preview all symposia

2021 Fall Meeting

Modelling and characterization


Computer-aided materials modelling: fundamental and applied insights merging physics and chemistry viewpoints at the atomic-scale

The scope of the present symposium is to offer a survey of the most advanced modelling approaches exploited to gain fundamental and practical insights for a wide range of functional materials for multiple technological applications (optoelectronics, energy, biomaterials). Special attention is devoted to the recent forefront applications of first-principles methods (static and molecular dynamics) as well as machine-learning techniques. The atomic-level knowledge provided by the combination of high-performance computing and advanced computational methods pave the route for a rational approach, based on an accurate assessment of materials’ chemistry and physics, to the design of novel materials with tailored properties for specific applications in next-generation technologies.


High-performance computing and advanced computational methods can offer nowadays an important contribution to the European Community Materials Research. Indeed, theoretical characterization from either first-principles or approaches that are more empirical are likely to provide important information in parallel to the experimental determinations. In that respect, the scope of this symposium aims at exploring the wide range of theoretical methods developed in the recent years. An important part will be devoted to theoretical and numerical developments to overcome nowadays-physical challenges.

This symposium will meet the challenge of assessing the role of chemical bonding in complex materials by employing as a theoretical endeavour, a survey of advanced computational approaches. The goal of these innovating methods is to investigate structural, dynamical, optical and electronic localization properties of specific materials of interest for next-generation devices (optoelectronics, energy harvesting and storage, spin or heat transport, thermoelectricity, biomaterials) coping with the current need for a sustainable technology.

Those approaches range from atomic-level first-principles methods to tight-binding models and molecular dynamics (MD) simulations. In addition, we will consider methods such as quantum Monte-Carlo or hybrid QM/MM methods for larger or biological systems. In parallel, machine-learning algorithms for materials screening would give a nice opening on future methodological perspectives in Materials Science.

Hence, another goal of the symposium is to present a general overview of theory and simulation contribution in the field. For example, we will consider applications in nanosciences and nanostructure materials, bulk, surfaces and interfaces, disordered and low-dimensional materials including graphene and bi-dimensional materials, organic molecules on metallic or oxide surfaces, magnetic and spin cross-over molecules, self-assembled molecular networks, and biological molecules.

In summary, this symposium will provide a wide and unique state of the art overview on the theoretical methods used to describe and characterize materials properties. It aims at having equilibrated contributions from important researchers in the community and young researchers to favor discussions and exchange, and draw some perspectives on the next challenges in the field.

Hot topics to be covered by the symposium:

  • methods and developments in first-principles and semi-empirical methods
  • molecular dynamics
  • machine learning
  • High Performance Computing
  • QM/MM simulations
  • graphene and 2D materials
  • molecular electronics and spintronics, magnetism
  • electronic transport and devices simulations, optical properties
  • electron-phonon coupling, thermoelectricity
  • metal/organic interfaces and frameworks
  • biological molecules and mechanisms
  • thermodynamics
  • renewable energies and storage
  • mass and heat transport
  • surfaces and interfaces
  • disordered, porous and hybrid organic-inorganic materials

List of invited speakers:

  • Raffaella Demichelis, Curtin Institute for Computation, The Institute for Geoscience Research, and School of Molecular and Life Science, Curtin University, Western Australia;
  • Julia Wiktor, Department of Physics, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden;
  • Evelyne Martin, ICube Laboratory, MaCEPV, University of Strasbourg, CNRS, UMR 7357, France;
  • Enrico Piccinini, Applied Materials Italy, Appleid AI | MDLx;
  • Assil Bouzid, Institut de Recherche sur les Céramiques (IRCER), CNRS UMR 7315, Université de Limoges, Centre Européen de la Céramique, 12 rue Atlantis, 87068 Limoges, France;
  • Laura Ratcliff, Imperial College London;
  • Gabriel Wlazlowski, Warsaw University of Technology;
  • Alister Page, University of Newcastle, Australia; Ulsan National Institute of Science & Technology, Korea Ulsan National Institute of Science & Technology, Korea; Aalto University, Finland;
  • Kerstin Falk, Fraunhofer IWM, Freiburg, Germany; Physics Department, University of Freiburg, Germany.
Start atSubject View AllNum.
08:25 Welcome message and introduction to the Symposium    
Modern computational approaches and methodological developments : Guido Ori
Authors : Laura E. Ratcliff
Affiliations : Imperial College London

Resume : Density-functional theory (DFT) is routinely used to simulate a wide variety of materials and properties, however, standard implementations of DFT are cubic scaling with the number of atoms, limiting calculations to a few hundred atoms. However, in recent years various linear scaling (LS) approaches have been developed, enabling simulations on tens of thousands of atoms. One key factor influencing the accuracy and cost of DFT is the basis set, where minimal, localized basis sets compete with extended, systematic basis sets. On the other hand, wavelets offer both locality and systematicity and are thus ideal for representing an adaptive local orbital basis which may be exploited for LS-DFT. One may also make further physically-motivated approximations, e.g. dividing a system into fragments or exploiting underlying repetition of local chemical environments, where each approximation may be controlled and quantified. This ability to treat large systems with controlled precision offers the possibility of new types of materials simulations. In this talk I will demonstrate the advantages of such an approach for large scale DFT calculations, as implemented in the wavelet-based BigDFT code. I will focus on the example of materials for organic LEDs, showing how this approach may be used to account for environmental and statistical effects on excited state calculations of disordered supramolecular materials.

Authors : Daniel Fritsch
Affiliations : Department Structure and Dynamics of Energy Materials, Helmholtz-Zentrum Berlin für Materialien und Energie, Hahn-Meitner-Platz 1, 14109 Berlin, Germany

Resume : First-principles calculations based on density functional theory have been established as the working horse for computational materials investigations. Depending on the size of the unit cell, for every material of interest a suitable choice for the unknown exchange and correlation functional has to be made; taking not only into account the desired accuracy, but also the available computational resources. As a rule of thumb, full structural relaxations based on so-called hybrid functionals require up to two magnitudes larger amounts of computing time compared to the standard functionals based on the various local density or generalised gradient approximations. In recent years, a promising combination of both approaches emerged, starting from a structural relaxation based on a cheaper functional, supplemented by a single shot hybrid functional calculation [1]. As many structural relaxations are performed by exploring the volume dependence of the total energy and a fit to a suitable equation of state, here we propose a new method for combining different levels of exchange and correlation functional for structural relaxations. In a first benchmarking step, this new method will be applied to various sets of promising energy materials, where full hybrid functional calculations are available [2,3], with a main focus on the performance of this new approach on the structural properties and the required computational resources. In a second step, this new approach will be applied to materials, which up-to-now have not been accessible to hybrid functional calculations due to the required computational resources. All the presented results on the structural, electronic, and optical properties will be critically discussed alongside experimental findings. This work made use of computational resources provided by the North-German Supercomputing Alliance (HLRN), and the Curta and Dirac HPC facilities of the FU Berlin and the Helmholtz-Zentrum Berlin, respectively. [1] D. Fritsch and S. Schorr, J. Phys. Energy 3, 015002 (2021). [2] D. Fritsch, B. J. Morgan, and A. Walsh, Nanoscale Res. Lett. 12, 19 (2017). [3] D. Fritsch, J. Phys.: Condens. Matter 30, 095502 (2018).

Authors : Michael Lorke, Peter Deak, Thomas Frauenheim
Affiliations : Bremen Center for Computational Materials Science, University of Bremen

Resume : Density functional theory is the workhorse of theoretical materials investigations. Due to the shortcoming of (semi-)local exchange correlation potentials, hybrid functionals have been established for practical calculations to describe surfaces, molecular adsorption, and defects. These functionals operate by mixing between semi-local and Hartree-Fock exchange semi-emprically. However, their parameters have to be optimized for every material separately. To treat materials with a more physics driven approach and without the need of parameter optimization is possible with many-body approaches like GW, but at an immense increase in computational costs and without the access to total energies and hence geometry optimization. We propose a novel exchange correlation potential[1] for semiconductor materials, that is based on physical properties of the underlying microscopic screening. We demonstrate that it reprocuduces the low temperature band gap of several materials. Moreover, on the example of defects in semiconductors, it respects the required linearity condition of the total energy with the fractional occupation number, as expressed by the generalized Koopman’s theorem. It is shown, that alloys can be treated semiconductors can be treated with a common choice of the functional. As the parameters of the calculation can in principle be determined from ab initio calculations, our approach can be seen as a non-emprical approximation. We also show that this novel functional can be used as a kernel in linear response TDDFT to reproduce excitonic effects in optical spectra [1] Physical Review B 102 (23), 235168

Authors : Riccardo Alessandri (1,2), Siewert J. Marrink (1)
Affiliations : (1) University of Groningen, the Netherlands; (2) University of Chicago, United States

Resume : The Martini model is one of the most popular coarse-grained molecular dynamics (MD) force fields. While originally developed for applications in structural biology and biophysics, the model has recently found an increasing number of applications also in the fields of nanotechnology and materials design [1]. We present how Martini-based coarse-grained MD simulations can be used to model soft matter blends for organic electronics. Such simulations are used to generate morphologies taking into account the processing conditions, such as spin-coating and thermal annealing [2]. Moreover, such coarse-grained models retain a sizable degree of chemical specificity and can be directly back-mapped to atomistic resolution. This allows not only to probe the impact of chemical modifications of the molecular structure on the resulting morphology, but also to gain access to electronic structure information while taking into account the large-scale self-organization process of the thin film [2,3]. As an application, we show how a multiscale modelling scheme – coupling Martini CG simulations to quantum chemical calculations – can be used to probe the impact of polar side chains on electronic and structural properties of organic semiconductor blends [3]. We find that the introduction of polar side chains on a similar molecular scaffold does not affect molecular orientations at interfaces. Such orientations are instead found to be strongly affected by processing conditions (i.e., thermal annealing) and polymer molecular weight. We find that polar side chains, instead, impact significantly the energy levels of the organic blend, causing broadening of these levels by electrostatic disorder. Finally, we will conclude with a brief discussion of extensions of the approach, including coupling to machine learning techniques to enable higher-throughput characterization of the simulations, one aspect that is paramount to enable computer-aided materials design. [1] Alessandri, R.; Grünewald, F.; Marrink, S.J. Adv. Mater. 2021, 2008635. [2] Alessandri, R.; Uusitalo, J.J.; de Vries, A.H.; Havenith, R.W.A.; Marrink, S.J. J. Am. Chem. Soc. 2017, 139, 3697-3705. [3] Alessandri, R.; Sami, S.; Barnoud, J.; de Vries, A.H.; Marrink, S.J.; Havenith, R.W.A. Adv. Funct. Mater. 2020, 2004799.

Authors : Stefano Leoni, Bo Hou
Affiliations : School of Chemistry, Cardiff University, Cardiff, CF10 3AT, UK; School of Physics and Astronomy, Cardiff University, Cardiff, CF24 3AA

Resume : Since the Materials Genome Initiative, computational and ab initio approaches are firmly part of the materials discovery process, which is profoundly advancing materials innovation. The ability to reliably simulate materials parameters, including structure and voltage, and anticipate properties of predicted materials, is a remarkable merit of computational approaches - one that can provide acceleration of technological innovation and boost its transfer to market. This cultural shift has facilitated joint efforts between experimentalists and theoretically oriented scientists, from industry to academia, to the extent that property calculations (geometry and stability) have become routine. However, the simulation of ion transport properties, a critical parameter for overall batteries operation and for catalysis processes, remains a considerable challenge. For instance, in Li-ion batteries, Li-ion migrates within the cathode, electrolyte, anode, and interfaces, requiring dedicated simulation techniques - not least because transport is a non-equilibrium property. Reliable and efficient computation of such properties would benefit automatic high-throughput protocols and machine learning approaches searching for battery materials, which are currently sought for. Here we provide such a computational scheme for ion dynamics and use it within a combined data mining and crystal structure searching framework to identify and fully characterise better materials with distinct mass transport properties, including battery materials. We will present specific techniques to accelerate ion dynamics, which would otherwise remain and elusive process in molecular dynamics simulations, towards an improved understanding of collective ion mobility phenomena and their response to defects, disorder and interfaces. Our results are paramount for a comprehensive understanding of ion mobility that is directly coupled with battery charging efficiency, which can now be put on firmer grounds than it has been possible so far. In addition, a precise un-derstanding of the connection between ion pathways within materials and their entanglement is a prerequisite for more powerful predictions based on machine learning frameworks.

10:30 Q&A Session / Break    
Predicting materials properties by first-principles and machine learning modeling : Yannick Dappe
Authors : Assil BOUZID, Philippe THOMAS and Olivier MASSON
Affiliations : Institut de Recherche sur les Céramiques (IRCER), CNRS UMR 7315, Université de Limoges, Centre Européen de la Céramique, 12 rue Atlantis, 87068 Limoges, France

Resume : Thanks to their outstanding optical properties, crystalline and amorphous TeO2-based materials are widely used in several fields such as laser light modulators, pressure sensors and optical switching devices. Interestingly, this family of materials exhibit a high third-order optical non-linearity several orders of magnitude larger than that of conventional silicate and borate glasses. Despite the large efforts undertaken to study these materials, their atomic scale structure and its impact on the macroscopic measurable properties remain a matter of debate. In this contribution, I will mainly focus on TeO2 glass and discuss the possible roots to achieve an accurate glassy structure by relying on first-principles molecular dynamics. In particular, effects of the exchange and correlation functional, model size, and vdW forces will be discussed. Furthermore, I will explore the possibility of combining first-principles modeling and machine learning in order to achieve large size-scale models produced over long time-scales without loosing the first-principles accuracy.

Authors : Saientan Bag, Manuel Konrad, Tobias Schlöder, Pascal Friederich, Wolfgang Wenzel.
Affiliations : Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz Platz 1, 76344 Eggenstein-Leopoldshafen, Germany; Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz Platz 1, 76344 Eggenstein-Leopoldshafen, Germany; Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz Platz 1, 76344 Eggenstein-Leopoldshafen, Germany; Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz Platz 1, 76344 Eggenstein-Leopoldshafen, Germany and Institute of Theoretical Informatics (ITI), Karlsruhe Institute of Technology (KIT), Am Fasanengarten 5, 76131 Karlsruhe, Germany; Institute of Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz Platz 1, 76344 Eggenstein-Leopoldshafen, Germany.

Resume : Adsorption and desorption of molecules are key processes in extraction and purification of biomolecules, engineering of drug carriers, designing of surface-specific coatings. To understand the adsorption process on the atomic scale, state-of-the-art quantum mechanical and classical simulation methodologies are widely used. However, studying adsorption using a full quantum mechanical treatment is limited to picoseconds simulation timescales, while classical molecular dynamics simulations are limited by the accuracy of existing force fields. To overcome these challenges, we propose a systematic way to generate flexible, application-specific force fields with chemical accuracy by training artificial NNs. As a proof of concept, we study the adsorption of the amino acid alanine on graphene and gold (111) surfaces and demonstrate the force field generation methodology in detail. We find that a molecule-specific forcefield with Lennard-Jones type two-body terms incorporating the 3rd and 7th power of the inverse distances between the atoms of the adsorbent and the surfaces yields optimal results, which is surprisingly different from typical Lennard-Jones potentials used in traditional force fields. Furthermore, we present an efficient and easy-to-train machine learning model that incorporates system-specific three-body (or higher order) interactions that are required, e.g., for gold surfaces. Our final machine learning based force field yields chemical accuracy at a speed-up of ~O(5) times compared to quantum mechanical calculation, which will have a significant impact on the study of adsorption in different research areas.

Authors : Christopher M. Andolina1, Marta Bon2, Daniele Passerone2, Wissam A. Saidi1
Affiliations : 1Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15216, USA; 2Empa, Swiss Federal Laboratories for Materials Science and Technology, Electron Microscopy Center, Überlandstrasse 129, CH-8600 Dübendorf, Switzerland

Resume : After many decades of intense research scrutiny, Au, Ag, and Au-Ag materials across many length scales are still of great interest. We interpret these efforts as an impetus for developing robust, accurate, and relatively fast computational methods for modeling these materials. To this end, we describe the training, development, and validation of a machine learning deep neural network potential (DNP) for improved modeling of Ag-Au materials. This DNP was iteratively trained using density functional theory (DFT) to produce a robust multi-length scale potential, which yields results comparable to DFT on a wide range of properties such as equilibrium and non-equilibrium lattices, mechanical properties, and defect energies. Further we describe adatoms (Ag or Au) energy barriers for diffusion on {100}, {110}, and {111} terminated surfaces (Ag or Au), which agrees with literature values. Ultimately, our DNP is a critical tool to study nucleation and growth of simulated seeded core-shell Ag and Au nanoparticles (NP), as much more costly to simulate by DFT and difficult to observe experimentally with the same level of detail. We show that both nanoalloys (Au201@Ag201 and Ag201@Au201) grow such that {111} facets significantly increase at the expense of the {100} ones. In contrast, the Ag core NP is found to have a more disordered inner structure than the Au one, and that Ag adatoms in Au@Ag NP have a more pronounced penetration power than Au in Ag@Au NP. These findings are rationalized in terms of adatom adsorption and diffusion energies.

Authors : Tan-Lien Pham, Assil Bouzid, Mauro Boero, Carlo Massobrio, Young-Han Shin, and Guido Ori
Affiliations : Department of Physics, University of Ulsan, Ulsan 44610, Republic of Korea; Institut de Recherche sur les Céramiques, UMR 7315 CNRS-Univesité de Limoges, Centre Européen de la Céramique, 12 rue Atlantis 87068 Limoges Cedex, France; Université de Strasbourg, CNRS, Institut de Physique et Chimie des Matériaux de Stras-bourg,UMR 7504, F-67034 Strasbourg, France.

Resume : Sodium rich solid electrolyte for Na ion battery, which is notoxic, nonflammabe, widely available, is a potential alternatives to traditional organic liquids. The alkali metal rich an- tiperovskite are among one of the most promising Li and Na solid electrolyte materials. There have been many efforts to increase its ionic conductivity and chemical stability. Li rich antiperovskite has shown excellent ionic conductivities as high as 1 ?1 at room temperature and ?1 at 250 ? C. Although, the Na-rich antiperovskite (NRAP) has a lower ionic conductivity of 0.22 ?1 at 240 ? C, researchers have improved the Na conductivity by elemental substitution [1]. Goodenough and Braga groups have reported that amorphous Na antiperovskites are superionic conductors with conductivity as high as 100 ?1 at around 150? C, far above the values obtained with other state-of-the-art solid electrolytes [2]. In this work, we shed light into the structural properties, electronic structure and ions dynamics properties of NRAP, firstly, by means first-principles molecular dynamics (FPMD) simulations. In particular, a series of Na 3 OCl-based systems will be the target of this work where we focus our attention on both their glassy and liquid states. The second part will be focusing on how to exploit the obtained FPMD database in order to build proper machine learned potentials to complement our work with the aim to enlarge the the size and time scales of our simulations.

Authors : Jianan Zhang*(1), Srilok Srinivasan (2), Aditya Koneru (1, 2), Subramanian K. R. S. Sankaranarayanan (1, 2), & Carmen M. Lilley (1).
Affiliations : (1) Department of Mechanical and Industrial Engineering, The University of Illinois at Chicago, 842 W. Taylor, Chicago, Illinois 60607, USA (2) Center for Nanoscale Materials, Argonne National Lab, Argonne, Illinois 60439, USA * lead presenter

Resume : Grain boundaries (GBs) in two-dimensional (2D) materials often have a profound impact on various material properties from mechanical to optical to electronic, yet searching all possible GB structures is a challenge. Here, we introduce a workflow based on an evolutionary algorithm for exploring possible defect structures formed at a lateral 2D interface. In a departure from conventional genetic algorithm based structure optimization methods, we perform genetic operations in the near interface region that allow us to be computationally efficient. We benchmarked our method using graphene, which is a well-studied 2D material with a wide range of point defects. Using an empirical potential as the surrogate of the evolutionary search, more than 11.5 million structures in total were evaluated for 128 GB orientations. For each orientation, a subset of low energy GBs predicted by the search was relaxed by first-principles calculations and used to validate the energetic rank order. With the validated formation energy, we rank-ordered the best 128 GB structures and performed a detailed statistical analysis of primitive rings to find the correlation between the ring distribution and the formation energy. We found that for low energy GBs (below 0.5 eV / Å) Stone–Wales defects will dominate while structures with a higher energy (0.5–1.1 eV / Å) show an increasing population of heptagons and nine-membered rings to form seven-nine atom ring defect pairs. For structures with energy higher than 1.1 eV / Å, the percentage of octagons and nine-membered rings increases, which indicates that these two types of rings are not energetically favorable. In addition to using an empirical potential as the searching surrogate, we found that it is natural to transfer the topological structures of 2D material into mathematical graphs, the atoms can be considered as the nodes, the bonds can be treated as the edges and the energy value corresponding to each structure is thus considered as a graph-level training label. Therefore, an attempt was made to train a graph neural networks (GNNs) as the searching surrogate. For 2D material blue phosphorene, which lacks reliable empirical potential that can correctly describe the energy for defective structures, we generated a dataset with the energy of each structure evaluated by first principle calculations. For the total energy of the system, the GNN model trained was able to reach mean absolute error and percentage error as low as 2.58 eV and 0.234 %, respectively. We believe that combining the GNN and the previously mentioned searching method will allow us to explore interface structures of more novel 2D material in the future. Our proposed methodology is broadly applicable to explore defective low dimensional materials and represents a powerful tool that enables a systematic search of GBs of lateral interfaces for 2D materials.

13:00 Q&A Session / Break    
Material design, comprehension and application by atomistic modeling : Michal Hermanowicz
Authors : Kerstin Falk, Andrea Codrignani, Lars Kruse, Daniele Savio, Michael Moseler
Affiliations : Fraunhofer IWM, Freiburg, Germany; Physics Department, University of Freiburg, Germany

Resume : Lubrication is the most relevant technique to reduce friction and wear in mechanical applications. The drive towards sustainable technologies has increased the demands on the performance of lubricants under the relevant tribological conditions, which can be very harsh. For instance, lubricants in roller element bearing and gear applications are subject to pressures of the order of Gigapascal, often combined with high operating temperatures. In addition, the high pressure can also lead to nanometer thin sizes of the tribo-gap and high shear rates in the remaining thin lubricant film. The design and optimization of liquid lubricants require a quantitative knowledge of their rheological properties in this so-called boundary lubrication regime, which is hardly accessible experimentally. Molecular dynamics (MD) simulations, in turn, are well suited to investigate the rheology of liquids under severe tribological conditions - especially in nanometer thin gaps. As will be illustrated throughout this talk, MD simulations can help in several ways to quantitatively model lubricated contacts in the boundary lubrication regime. For example, relevant liquid properties, most of all viscosity, can be calculated explicitly under well-controlled conditions (pressure, temperature, gap size, shear rate). The microscopic information about the molecular dynamics inherent in MD simulations allows extracting structure-property relations. Such relations between the lubricant’s molecular structure and its rheological behavior are an important step towards the optimization and targeted design of lubricants with specific properties. The first part of the talk focuses on the prediction of Newtonian viscosities under high pressure and temperature conditions for typical base oil constituents, namely linear and branched alkanes [1]. Results from theoretical modeling based on an extensive structure evaluation of equilibrium MD trajectories are presented: the Stokes-Einstein relation is combined with free volume and random walk concepts for the molecular self-diffusion, which results in a viscosity scaling law for the considered P and T. The model parameters (hydrodynamic radius, random walk step size, and step frequency) were calculated from equilibrium MD via microscopic ensemble averages. The talk then moves on to consider nanometer thin tribo gaps, where the rheological properties can be influenced by the gap size, or continuum hydrodynamics concepts may fail altogether. MD simulations describe the liquid’s behavior even in such highly confined gaps. Thus, the knowledge from MD simulations can be used to “correct” continuum modelling of friction contacts via improved constitutive laws. Such a correction scheme will be discussed for the example of an asperity slider on a plane surface in the boundary lubrication regime. Results of a large scale MD parameter study will be compared to a continuum prediction (Reynolds calculation). [1] K. Falk, D. Savio and M. Moseler, “Nonempirical Free Volume Viscosity Model for Alkane Lubricants under Severe Pressures”, Phys. Rev. Lett. 124, 105501 (2020).

Authors : Udoka NWANKWO, Nicolas ONOFRIO, Chi-Hang LAM
Affiliations : Department of Applied Physics, The Hong Kong Polytechnic University, Hong Kong SAR China

Resume : Contemporary energy storage and electronic devices are often governed by electrochemical processes. Despite their significance and progresses made towards unveiling their properties, the mechanisms that govern their operation are not fully understood, especially at the atomic level. Beyond experimental limit, atomistic simulation can provide unprecedented insights and details about the processes and improve the devices’ development. Atomic description of these systems requires reactive interaction potential to be able to describe (i) the potential chemistry between atoms and molecules, and (ii) the evolving charge distribution and polarization effects. Calculating Coulomb electrostatic interaction and polarization effects require a good estimate of partial charge distribution in molecular systems, at least for proper prediction of the systems’ stability and solubility. However, models such as ReaxFF as well charge equilibration (QEq) method include Coulomb interaction up to a “short-range” distance cutoff for computational speed, limiting its application to nanodevices with unrealistic small sizes. Ignoring long-range distance electrostatic interactions affect the ability to describe electrochemistry in large systems. In this work, we studied the Coulomb long-range effect between charged particles including the computation of dynamical charges and forces. By extending the QEq method to include long-range effect, we expect proper account of Coulombs interaction in reactive molecular dynamic simulation. We validated our approach to compute charges on a series of metal organic frameworks and some simple systems compared to regular QEq and quantum mechanics (QM). The results show that without long-range, the charges are slightly overestimated. We combined our QEq method with Ewald to compute forces and evaluated the long-range effect in some simple capacitor configurations. The results show some noticeable difference between charges as well forces computed with and without long-range Coulomb. The difference stems from long-range influence on ions of the capacitor. This indicate that Ewald forces are key to describe accurate electrostatic interaction between separated charged electrodes. Moreover, we applied our new framework to graphene/water system and compare with QM. We found that there are more structures in the number and charge distribution of the system compared with regular QEq. Our combined approach would enable the atomic description of electrochemical systems with realistic electrolyte thickness while conserving the electrostatic effect of charged electrodes throughout the dielectric layer such as in battery and emerging solid-state memory.

Authors : Caglanaz Akin, Busra Demir, Hashem Mohammad, Joshua Hihath, M. P. Anantram, Ersin Emre Oren
Affiliations : Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey. ; Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey. ; Department of Electrical Engineering, University of Washington, Seattle, WA, USA. ; Electrical and Computer Engineering Department, University of California Davis, Davis, CA, USA. ; Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey.

Resume : DNA, known as genetic material, is one of the promising molecules for charge transport studies due to its unique electronic properties. DNA’s molecular structure and electrical conductivity depends on various parameters including its sequence and doping with small molecules called intercalators [1]. Electron transmission probability of DNA can be controlled with molecular doping to use in molecular electronics applications such as data storage, biosensors, rectifiers, transistors [2-4]. In this research, we investigated the molecular conformations of different short DNA sequences (GC-rich, AT-rich and mixed) intercalated with anthraquinone and acridine type intercalators with molecular dynamics (MD) simulations. Then, we selected representative structures of simulations via RMSD based clustering algorithms and calculated their electron transmission probabilities using density functional theory (DFT) and Green’s function-based transport model. We showed that the structural stability and electrical conductivities of DNA-intercalator systems highly depends on the type of intercalator molecule and DNA sequence, also the intercalation position. [1] H. Mohammad, B. Demir, C. Akin, B. Luan, J. Hihath, E. E. Oren, M. P. Anantram (2021) Role of intercalation on electrical properties of nucleic acids for use in molecular electronics. Nanoscale Horiz., Accepted Manuscript [2] T. Harashima, C. Kojima, S. Fujii, M. Kiguchi, T. Nishino (2017) Single-molecule conductance of DNA gated and ungated by DNA-binding molecules. Chem. Commun., 53, 10378–10381 [3] C. Guo, K. Wang, E. Zerah-Harush, J. Hamill, B. Wang, Y. Dubi, B. Xu (2016) Molecular rectifier composed of DNA with high rectification ratio enabled by intercalation, Nat Chem., 8(5), 484-90 [4] X. Wang, L. Gao, B. Liang, X. Li, X. Guo (2015) Revealing the direct effect of individual intercalations on DNA conductance toward single-molecule electrical biodetection. J. Mater. Chem. B., 3, 5150

Authors : Werner Dobrautz, Giovanni Li Manni, Nikolay Bogdanov, Ali Alavi
Affiliations : All: Max Planck Institute for Solid State Research, Stuttgart, Germany Ali Alavi additionally: Yusuf Hamied Department of Chemistry, University of Cambridge, UK

Resume : We present how to compute the one- and two-body reduced density matrices within the spin-adapted full configuration interaction quantum Monte Carlo (FCIQMC) method, which is based on the graphical unitary group approach (GUGA). This allows us to use GUGA-FCIQMC as a spin-pure configuration interaction (CI) eigensolver within the complete active space self-consistent field (CASSCF) procedure, and hence to stochastically treat active spaces far larger than conventional CI solvers whilst variationally relaxing orbitals for specific spin-pure states. We apply the method to investigate the spin-ladder in iron-sulfur dimer and tetramer model systems. We demonstrate the importance of the orbital relaxation by comparing the Heisenberg model magnetic coupling parameters from the the CASSCF procedure to those from a CI-only procedure based on restricted open-shell Hartree-Fock orbitals. We show that orbital relaxation differentially stabilizes the lower spin states, thus enlarging the coupling parameters with respect to the values predicted by ignoring orbital relaxation effects. Moreover, we find that while CI eigenvalues are well fit by a simple bilinear Heisenberg Hamiltonian, the CASSCF eigenvalues exhibits deviations that necessitate the inclusion of biquadratic terms in the model Hamiltonian. These clusters are of major importance in organometallic chemistry and as cofactors in biology and are involved in a multitude of processes, including photosynthesis, respiration and nitrogen fixation, being responsible for redox reactions and electron transfer, act as catalytic agents and even provide a redox sensory function. A theoretical understanding of the intricate interplay of the energetically low-lying spin states of these systems, guided by accurate numerical results, could provide insights towards the synthetic realization of these processes.

Authors : Graeme M. Day, Chi Y. Cheng
Affiliations : School of Chemistry, University of Southampton; School of Chemistry, University of Southampton

Resume : We present a computational approach for the discovery of high charge mobility, small-molecule organic semiconductors. This application requires a methodology that not only optimizes the molecular structure for specific electronic properties but also for specific solid-state packing preferences that will increase the electronic coupling between its molecular units. We present an approach which tackles both requirements by carrying out an evolutionary search over a specified chemical space, using molecular mutation and cross-over, to identify molecules with low reorganization energies and high electron affinities. The best performing molecules are then taken through a crystal structure prediction (CSP) [1] and carrier mobility evaluation stage to effectively couple the molecular and crystal structure searches. With this methodology, we optimized the chemical structures of a population of molecules within a chemical space of azapentacene molecules [2]. By carrying forward the best performing molecules through the CSP and mobility evaluation stage we produce a list of promising molecules each with an associated level of risk and reward for their abilities to form high carrier mobility organic semiconductors. More recently, we have extended the approach to the exploration of a wider chemical space, performing CSP on several hundred molecules. [1] “Convergence properties of crystal structure prediction by quasi-random sampling” Journal of Chemical Theory and Computation, 12, 910-924 (2016). [2] “Evolutionary chemical space exploration for functional materials: computational organic semiconductor discovery” Chemical Science, 11, 4922-4933 (2020).

16:00 Q&A Session / Break    
Poster session I : Guido Ori, Elena Levchenko
Authors : Grzegorz Matyszczak
Affiliations : Warsaw University of Technology, Faculty of Chemistry, Department of Chemical Technology

Resume : Wide practical applications of materials is the driving force of materials science. Computational methods of investigation of materials utilize ab initio calculations, as well as machine learning algorithms creating new discipline named materials informatics. Knowing the structure or chemical composition of one interesting material it is possible to derive its analog with similar or better properties. This study presents the chemico-theoretical derivation of four new analogs of well-known kesterite Cu2ZnSnS4 – a material with promising properties for applications in photovoltaics - and their investigation based on computational methods. An evolutionary algorithm Uspex is used for the prediction of crystal structures of these materials, while the ab initio calculations are exploited to determine their band structures. For comparison, in DFT calculations two approximations are used: GGA-PBE and HSE06. Obtained crystal structures are compared and discussed, and predicted values of band gaps are assessed in context of potential applications of the title materials in photovoltaics.

Authors : Raghvender, Assil BOUZID, David HAMAN, Philippe THOMAS and Olivier MASSON
Affiliations : Institut de Recherche sur les Céramiques (IRCER), CNRS UMR 7315, Université de Limoges, Centre Européen de la Céramique, 12 rue Atlantis, 87068 Limoges, France

Resume : Experiments have shown that TeO2 glasses exhibit high third-order non-linear optical characteristics [1], making them promising candidates for frequency conversion mechanism, optical switches, and other applications. TeO2 is a conditional glass former, which necessitates the addition of a modifier oxide (MO) to achieve better stability of the glassy phase. The addition of MO to TeO2 glass leads usually to a decrease of its non-linear properties. Interestingly, opposite trend has been observed for Tl2O modifier [2,3], as its addition, contrary to other MO leads to maintain the high optical non-linearity of TeO2 based glasses. The present work [4] aims at studying the structural properties of Tl2O-TeO2 glasses at the atomic level by developing Buckingham interatomic potential parameters for Tl^(+) -O^(2-) and Tl^(+) -Tl^(+) interactions. To account for polarization effects, we complement the Buckingham potential with a core-shell description of the atoms [5]. The potential parameters were fitted on Tl2TeO3, Tl2Te2O5 and Tl2Te3O7 crystalline structures simultaneously and the transferability of the potential was tested against various Tl^(+) oxide based crystalline structures by evaluating error in calculating structural parameters. The obtained potential was then used to perform molecular dynamics simulation of Tl2O-TeO2 binary amorphous system with varying concentrations. The structural properties of the obtained atomistic models were analyzed and compared to available experimental X-ray diffraction data [6] showing a good agreement, supporting thereby the viability of the derived potential. 1. S.-H. Kim, T. Yoko, S. Sakka, “Linear and Nonlinear Optical Properties of TeO2 Glass”, J. Am. Ceram. Soc. 76 (10) (1993) 2486–2490 2. B. Jeansannetas, S. Blanchandin, P. Thomas, P. Marchet, J. C. Champarnaud-Mesjard, T. Merle Méjean, B. Frit, V. Nazabal, E. Fargin, G. Le Flem, M. O. Martin, B. Bousquet, L. Canioni, S. Le Boiteux, P. Segonds, L. Sarger, “Glass Structure and Optical Non-linearities in Thallium(I) Tellurium(IV) Oxide Glasses”, J. Solid State Chem. 146 (2) (1999) 329–335 3. M. Dutreilh-Colas, P. Thomas, J. Champarnaud-Mesjard, E. Fargin, “New TeO2 based glasses for nonlinear optical applications: study of the Tl2O-TeO2-Bi2O3, Tl2O-TeO2-PbO and Tl2O-TeO2-Ga2O3 systems”, Phys. Chem. Glass. 44 (5) (2003) 349–352 4. Raghvender et al., “Buckingham interatomic potential for thallium oxide (Tl+-O2-); application to the case of thallium tellurite glasses.”. submitted to Comput. Mater. Sci. , (2021) 5. L. Torzuoli, A. Bouzid, P. Thomas, O. Masson, “An enhanced core–shell interatomic potential for Te–O based oxides”, Mater. Res. Express 7 (1) (2020) 015202. 6. Torzuoli, Lyna, Etude de la structure des verres des systèmes TeO2-MxOy (M = Ti, TI) par diffusion totale des rayons X et dynamique moléculaire, Ph.D. thesis, Thèse de doctorat dirigée par Masson, Olivier et Hamani, David Matériaux, Institut de Recherche sur les Ceramiques, Limoges (France), (2020).

Authors : Ms Arpita Mukherjee(1), Dr. Tushar Rana(2)
Affiliations : SRM Institute of Science and Technology, Kattankulathur, Tamilnadu, India, 603203

Resume : The discovery of graphene and its intriguing properties gave rise to a new class of materials called "2D magnets". Since then there was a urge to investigate the important physical properties of the 2D materials. It is well known that two-dimensional (2D) materials provide a unique platform for spintronics and magnetism, where the atomic thinness of the layers leads to strong tenability in the physical properties of the materials. B20 Silicides are intrinsically compatible with silicon technologies due to their high carrier spin polarisation. Thus, they may prove ideal for spin injection and into silicon devices. Given of all this interest, it is natural to ask if there are other isostructural materials that might also yield interesting magnetic ground state. Recently, a lot of studies have been done on CoSi, MnSi, MnGe, FeGe, FeSi to tune their magnetic properties using appropriate doping or by creating their nanostructures. Later, it was found that most Co - rich alloys are rich in magnetic order. Among them, CoSi and CoGe are an exception as it does not exhibit any kind of magnetic order. However, cobalt generally exhibits a trend towards ferromagnetism and CoSi nanowires having diameters of the order of 50 nm actually exhibit a small net moment due to spin-polarized Co atoms near the surface. Moreover, the magnetism in the CoSi thin film has been investigated experimentally and theoretically. The magnetism in CoGe thin film and nanowires has not been explored intensively. By motivating through a recent study on magnetic properties of CoSi thin film, we will be exploring the structural, electronic, and magnetic properties of CoGe (111) thin film. The emphasis of this work is on the interplay between the geometries and the different dimensional nanostructures using DFT that will be very useful to prepare the backbone of future spintronics and optoelectronic properties. This study could provide a new platform for further advancement in material science.

Authors : Christoph Hauenstein, Stefano Gottardi, Engin Torun, Ruud Gijsen, Arthur Vauzelle, Siebe van Mensfoort, Harm van Eersel
Affiliations : Simbeyond B.V., Het Eeuwsel 57, 5612AS Eindhoven, The Netherlands

Resume : At present, the development of organic electronic devices largely occurs by following empirical recipes and by trial-and-error, while rational design is still in an early stage. The reasons for this are the relative novelty of the technology and the complexity of organic semiconductors. It is expected, however, that in the next decade rational design based on predictive modeling will take a more central stage and will accelerate the research and the development of organic semiconductor-based devices. In this study, we will discuss the capabilities of Bumblebee [1], Simbeyond's predictive simulation software that is based on a 3D kinetic Monte Carlo engine and simulates the processes happening in the device mechanistically. With the help of predictive virtual experiments we show how R&D development cycles can be effectively reduced. Featuring an intuitive workflow, user interface and data visualization Bumblebee provides precious insight into the operation of a device down to the nanoscale details. Experimentally inaccessible processes can be easily studied with only minimal effort required by the modeler. A wide range of possibilities for stack optimization become available thanks to the support for complex 3D material morphologies. Bumblebee gives a molecular-scale and nanosecond-resolved view of the optoelectronic processes occurring in the device, including the actual spatial non-uniformity of the current density and final emission [2], a decomposition of the efficiency loss processes such as exciton-polaron quenching and exciton-exciton annihilation, and even molecular degradation scenarios [3,4]. Bumblebee unprecedentedly allows the study of the functioning of a device at the molecular scale in 3D, which is experimentally impossible. Using Bumblebee, both the material and stack properties can be optimized simultaneously, reducing R&D costs and time-to-market. [1] Bumblebee is the state-of-the art kinetic Monte Carlo simulation code of Simbeyond B.V, [2] M. Mesta, et al. Nat. Mater. 12, 653 (2013) [3] R. Coehoorn, et al. Adv. Funct. Mater. 25(13), 2024 (2015) [4] S. Gottardi, et al. Appl. Phys. Lett. 114, 073301 (2019)

Authors : Meysam Esmaeilpour, Mariana Kozlowska, and Wolfgang Wenzel
Affiliations : Institute of Nanotechnology, Karlsruhe Institute of Technology (KIT), Germany

Resume : Chemical vapor deposition (CVD) is the most promising method for high quality, large area graphene synthesis. Optimization of this chemical process will enable control over crucial properties, such as graphene quality and domain size. This requires the development of a detailed atomistic understanding of the underlying processes guiding the growth mechanism. In particular there is a need to understand the mechanism behind graphene nucleation and growth during CVD and its dependence on the synthetic parameters: temperature, CVD pressure, catalyst type, facet etc.The complexity of CVD prohibits a complete description of all reaction mechanisms at the DFT level. Using the library of surface reac-tion rates, we have developed a Kinetic Monte Carlo (KMC) method to study the process of CVD of graphene from methane on Cu(111)under different synthesis conditions. It explains how synthesis parameters affect the quality and domains size of graphene. The results are compared with experimental measurements, enabling better understanding of the CVD mechanism.

Authors : Saibal Jana, Wolfgang Wenzel
Affiliations : Institut für Nanotechnologie (INT), Karlsruher Institut für Technologie (KIT), 76344 Eggenstein-Leopoldshafen, Germany

Resume : One of the most efficient rechargeable, high power density, long life cycle, portable energy devices for clean energy storage technology is lithium-ion batteries (LIBs). However, the Li content in Earth?s crust is limited. Therefore, Sodium-ion batteries (SIBs) have received a great deal of attention as an alternative to LIBs. One of the major scientific challenges for competitive SIBs is the development of highly efficient anode materials. The stability of carbon materials and the high specific capacity of phosphorus materials motivate us to examine carbon?phosphorus solid solutions as anode materials. Here, I will present a rationally designed carbon-phosphorus (3:1) solid solutions 2D material PC3 for the application in the field of SIBs technology.

Authors : V. A. Yuriev1, E.V. Levchenko2, Valery Nebol'sin1
Affiliations : 1 Department of Radio Engineering and Electronics, Voronezh State Technical University, 14 Moskovsky Pr, 394026 Voronezh, Russia 2 College of Engineering, Science and Environment The University of Newcastle, NSW2308, Australia

Resume : In this work we develop an innovative concept of the crystal-melt phase transition mechanism. This concept takes into account not only the reciprocating dynamics of the crystal lattice, but also the rotational degree of freedom of atoms. The proposed dynamics of the crystal lattice complements the phonon model and allows explaining (i) the presence of premelting of atomic crystals at temperatures below the temperature of thermodynamic equilibrium of the solid and liquid phases, as well as (ii) the phenomenon of supercooling of the liquid phase during crystallization. In this novel model of melting an atom located near a crystal lattice defect loses directional bonds and acquires a rotational degree of freedom. This atom is dynamically different from the crystal lattice atoms, i.e. thermodynamic features of another phase appear both in the values of internal energy and in entropy. The continuous transition from a solid crystalline to a liquid state can be associated with a gradual increase of the number of atoms acquiring a rotational degree of freedom. Supercooling can be explained by the fact that crystallization, i.e. the emergence of a rigid framework will appear when the directed bonds are strengthened, the rotation of atoms turns into oscillations, and the reciprocating dynamics of the lattice will prevail. In this case, rotational oscillations will not be reduced to zero, especially near defects, where the coordination number is reduced. Apparently, the rotational degree of freedom of atoms located around the core of a dislocation is associated with the fluidity of crystals or in other words with the stage of easy glide of dislocations.

Start atSubject View AllNum.
Chemistry and physics of interfaces and materials : Guido Ori
Authors : Alicia Schuitemaker, Paolo Raiteri, Julian D. Gale, Raffaella Demichelis
Affiliations : Curtin Institute for Computation, The Institute for Geoscience Research, and School of Molecular and Life Science, Curtin University, Western Australia

Resume : Living organisms have the powerful ability to synthesize minerals with specific crystal structures, textures, shapes, sizes and compositions as part of their functional hard tissues. Vaterite (CaCO3) is one such mineral. Despite being rare and metastable with respect to other forms of calcium carbonate, it is not uncommon in sea creatures1 and recently has been found to occur in plants.2 Biological molecules are the key to directing its growth and stabilisation; for example, chiral acidic amino acids can modify the organisation of vaterite crystallites depending on the enantiomer added during their formation and aggregation.3 Computer modelling can provide insights into the atomic scale processes that lead to the formation of such specifically designed materials, both via helping to interpret experiments and making predictions that can direct experiments. The main challenge here is developing a realistic, thermodynamically and kinetically predictive model able to access significant size and time scales at an affordable computational cost.4 This presentation will focus on understanding the interaction between vaterite and the enantiomers of aspartic acid. The structure determination of the solid phase,5 the development of an accurate model for both the mineral and organic molecule in water,6 and the results of ion pairing, cluster formation and organic-mineral attachment will be discussed. References: [1] Jacob et al.; Nat. Commun. 2017, 8, 1265. [2] Wightman et al.; Flora 2018, 241, 27. [3] Jiang et al.; Nat. Commun. 2017, 8, 15066. [4] Demichelis et al.; Nat. Commun. 2011, 2, 590; Raiteri et al.; J. Phys. Chem. C 2015, 119, 24447. [5] Demichelis et al.; Cryst. Growth Des. 2013, 13, 2247 and references therein. [6] Jiang et al.; Nat. Commun. 2019, 10, 2318; Schuitemaker et al. J. Chem. Phys. 2021, 154, 164504

Authors : Qian Wang 1,2, Yuanyuan Li 2, Xiaomin Liu 2, Mingliang Wang 1*, Hong Zhu 2, Haowei Wang 1,2
Affiliations : 1. State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, China; 2. School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Resume : The stability and wettability of ceramic/metal interface directly affect the mechanical properties of composites. Alloying as an effective experimental method can improve the interface properties. Through the first-principles study based on density functional theory (DFT), the alloying design of ceramic/metal interface can be accelerated via exploring the interface alloying mechanism. Taking TiB2/Al composite as an example, the thermodynamically most stable Al(111)/TiB2(0001) and Al(001)/TiB2(0001) interface models with the Ti terminal and center stacking sequences are established in this study to analyze the alloying behavior. The interfacial abilities of the initial and alloyed interfaces with 37 doping atoms are systematically compared via the formation energy, interface energy, and work of adhesion. According to the calculation results, it is found that the addition of 11 effective alloying elements (i.e., Mg, Ca, Ag, Ce, Au, Pd, Y, Sc, Pt, Hf, and Zr ) can improve the stability and wettability of the Al(111)/TiB2(0001) interface, and the interfacial performance of the Al(001)/TiB2(0001) interface is improved by the introduction of 13 alloying elements ( i.e., Si, Co, Ni, Cu, Zn, Ru, Rh, Pd, Ag, Os, Ir, Pt, and Au). Combined the Voronoi volume and electronic structure analysis, the alloying effect of some solute atoms (Si, Ge, Pt, etc.) mainly depends on the hybridization between Al and alloying atoms, the other (Mg, Sc, Ni, etc.) mainly relies on the distribution of interfacial strain energy. Thus, the alloying effect of solute atoms on the Al/TiB2 interface depends not only on the properties of solute atoms but also on the interface environment. Through further relevant experimental characterization, it was found that the addition of effective alloying elements can effectively improve the dispersion of TiB2 particles in Al-based composites by promoting the formation of the Al/TiB2 interface and improving the interfacial wettability. Generally, our calculation guides the interface alloying strategy to enhance the performance of ceramic/metal interface and provides a fundamental explanation for the related interfacial mechanisms.

Authors : V.L. Karbivskyy, N.A. Kurgan, A.O. Romansky, L.I. Karbivska
Affiliations : G. V. Kurdyumov Institute for Metal Physics of the N.A.S. of Ukraine

Resume : Player et al. [1] suggested that Posner clusters may be the key to the realization of thinking quantum component. Although this assumption still leaves a number of questions, and calculating quantum entanglement using density functional theory is a quite difficult task; understanding the influence of the structural components of hydroxyapatite on the electron density of the crystal as a whole, as well as changes in the DOS with various kinds of substitutions that are possible in the body as a result of biochemical reactions, is an important initial basis both to study the role of Posner clusters in the body, and to deepen knowledge about metabolic processes in bone tissue and the role of hydroxyapatite in them. Within the framework of the density functional theory, the total and partial densities of electronic states were calculated for: calcium hydroxyapatite (Ca-HAP) with various types of the most characteristic substitutions, as well as the Posner cluster, a phosphate tetrahedron and calcium ions of all non-equivalent positions in the corresponding oxygen environment. We also have described the Posner cluster by EXAFS spectroscopy. It is shown that the covalent interaction of the 3s states of phosphorus atoms with the 2p states of oxygen atoms in Ca-HAP is stronger than in an isolated phosphate tetrahedron. A characteristic feature of calcium apatites is the energetic localization of the Ca 3p core electronic states in the valence band region. We substantiate anomalous structural stability and some properties of calcium apatites due to the double-valley effective potential of 3d electrons of calcium. Calcium can be an element with possible d-electron collapse in certain compounds. During the collapse of the d-electron, the overlap of the 3d-radial wave function with the 3p-electron function significantly increases. This effect leads to an increase in the electrostatic interaction between the above-mentioned shells. Thus, the collapse of the d-electron in the 3р53d isoelectronic series leads to an increase in the role of the electrostatic interaction in comparison with the spin–orbit interaction. In Ca-HAP atomic effects play a significant role in forming the shape of the calcium Lα-spectra and, as the result, the role of d-states of calcium in bond formation is levelled by their significant localization, apparently, in the inner valley of effective potential. Studies of the quantum yield spectrum in the region of the LII, LIII absorption edges of Ca in CaF2 revealed the presence of narrow selective maxima, which are not explainable within the framework of theoretical calculations. Apparently, these maxima reflect transitions to vacant d states localized near calcium atoms. [1] Player Th. C., Hore P. J. Posner qubits: spin dynamics of entangled Ca9(PO4)6 molecules and their role in neural processing // Journal of the Royal Society Interface. – 2018. – 15, 147.

Authors : Irene Amiehe Essomba Berenger-1, Guido Ori-1, Mauro Boero-1,2
Affiliations : 1-Université de Strasbourg, CNRS, Institut de Physique et Chimie des Matériaux de Strasbourg, UMR 7504, F-67034 Strasbourg, France 2-Institute of Materials and Systems for Sustainability, Nagoya University, Nagoya 464-8601, Japan

Resume : This study focuses on the set-up of a computational procedure combining classical and first-principles molecular dynamics to explore the intricate interactions involved at the interface made by an ionic liquid (IL) deposited on top of a 2D semiconducting material layered MoS2. The combination of IL with such 2D materials has been found to promote exotic phenomena that could revolutionize the next generation of electronics devices (triboiontronics, superconductivity and quantum interference transistors) [1-3]. However, key questions remain unanswered about the specific interactions occurring between the ILs and the 2D materials. In this work, we provide microscopic insight into a series of imidazolium-based IL in the bulk phase and at the interface with MoS2. A detailed assessment of the structural properties, electronic structure and ion dynamics will be provided in order to explain the complex interplay of the interactions occurring in these systems. Whenever possible, our theoretical results will be discussed and combined with available experimental findings. References [1] Gao et al. Adv. Mater. 1806905, 2018. [2] Costanzo et al. Nature Nanotech. 11, 339 2016. [3] John et al. Adv. Mater. 1800220 2018.

Authors : Reza Mahjouba and Nikki Stanforda
Affiliations : Future Industries Institute, University of South Australia, Mawson Lakes SA 5095, Australia

Resume : Grain boundaries can be divided into two major groups of symmetrical and asymmetrical boundaries. While the former corresponds to the plane of mirror symmetry of two adjacent grains, all others constitute asymmetrical grain boundaries. While first principles simulations have played a remarkable role in shedding light on the property-structure relationships of grain boundaries, they have been restricted to symmetrical grain boundaries due to the challenges originated from the larger number of atoms required to represent asymmetrical boundaries. Nevertheless, such challenges may be confronted, and the character of asymmetrical grain boundaries can be simulated and their interfacial energy barrier, local topology, effect of segregants on the structure, mechanical strength, and thermodynamic stability can be computed combined with their electronic structure. Also, such properties can be compared with those of symmetrical grain boundaries and the contrasting and common features are delineated. Such findings have been compared and corroborated with experimental observations.

10:05 Q&A Session / Break    
Disordered, porous, low-dimensional and hybrid organic/inorganic materials I : Elena Levchenko
Authors : Evelyne Martin
Affiliations : ICube

Resume : In the present talk, I will report on the study of heat propagation in materials at the nanoscale by atomic scale modelling. Thermal properties are determined using the approach-to-equilibrium molecular dynamics (AEMD) strategy [1]. This approach aims at exploiting affordable time trajectories corresponding to transient regimes. The trajectories are modeled by first-principles molecular dynamics (FPMD). After describing the methodology and the informations on thermal transport to which it gives access, I will focus on two applications. The first application of the AEMD is an interfacial molecular layer in between heat reservoirs. It is observed that the heat transport is driven by the transfer at the interface in this case, and this fac tenables to obtain the thermal resistance of the layer from the transient time of the AEMD. The thermal resistance has two contributions, the first one corresponding to the bond between the molecules and the reservoirs, and the second attributed to heat conduction in the diffusive regime along the molecular chains [2]. Then, I will consider the study of amorphous materials and the search for propagative modes using the AEMD. To this purpose, the thermal conductivity of three glasses, GeTe4 [3], Ge2Sb2Te5 [4] and SiO2 [5] has been determined as a function of the length in the direction of the heat transport. Our results are substantiated by quantitative agreement with experiments. A length dependence is observed, that is a compelling evidence of the existence of propagative modes. These results have profound implications in the thermal management of nanodevices. [1] E. Lampin et al, J. Appl. Phys. 114, 033525 (2013) [2] T.-Q. Duong et al, J. Chem. Phys. 153, 074704 (2020) [3] T.-Q. Duong et al, Phys. Rev. Mat. 3, 105401 (2019) [4] T.-Q. Duong et al, RSC Adv. 11, 10747 (2021) [5] E. Martin et al, submitted to Phys. Rev. B

Authors : Alister J. Page, Ben D. McLean, Izaac Mitchell, Esko Kauppinen
Affiliations : University of Newcastle, Australia; Ulsan National Institute of Science & Technology, Korea Ulsan National Institute of Science & Technology, Korea; Aalto University, Finland

Resume : Over the last few decades, catalytic chemical vapor deposition (CVD) has matured as a synthetic technique for producing many low-dimensional inorganic nanomaterials, such as carbon nanotubes (CNTs), graphene, boron nitrides and transition metal dichalcogenides. The general mechanism of graphene and CNT formation during CVD is now well established [1]. However, by and large this picture of nucleation has been developed by considering the chemistry of carbon by itself, when, in reality, there are many other chemical species present in a CVD reaction chamber. In this lecture I will discuss our recent quantum chemical simulations that show the influence of such species on the nucleation and growth mechanisms of carbon nanomaterials during CVD (e.g. H2 [2,3], H2O [4], NH3 [5], etc.). In contrast to carbon nanomaterials, little is known regarding the catalytic pathways underpinning CVD synthesis of boron nitride nanomaterials [1]. I will present the first mechanism explaining the nucleation of boron nitride nanotubes (BNNTs) via CVD of boron oxide and ammonia borane, based on reactive molecular dynamics simulations [6]. Strikingly, BNNTs nucleate via a ‘network fusion’ mechanism, by which distinct BN fragments first form before ‘clicking’ together on the nanoparticle surface. We also reveal key roles played by H2O and H2 partial pressures and the presence of solid-phase catalytic nanoparticles on this mechanism. References [1] B. McLean et al., Phys. Chem. Chem. Phys. 19, 26466-26494 (2017). [2] I. Mitchell et al., Carbon 128, 215 (2018). [3] A. Saeed et al., Adv. Func. Mater., (2020) DOI: 10.1002/adfm.202005016 [4] Hussein, A. et al., Nanoscale. 12, 12263-12267 (2020). [5] C. A. Eveleens et al., Nanoscale 9, 1727 (2017). [6] B. McLean, G. Webber, A. J. Page. J. Am. Chem. Soc. 141, 13385-13393 (2019).

Authors : Mohammed Guerboub(1), Assil Bouzid(2), Evelyne Martin(3), Mauro Boero(1,4), Carlo Massobrio(1), Guido Ori(1)
Affiliations : (1) Université de Strasbourg, CNRS, Institut de Physique et Chimie des Matériaux de Strasbourg, UMR 7504, F-67034 Strasbourg, France; (2) Institut de Recherche sur les Céramiques, UMR 7315 CNRS-Univesité de Limoges, Centre Européen de la Céramique, 12 rue Atlantis 87068 Limoges Cedex, France; (3) Université de Strasbourg, CNRS, ICube, UMR 7357, Strasbourg 23, rue du Loess 67037 Strasbourg, France; (4) Institute of Materials and Systems for Sustainability, Nagoya University, Nagoya 464-8601, Japan;

Resume : Phase-change materials (PCMs) are nowadays widely investigated systems for a wealth of applications, from non-volatile memory to neuromorphic computing and storage applications [1]. Most of these applications are based on their different behavior in terms of optical contrast and resistivity properties between the amorphous and crystalline phases. In this work, by employing first-principles molecular dynamics simulations (FPMD), I will highlight a detailed quantitative assessment of the structural properties and electronic/bonding features of a series of amorphous compositions within the ternary Ge-Sb-Te diagram [2]. In particular, x-ray/neutron structure factors, radial distribution functions, atomic coordination and structural order parameters as well as rings statistics will be reported and discussed in light of the available experimental quantities. This set of results will also be complemented with a detailed account of their electronic properties (electronic density of states) and chemical bonding properties by means of maximally localized Wannier functions. References : [1] Zhang et al. Nature Rev. Mater. 4, 150 2019. [2] Bouzid et al. Phys. Rev. B 96, 224204 2017.

Authors : Gabriele Boschetto, Stefania Carapezzi, Aida Todri-Sanial
Affiliations : LIRMM, Universitè de Montpellier, CNRS, Montpellier 34095, France

Resume : Atomically thin two-dimensional (2D) materials have been —and are still currently being— extensively studied due to their unique mechanical, electrical, and optical properties, which enable the development of innovative devices and technologies. Within the vast chemical space of transition metal dichalcogenides (TMDs), single-layer molybdenum disulphide (MoS2) is with no doubt one of the most studied material due to its stability and its direct optical band gap of 1.8 eV, which make it the ideal candidate to be used in a wide range of nanoelectronic devices, going beyond conventional CMOS technology.(1) Here we look at MoS2 in the context of biosensing, and we study such material as the core component of field-effect biosensors for the detection of cortisol. We want to bridge the gap between materials’ properties and device physics and to do so, we carry out first-principles atomistic computer simulations in the framework of density functional theory (DFT). Recently, MoS2 has been studied as a sensing platform for detecting mainly gas and small biological molecules, such as glucose.(2) Enzymatic biosensing is the most common approach, however, non-enzymatic sensing can provide higher sensor stability and prompt response at the expense of chemical selectivity. Here, we are interested in the non-enzymatic detection of cortisol in human sweat as a mean to monitor the risk of cardiovascular diseases. However, it is not known if such analyte interacts in a suitable way with MoS2, and if so, how. Thus, we thoroughly explore the MoS2/cortisol interaction in terms of both structural, electronic, and charge transfer properties to assess viable sensing mechanisms. We also study the impact of some of the most used metal dopants employed in lab-scale experiments, such as Ni, Pt, Pd, in order to modulate the sensing platform with respect to bare MoS2. Overall, our work ultimately aims to obtain a deep understanding of the properties of MoS2 when used as a sensor to drive the design of devices towards better performance. References (1) Y. Qiao et al., “Fabricating molybdenum disulfide memristors,” ACS Appl. Electron. Mater., 2, 346-370, 2020. (2) G. Jeevanandham et al., “Nickel oxide decorated MoS2 nanosheet-based non-enzymatic sensor for the selective detection of glucose,” RSC Adv., 10, 643-654, 2020.

12:35 Q&A Session / Break    
Materials for energy storage and conversion devices : Michal Hermanowicz
Authors : Julia Wiktor
Affiliations : Department of Physics, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden

Resume : Functional materials used for solar-to-energy conversion have a wide range of complex characteristics. Theoretical modeling of these classes of materials is hence far from straightforward, and requires not only complex computational methods, but also attention to a number of effects that can influence their functioning. There are numerous complications to be considered at the atomic scale in order to create a predictive ab-initio model, for instance temperature effects, electron-hole and electron-phonon interactions, defect formation or interface effects. Additionally, the influence of these effects on each other must be assessed. On the example of complex metal oxides and halide perovskites, I will discuss how some of these effects influence material modeling and efficiencies of devices based on these materials. I will also show which computational methods and schemes are necessary to describe these effects and discuss the interactions between them.

Authors : Rebekka Tesch [1, 2, 3], Michael H. Eikerling [1,2,3], Piotr M. Kowalski [1,2]
Affiliations : [1] Theory and Computation of Energy Materials (IEK-13), Institute of Energy and Climate Research, Forschungszentrum Jülich, Wilhelm-Johnen-Straße, 52425 Jülich, Germany; [2] Jülich Aachen Research Alliance, JARA-CSD and JARA-ENERGY, 52425 Jülich, Germany; [3] Chair of Theory and Computation of Energy Materials, Faculty of Georesources and Materials Engineering, RWTH Aachen University, 52062 Aachen, Germany

Resume : Modeling of electrode/electrolyte interfaces at the atomic scale under applied potential is essential for the understanding of electrocatalytic phenomena and the design of enhanced materials for energy storage and conversion devices. However, a computationally efficient and sufficiently accurate simulation approach to self-consistently model the coupled phenomena at the solid/liquid interface does not yet exist [1]. A promising approach is the Effective Screening Medium Reference Interaction Site Method (ESM-RISM) [2] that combines a density functional theory (DFT) description of the metal electrode with a classical theory of liquids description of the liquid electrolyte. We tested the ability of this method to predict the properties of the charged interface between a partially oxidized Pt(111) surface and an aqueous electrolyte. A thorough assessment of the parameterization of the model and the importance of computing the near-surface water layer explicitly will be discussed. We will demonstrate that ESM-RISM yields correct structural, electrostatic and charging properties of the interface. In particular, it is able to reproduce the peculiar, non-monotonic charging relation of the Pt(111)/electrolyte interface. By comparing to independent theoretical models and explicit simulations, we show strengths and limitations of ESM-RISM for the modeling of electrochemical interfaces. [1] M. J. Eslamibidgoli and M. H. Eikerling, Curr. Opin. Electrochem. 9, 189, 2018. [2] S. Nishihara and M. Otani, Phys. Rev. B 96, 115429, 2017.

Authors : a. Mauricio Rincón Bonilla, b. Fabián García Daza, c. Javier Carrasco, a, d. Elena Akhmatskaya
Affiliations : a. BCAM - Basque Center for Applied Mathematics, Alameda de Mazarredo 14, E-48009 Bilbao, Spain; b. Department of Chemical Engineering and Analytical Science, The University of Manchester, Manchester M13 9PL, United Kingdom; c. Centre for Cooperative Research on Alternative Energies (CIC energiGUNE), Basque Research and Technology Alliance (BRTA), Alava Technology Park, Albert Einstein 48, 01510, Vitoria-Gasteiz, Spain d. IKERBASQUE, Basque Foundation for Science, Plaza Euskadi 5, 48009 Bilbao, Spain

Resume : Developing inexpensive, efficient and scalable all-solid-state batteries (ASSBs) is key to enabling safer and more-energy dense technologies than today's Li-ion batteries. The solid-state electrolytes embedded within ASSBs must be highly conductive, chemically stable and mechanically flexible, while providing intimate contact with the electrolytes and inhibiting dendrite growth. Composite solid-state electrolytes (CSSE) materials comprising a conductive, flexible polymer matrix and either inert or conductive ceramic inclusions have emerged as a promising alternative to fulfill these requirements. Yet, the electrochemical activity of these materials is highly dependent on the interfacial Li-ion dynamics, whose molecular underpinnings remain elusive. In this work, we investigate a CSSE integrating polyethylene oxide (PEO) plus lithium bis(trifluoromethanesulfonyl) imide (LiTFSI) and Li-ion conductive Li7La3Zr2O12 (LLZO) garnet fillers. By combining Molecular Dynamics and a generalized hybrid Monte Carlo approach, we are able to gain unprecedented insights into the Li-ion dynamics at the PEO/LLZO interface. Our simulations verify that interfacial ionic conduction does not occur within atomistic time-scales. Using umbrella sampling calculations, we find that the steep free energy barrier for Li-ion transfer from the garnet surface to the PEO(LiTFSI) phase is the result of strong interactions between the coordinating O2- atoms in the garnet and the Li+ ions in stable surface Li-sites. However, this barrier is asymmetric and Li-ion transfer from the PEO(LiTFSI) phase to the garnet is in fact energetically favorable but limited due to (i) site availability and (ii) path availability (i.e., the availability of a path of coordinating polymer oxygens leading to the available site). We also examine the effect of increasing confinement of the PEO(LiTFSI) phase between the two garnet surfaces, which has been speculated to lead to fast ionic diffusion. We do not find evidence for such a phenomenon, even when the PEO(LiTFSI) phase is confined to a 1 nm thin layer. Instead, we report the complete depletion of Li+ from the polymer phase at the examined concentrations, due to adsorption on the garnet surfaces. Consequently, a significant drop in conductivity is expected to occur at concentrations below particle close-packing, as recently shown in experimental work. Our findings provide clues for the formulation of strategies aimed at increasing interfacial ion exchange in PEO:LLZO CSSEs and confirms the absence of conduction enhancement across the garnet surface at temperatures above the glass transition, when the PEO phase is fully amorphous.

Authors : Tanguy Picard, Long Hoang Bao Nguyen, Jean-Sébastien Filhol, Marie-Liesse Doublet, Nicolas Sergent, Cristina Iojoiu, Fannie Alloin
Affiliations : Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, Grenoble INP, LEPMI, 38000 Grenoble, France ; ICGM, University of Montpellier, CNRS, ENSCM, Montpellier 34095, France ; RS2E French Network on Electrochemical Energy Storage, FR5439, Amiens, France

Resume : Calcium batteries are considered among the most suitable candidates for post-lithium batteries due to their high charge density, theoretical specific capacity for Ca0 of 1,340 mAh/g and a volumetric capacity similar to that of Li i.e. 3,833 mAh/mL. Considering its reactivity and properties, calcium metal is safer than lithium as it is less prone to burst in flame in contact with air. Additionally, Ca plating-stripping, reported so far, did not show any dendrites growth, therefore if this anode can be combined with a solid polymer electrolyte (SPE), the battery challenges related to the safety can be overcome. On the other hand, calcium plating is impaired by sloppy surface reactivity due to large passivation layer. To date, the electrolyte is still the weak link to develop properly Ca batteries. Lot of studies are ongoing to find the most suitable candidate with good chemical and electrochemical stabilities, high ionic conductivity and controlled reactivity with Ca metal in order to prevent the formation of non-conducting passivation layer. To design high performance electrolyte, a fundamental understanding of the earth alkaline ions transport mechanisms, by combining modelling and experimental results, is of prior importance. The ether oxide electrolytes are among the most studied but their behaviour for Ca application, in term of conductivity, dependency of conductivity with temperature and salt dissociation are very different as compared to Li electrolytes. Therefore, in the frame of FET-open VIDICAT project, DFT and MD are used to both i) go deeper in the understanding of the unexpected behaviour of electrolytes, ii) study mass transport in divalent based solid polymer electrolytes. Such work would allow us to guide experimental work towards promising polymer structures. Experimentally, with the ether Ca electrolytes we observed a decrease in conductivity after a certain temperature which is depending on the concentration. By using DFT models, we are able now to explain this behaviour. To do so, thermodynamics of salt dissociation has been investigated and it has been proved that, considering solvent structuration and salt dissociation, the global entropy of solubilisation is strongly negative, making the entire solubilisation process impaired by increasing temperature. In parallel, a new conductivity model has been proposed based on weak electrolyte and free volume theories, correlative to the common VTF model. This model is not parametric, meaning that all values have physical meaning and can be measured and/or computed, its formalism depends on transport mechanism (perfect gas, gas kinetics, Brownian motion…). Therefore, by comparison to experimental data, transport mechanisms can be identified. Furthermore, as all parameter can be computed, ionic conductivity of materials can be computed with no need of prior experiments. A benchmark of different chemistries is ongoing and is probed by comparison with electrolytes synthesised by VIDICAT teams that would allow us to find new SPE structure with lower laboratory effort.

Authors : Carlos de la Cruz, Rebeca Marcilla, Andreas Mavrantonakis*
Affiliations : IMDEA ENERGY

Resume : Electrical energy storage and conversion is crucial to a clean, sustainable, and secure energy future. Redox-Flow Batteries (RFBs) are considered a promising energy storage technology, which can exhibit high potential, high efficiency, and the ability to decouple the power and energy density. During the last few years there has been a an increased interesest in the field of organic-based redox flow batteries, because of their structural diversity, abundancy, environmental friendliness and price compared to the vanadium compounds. High-throughput computational screening offers the possibility of exploring thousands of molecules for desirable properties without the need for experimental trial and error.[1] In this study, we will show how the application of computational chemistry methods allows us to gain deep understanding in the redox chemistry of phenazines, which are a new family of organic molecules resently used in RFBs.[2] Density Functional Theory (DFT) calculations are performed in order to explore the redox behaviour of phenazine molecules functionalized with single and multiple electron-donating and ?withdrawing functional groups (FGs) in aqueous and non-aqueous media. The DFT results provide key insights into their structure?activity properties that may be used in the design and tuning of new molecules for electrochemical energy storage in aqueous or non-aqueous media[3]. Moreover, the DFT calculations reveal how the number and the position of the FGs can modify the redox potential, and how the FGs can also be related with side reactions or stability. References [1] S. Er, C. Suh, M. P. Marshak and A. Aspuru-Guzik, Chemical Science, 2015, 6, 885-893. [2] A. Hollas, X. Wei, V. Murugesan, Z. Nie, B. Li, D. Reed, J. Liu, V. Sprenkle and W. Wang, Nature Energy, 2018, 3, 508-514. [3] C. de la Cruz, A. Molina, N. Patil, E. Ventosa, R. Marcilla and A. Mavrandonakis. New insights into phenazine-based organic redox flow batteries by using high-throughput DFT modelling. Sustainable Energy Fuels, 2020,4, 5513-5521.

15:35 Q&A Session / Break    
Poster session II : Yannick Dappe, Michal Hermanowicz
Authors : Debolina Misra, Satyesh K. Yadav
Affiliations : Department of Physics, Indian Institute of Information Technology Design and Manufacturing Kancheepuram, Chennai, 600127, India; Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai, 600036, India.

Resume : In heterogeneous catalysis, noble metals such as Pt, Pd and Au have shown the highest activity till date. However, the high cost and low abundance of noble metals pose a threat in designing cost effective and efficient catalyst materials and led to the quest for designing non-noble materials as catalysts. Meanwhile, single atom catalysts (SAC) consisted of atomically dispersed metal atoms on a support came as a potential rescue from the above-mentioned problems because of their high surface-to-volume ratio, which make them potentially more efficient catalyst. In addition, using the dispersed noble metal atoms as catalysts can drastically reduce the cost of catalyst fabrication. Nevertheless, agglomeration of the noble metal single atoms (SAs) into clusters surfaced as a practical issue during fabrication of SAC. Using first-principles modelling based on density functional theory calculations we show that SACs can indeed be stabilized in their atomic form on suitable oxides through interstitial doping via ion-implantation. Earlier we have shown that transition metal (TM) atoms if implanted in a rock-salt structured oxides (such as MgO or BaO) can be stabilized as interstitials, depending on the charge state, relative ionic radii and oxygen affinity of the TM dopant in the host. In this work, we show that if TM dopants such as Nb is implanted in BaO, it will be stable at subsurface interstitial, and the Nb-doped BaO acts as an excellent support for noble metal single atoms such as Au. We have carried out an in-detail investigation of the preferred binding site of the Au single atom on the Nb-doped BaO (001) support and our calculations showed that near-Nb surface O-top sites act as the strongest binding sites for Au single atoms with binding energy as high as -3.83 eV. In addition, our calculations also revealed that if number of Au atoms is increased to a maximum of 5 (per Nb dopant) on the BaO oxide support, all the 5 Au atoms are anchored strongly on the doped oxide. To find out the possible cause behind this strong binding of the Au single atoms on BaO in presence of Nb dopant, we performed the Bader charge analysis calculations which revealed that it is the charge transfer between the Nb and the Au atoms which is responsible in binding the Au atoms strongly on the surface and this charge transfer also prevents cluster formation among the Au single atoms. Hence, a charge transfer between the Nb atoms implanted in BaO and the Au adatoms helps in binding the SACs on the support and also prevent agglomeration. The maximum number of the noble metal SA that can be anchored on the support in this way depends on the preferred charge state of the implanted TM dopant in the host. Our work indicates an altogether new way of stabilizing noble metal single atoms on oxide supports and can lead to further studies involving a range of dopants and a variety of oxides as supports.

Authors : Oskar Cheong [1][2][3], Rebekka Tesch [1][2][3], Thomas Bornhake [1][2], Alison Shad [1][2][4], Michael H. Eikerling [1][2][3], Piotr M. Kowalski [1][2]
Affiliations : [1] Theory and Computation of Energy Materials (IEK-13), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; [2] Jülich Aachen Research Alliance, JARA-CSD and JARA-ENERGY, 52425 Jülich, Germany; [3] Chair of Theory and Computation of Energy Materials, Faculty of Georesources and Materials Engineering, RWTH Aachen University, 52062 Aachen, Germany; [4] Walter Scott, Jr. College of Engineering, Colorado State University, Fort Collins, CO 80523, USA

Resume : Development of effective electrocatalysts for reduction of anthropogenic CO2 emission could be facilitated by effective simulation-aided screening of catalyst materials. However, since most electrocatalytic reactions take place in the presence of aqueous solution, to correctly compute surface chemistry the solvation effects must be accounted for. Previous density functional theory (DFT) studies have shown significant impact of the addition of explicit water molecules on the computed selectivity of CO and HCOOH for the CO2 reduction reaction on the Pb(100) surface [1]. While explicit computation of water molecules is widely used, albeit computationally intensive, implicit solvation methods (e.g. SCCS [2], ESM-RISM [3]) represent an attractive alternative in computation of surface chemical reactions at solid-aqueous phase interface. Here, we discuss results of our test studies on the performance of explicit and implicit solvation computational approaches on the selectivity of the CO2 reduction reaction pathway towards HCOOH/CO on Pb(100) surface under applied potential. The advantages and limitations of both approaches to the computation of electrochemical reaction pathways will be highlighted. [1] Fan et al., ACS Catal. 10, 10726 (2020). [2] Andreussi et al., J. Chem. Phys. 136, 064102 (2012). [3] Nishihara and Otani, Phys. Rev. B, 96, 115429 (2017).

Authors : D. F. Carvalho (1), M. A. Martins (2), M. R. Correia (1)
Affiliations : (1) Department of Physics and i3N, University of Aveiro, Portugal; (2) Department of Materials and Ceramics Engineering and CICECO, University of Aveiro, Portugal

Resume : Metallic nanoparticles (MNPs) exhibit very interesting optical properties for applications in several areas of science, related with the plasmonic phenomena [1]. They are widely used in photovoltaic devices, light emitting devices, chemical sensors, development of substrates for surface-enhanced Raman spectroscopy (SERS), etc. The determination of the localized surface plasmon (LSP) resonance frequency of the MNPs is fundamental in the comprehension of the interaction mechanisms with other materials. The LSP resonance frequency of the MNPs depends on the chemical composition, size, shape, and the surrounding medium. However, both the interaction between MNPs and the presence of a substrate are also important in changing the resonance frequency and the near-field distribution, being many times neglected. Understanding the interaction between a set of MNPs in a 2D array on a substrate is essential for the development of optimized structures. Experimentally, the random distribution of MNPs in 2D arrays is the most common. Some studies of the coupling of MNPs with gallium nitride (GaN) to increase the emission efficiency and as SERS substrates have been published [2,3], but the influence of the organization of MNPs has not been sufficiently explored. In this study, an analytical simulation of the optical properties of random arrays of MNPs is developed, considering the interaction between MNPs and the presence of a GaN substrate, through the discrete dipole approximation (DDA) and the image charge method [4]. The DDA allows to obtain the scattering and absorption of a set of particles, small enough to be described by a dipole. In this model, the dipole moment of the MNP is calculated using the empirical polarizability of Kuwata [5], that allows the study for larger MNPs (maximum dimension of ~100 nm) in comparison with the quasistatic polarizability. The effect of the substrate is considered through the image method, where the MNP will induce the presence of charges on the substrate, which can be represented by a new dipole moment. It is demonstrated that the increasing of the surface density of MNPs in the arrays and the GaN substrate results in a shift of the resonance peaks of the plasmons to higher wavelengths, a decrease in extinction cross sections, and an increase of the near-field enhancement. This study allows to theoretically predict the optical properties of MNPs arrays on a GaN substrate through the input of topographic information from SEM or AFM, providing important information concerning the coupling plasmonic phenomena with semiconductors. [1] H. Yu et al., Npj Comput. Mater., 5, (2019); [2] J. Henson et al., Appl. Phys. Lett., 95, (2009); [3] M. Zhang et al., Sensors Actuators B, 253, (2017); [4] W. Yang et al., J. Chem. Phys., 103, (1995); [5] H. Kuwata et al., Appl. Phys. Lett., 83, (2003). This work was developed within the scope of the project i3N, UIDB/50025/2020 & UIDP/50025/2020, financed by national funds through the FCT/MEC.

Authors : Ziga Casar, Aslam Kuhni Mohamed, Karen Scrivener, Paul Bowen
Affiliations : Laboratory of Construction Materials Institut des Matériaux Ecole Polytechnique Fédérale de Lausanne (EPFL), Institute for Building Materials ETH Zurich, Laboratory of Construction Materials Institut des Matériaux Ecole Polytechnique Fédérale de Lausanne (EPFL), Laboratory of Construction Materials Institut des Matériaux Ecole Polytechnique Fédérale de Lausanne (EPFL)

Resume : Cement is the most manufactured product on earth by mass. While being a low impact material per kg, due to the colossal mass of cement being produced it still accounts for 5-8% of anthropogenic CO2 emissions, the highest industrial share after power production [1]. The main hydration product is calcium silicate hydrate (C-S-H), which comes from the reaction between tricalcium silicate (main component of Portland cement) and water. This nanocrystalline material has a variable stoichiometry, as a variable Ca/Si ratio, and grows in sheet like structures. In recent years a combination of experimental techniques, as NMR, and advanced molecular modelling techniques (Molecular Dynamics and DFT), brought a better insight into the underlying structure [2]. The recently developed brick model [3] is the first computational model of C-S-H, which shows an excellent agreement with experimental data, and therefore provides a systematic approach in designing computational structures, which are observed experimentally.  Due to the colossal amount of cement being produced and being needed, an alternative which could lead to a reduction in CO2 emissions is needed. Such an alternative material needs to be widely available. It was shown that the most realistic alternative is in replacing part of the Portland cement with so-called SCMs (supplementary cementitious materials) [4]. While those materials perform well in reducing the CO2 emissions, they result in a lower compressive strength in the first days, which comes from the lower reactivity. Luckily, researchers have shown that with additional of zinc we can increase the reactivity in this first hydration period [5]. But this is only true for tricalcium silicate, since if used in Portland cement it retards the reaction. While there is evidence that zinc is indeed incorporated into the C-S-H structures, the underlying mechanism of hydration is still unclear. The Brick Code and EricaFF2 force field have shown great accuracy when predicting the C-S-H structure. We further developed both to accommodate C-S-H structures with incorporated zinc, meaning we extended the force field with the needed zinc interactions for cementitious environments, and carefully validated them. With this we can compare favorable zinc sites in C-S-H, and compare them with NMR data, in order to obtain knowledge of the exact location of zinc in C-S-H. With this we can start building a hypothesis on the underlying mechanisms of hydration, and why C-S-H sheets grow twice as fast in presence of zinc, which would open a door in-to engineering the reactivity of cementitious systems, and reduce the global CO2 footprint by a significant amount.   [1] U. Environment, K.L. Scrivener, V.M. John, E.M. Gartner, Eco-efficient cements: Potential, economically viable solutions for a low-CO2, cement-based materials industry, Cem. Concr. Res. 114 (2018) 2–26  [2] A. Kumar, B.J. Walder, A. Kunhi Mohamed, A. Hofstetter, B. Srinivasan, A.J. Rossini, K. Scrivener, L. Emsley, P. Bowen, The Atomic-Level Structure of Cementitious Calcium Silicate Hydrate, J. Phys. Chem. C. 121 (2017) 17188–17196.  [3] A. Kunhi Mohamed, S.C. Parker, P. Bowen, S. Galmarini, An atomistic building block description of C-S-H - Towards a realistic C-S-H model, Cem. Concr. Res. 107 (2018) 221–235.  [4] B. Lothenbach, K. Scrivener, R.D. Hooton, Supplementary Cementitious Materials, Cem. Concr. Res. 41 (2011) 1244–1256  [5] A. Bazzoni, M. Suhua, Q. Wang, X. Shen, M. Cantoni, K.L. Scrivener, The effect of magnesium and zinc ions on the hydration kinetics of C3S, J. Am. Ceram. Soc. 97 (2014) 3684–3693. 

Authors : Christopher M. Andolina, Jacob G. Wright, Wissam A. Saidi
Affiliations : Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15216, USA

Resume : arious industrial/commercial applications use Al-Mg alloys, yet, the Mg added to Al materials to improve strength is susceptible to surface segregation and oxidation, leaving behind a softer and Al-enriched bulk alloy. To better understand this process and provide a systematic methodology for investigating dopants that can mitigate corrosion, we have established a robust atomistic deep neural net potential (DNP) using a dataset generated with first-principles density functional theory (DFT). Our DNP was validated against DFT values and iteratively trained to produce results that have high fidelity reproducing elemental and intermetallic Al-Mg systems' general bulk properties such as point defects, non-ground state lattice configurations, mechanical properties, various surface terminations for miller indices <4, and planar defects. We applied our DNP to study surface segregation enthalpies in Al-Mg slabs of Mg molar ratios from 0 to 20% using a hybrid Monte Carlo and molecular dynamics (MC/MD) approach at temperatures from 200 to 800 K. Our calculations show a linear trend in the formation energy and Al-Mg alloy density as a function of temperature consistent with experimental literature. We also observe that the surface segregation for Al-Mg alloys is anisotropic such that (111) < (100) < (110), with (111) surfaces displaying the lowest segregation enthalpies and Mg enrichment. Furthermore, we model the segregation tendencies obtained from the MC/MD simulations by adapting a recently introduced isotherm model for grain boundary segregation. Our results show that this new model describes the MC/MD with higher fidelity than the McLean and Fowler-Guggenheim isotherm models.

Authors : I.-B. Lin, D. Bocharov, I. Isakovica, S. Piskunov
Affiliations : Institute of Solid State Physics, University of Latvia, 8 Kengaraga str., Riga LV-1063, Latvia

Resume : Hydrogen production from water in photoelectrochemical cells is an important step toward carbon-free fuel. The most convenient process for hydrogen production is the splitting of water, which demands a catalytic reaction involving a semiconducting photocatalyst. In this study we discuss an approach allowing us the modelling of photocatalyzed reactions on surfaces of semiconducting TiO2 nanotube. Our approach is based on time-dependent density functional theory (TDDFT) that explicitly accounts for the evolution of one-electron excited states. TDDFT for excited-state calculations involving single excitations to define absorption spectra of catalyst, as well as for simulation of charge transfer processes at the surface of TiO2 nanotube. Our calculations describe a large fraction of the processes occurring catalyst/water interface. These processes initiate at excitation and eventually result in charge separation, delocalization, and relaxation. They are accompanied by electron/hole trapping near the surface and their transfer back to the electrode or to the electrolyte. From our calculations we predict the possibility of hole-mediated splitting of molecularly adsorbed water on the surface of TiO2 nanotube. In our study we show a general computational strategy for describing photoexcited oxide/adsorbate interfaces and also demonstrate that the occurrence of water dissociation on the TiO2 nanotube depends sensitively on the local atomic environment and external parameters such as temperature. Funding from M-ERA.NET project CatWatSplit is greatly acknowledged.

Authors : A. Zachinskis, D. Bocharov, S. Piskunov, J. Purans
Affiliations : Institute of Solid State Physics, University of Latvia, Riga, Kengaraga 8, LV-1063

Resume : Gallium oxide has become one of the most attractive materials of today. The reason for this large interest is the extremely promising properties for electronic and optical applications of this wide bandgap material together with high thermal and chemical stability. The unique stability of the electrical conductivity under thermal and ionizing radiation stress, combined with high transparency and the toughness under high electrical field opens up new possibilities for applications as extended spectral range transparent electrodes in ultraviolet (UV) optoelectronics (e.g. UV LEDs or photovoltaics) and robust Ohmic contacts for use in extreme environments. In this study, we present ab initio calculations of electronic structure for different existing phases of Ga2O3, as well as the impact of F-center formation on electronic properties predicting the position of induced level in band gap. The obtained results allow us to explain the applicability of Ga2O3 polymorphs for photovoltaic applications. Financial support provided by Latvian Council of Science project No. lzp-2020/1-0345 is greatly acknowledged.

Authors : Valery Nebol'sin1, Vladimir Yuryev1, Elena V. Levchenko2
Affiliations : 1 Department of Radio Engineering and Electronics, Voronezh State Technical University, 14 Moskovsky Pr, 394026 Voronezh, Russia 2 College of Engineering, Science and Environment The University of Newcastle, NSW2308, Australia

Resume : The mechanism of interaction of oxygen with graphene is not yet completely clear and requires further investigation. Multi-stage interactions between the oxygen and graphene are considered in this work. There is no unambiguous chemical formula of this oxygen-containing compound in the literature. At the same time, there are many models for the structure of “graphene oxide”. Most recent studies on the local structural characterization of graphene oxide sheets show a large degree of structural disorder in the carbon skeleton due to the random arrangement of oxygen functional groups. In terms of oxidation "graphene oxide" can be very different and may contain from 3% to 40% oxygen in weight. Furthermore, there are no precision analytical methods to accurately phase-characterize this two-dimensional material. Moreover, the conditions of "graphene oxide" synthesis and its "reduction" have a strong effect on the structure and chemical composition of this 2D material. In this work, we claim that “graphene oxide” cannot be considered as an independent phase of a substance, in fact, an oxide. An oxide, by definition, is a compound of an element with oxygen in the "-2" oxidation state. Therefore, for the realization of the oxidation state of "-2", oxygen needs to form two chemical bonds with two neighboring graphene atoms. On the surface of graphene, however, a large number of oxygen atoms share only one single bond with carbon atoms, which contradicts the definition of oxide. Furthermore, one of the criteria for the existence of an independent phase of a substance is its stability under wide conditions, which, in turn, is determined by the strength of the chemical bond. The greater the multiplicity and the shorter the length of the chemical bond, the stronger it is. But the chemical bond between C and O, which is formed in "graphene oxide", cannot be compared with the chemical bond in CO and CO2 molecules neither in multiplicity, nor in length, and, consequently, neither in energy nor in strength. The interaction of graphene with oxygen can be considered a chemical reaction with the formation of an oxide phase only in the case of detachment of an atom from the graphene lattice and the formation of two or three units of valence (and strong) bonds of a carbon atom with oxygen, i.e. CO and CO2 molecules. And what is called in the literature as "graphene oxide" is in fact the result of the first stage of graphene oxidation, where we can observe an adsorption absorption of oxygen atoms by the carbon surface of graphene. It is obvious that chemisorption of oxygen-containing complexes occurs on the graphene surface. We show that the processes of thermal reduction of "graphene oxide" described in the literature, resulting in the formation of CO and CO2 molecules, are in fact the final stage of the graphene oxidation process. Hence, it follows that "graphene oxide" as an independent phase does not exist and cannot exist, and the existing models of the structure and chemical formulas of "graphene oxide" are speculative.

Start atSubject View AllNum.
08:30 Break    
10:30 Three-min Thesis competition | Plenary    
13:30 Lunch break    
Multiscale and hybrid approaches : Yannick Dappe
Authors : G. Calogero, D. Raciti, A. Sciuto, G. Fisicaro, I. Deretzis, A. La Magna
Affiliations : CNR Institute for Microelectronics and Microsystems, Catania, Italy

Resume : Nanosecond laser annealing (LA) is a powerful tool for micro- and nano-electronic manufacturing processes where strongly confined heating is desirable, e.g., low-temperature processing of 3D sequentially integrated devices. Optimizing the LA process along with the device design is challenging, especially for complex 3D nanostructured systems with various shapes and phases. To this purpose, modelling critical LA phenomena at the nanoscale with atomic resolution is essential. Most state-of-the-art simulators are based on continuum models, which cannot capture the crystal-orientation dependent kinetics of solid-liquid interfaces, or the formation and evolution of point and extended defects, which can be crucial to corroborate experiments. We present an open-source, multi-scale LA simulation tool for group IV semiconductors and alloys, based on the parallel coupling of a continuum, finite-elements, µm-scale thermal solver with a super-lattice Kinetic Monte Carlo atomistic model. Benchmarks against continuum phase-field models and experimental data validate the approach. LA of a Si(001) surface is studied varying laser fluence and pulse shape, assuming both homogeneous and inhomogeneous nucleation, revealing how liquid Si nuclei grow and coalesce during irradiation, and how defects affect their evolution. The formalism is applicable to any system where the atom kinetics is space- or time-dependent, due to temperature or strain.

Authors : Artem Fediai, Jonas Armlder, Wolfgang Wenzel
Affiliations : Karlsruhe Institute of Technology

Resume : Doping is a key technology in both organic and inorganic electronics, allowing to control the type and the magnitude of the conductivity over many orders of magnitude. In organic electronics, it is primarily used to reduce injection barriers and voltage drop on the electron/hole transporting layers of organic light-emitting diodes (OLEDs). Recently, the efficiency of doping has been correlated with the properties of underlying molecules, but a multiscale method that would start from the atomistic level and go all scales towards mesoscopic level, is still missing. The multiscale method will be presented that allows to establish a direct link between the type of dopant and host molecules and the macroscopic properties of the constituent doped material. In this method, the kinetic Monte-Carlo method acts as the last component of the multi-scale approach, in which first the morphology of doped materials is generated using the Deposit method (Monte-Carlo protocol that mimics the vapor deposition of thin organic films [J. Comput. Chem. 2013, 34, 2716-2725]) and then the Coulomb interactions in host-dopant charge-transfer complexes are computed using ab-initio methods, based on the recently improved Quantum Patch method [J. Chem. Theory Comput. 2014, 10, 9, 3720-3725, J. Chem. Theory Comput. 2021, doi: 10.1021/acs.jctc.1c00036]. The latter has been extended here to also treat charge-transfer complexes embedded in a material [Proc. of SID Conf. “Display week 2020” 2020]. As a test case, we have applied this approach to three amorphous doped organic materials: alpha-NPD:F6TCNQ, SpiroTAD:F6TCNQ and SpiroOMeTAD:F6TCNQ. Due to well-balanced assumptions, our multiscale approach allows us to calculate the Coulomb interaction energies of several tens of host-dopant pairs and obtain the distribution of the charge-transfer states energies at the relevant distances. This information along with the energy levels distribution was used to compute the density of states and the doping efficiency. We then discuss the role of the near-field Coulomb interactions in enhanced doping efficiency. Our method can be used for in silico design of new dopant molecules and establishing molecular origin of doping efficiency.

Authors : Silvio Osella
Affiliations : Chemical and Biological Systems Simulation Lab, Centre of New Technologies, University of Warsaw, Banacha 2C, 02-097 Warsaw, Poland

Resume : In recent years, research effort has been devoted to the generation of hybrid materials which change the electronic properties of one constituent by changing the optoelectronic properties of the other one. The most appealing and commonly used approach to design such novel materials relies on combining organic materials or metals with biological systems like redox-active proteins. Such hybrid systems can be used e.g. as bio-sensors, bio-fuel cells, biohybrid photoelectrochemical cells and nanosctuctured photoelectronic devices. Although experimental efforts have already resulted in the generation of a number of hybrid bio-organic materials, the main bottleneck of this technology is the formation of a stable and efficient (in terms of electronic communication) interface between the biological and the organic/metal counterparts. In particular, the efficiency of the final devices is usually very low due to two main problems related to the interfacing of such different components: charge recombination at the interface and the high possibility of losing the function of the biological component, which leads to the inactivation of the entire device. Here, we present a multiscale computational design which allows the study of the direct electron transfer mechanism in complex interfaces for stable and highly efficient hybrid materials for biomimetic application, consisting of single layer graphene (SLG) as organic material/metal and small light harvesting protein complex as biological counterpart, linked together via a self-assembly monolayer (SAM), in order to create novel biomimetic materials for solar-to-fuel, bio-transistors or bioorganic electronic applications.

Authors : Gerhard Goldbeck, Gabriele Mogni, Welchy Leite Cavalcanti
Affiliations : Goldbeck Consulting Ltd.; Fraunhofer Institute for Manufacturing Technology and Advanced Materials (IFAM)

Resume : VIMMP is an EU-funded project aimed at establishing the web's first marketplace platform dedicated to facilitating the uptake and execution of scientific and engineering materials science modeling simulations. VIMMP will thus streamline and promote the exchange of data and modeling software information between all materials modelling stakeholders, for the benefit of increased innovation in European manufacturing industry. The project is scheduled to become operational as a commercial platform towards the beginning of 2022, and will offer among other services an advanced ontology-driven search engine, capable of retrieving appropriate materials modeling software and workflows solutions offered by various independent providers. These solutions will be at the disposal of industry and academic end-users willing to embrace the power of materials science simulations as part of their R&D efforts. In addition, the VIMMP marketplace will also offer a "translation" service, in the form of expert consultants available to assist marketplace users in assembling and performing their modeling workflows and simulations, as well as to conduct more general contract-research activities. Finally, the marketplace will also include plenty of training and documentation material, interactive user forums, data and software repositories, and also access to external providers of high-performance computing facilities and resources for executing the desired simulations. Please find more information about the VIMMP marketplace project on the following website:

Authors : Didem Ozkaya, Caglanaz Akin, Busra Demir, Ersin Emre Oren
Affiliations : Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey. ; Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey. ; Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey. ; Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey. Department of Materials Science & Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey.

Resume : Proteins are known to have a role in mineral nucleation, growth, and structure creation, as well as providing molecular scaffolding for the formation of hard tissues such as bone and dental tissues [1]. Using biomimetic approaches, we can mimic this scenario by generating genetically engineered peptides/proteins for inorganic material synthesis, formation, and assembly [2]. A deeper understanding of the interaction between peptide and inorganic surfaces, also binding affinities or specificities may aid in the development of novel peptides with desired features in engineering and medicine [3,4]. Designing new gold peptide interaction is crucial for applications in biosensors, drug delivery, imaging and diagnostics, implants, and other biomedical and biotechnology applications. In this research, we use atomistic methods to understand how individual amino acids and small peptides interact with Au (111) surface that will allow us to design novel moieties for gold functionalization. Peptides which are used in this research are chosen systematically by induction starting from the best binding amino acids among 20, up to the chain of 7 amino acid peptides. Structures of chosen peptides were optimized on the gold surface by conformational search. Then, best conformation(s) used as starting structure for molecular dynamics (MD) simulations. Peptide stability and binding affinities were analyzed with the MD trajectories. Also, binding free energy calculations were done by using MM-GBSA method [5]. [1] Mann,S. (1988) Molecular recognition in biomineralization. Nature, 332, 119–124. [2] Ball,P. (2001) Life’s lessons in design. Nature, 409, 413–416. [3] Hnilova, M. (2008) Effect of molecular conformations on the adsorption behavior of gold-binding peptides, Langmuir, 24, 12440-12445. [4] Oren, E.E. (2007) A novel knowledge-based approach to design inorganic-binding peptides, Bioinformatics, 23, 2816-2822. [5] Tsui, V. (2000) Theory and applications of the generalized Born solvation model in macromolecular simulations, Biopolymers, 56, 275-291.

15:20 Q&A session    
15:35 Symposium G GSA and Poster awards    
Start atSubject View AllNum.
Materials modeling for nanoelectronics : Elena Levchenko
Authors : Enrico Piccinini
Affiliations : Applied Materials Italy, Appleid AI | MDLx

Resume : The characteristic size of state-of-the art electronic devices is in the 10-nm range or so. At this size every single atom, ion or defect counts and impacts the device performance. On the one side, the production process should be highly optimized and controlled to reduce device-to-device variability and increase reliability; on the other side, standard TCAD simulations that rely on the Boltzmann equation must be enhanced to account for nearly single-carrier phenomena. After the huge increase of the computational power and the shrinkage of the device size, atomistic simulations have shown their potential for investigating the material properties at the nanoscale. Nevertheless, full atomistic simulations of nanoelectronic devices are possible only for concepts in academic research, and engineers cannot rely on them for the optimization of mass products. Fully automated, high-throughput material-to-device toolchains have been sought and designed by process and device engineers. The proprietary software Ginestra® fills the gap between first-principle simulations and standard TCAD solutions. Data from ab-initio and atomistic simulations are fed into Ginestra® to let it be predictive or to perform material screenings with respect to some target specifications of a given device under test. Some examples in the realm of memory devices are presented, and the importance and the accuracy of some key parameters on the outcomes is assessed and discussed.

Authors : A. Deniz Özdemir (1), Pramit Barua (1), Feliks Pyatkov (1), Frank Hennrich (1), Yuan Chen (2), Wolfgang Wenzel (1), Ralph Krupke (1), Artem Fediai (1)
Affiliations : (1) Institute of Nanotechnology, Karlsruhe Institute of Technology, Hermann-von-Helmholtz-Platz 1, 76344, Germany (2) School of Chemical and Biomolecular Engineering, The University of Sydney, Camperdown, NSW 2006, Australia

Resume : Low-dimensional all-carbon nanostructures are promising candidates for future flexible and transparent transistors, as well as light-emitters [Pyatkov et al, Nature Photonics 10 , 420–427 (2016)]. Carbon nanotube field-effect transistors with graphene electrodes offer high thermal stability and allow for ultra-flat devices able to carry high current densities. Understanding quantum transport in such systems is essential for device manufacturing. However, conventional simulation methods are limited to local contacts with an overlap width of only a few angstroms, and first-principle quantum transport calculations in systems with realistic contact geometries remained unfeasible. Simulation methods for extended CNT/metal contacts were presented in prior theoretical studies [Fediai et al, Nanoscale, 8, 10240 (2016)]. Here, we report quantum transport calculations based on the non-equilibrium Green function formalism and density functional theory, which are capable of treating ~100nm long CNT/graphene overlap regions and electrostatic gating at a pure ab-initio level. These calculations are compared with experiments where graphene electrodes were fabricated and dielectrophoretically bridged with single semiconducting CNTs followed by an electrical characterization over a wide temperature range. The simulation results are in agreement with the experimental data and the fabricated devices exhibit ambipolar behavior, consistent with the theory. Due to large CNT/graphene spacing and small CNT/graphene overlap, only a fraction of the quantum conductance can be achieved for the on-state conductivity. A remarkable feature of the all-carbon transistor with graphene electrodes is electrostatically induced doping of the carbon-system in the contact regions caused by the gate electrode. This is made possible by the low density of states in graphene near the K-point. We hope that this newly developed and verified simulation approach will accelerate progress in fabrication as well as the understanding of advanced all-carbon devices.

Authors : Kana Ishisone1, Mauro Boero1,2
Affiliations : 1 Université de Strasbourg, CNRS, Institut de Physique et Chimie des Matériaux de Strasbourg, UMR 7504, F-67034 Strasbourg, France 2 Institute of Materials and Systems for Sustainability, Nagoya University, Nagoya 464-8601, Japan

Resume : Our study aims at providing a fundamental insight into the multifold interactions occurring in ionic liquids (ILs), and at the interface with 2D nanomaterials. To date, ILs are used as electrolytes in the three-terminal nano-devices leading to major experimental observations such as electric-field induced superconductivity and quantum interference. These systems are prone to have applications in next-generation memory devices for neuromorphic computing. Yet, their integration in real devices is hampered by the lack of fundamental understanding of the interactions occurring inside the IL and at the interface with a 2D WSe2 substrate. By first-principles simulations, we provide a microscopic picture of 1-Ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([EMIM] [TFSI])[1], a currently targeted IL. Dynamical simulations provide clear picture of the structural and dynamical properties of this IL, not yet unraveled. Moreover, an analysis of the electronic structure and partial charges of the two components, cation and anion, allow to rationalize the nature of the electrostatic interactions and hydrogen bonding properties. Combining with complementary studies of the WSe2 substrate, our work has paved the route to simulate the real experimental composite devices. The microscopic details unraveled in this ongoing project are of primary importance for boosting this innovative research line and to unravel all the fundamental microscopic features escaping experimental probes. References [1] ACS Nano vol. 8, page 923 (2014)

Authors : Jonas Armleder (1), Timo Strunk (2), Franz Symalla (2), Pascal Friederich (2,3), Jorge Enrique Olivares Peña (1), Tobias Neumann (2), Wolfgang Wenzel (1), Artem Fediai (1)
Affiliations : 1) Institute of Nanotechnology, Karlsruhe Institute of Technology, 76344 Karlsruhe, Germany; 2) Nanomatch GmbH, Griesbachstraße 5, 76185 Karlsruhe, Germany; 3) Institute of Theoretical Informatics, Karlsruhe Institute of Technology, 76344 Karlsruhe, Germany

Resume : The ionization potential (IP), electron affinity (EA), and cation/anion polarization energies of organic molecules determine injection barriers, charge carriers balance, doping efficiency, and light outcoupling in organic electronics devices, such as organic light-emitting diodes (OLEDs). Computing IP and EA of isolated molecules is a common task for quantum chemistry methods. However, once molecules are embedded in an amorphous organic matrix, IP and EA values change due to polarization effects and accurate predictions become challenging. Using a revised quantum embedding method [1] that computes the charge density of molecules in an amorphous solid, we determine the (de)polarization energy contributions to the IP and EA which allows straightforward interpretation of their nature. [2] The dielectric permittivity and ionization potentials are accurately predicted in three test materials, NPB, TCTA, and C60. The method paves the way towards reliable virtual screening of amorphous organic semiconductors with targeted IP/EA, polarization energies, and relative dielectric permittivity. [1]: P. Friederich, F. Symalla, V. Meded, T. Neumann, W. Wenzel, J. Chem. Theory Comput. 2014, 10, 9, 3720–3725: [2]: J. Armleder, T. Strunk, F. Symalla, P. Friederich, J. E. Olivares Peña, W. Wenzel, A. Fediai, J. Chem. Theory Comput., 2021:

Authors : Jorge Enrique Olivares Peña Wolfgang Wenzel Artem Fediai
Affiliations : Karlsruhe Institute of Technology

Resume : Molecular spintronics uses the spin degree of freedom to develop technology that can control electrical currents in nanodevices. A good understanding of the fundamental physics in nanoscale systems and reliable technical tools for simulating them are required to exploit the full capacity of the spin in molecules. Experimental setups have shown a remarkable spin-dependent behaviour of the current (Magnetoresistance) through a carbon nanotube decorated with single molecular magnets (SMMs). Empirical models have been proposed to explain this effect, but none of them based on first-principles calculations. Here we present results of ab-initio simulations of the CNT decorated by two terbium phthalocyanine SMMs depending of the relative SMM spin. The current-voltage and differential conductance maps were obtained using the Landauer fomalism and DFT. Our results show how the relative spin of two SMMs affects the conductance through a CNT and the advantages and limitations of an non- interactive as well as an interactive approach. The system under study, allows us to extend our approach to any periodic sytem and different types of SMMs, opening a field of ab-initio studies of nano-electronic spintronic devices in the non-interacting as well as in the interacting regimes.

10:05 Q&A Session / Break    
Disordered, porous, low-dimensional and hybrid organic/inorganic materials II : Guido Ori
Authors : Gabriel Wlazłowski
Affiliations : Warsaw University of Technology

Resume : Superfluidity is a generic feature of many quantum systems at low temperatures. It has been experimentally confirmed in condensed matter systems like 3He and 4He liquids, in nuclear systems including nuclei and neutron stars, in both fermionic and bosonic cold atoms in traps. Superfluids exhibit fascinating dynamical properties, in many cases very different from dynamics observed in classical analogs. Capabilities of nowadays supercomputers allow for modeling dynamics of such systems using a microscopic framework based on the time-dependent density functional theory (TDDFT). In this talk, I will review the most relevant applications of the superfluid-TDDFT framework achieved recently with the help of computer systems like Summit (ORNL, USA) and Piz Daint (CSCS, Switzerland) together with the presently utilized numerical and technical solutions. In particular, the dynamics of quantum vortices, which lie at the heart of the quantum turbulence phenomenon, will be presented. Finally, challenges for future exascale systems in the context of modeling superfluidity/superconductivity will also be highlighted.

Authors : Marios Zacharias and Pantelis C. Kelires
Affiliations : Department of Mechanical and Materials Science Engineering, Cyprus University of Technology, P.O. Box 50329, 3603 Limassol, Cyprus

Resume : Quantum confined nanostructures (QCNs) have many potential applications in the field of nanomaterials and optoelectronics, including light-emitting diode and solar cell devices. One of the fundamental properties of interest in these nanostructures is their quantum-confined bandgap, determining their photoluminescence properties [1,2]. Current density functional theory (DFT) calculations on free-standing, or embedded, QCNs bandgaps are performed by describing the nuclei as classical particles at 0 K, disregarding quantum zero-point motion and thermal effects. In this talk, I will show how these effects can be incorporated in ab-initio calculations of QCNs using the special displacement method (SDM) [3,4]. I will present DFT-SDM calculations of temperature-dependent band gaps of free-standing and embedded silicon and graphene quantum dots. I will also discuss the role of quantum confinement on the phonon-induced zero-point renormalization and present full band structures at finite temperatures. As a final point, I will demonstrate the effect of quantum confinement on the electron-phonon matrix elements of these nanostructures by presenting calculations of individual phonon contributions to the Eliashberg function. [1] D. Chen et al., Nanoscale, 11, 4226 (2019). [2] M. Zacharias and P. C. Kelires, Phys. Rev. B 101, 245122 (2020). [3] M. Zacharias, and F. Giustino, Phys. Rev. B 89, 075125 (2016). [4] M. Zacharias, and F. Giustino, Phys. Rev. Res., 013357 (2020).

Authors : Mula, S.*(1), Donà, L.(2), Civalleri, B.(2), & van der Veen, M. A.(1)
Affiliations : (1) Department of Chemical Engineering, Delft University of Technology, The Netherlands (2) Department of Chemistry, University of Torino, Italy

Resume : Metal-organic frameworks (MOFs) are 3-dimensional porous frameworks with metal ions as inorganic nodes connected by organic linkers. MOFs possess a very low dielectric permittivity εr due to their porosity, favorable for piezoelectric energy harvesting. Moreover, due to their highly tunable structure, it should be possible to modify them to obtain a high piezoelectric coefficient. Even though they have a huge potential as piezoelectric materials for energy harvesting, a detailed analysis of the piezoelectric properties of MOFs is lacking so far. We focus on a class of cubic non-centrosymmetric MOFs namely ZIFs (Zeolitic Imidazolate Frameworks), which have a divalent metal ion connected by imidazolate organic linkers. We calculated the piezoelectric coefficients (e14 and d14) through quantum mechanical calculations using the ab initio periodic code CRYSTAL17. This theoretical work discusses how piezoelectric coefficients vary in Zn/Cd based ZIFs and compares them to already existing Zn/Cd based inorganics and some organic piezoelectric polymers. To understand the change in piezoelectric coefficients due to the metal ion and linker substituents, we vary the metal ion Zn /Cd and 4 substituents -CH3, -NO2,-Cl,-CHO on the imidazolate linker. Our results indicate that e14 is the same order of magnitude for all ZIFs and lower than Zn/Cd based II-VI inorganic piezoelectric. Although final e14 is similar for all ZIFs, we see contributions to e14 i.e., the clamped ion (e014) and internal strain (eint14) are distinct for each ZIF based on the linker substituent. The piezoelectric coefficient d14, calculated from e14 and elastic compliance constant s44, is the largest for CdIF-1 (Cd2+ and -CH3 linker substituent). This is mainly due to the higher elasticity of the framework. Interestingly, the magnitude is actually higher than II-VI inorganic piezoelectrics, and of a similar magnitude as the quintessential piezoelectric polymer PVDF.

Authors : Zhenzhen Yan, Howard Sheng, Evan Ma, Bin Xu, Jinfu Li, Lingti Kong
Affiliations : Zhenzhen Yan, School of Materials Science and Engineering, Shanghai Jiao Tong University; Howard Sheng, Department of Physics and Astronomy, George Mason University; Evan Ma, State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University; Bin Xu, Department of Materials Science and Engineering, Johns Hopkins University; Jinfu Li, School of Materials Science and Engineering, Shanghai Jiao Tong University; Lingti Kong, School of Materials Science and Engineering, Shanghai Jiao Tong University;

Resume : Understanding how the interfacial atoms in the liquid side are rearranged into the growing crystal remains unclear so far. The energy barrier associated with the interfacial structural evolution of crystal growth, thermally activated or barrier-less, is a critical indicator but not well understood, even for simple pure BCC and FCC metals. Here, we use a new defect-mediated approach to monitor the intermediate dumbbell-interstitial-like pairs (DILPs) preceding the growing BCC crystal with molecular dynamics simulations. This approach is different from the classical approach to crystal growth, in that the interfacial liquid is now perceived as evolving towards a disordered solid from the fluid, such that ideas can be drawn from solid-state transformations using defect mediated mechanisms as opposed to the notions of single-atom addition that is completely individual and incoherent. We found that liquid atoms next to the interface first rearrange into densely packed patterns, forming intermediate configurations, and they take a specific structural pathway to attach onto lattice sites, involving collective ordering rather than doing it randomly one atom at a time. For BCC, the interfacial intermediate structures are 3D-connected DILP motifs, while for FCC, they are mostly 1D chains. The differences, by analyzing their distinctly different structural evolution and associated activation energy barrier, shed light on the contrasting crystal growth behavior for BCC versus FCC metals in deeply undercooled liquids.

Authors : Ghada Belhadj Hassine, Marek Sierka
Affiliations : Otto Schott Institute of Materials Research, Friedrich Schiller University Jena, Germany.

Resume : Silica, in its crystalline and glassy forms, is of immense technological importance, with applications ranging from integrated circuits to supports for heterogeneous catalysis. A detailed understanding of the silica structure and its crystal-vitreous phase transformation is therefore of utmost importance. Experimental studies have reported the observation of the transformation of a 5-7-5-7 to a 6-6-6-6 Si-O-Si ring system in a crystalline bilayer silica [1]. In addition, apparent activation energies for a crystalline to vitreous transformation were obtained in UHV and O2 atmosphere for a silica bilayer supported on Ru(0001) system [2]. In this work, DFT calculations were performed to investigate the formation mechanism of Stone-Wales type defect starting from a perfectly hexagonal silica bilayer structure. The process found shows a complex formation mechanism, in which the two layers exhibit decoupled behavior in terms of chemical bond rearrangements [3]. Charge density analysis is used to rationalize the influence of the metal support and to investigate the effect of interfacial O/Ru(0001) on the energetics of the transformation. Literature: [1] P. Y. Huang, S. Kurasch, J. S. Alden, A. Shekhawat, A. A. Alemi, P. L. McEuen, J. P. Sethna, U. Kaiser, D. A. Muller, Science 2013, 342, 224. [2] H. W. Klemm, PhD Thesis 2018, TU Berlin. [3] H. W. Klemm, M. J. Prieto, F. Xiong, G. B. Hassine, M. Heyde, D. Menzel, M. Sierka, T. Schmidt, H. J. Freund, Angew. Chem. Int. Ed. 2020, 59, 10587.

12:35 Q&A Session / Closing Remarks    

Symposium organizers
Elena LEVCHENKOUniversity of Newcastle

School of Mathematical and Physical Sciences, Faculty of Science, University Drive, Callaghan NSW 2308 Australia
Guido ORIInstitut de Physique et Chimie des Matériaux de Strasbourg

IPCMS, CNRS - University of Strasbourg, 23 Rue du Loess, F-67034 Strasbourg, France
Michał HERMANOWICZ (Main organizer)University of Warsaw

Interdisciplinary Centre for Mathematical and Computational Modelling, ul. Tyniecka 15/17, 02-630 Warsaw, Poland
Yannick J. DAPPEService de Physique de l’Etat Condensé (SPEC – CNRS – CEA Saclay)

Bât. 771 Orme des Merisiers F-91191 Gif-sur-Yvette, France