University of Minnesota
School of Physics & Astronomy

Physics and Astronomy Colloquium

Thursday, April 9th 2009
4:00 pm:
Speaker: Albert Fert, 1Unité Mixte de Physique CNRS/Thales, 91767 Palaiseau, and Université Paris-Sud
Subject: Carbon nanotubes, graphene, molecules : promising materials for spintronics
Note different day for the colloquium, this week only. Refreshments served in Room 216 Physics after colloquium

The usual materials of classical spintronics are magnetic and nonmagnetic metals, magnetic and nonmagnetic semiconductors and, for tunnel junctions, insulating materials like MgO or alumina. However, nowadays, promising results begin to be obtained with a new family of materials which includes carbon nanotubes, graphene and several types of magnetic or non-magnetic molecules. The general advantage of carbon-based materials is mainly their long spin lifetime related to the small spin-orbit coupling of carbon, but, as we will see, the very high electron velocity of some of them is also of great interest for spintronics. The first part of the talk will be an introduction on classical spintronics and a review of what can be done with molecular materials for TMR, spin transport in lateral structures, magnetic switching or microwave generation by spin transfer. In the second part of the lecture I will focus on the general problem of spin transport in a nonmagnetic lateral channel between a spin-polarized source and a spin-polarized drain, a structure which is at the basis of several concepts of logic devices or spin transistors. The main difficulty is related to the transformation of the spin information – related to the magnetic configuration of the electrodes- into a large electrical signal, ideally Δ V/V ≈ 1 or larger, if V is the bias voltage and Δ V some voltage variation induced by a change of the magnetic configuration. In experiments on structures in which the lateral channel is a metal or a semiconductor, Δ V/V does not exceed a few 1% and the electrical signal Δ V is generally in the μ V range. In contrast, in the experiments on carbon nanotubes between ferromagnetic contacts we will present, high values of Δ V/V ( above 70%) and large Δ V (of the order of 100 mV) can be obtained. After a description of the theoretical background, we will discuss the origin of the difficulties for semiconductors and explain why large values of Δ V/V and Δ V can be easily obtained with carbon nanotubes. We will emphasize the potential of carbon nanotubes, graphene and other molecules for spintronics, and conclude by presenting some next challenges for molecular spintronics.

The weekly calendar is also available via subscription to the physics-announce mailing list, and by RSS feed.