Au materials. It is unexpected from the scientific community since these materials until now were described using the “free electron model” which is isotropic. A calculation based on the realistic “band structure” of these materials had revealed this anisotropy. We have made a proposal how can this be observed experimentally using the so-called Andreev reflection technique.

Did you write publications about this?

Yes, writing publications is a major part of our work. It is the barometer of the research that has been done or ongoing. We had some publications that appeared in Physical Review B, Applied Physics Letters, European Physical Journal B and the renowned Physical Review Letters.

I do expect that in your field you have a close collaboration with experimentalists?

Yes, that is so. You cannot do it on your own. You need a strong interaction with other groups, experimentalists as well as theorists. We have a very nice and fruitful interaction with the group of Prof. Cock Lodder and Dr. Ronnie Jansen of the SMI group.

Where are you from?

I am from Algeria where I have done my Master. This was carried out in close collaboration with IPCMS-Strasbourg University, in France. I was awarded two scholarships there for short periods of almost four months each. It was in fact a very good preparation for the PhD work I am doing now – which (my Master) was on magnetism in transition metals and their alloys. I applied to a position that professor P.J. Kelly and Mesa+ advertised and was invited for a short visit of three days. I gave a presentation of my Master work and got to know what is done in the group of Dr. Ronnie Jansen and Prof. Dave Blank. At the end I was offered a PhD position to study electronic transport in hybrid nanostructures.

What did you like most about your research?

Besides broadening my knowledge in science, I liked very much the international environment where I worked, the interaction with various disciplines of research either by talking to people, attending conferences and meetings, or simply going out together and sharing moments of distraction, and sporting too. I also liked the fact that I have done something as my humble contribution to the field of magnetotransport, which field has a direct impact on our daily life. For example, have a quick look back at hard disk drives history in computers. Few years back people were talking about megabytes storage, or at most one gigabyte, but nowadays they talk about terabytes of storage and memory. Allow me to put this proposition: “Electrons have spin as well as charge, and this may make all the difference in future electronics”.

I don’t understand, it all sounds very fundamental to me.

Yes, that may be so, but the outcome of the measurements, calculations and predictions is of great practical value. They tell you what materials are expected to perform nicely, which one may be to use and what currents, voltages or magnetic fields to apply for a specific and desired outcome. It's all a matter of tuning, desire and what do you want to achieve.

Did you like your stay in the Netherlands?

(Laughs) Now everybody expects me to say something negative about the weather or

something like that. But I was not really bothered by that. When I arrived here in Enschede from Strasbourg by train and bus, I did not even realize I was in the Netherlands. Everybody spoke English with me. The openness of people and the atmosphere in the group really was a major factor that made me decide to do a PhD here.

For the summary of the thesis, click here. (In English, French, Netherlands)