Charge, Spin and Thermal transport in nanostructures

Charge transport properties of materials at the macro- and micro-scale are well described assuming that electrons can be treated much like classical particles in a pinball machine instantaneously bouncing off each other. In contrast, quantum effects become dominant at the nano-scale and a full quantum mechanical description, which accounts for the wave-like nature of electrons, is required. In collaboration with the Computational Spintronics Group we develop the Smeagol software, which combines Density Functional Theory with the non-equilibrium Green’s functions scheme for quantum transport. Smeagol is used to compute current-voltage characteristics of nano-devices such as single-molecule junctions [1, 2].

Beside charge transport we also investigate spin transport. The spin is the electron intrinsic angular momentum, which has two possible states, called up and down. It can be transported either by spin-polarized charge currents or by pure spin-currents. Spin-polarized charge currents result from an unbalance number of up-spin and down-spin conduction electrons. In contrast, pure spin-currents emerge when up-spin electrons and down-spin electrons move towards opposite directions. Our research employs Smeagol to study spin-polarized currents and pure spin-currents across a number of nano-devices [3]. We predict effects such as the giant and the tunnelling magnetoresistance, which are used in sensing technologies and in the read-heads of hard-disk drives. Moreover, we are interested in the so-called spin-charge conversion effects, where charge currents are converted into either spin currents or spin-polarized electrons populations in materials with large spin-orbit coupling. In particular we study the basic mechanisms for spin-charge conversion effects at interfaces, where they results drastically enhanced. Spin-charge conversion effects enable the control of the electron spin with no need for a magnetic field. They have been widely explored for information storage and processing devices.

Electrons moving through a material carry also energy resulting in heat transfer. When a voltage is applied to a device, heat is transferred from one side to the other, creating a temperature difference. Conversely, a temperature difference across a device creates a voltage. These are generally called thermoelectric effects. We study thermoelectricity in nano-scale systems, in particular molecular junctions [4], by using the quantum theory for electron transport and the Smeagol code. Yet, we note that electrons are not the only energy carries. Energy is also transported by crystal vibrations and, in magnetic materials, by spin-waves, that are propagating disturbances of the align electron spins. Our research plan aims at extending Smeagol to compute energy transport in nanoscale devices including the combined contribution of electrons, vibrations and spin-waves.

[1]A.V. Rudnev, V. Kaliginedi, A. Droghetti, H. Ozawa, A. Kuzume, M.-aki Haga, P. Broekmann, I. Rungger, Science Advances 6, e1602297 (2017).

[2]F. Bejarano, I.J. Olavarria-Contreras, A. Droghetti, I. Rungger, A. Rudnev, D. Gutiérrez, M. Mas-Torrent, J. Veciana, H.S.J. van der Zant, C. Rovira, E. Burzurı́, N. Crivillers, Journal of the American Chemical Society 5, 1691 (2018).

[3]Non-equilibrium Green’s Function Methods for Spin Transport and Dynamics, I. Rungger, A. Droghetti, M. Stamenova, in Handbook of

[4]Materials Modeling: Methods: Theory and Modeling, Springer International Publishing (2018).
A. Droghetti and I. Rungger, Physical Chemistry Chemical Physics 22, 1466 (2020).

Charge and spin transport in organic materials

Organic materials (molecular crystals, polymers) can be fabricated by cheap chemical methods on flexible substrates for disposable, flexible and wearable electronic devices. At the same time, organic materials are generally considered ideal media for spintronics owing to their long spin-relaxation time, which means that a spin-polarized electron population remains so for a long time (up to 1 s). In spite of this, experiments are often hard to interpret. Charge and spin transport in organics have been theoretically described in terms of two fluids [1]. The charge current is due to electrons hopping between molecular units or impurities.

Concurrently, unpaired electrons can be trapped at molecular units, which therefore present a magnetic moment. The spin current is eventually enabled by the exchange coupling between these magnetic moments [2] and is carried by spin-waves. Charge and spin propagate through different regions as two fluids with different diffusion constants. Recent experiments claimed some evidence for this two-fluid behaviour. Our research program develops a computational multiscale approach to simulate charge and spin-transport in organics. The goal is understand how the two fluids emerge in real materials and, in particular, what the key material properties affecting spin transport are. Ultimately, we aim at introducing conceptually new ways to enhance and control the spin-current paving the way for new organic-based spintronic devices.

[1]A. Droghetti and S. Sanvito, Physical Review B 99, 094413 (2019).

[2]A. Droghetti, Journal of Magnetism and Magnetic Materials 502, 166578 (2020).

Correlated systems

Electron correlation in quantum systems is a measure of how much the motion of one electron is influenced by the presence of all the other electrons. Common metals (Au) or semiconductors (Si) are considered weakly correlated: an electron can be described as a free particle affected only by an average electrostatic potential generated by all other electrons. However, there are many materials, called strongly correlated, where this picture is not valid anymore. Electrons avoid each other by occupying excited states or remaining far apart. Electron correlation is responsible for many properties of materials and phenomena such as (anti)ferromagnetism and superconductivity.

We develop theoretical methods to describe correlated materials and nanostructures. In particular, we follow three different research directions. First, we improve or develop novel schemes within Density Functional Theory (DFT). These include self-interaction corrections [1] and constrained-DFT [2] applied to systems with states, which are localized in space and, as such, large electron correlation effects emerge to keep electrons apart. Second, we employ Quantum Monte Carlo techniques where the many-electron quantum mechanical wave-function is stochastically sampled. We have studied how the electron correlation inside the d orbitals of transition metal ions determine their magnetic moments [3]. Finally we combine Density Functional Theory with the so-called Dynamical Mean-Field Theory (DMFT). We carried out the implementation of DMFT in Smeagol [4, 5]. The goal is to understand how electronic correlation affects the charge and spin transport properties of nano-junctions.

[1]A Droghetti, CD Pemmaraju, S Sanvito, Physical Review B 78, 140404 (2008).

[2]A. Droghetti, I. Rungger, C.D. Pemmaraju, S. Sanvito, Physical Review B 93, 195208 (2016).

[3]M. Fumanal, L.K. Wagner, S. Sanvito, A. Droghetti, Journal of Chemical Theory and Computation 12, 4233 (2016).

[4]A Droghetti, I Rungger, Physical Review B 95, 085131 (2017).

[5]L. Chioncel, C. Morari, A. Östlin, W.H. Appelt, A. Droghetti, M.M. Radonjić, I. Rungger, L. Vitos, U. Eckern, A.V. Postnikov, Physical Review B 92, 054431 (2015).