*arrow_back*Back

## ETSF Young Researchers' Meeting 2014: evolution of ab-initio methods for condensed matter - connection with experiments and industry

#### University of Rome La Sapienza - Italy

#### Organisers

*FOR IMPORTANT PRACTICAL INFORMATION, PLEASE VISIT:*

http://etsf_yrm2014.roma2.infn.it/home.html

**DEADLINE** for application and abstract submission:** 7th MARCH 2014**

The European Theoretical Spectroscopy Facility (ETSF) [1] (formerly known as NANOQUANTA) is a network carrying out state-of-the-art research on theoretical and computational methods for studying ground state, electronic and optical properties of materials. This network gathers the experience and the know-how of more than 200 researchers, including professors, postdocs and Ph.D. students, in Europe and in the U.S.A., facilitating collaborations and rapid transfer of knowledge. The main objective of ETSF is to provide knowledge and expertise in the field of theoretical spectroscopy and to broaden access to them across the public and private sector.

Starting from 2004, the young researchers within the ETSF have organized, in different European cities, the so-called Young Researchers' Meeting (YRM) [2-8].

This is an annual conference, organized and attended only by Ph.D. students and postdocs. Apart from these young researchers, who organize the meeting, the "supporting" organizers, that are experienced professors of the network, assist and guide the organizing committee.

The main purpose of the YRM is to offer a precious chance to young researchers to interact, to start new collaborations, to exchange ideas and new techniques and to improve their presentation, communication and organizational skills. The importance of these aspects in the formation of a scientist can be gathered from the fact that many previous young researchers have obtained excellent positions in prestigious institutions and private companies.

It is noteworthy that the YRM meetings are not funded by ETSF directly, but only indirectly. Most participants are employed in ETSF projects and the conference fee is supported from the project budget. In this sense, external funding is of paramount importance for organizing this conference. Nonetheless, the conference is highly regarded by the ETSF senior staff and considered an excellent opportunity for promoting the ETSF values and for extending the network.

In this network, Rome, through the Condensed Matter Theory Group (CMTG) of University of Tor Vergata, is a historically important node, among the founders of ETSF. Within the CMTG, Prof. Dr. Olivia Pulci leads the ab-initio simulation section [9]. She is also the coordinator of the ETSF Optics Beamline.

Rome is the city that will host the YRM conference in 2014. The meeting will be held at La Sapienza University, (hosting the italian CECAM node), one of the most prestigious and ancient universities in Italy [10].

During the years, La Sapienza hosted many outstanding physicists (such as Enrico Fermi and Edoardo Amaldi) and also nowadays the level of the scientific research activity is very high: as an example the Boltzmann medal in 2013 was won by Giovanni Jona-Lasinio, professor in La Sapienza university.

Then the location of the meeting offers an important opportunity to the participants to come in direct contact with the italian educational system and scientific life, favouring collaborations and exchanges of ideas in a stimulating environment.

As ETSF is a large network with researchers of more than 15 nationalities, hosting the 11th edition of the YRM conference is also a great opportunity for Rome and in particular for La Sapienza University. It is a unique chance for social, cultural and scientific exchange. So both sides can greatly benefit from this purposeful experience.

The University of Rome La Sapienza hosts every year many other conferences and it is well prepared to deal with these events, having special agreements with hotels, restaurants, catering companies and a canteen. All this will facilitate the organization and will help to reduce the costs considerably. La Sapienza University, through the administrative secretary of the Physics' department, has already approved the use of a conference room with projection equipment for the YRM 2014. The involved costs of the room and the projection equipment are hence supported by University La Sapienza.

The neighborhood of the university is not touristic, it is mainly inhabited by students and it is cheaper than other parts of Rome. At the same time it is in a central location, near the main train station, and it is then very well connected to both the main airports of Rome.

The YRM that will be organized in Rome in 2014 will have two main very important novelties: the first one is the introduction of an experimental-industrial day.

The organization of an "industry day" with speakers from various Research & Development (R&D) departments was successfully introduced in the 9th edition of YRM, with the final goal of encouraging future collaborations and giving the opportunity to initiate new projects of industrial interest.

Starting from this idea, we decided to integrate the industrial day with the inclusion of experimental theme talks. The main objective is to underline and strengthen communication between experimentalists and theoreticians, an essential link to exploit in the best possible way all the potentialities of research activities in condensed matter physics.

Besides showing their work, experimental speakers will be encouraged to highlight the possible fields of their research that could deeply benefit from a comparison between experiments and theoretical computational simulations. One of the ETSF goals is to describe, from a theoretical point of view, experimental results, in order to formulate predictions for new experiments. For this reason the ETSF has always been opened to collaborations with experimentalists and it has always supplied theoretical support to them. It is important that this activity continues, encouraging the youngest members of the network to communicate and start new collaborations with experimentalists from both private and academic sectors.

The second novelty is the addition of a half-day meeting with some of the best students coming from High Schools in Rome. This meeting will be organized in the last day of the conference.

In organizing this event we have the support of Prof. Arturo Marcello Allega who, besides being the principal of one of the largest high schools in Rome, is also very active in the implementation and development of national projects related to education. The main objective of this activity is, in accordance with the mission of the ETSF, to spread the meaning and the importance of the scientific research in condensed matter physics, in particular related to ab-initio simulations.

This is a fabulous chance for young Italian students and, as well, for YRM participants. Indeed, once more, young researchers will have the opportunity to improve their presentation and scientific communication skills, answering the questions of the students and explaining the importance of their research to a general audience.

The increasing interest in the study of nanoscale electronic phenomena is motivated by the imperative of constructing faster and smaller electronic devices able to work in a broad set of environments. This new research field, denoted as molecular electronics, includes, for example, electron transport through nanowires, single molecules or organic molecular structures. Future applications of the basic concepts for building electronic devices aim to very fast switching, in order to build, for example, faster computing units. The ultimate goal of this research field is to replace the conventional semiconductor devices by structures with components at the molecular level. Commercially available transistors, e.g., in computers, flat plasma screens, mobile telephones etc., have reached nowadays the thickness of approximately 350 atoms. A further reduction of about 50 to 70 atoms would still be possible with the present techniques, but beyond this miniaturization, a radical change in the device functionality is required. While the electronic components available today can be well described by (semi-)classical physics, in a molecular device one must account from its fabrication to its working principles for the quantum effects.

Understanding the functionality of such devices, or the factors contributing to their stability, is a great scientific challenge. The two main directions are (i) the understanding of the electronic structure of the nanoscale materials and (ii) the understanding of the fundamental processes taking place on very short time scales. Understanding the electronic structure (within the field of theoretical optical spectroscopy) of materials would help designing new materials in order to fit the requirements of stability and functionality needed in fast switching processes. On the other hand, understanding the fundamental processes is of crucial importance for designing the functionality of such devices. This conference aims to bring together the understanding and the know-how from both research directions, in order to better identify the problems and the challenges.

In order to reach this goal, we will set up the following theoretical sessions:

- Ground state properties

- Photoemission

- Optical properties

- Electron-phonon interactions

- Quantum transport

- New theoretical methods

GROUND STATE PROPERTIES

The Born-Hoppenheimer approximation is the starting point of most theoretical condensed matter simulations. Many of these simulations concern the calculation of ground state properties (for example forces, formation energies, geometries). Density Functional Theory [1,2] is well-known to be successful in reproducing and predicting ground state properties of solid state matter, ranging from surfaces to bulk crystals and molecules. In most cases the comparison with experiments is excellent: structural details of a wide variety of solids and molecules are predicted within an error of 1% compared to the experiments [3]. Many other properties such as vibrational spectrum, bulk modulus and dielectric constants can be accurately calculated.

Nevertheless, the Born-Hoppenheimer approximation, substantially consisting in neglecting the quantum behavior of material nuclei, is not a good approximation for metals and for systems in which light atoms (like hydrogen) play an important role (ice, water). For this reason, nuclear quantum effects were introduced in simulation of water in a work by D. Marx et al. [4], using the path integral formulation of quantum mechanics. The computational cost of this simulation discouraged the application of the method to more complex systems. Nevertheless Michele Ceriotti et al. [5] in 2009 set up a method, based on a non-Markovian Langevin equation, to include quantum corrections to the classical dynamics of ions in a quasi-harmonic system. They found that results in agreement with path-integral methods can be obtained using only a fraction of the computational effort. This is the reason why Michele Ceriotti has been invited to give a talk on ground state properties.

PHOTOEMISSION

One of the main purposes of theoretical spectroscopy is to be able to reproduce and interpret experiments, in order to deeply understand and describe a large variety of systems.

Many experimental techniques, such as photoemission spectroscopy (XPS, UPS) or inverse photoemission spectroscopy (IPES) deal with charged electronic excitations.

Density Functional Theory is generally not appropriate to describe this kind of excitations, even if using ad-hoc exchange-correlation functionals can improve the description of excited states properties.

One of the most suitable approaches to study excited state properties of semiconductors, insulators and molecular systems [6] is the Many Body Perturbation Theory (MBPT) based on Green's functions.

In particular, the GW approximation [7] for the Self-Energy is the most widely used and reliable. A recent work [8], not yet published, by Falk Tandetzky et al., showed that the GW equation gives a multiplicity of solutions and discussed which of them are physical ones.

Falk Tandetzky has been invited to give a talk about his specific work and, more generally, GW approximation.

OPTICAL PROPERTIES

Neutral electronic excitations are strictly related to optical properties. These are among the most complex and difficult properties to describe and predict by ab-initio methods. The greatest difficulty is to evaluate the contribution of the electron-electron and electron-hole interaction in the shape and intensity of an optical spectrum.

In condensed matter community the most used methods for optical properties simulations are based on MBPT (by the solution of the Bethe-Salpeter equation [9]) or on Time Dependent Density Functional Theory (TDDFT) [10]. Both theories are exact, but their application in a real many-electron interacting system requires several approximations. The suitability in using one or the other theory depends on geometrical and chemical system properties and on the kind of information that one would like to extract from these calculations.

The extreme difficulty in performing these calculations considerably reduced the number of possible applications of many body perturbation theory to fluorescence, that is very important to understand the chemical structure of a system and it is of the most measured properties in experiments. There are only two applications in literature, both by S. Louie [11,12], to very small systems.

In order to give an overview of neutral excitations, we invited Daniele Varsano who is an international expert in many body perturbative techniques, TDDFT, and applied them to a great variety of systems ranging from condensed matter to biological physics.

ELECTRON-PHONON INTERACTION

The electron-phonon coupling is the main responsible of most of the amazing phenomena occurring in material science, going from superconductivity to piezoelectric materials, and from its relations with electron transport to termo-electric effects, such as the spin-seebeck effect [13] whose mechanism is based on the magnon-phonon interaction. Most of the theoretical studies of these physical processes are based on phenomenological approaches. Although these approaches qualitatively well describe the nature of these phenomena, prediction and design of new materials still remain a great challenge. This is the reason why the use of ab-initio calculations for the study of the role of the electron-phonon coupling on physical properties of condensed matter could induce a great improvement in this research field. A very important step in this direction has been done by M. Cardona [14] and more recently by E. Cannuccia and A. Marini [15], that demonstrated the importance of the electron-phonon coupling in spectroscopy by ab-initio calculations. In particular, E. Cannuccia and A. Marini proved that, besides its importance in metals and in materials at finite temperature, the electron-phonon coupling covers a very important role also in insulators at very low temperature. They showed that the quantum zero-point motion of the carbon atoms induces strong effects on the optical and electronic properties of diamond and trans-polyacetylene. Because of the importance of this subject, one of the invited speakers to the conference is Elena Cannuccia, who already accepted to come and speak about this topic to the other young researchers.

QUANTUM TRANSPORT

Further miniaturization of electronic devices is limited by power consumption, heat dissipation and

quantum effects. New materials and nanoscale devices have been proposed to circumvent the limits of

conventional technology. For example, IBM and Intel have started to use high-k materials for some devices [16], a high-frequency transistor based on graphene was developed by IBM [17] and the first functional transistor made of a single molecule was produced in Bell Labs [18]. Modeling of electronic transport is expected to play a crucial role in the design of such nanoscale devices.

For several decades, quantum transport has been studied with tight-binding methods. Recently, the

community has moved to DFT calculations with great success. Nowadays this is the standard theory for quantum transport modeling. However, electron correlation can have a big impact in the calculated current [19], and some quantum phenomena cannot be described by DFT. Therefore, TDDFT and MBPT have been explored in the context of electron transport. In particular, TDDFT can predict Coulomb Blockade phenomena, in which electron correlations induce periodic current oscillations and the quantization of conductance [20].

The most popular transport methods are based on the Kubo formalism and the Landauer-Buttiker approach. Beyond these theories, the Keldysh formalism, which includes correlations and non-equilibrium phenomena, has not been implemented yet for real systems but there are ongoing efforts in this direction. Nevertheless, recent developments and implementations of novel ab-initio techniques based on GW method allowed the inclusion of non local field effects in quantum transport, and were applied with success to molecules, junctions, graphene and several other systems [21,22].

Interesting results are also expected in magnetic transport. Since its observation in 2008 [23], more and more scientific publications are treating the recently discovered phenomenon named spin Seebeck effect, that has several possible applications in spin-tronics, and the meeting will be an occasion to verify progress done in ab-initio calculations of this effect.

An invited talk for this session will be given by Elham Khosravi, an international expert in time-dependent phenomena in quantum transport that published several important works on this topic [24,25].

THEORETICAL DEVELOPMENTS

The continuous development of new materials and the emerging technical challenges at quantum level, direct the necessity of new theoretical algorithms and new approximations for the calculation of physical properties. In this field, theoreticians are trying to find a synthesis between time-dependent density functional theory (TDDFT) and many-body perturbation theory (MBPT). The TDDFT calculations are essentially dependent on space and time only one time and are therefore computationally feasible for large systems, whereas the MBPT calculations involve 2 or 4 point functions but with the advantage of a more detailed description of the evolution of the system. The new algorithms and approximations are of paramount importance at the forefront of computational solid-state physics.

In our conference, Dr. Adrian Stan [26,27] has been invited to give a talk on these topics. Over the years he developed algorithms for converging GW approximations [28] and for the time-evolution of extended systems [29].

## References

**Belgium**

Matthieu Verstraete (University of Liege) - Organiser

**France**

Lucia Reining (CNRS-Ecole Polytechnique) - Organiser

**Germany**

Friedhelm Bechstedt (University of Jena) - Organiser

**Italy**

Giovanni Ciccotti (University "La Sapienza") - Organiser

Fernanda Lupinacci (University of Rome "La Sapienza") - Organiser

Simone Melchionna (ISC - Consiglio Nazionale delle Ricerche) - Organiser

Adriano Mosca Conte (University of Rome Tor Vergata) - Organiser

Olivia Pulci (University of Rome Tor Vergata) - Organiser

Claudia Violante (University of Rome Tor Vergata) - Organiser