We would like to thank all applicants for computation time within the 24th Open Access Grant Competition.

A total of 1 574 002 node hours were requested. Considering productive use of the resources, the Allocation Commission partially redistributed the computational resources, and several projects were subject to minor allocation reductions.

All projects underwent technical evaluations and evaluation of the registered publications per project ratio.

The Allocation Commission based the allocation decisions on scientific excellence, computational readiness, and socioeconomic impact. In addition to the score assigned by the reviewers, the Commission also took into account (with a weight of 50%) the registered publications per project ratio. The registered publications per project ratio are defined as the number of publications registered in the last three years divided by the number of past OPEN projects that were concluded more than one year ago but less than four years ago. On average, 1.0 publication is expected per completed project. Projects in the extent of 10% of requested resources were not subject to peer review. In such a case, the committee assumed the highest score on scientific excellence, computational readiness, and socioeconomic impact. In the case of new users, the committee assumed the highest score on the registered publications per project ratio. Low registered publications per project ratio were the primary cause of significant allocation cuts.

The Allocation Commission acknowledged the high scientific and technical level of the submitted projects. From the maximum score of 30 points, the projects averaged 24.2, at a minimum of 18 points. Nine of the projects exceeded 25 points.

Among the 58 projects, including two multi-year in the first period, a total of 1 497 623 node hours were allocated. Projects that were not subject to peer review represent 286448 (19%) of the allocated resources. The relative increase in the share of allocations without peer-review is an artifact of the redistribution for more even allocations of computing resources. 

 

---

The Allocation Commission decided on the allocations within the 24^th Open Access Grant Competition as follows:

Researcher: Chieh-Jen Yang

Project: A new ingredient of effective field theory power counting: particle-number-dependent description of nuclei

Allocation: Barbora CPU 4600, Barbora FAT 1000, Barbora VIZ 40, Karolina CPU 25000, Karolina VIZ 40

Abstract: Understand nuclear structure and reactions in terms of the basic forces between nucleons has been one of the outstanding problems in physics. Answering this fundamental problem will empower us to have a better control of nuclear energy and other applications. A modern description of nuclear force is achieved through chiral effective field theory (EFT), where reliable forces between nucleons can be derived. Although two-nucleon forces dominate at light nuclei, recent investigations show that many-body forces play an important role for medium and heavy systems. However, a study of this new ingredient beyond the lowest order is missing in essentially all analyses performed up to date. The objective of this project is to perform a thorough study of the above missing ingredient. To enable a clear renormalization group (RG) and power counting analysis, we will develop a new chiral EFT power counting schemes which generate RG invariant results, and perform ab-initio calculations of finite nuclei and infinite nuclear matter with three- and four-body forces to test our assumptions. Upon completion, it will provide a solid foundation for an efficient and model-independent description of all low-energy nuclear processes.

 

Researcher: Stepan Sklenak

Project: Periodic DFT studies of zeolite-based catalysts

Allocation: Barbora VIZ 40, Karolina CPU 28000,  Karolina VIZ 40

Abstract: Zeolite based catalysts are the most important industrial catalysts used mainly in petrochemistry (production of fuels, various hydrocarbons), and furthermore, for N₂O decomposition and N₂O selective catalytic reduction (SCR), for direct NO decomposition as well as for SCR of NOx using various reducing agents. In addition, zeolites are also employed as catalysts in synthesis of fine chemicals and as well as in the processing of biomass. Zeolites are also used as ion exchange agents (e.g., they replaced harmful phosphates in washing powder detergents). Zeolites are crystalline microporous aluminosilicates with a unique microporous nature, where the shape and size of a particular pore system exerts a steric influence on the reaction, controlling the access of reactants and products. Thus, zeolites are often said to act as shape-selective catalysts. Increasingly, attention has focused on fine-tuning the properties of zeolite catalysts in order to carry out very specific syntheses of high-value chemicals. Periodic DFT methods permit investigations of properties of zeolite-based catalysts which are needed for their fine-tuning. DFT calculations are complementary to experimental examinations and together they can provide more complex knowledge of the properties of the studied catalysts and the reactions they catalyze.

 

Researcher: Miroslav Jícha

Project: Holistic approach to Rotating Packed Bed (RPB) Carbon Capture process with the use of 3D CFD

Allocation: Barbora VIZ 40, Karolina CPU 780, Karolina VIZ 40

Abstract: The project will study the mechanisms that take place in a rotational absorber for CO2 capture (Rotational Packed Bed RPB). The RPB combines hydrodynamic and process mechanisms. The major issues are liquid spray in the rotor eye, in the rotating packing made of knitted wire mesh or a zigzag geometry and in the outer cavity, incl. liquid spray on the outer surface of the packing and formation of a liquid film on casing walls. So far attention has been mostly paid to the effect of rotation or type of the packing on the overall efficiency of gas capture, often supplemented by CFD. However this engineering approach doesn´t elucidate the role of individual elements of the RPB and their contribution to CO2 capture. Moreover there is no relevant literature that illuminates the abovementioned mechanisms in context, and which studies in detail the interaction between liquid distribution and the efficiency of gas capture. Therefore the main goal of this project is to consider the RPB as a whole and using 3D computational modelling with subsequent limited validation to unveil the interconnection and the influence of individual RPB segments on CO2 capture and to optimize their capture efficiency. Due to practically impossibility of experiments inside the packing, 3D CFD simulations using Volume of Fluid technique with a limited validation by own experiments is the only way how to recognize and optimize processes in the entire RPB.

 

Researcher: Robert Vacha

Project: Peptide Killers of Bacteria 1

Allocation: Barbora CPU 48700, Barbora VIZ 40, Karolina CPU 75000, Karolina GPU 5000, Karolina VIZ 40

Abstract: Antibiotic-resistant bacteria cause more than 700 000 deaths per year, and the forecast is 10 million per year in 2050. Moreover, emerging strains of bacteria resistant to all available antibiotics may lead to a global post-antibiotic era. Because of this threat, the WHO and the UN are encouraging the research and development of new treatments. This proposal aims to design unique peptides that selectively target and disrupt the membranes of pathogens but not those of human cells. To obtain such peptides, we will develop an innovative coarse-grained model of membranes, which will enable us to establish the relationship between peptide sequence motifs and their affinity to membranes with specific lipid compositions. Our computational results will be experimentally verified with peptidemembrane affinity measurements using the quartz crystal microbalance method. The results and knowledge obtained within this project will not only enable us to determine new peptides selectively killing bacteria but will also enable the development of peptides targeted against membranes of enveloped viruses, cancer cells, or even cellular organelles with potential application as sensors, biomarkers, or therapeutics.

 

Researcher: Jan Simkanin

Project: Thermochemically-driven convection and dynamos operating at low Ekman numbers – Part III (follow up to OPEN-21-1 and OPEN-22-28)

Allocation: Barbora VIZ 40, Karolina CPU 25000, Karolina VIZ 40

Abstrakt: The Earth's magnetic field is one of the most variable geophysical fields and provides us with effective protection against high-energy, electrically charged particles from the solar wind, solar flares and is a useful tool for navigation, not only for us, humans but also for animals. The Earth's magnetic field is generated by convective motions of an electrically conductive melt in the Earth's outer core and penetrates the surface of the Earth. These generation processes are called Geodynamo for short. On the Earth’s surface, a large-scale, dipole-dominated field is observed. However, we have no direct information about the magnetic field in the Earth'score. For this reason, we model magnetohydrodynamic processes in the Earth's core numerically. Numerical modelling of Geodynamo has made enormous progress during last years thanks to the development of computer technology. Numerical models produce magnetic fields that are close to the observed geomagnetic field.  We are also able to reproduce its temporal changes, whether short-term or long-term ones (so-called secular variations). However, we cannot use parameters typical of the Earth's core in our models. The real parameter values cannot be used for computational reasons – so far there is no supercomputer in the world that could solve a Geodynamo model with such values. However, we are gradually approaching these values as the computers become more and more powerful.

 

Researcher: Jiří Kolafa

Project: MD simulation of supersonic expansion on GPUs

Allocation: Barbora VIZ 40, Karolina GPU 1000, Karolina VIZ 40

Abstract: The upper part of Earth atmosphere contains small frozen particles consisting of a few hundred water molecules. These particles are called water clusters and are much more interesting than it might seem. Other molecules can adsorb on the cluster surface and engage in a sequence of chemical reactions initiated by solar radiation. Reactions on the cluster can have significant impact on ozone layer damage, global warming and weather forecast. Since it is difficult to study the clusters in situ, we have to resort to laboratory models. One of possible methods is based on supersonic expansion of water vapor to a vacuum. Vapor expands through a small orifice, reaches supersonic speeds and cools down. However, there is contradictory evidence about the shape of these particles: Cross-section measurements indicate irregular shapes, which is a bit against the common sense. In 2018, we published a paper about a new method to simulate supersonic expansions directly at molecular dynamics level.[1] The method was parallelized for GPUs and run on the IT4I supercomputers. The results were recently published.[4]  We hope that with sufficient resources we can perform a more comprehensive study of expanding systems under different conditions and analyze a large number of clusters for their thermodynamic and kinetic properties.

 

Researcher: Sergiu Arapan

Project: A DFT study of the magnetic phase ordering near quantum critical point in CeFe(x)Co(1-x)Ge3 systems

Allocation: Barbora VIZ 40, Karolina CPU 28000, Karolina GPU 2800, Karolina VIZ 40

Abstract: CeCoGe₃ is a unique antiferromagnetic material (AFM) with 3 different low temperature magnetic order phases. In addition, it exhibits unconventional superconductivity (SC) that can be potentially mediated by magnetic fluctuations. CeFeGe₃ is a paramagnet with a typical heavy fermion (HF) system behavior.  Dilution of CeCoGe₃ with CeFeGe₃ results in the change of magnetic ordering with occurrence of ferromagnetism (FM) for certain concentration x in the CeFe_xCo_1-xGe₃. This fact implies a complexity of magnetic interactions and the existence of a quantum critical point (QCP). The plethora of different magnetic phases suggests different types of electronic structures, which can be “probed” computationally within the density functional theory (DFT). The ultimate goal is to achieve the co-existence of the ferromagnetism and superconductivity with future outlook to promote such systems towards room temperatures for many industrial applications (Maglev train, magnetic bearings etc.)

 

Researcher: Dagmar Zaoralová

Project: Structure-activity relationships of transition-metal particles anchored to nitrogen-doped graphene for designing efficient single-atom catalysts

Allocation: Barbora VIZ 40, Karolina CPU 20000, Karolina GPU 5000, Karolina VIZ 40

Abstract: Single-metal atoms anchored on substrate materials have attracted considerable interest in recent years as so-called single-atom catalysts (SACs). Because of the maximally reduced size of the metal particles, the SACs exhibit exceptional catalytic performance. On the other hand, SACs often suffer from undesirable processes such as aggregation into larger clusters and leaching during reactions.  Therefore, a sufficiently strong interaction of the metal particle with an anchoring site in the structure of a supporting material is needed. Nitrogen-doped graphenes are a class of substrate materials that offer a variety of anchoring sites. Even with the same metal element, the anchoring site significantly influences the stability and catalytic activity of the SAC. Despite interest in SACs, the in-depth understanding of the structure-activity relationships at the atomic level is still missing. To gain a general view of anchoring sites (vacancies, defects, etc.) that are naturally present in the structure of nitrogen-doped graphenes, their ability to immobilize metal atoms, and the catalytic activity of the resultant SAC, we need to optimize a large number of structures using methods of density functional theory (DFT). To this end, we plan to utilize program packages for calculations in periodic boundary conditions, namely the Vienna Ab initio Simulation Package (VASP), as well as for calculations on cluster models, namely Gaussian. We believe that the results can guide experimentalists in designing maximally efficient SACs.

 

Researcher: Alžběta Špádová

Project: Nanoparticle-assisted laser-wakefield electron acceleration

Allocation: Barbora VIZ 40, Karolina CPU 9000, Karolina VIZ 40

Abstract: The laser-wakefield acceleration has drawn a wide attention of scientific community as a promising way of constructing compact high-energy electron accelerators. This mechanism uses plasma waves generated via the interaction of intense ultrashort laser pulses with gas targets to capture and accelerate electrons to relativistic velocities. Although these accelerators have high potential to replace the conventional accelerators in future in many interesting industrial and medical applications as well as in various areas of fundamental research, the quality of generated beams has to be further improved to enable efficient use in the above mentioned fields. Since the electron injection is one of the key features determining the beam characteristics, several injection mechanisms yielding better electron beam parameters were proposed over the last twenty years. The most recent and very promising scheme utilizes a gas target containing nanoparticles. These targets could allow us to precisely control the location where the electrons are being injected into the wakefield, as well as to improve some other important beam parameters. The goal of our project is to study this novel electron injection mechanism based on the gas targets with nanoparticles. For this purpose, numerous large-scale particle-in-cell simulations in two- and three-dimensional geometries are necessary to be performed.

 

Researcher: Marek Pecha

Project: ml4py – distributed machine learning tools (dev/stage) II

Allocation: Barbora CPU1600, Barbora FAT 16, Barbora GPU 200, Barbora VIZ 40, DGX-2 200, Karolina CPU 4600, Karolina FAT 16, Karolina GPU 400, Karolina VIZ 40

Abstract: This project continues with the implementation and benchmarking of emerging tools for distributed machine learning called ml4py. Unlike standardly used machine learning kits, ml4py implements a mechanism for training predictors over cluster in its runtime using Message Passing Interface (MPI). Thus, no other frameworks such as Horovod or orchestration based on running multiple container instances of an application are required. Accelerators such as graphic cards are utilized using the OpenCL and CUDA technologies through the ml4py runtime.

Current research related to machine learning commonly prefers deep learning techniques using (deep) neural networks. An associated training phase could be considered as a stochastic process that converges to an expected model. Our previous research on machine learning shows that determinism allows us to design training pipelines directly and explain the qualities of an attained learning model and numerical behaviour of the underlying solver more straightforwardly. However, these approaches are expensive in the sense of computational cost, and they can efficiently solve up to medium scale tasks by the state-of-the-art. Since ml4py is designed for massively parallel approaches, determinism can be incorporated into training pipelines for big data analysis. To speed the deterministic training phase up, stochastic optimization methods can be exploited. Basically, we find a good candidate for a solution using stochastic approaches expeditiously, and this candidate is used as a hint for deterministic and parallel predictor training subsequently. It helps to attain high-quality explainable models within a reasonable time.

The ml4py framework is tested on datasets ranging from bioinformatics, geoscience to computer vision.

 

Researcher: Pavel Ondračka

Project: Retrieving data. Wait a few seconds and try to cut or copy again.

Allocation: Barbora CPU 12600,  Barbora VIZ 40, Karolina VIZ 40

Abstract: The bench-mark protective coating material utilized in the cutting and forming applications is TiAlN exhibiting high hardness as well as stiffness. Unfortunately, these favorable properties are associated with unwanted brittle deformation behavior resulting upon mechanical loading in the formation of cracks which limits the performance and lifetime of the coated tools. From a materials design point of view a rather unusual combination of properties - high hardness and stiffness together with moderate ductility - is therefore required for the next generation of protective coating materials. This project aims to design coating materials with high hardness and enhanced fracture toughness for cutting and forming applications with extended lifetime and performance by ab initio calculations on a novel class of orthorhombic boron-carbide materials.

 

Researcher: Jiri Klimes

Project: Accuracy and precision for extended systems

Allocation: Barbora CPU 19100, Barbora VIZ 40, Karolina CPU 25000, Karolina GPU 10, Karolina VIZ 40

Abstract: Computer simulations have become indispensable for understanding the majority of experimental observations and thus understanding our world. This is especially true at the atomic and molecular level where experimental observations can become very difficult to interpret. However, if simulations are to be useful, they need to model the system reliably. For example, if there are different pathways for a chemical reaction, the simulation needs to be able to correctly predict their relative importance so as to increase our understanding of the system. The world at the atomic level is governed by the complex laws of quantum mechanics. In practice, this means that increasing the reliability of simulations, i.e., using less approximations, increases the computational cost substantially. While we understand well the impact of some of the approximations, there are some for which our knowledge is limited. In some cases, these can lead to unexpectedly large loss of the reliability of calculations. In our project we analyse some of the problematic cases occurring for calculations of intermolecular interactions, either in small clusters, or in molecular crystals.

 

Researcher: Jaroslav Resler

Project: Tests of large simulations of urban climate with turbulent resolving model.

Allocation: Barbora VIZ 40, Karolina CPU 48000, Karolina VIZ 40

Abstract: The newly developed urban climate model PALM (www.palm4u.org) allows to perform detailed simulations of conditions in urban areas, mainly with respect to phenomena of the thermal comfort and air quality. We have significantly contributed to the model development and validation and we are using it for testing the efficiency of urban climate adaptation measures. For a project TURBAN - Turbulent-resolving urban modelling of air quality and thermal comfort (Norway Grants), we prepared a large domain for model validation mainly with respect to the turbulent air flow. This domain is about four times larger than our largest simulation to date and it will simulate all relevant sizes of eddies above the urban area. We will prepare and test different configurations of the model and optimize these large simulations to enable us to efficiently compute multiple runs of the final simulations for model validation. Results of this testing are necessary for the following long-term IT4I call application.

 

Researcher: Michal Otyepka

Project: Following the path of photoexcited electrons in the complex structures of carbon dots (follow up to OPEN-22-11)

Allocation: Barbora VIZ 40, Karolina CPU 50313, Karolina VIZ 40

Abstract: Carbon dots (CDs) represent low toxic, biocompatible, chemically stable nanomaterials. Most importantly, they possess bright, intense and tunable photoluminescence (PL) properties. This makes CDs immensely interesting and applicable in bioimaging, sensing, photocatalysis, theranostics and other diverse areas. Although the PL and structural properties of CDs have been theoretically examined on small ad hoc models, the so far published studies are still not successful in explaning the mysterious and complex field of structural-dependent PL of real CDs. This project shall be the continuation of our previous Open-22-11 project, where we seek to address dynamical processes upon photoexcitation in CDs. More specifically, we are trying to capture structural relaxation and energy transfer between multi-luminescent centers of CDs using quantum methods applied to realistic structural models. To this end, we plan to utilize program packages simulating the dynamics of electronically excited molecules and molecular assemblies, in combination with ab initio computational chemistry software packages. We are convinced that these calculations will decipher the specificities of core, surface and molecular states in PL of CDs. Such knowledge may guide experimentalists to efficiently and controllably synthesize CDs with defined PL by modifying their chemical composition and structure.

 

Researcher: Alberto Marmodoro

Project: CHIRSPIN

Allocation: Barbora CPU 6250, Barbora VIZ 40, Karolina VIZ 40

Abstract: Spintronics investigates solid state physics problems, with emphasis on possible future devices based on the spin of electrons, beside their charge. In previous years, this research line has studied for instance ordered 2D heterostructures composed by magnetic and non-magnetic layers (Duine et al.,2018) or randomly doped semiconductors (Jungwirth et al., 2006). More recently, interest has grown toward ways to exploit basic features of the bulk lattice geometry, to achieve spin-selectivity even in the absence of a net magnetic moment, such as it the case of antiferromagnets (Baltz et al., 2018; Gomonay et al., 2017). Nowadays, mounting experimental and theoretical evidence is also attracting attention to chirality-induced-spin-selectivity (CISS) materials (Shiota et al., 2021). In these crystals, the arrangement of sublattices in a winding or corkscrew-like fashion results in magnified relativistic effects, that produce spin-polarization even in the absence of any magnetic atom within the unit cell. We intend to study at a first-principles level some representative compounds from the above CISS family. Our goal is a better understanding of the connection between chiral geometry and spin-selectivity, through its manifestations in the electronic band structure, transport properties, and other magneto-electric effects. We will explore theoretical robustness of their spin selectivity with respect to different types of disorder, as given by lattice defects and finite temperature.

 

Researcher: Stanislav Polzer

Project: Abdominal Aortic Aneurysm Rupture Risk Assessment in the Pandemic Era

Allocation: Barbora VIZ 40, Karolina CPU 27800,  Karolina VIZ 40

Abstract: Abdominal aortic aneurysm (AAA) is a permanent diameter enlargement of abdominal aorta. Its rupture is a life-threatening event. On the other hand the non-ruptured AAA does not bring any complication to patient in most cases, therefore operations should be done only in cases with AAA close to rupture. Mostly used criterion for operation is maximum diameter. However it is not very accurate since not all large AAAs rupture while some small does. Therefore new criterions have been searched by scientific teams. One of the criterions is based on a wall stress analysis on the AAA. This project aims at use more complex criterion to evaluate risky AAAs so that medical doctors could use again our advice to repair critical AAAs and postpone unnecessary surgeries during covid-19 lockdown.

 

Researcher: Piotr BLONSKI

Project: Tuning Magnetic Anisotropy Energy in Graphene-based Materials for Data Storage

Allocation: Barbora CPU 8000, Barbora VIZ 40, Karolina VIZ 40

Abstract: Ever-increasing amounts of data place an urgent need for new magnetic materials with increased storage capacity. To this end, at the heart of magnetic materials research is the quest for substrates hosting an ordered array of atom-sized magnets, each of which can store a single bit of information. One of the great challenges in this field is preventing thermally induced reorientation of the magnetic moments between the easy and hard magnetization axis or, in other words, increasing the energy barrier for magnetic moments to flip their directions. The Project will focus on the theoretical development of new materials that can store much greater amounts of data and do so in a more energy-efficient way compared to the currently used materials. To this end, small transition-metal clusters deposited on a solid-supported functionalized-graphene will be studied by using spin-polarized density functional theory including spin-orbit coupling, towards generation of magnetic centers with a superbly high magnetic anisotropy energy (MAE) allowing for the data retention time of 10 years at a temperature of at least liquid nitrogen of 77 K. The possibility to manipulate magnetic anisotropy via electric fields will be addressed: A large MAE is needed to stabilize a magnetic bit against thermal agitations; a low MAE is desired during writing information. The proposed work will offer the guidance for the synthesis of materials with magnetic properties tailored on the level of individual atoms.

 

Researcher: Pavel Balaz

Project: Neural Network Quantum States of Antiferromagnetic Spin Systems

Allocation: Barbora CPU 400, Barbora FAT 100, Barbora GPU 500, Barbora VIZ 40, Karolina CPU 1000, Karolina FAT 80, Karolina VIZ 40

Abstract: Understanding many body frustrated spin systems and their phases remains a daunting challenge of quantum mechanics and condensed matter physics. Even simple models lead to complex states which are extremely difficult to analyse by means of ordinary methods. Ambiguous results of various approximative methods intensify the need for novel approaches. One of the most promising attempts to deal with this problem comes with the advent of machine learning in quantum mechanics. It has been recently demonstrated that models based on neural networks can describe complex many body spin systems more accurately than many of the previously used approaches. The main reason for this is a high expressivity of the neural networks, which can be used as universal approximators of multidimensional functions, such as many body wave function. The model expressivity can be further increased by taking into account correct symmetry properties of the system as well as by employing proper training methods. In this project, we shall study possibilities of enhancing the neural network-based models of realistic highly frustrated two-dimensional spin systems. We will concentrate on the symmetry analysis of the quantum states and improvement of the learning algorithms. The main goal is to mitigate the infamous sign-problem of quantum Monte Carlo simulations and theoretically investigate the phases of quantum magnets such as SrCu2 (BO3 )2 .

 

Researcher: Mauricio Maldonado Dominguez

Project: Transduction of chemical energy into motion. The case of C-H activation.

Allocation: Barbora CPU 3400, Barbora VIZ 40, Karolina CPU 3633, Karolina VIZ 40

Abstract: The translation of motion from chemical reactions into useful energy is becoming increasingly recognized as fertile ground to increase the efficiency of such transformations, yet it remains vastly unexplored. In this project, we will investigate a series of C-H cleavage reactions in solution where we have pinpointed the existence of nonequilibrium energy redistribution (J. Am. Chem. Soc. 2020, 142, 3947, supported by project OPEN 18-2). In this project, we will study to which extent this excess is funneled into the translational velocity of the nascent radical products. We recently demonstrated that this kinetic energy excess is quantitatively correlated with the product ratio in bifurcating reactions (Chem. Sci., 2021, 12, 12682, project OPEN 22-17). The projected results will pave the way to the design of processes where chemical reactions can be coupled to maximize their energy efficiency.

 

Researcher: Vojtech Kostal

Project: Rationalizing of charge scaling approach for molecular dynamics aided by ab initio calculations

Allocation: Barbora VIZ 40, Karolina CPU 14100, Karolina VIZ 40

Abstract: Molecular dynamics is nowadays a well-recognized class of methods providing molecular insights into the behavior of, e.g., complex systems of biological relevance in their natural environment — aqueous solution. To this end, numerous empirical force fields were developed as an efficient and affordable approach. However, currently available methods suffer from several drawbacks. For example, unphysical behavior of highly charged species like ions and side-chains of proteins is observed due to the absence of electronic polarization of the surrounding environment in non-polarizable classical molecular force fields simulations. Recently, the so-called electronic continuum correction (ECC) [1], via scaling of partial charges, was shown to account for this issue in an effective yet rigorous way in non-polarizable classical molecular force fields. Still, it is unclear why the scaling factor proposed by the theory is larger than the optimal one found by many ECC projects. To shed light into this issue, here, we aim to characterize ion pairing in a low-dielectric environment based on state-of-the-art ab initio molecular dynamics (AIMD) and its comparison with force field molecular dynamics (MD). This comparison should allow us to disentangle the effect of electronic polarization from the nuclear one, which is hard to quantify in more polar solvents due to their more complex molecular structure and quantitatively smaller effect. Moreover, we aim to answer whether the charge scaling, a mean field approximation, is justified only at long intermolecular distances where particles are well solvent-separated or is also applicable at short or contact distances as well. This has deep implications when attempting to improve classical non-polarizable force fields.

 

Researcher: Marco Vitek

Project: Investigating nonmetal-metal transition in liquid ammonia solutions in the context of the percolation theory

Allocation: Barbora VIZ 40, Karolina CPU 8000, Karolina VIZ 40

Abstract: This project focuses on the investigation of solvated electrons originating from alkali metals dissolved in bulk liquids. Our goal is to use electronic structure calculations and ab initio molecular dynamics (AIMD) to provide insight into the insulator-metal transition from individual solvated electrons, through dielectrons, to the onset of more delocalized states. In particular, we would like to utilize percolation theory to describe the transition from the blue to the bronze-coloured metallic solution and the associated increase of conductivity.^1-4 This will be accomplished by describing the formation of delocalized states from distinct individual electrons and dielectrons as a percolation transition. The power of percolation theory lies in the fact that it is capable of describing critical phase transitions of rich physical content, yet it may be formulated in terms of very simple geometrical concepts. It is a versatile model, with applications to such diverse problems as supercooled water, galactic structures, fragmentation, porous materials, earthquakes, and many more.⁵ In percolation theory, a random system is described by a percolation threshold P_c, which  represents the critical value of connectivity needed for the emergence of clusters spanning the whole system. Below the threshold, such connected components do not exist, while above it, a giant cluster of the order of system size can be observed.⁵ In this project, we are proposing a hypothesis that the delocalized states in metallic ammonia solutions should differ in the percolation threshold value from the states of the distinct individual electrons and dielectrons. For this purpose, we plan to perform an extensive sampling of the electron density of alkali-metals ammonia solutions through ab initio molecular dynamics simulations on a hybrid DFT level utilizing advanced and heavily parallelizable simulation algorithms.

 

Researcher: Marie Behounkova

Project: Effect of localized melt pockets on Europas tidal deformation

Allocation: Barbora VIZ 40, Karolina CPU 2400, Karolina VIZ 40

Abstract: Jupiter's moon Europa harbors underneath a young and tectonically modified ice shell a salty ocean interacting with possibly still active rocky interior. Such an oceanic environment makes Europa a primary target in the search for a habitable world beyond the Earth. However, the Europa's ocean habitability is conditioned by the intensity of magmatic activity and by the heat released from the rocky mantle to the seafloor. Similar to the volcanic moon, Io, tidal energy dissipated in Europa's interior is expected to influence its thermal state. Our previous work studied the evolution of Europa's rocky mantle, including tidal dissipation. We have shown that, even though silicate volcanism is strongly reduced compared to Io, significant amounts of silicate melt hidden below Europa's hydrosphere can be produced during most of its history due to the limited efficiency of internal cooling. We have also demonstrated that melt production is concentrated in pockets in the polar areas at present due to the increased tidal dissipation near the poles. The melt retention and its influence on tidal dissipation were not included, even though partially melt areas embedded in the rocky mantle can severely change the material properties above a critical melt fraction. In the frame of this project, we suggest studying the tidal deformation of Europa's rocky mantle, including melt pockets, which will allow us to evaluate the available heat sources controlling the habitability potential of Europa.

 

Researcher: Barnabas Barna

Project: Bringing balance to the Force: studying the origin of equilibrium near the Galactic Center

Allocation: Barbora VIZ 40, Karolina CPU 4900,  Karolina VIZ 40

Abstract: At a distance of 25.500 lightyears, the Galactic Center of the Milky Way is subject of particular interest to both the observations and theoretically oriented astronomical studies. Despite the decade-long intensive observational campaigns, we still know only a little about the exact structure of the inner ~100 ly, because observations are limited by the line of sight effects and the extinction from the interstellar matter (ISM). Thus, numerical methods are required to simulate the contribution of various energy and mass injections, e.g. stellar wind or inwardly moving molecular clouds.

We are planning multiple hydrodynamic simulations with FLASH in 3D to investigate the impact of stellar wind and supernova explosions on the interstellar environment. The results will be matched with recent observations, e.g. the current distribution, velocity, and turbulence of the ISM. The most realistic setup will serve as a model environment in future simulations for studying whether supernovae can deliver mass to the supermassive black hole of the Galaxy.

 

Researcher: Marek Stepan

Project: Conductance of Molecular Junctions - DFT+GW Approximation

Allocation: Barbora FAT 200, Barbora VIZ 40, Karolina CPU 2700, Karolina VIZ 40

Abstract: Continuous miniaturisation of the electronic devices brought about the current digital revolution. However, further progress is limited by the fundamental properties of such devices. At the nanometer scale, the quantum-mechanical nature of the electrons becomes sufficiently pronounced and the classical description is no longer accurate. In order to further miniaturize the devices, a new approach which uses the quantum properties of the electrons instead of suppressing them needs be developed. In order to achieve this, a reliable simulation of properties of the nanoscale devices needs to be available. The field of molecular electronics tries to solve some of these problems in the context of systems of single molecules bridging the gap between two wires. The quantum nature of the molecule itself and of the bonding between the molecule and the wires makes this problem particularly exciting, but also very computationally demanding. This project aims to use the density functional theory (DFT) refined by the GW approximation to demonstrate the reliability in prediction of the molecular junction conductance.

 

Researcher: Dana Nachtigallova

Project: Computational modelling of photocatalytic water splitting and hydrogen formation on TiO2 surfaces

Allocation: Barbora CPU 9400, Barbora FAT 200, Barbora GPU 1000, Barbora VIZ 40, Karolina CPU 50000,  Karolina FAT 200, Karolina GPU 20000, Karolina VIZ 40

Abstract: The project aims to improve the understanding of photocatalytic water splitting on pure or doped TiO2 semiconductors through computational approaches. These studies will be done in close collaboration with experimental partners which use state-of-the-art experimental techniques This collaboration has the potential to contribute to the realization of the hydrogen economy with great applicability in various fields. The DFT calculations, employing selected DFT functionals, will be performed to calculate the electronic, optical, and mechanical properties of TiO2 polymorphs. These studies will be carried using crystal models to mimic experimental conditions. From the crystal model we cut the cluster model and perform expensive single and multi-reference calculations.

 

Researcher: Frantisek Karlicky

Project: Physics of MXenes – Promising 2D Materials

Allocation: Barbora CPU 77800, Barbora VIZ 40, Karolina CPU 31000, Karolina VIZ 40

Abstract: This project is focused on study of physical properties of a new class of 2D materials, MXenes. These materials are now regarded as highly suitable candidates for numerous technological applications for their robustness and wide range of physical properties achievable within a single materials class (e.g., magnetic properties, tunable band gap from metals to semiconductors) by virtue of the variable composition and surface functionalization. Various MXenes may be well combined to form heterostructures, further extending application potential of these promising materials. On the other hand, accurate theoretical description of MXenes remains challenging on the methods side, as well as for prohibitive computational cost. Due to the presence of delicate interplay of multiple effects in 2D materials, accurate predictions of MXene properties require accurate and costly many-body methods instead of usual density functional theory. The project will contribute to the fundamental understanding of MXene physics and boost their experimental research as well as technological applications.

 

Researcher: Jiri Jaros

Project: Modeling of Low Intensity Focused Ultrasound Using Artificial Neural Networks

Allocation: Barbora CPU 4800, Barbora GPU 500, Barbora VIZ 40, Karolina CPU 1000,  Karolina GPU 500, Karolina VIZ  40

Abstract: Transcranial low-intensity focused ultrasound (LIFU) therapy is increasingly used for the non-invasive treatment of brain disorders. However, conventional numerical wave solvers are currently too computationally expensive to be used online during treatments to predict the acoustic field passing through the skull.

As a step towards real-time predictions, we developed a fast iterative solver for the heterogeneous Helmholtz equation in 2D using a fully-learned optimizer. The lightweight network architecture is based on a modified UNet that includes a learned hidden state. The network is trained using a physics-based loss function and a set of idealized sound speed distributions with fully unsupervised training (no knowledge of the true solution is required). The learned optimizer shows excellent performance on the test set, and is capable of generalization well outside the training examples, including to much larger computational domains, and more complex source and sound speed distributions, for example, those derived from x-ray computed tomography images of the skull.

However, the solver is currently able to work only with a single 2D slice of the brain. In order to extend the usage of the solver, we are going to extend it to 3D in this project. This will require further improvement of the UNet structure, more robust evaluation of the loss function, improved training procedure and almost an order of magnitude higher computational cost.

 

Researcher: Roman Divis

Project: Reflective nested simulations in railway transport used as training data set to improve the decision-making process in simulators

Allocation: Barbora VIZ 40, Karolina CPU10500, Karolina VIZ 40

Abstract: Railway transport is used as one of the dominant types of passenger and freight transport. Compared to road transport, the rail infrastructure represents a more restricted environment with limited infrastructure. Unexpected events, train delays, and malfunctioning railway infrastructure devices can affect the quality and continuity of rail traffic operation. Computer simulations of railway transport usually need to consider these unexpected (stochastic) events. In reality, there is usually an employee who is responsible for resolving issues that come from possible conflicts (e.g., occupied station track due to train delay). In simulation, to solve these conflicts, there are decision-making algorithms. These algorithms are based on many different approaches - on simple priority lists, on multi-criterial evaluation, on artificial neural networks, ... On the basis of our previous research, we applied the reflective nested simulations (RNS) method. The method provides good results but is burdened by high computational costs. The aim of this project is to use reflective nested simulations method to gather training data set for other decision-making methods that are based on learning (such as artificial neural networks, decision trees, etc.). These methods may provide results of the same quality as reflective nested simulations without expensive computations during simulation time. The possibilities and limitations of this approach are subject of future research.

 

Researcher: Fabien Jaulmes

Project: Computational modelling of fast ion orbits and their consequences in tokamak plasmas

Allocation: Barbora CPU 1100, Barbora VIZ 40, Karolina CPU 780, Karolina VIZ 40

Abstract: Nuclear fusion technology might enable us to generate energy without releasing large amounts of greenhouse gases into the atmosphere or leaving behind us long lived radioactive waste. The tokamak concept involves the use of magnetic fields to confine plasma hot enough to sustain fusion within itself. Today, as a part of international project under the title ITER, a new tokamak is built in southern France. If successful, the device would be the first one of its kind to produce net energy. Fusion is now supported as a way to revert climate change and was discussed at the COP26 : https://unfccc-cop26.streamworld.de/webcast/iter-organization-fusion-energy-the-state-of-art. COMPASS Upgrade (COMPASS-U) will be a large magnetic field (5T) tokamak that will allow the scientific investigation of various physics issues related to the operation of the future ITER. In particular an 80keV Neutral Beam Injection system is planned to heat up the plasma with 4MW of external power. The study and modelling of NBI-born particle behavior is of great relevance and might impact future design of the system and its integration in the overall design. We request here computational time for the modelling of the interaction of the fast particles with the background plasma. Our modelling tool, the EBdyna code with its new collisional features, has been benchmarked against the NUBEAM code on several test-cases. A publication in the Nuclear Fusion journal summarizes the results of our initial modelling effort [1].

 

Researcher: Diego Nicolas Calderón Espinoza

Project: Moving-mesh Radiation-hydrodynamic Modelling of Astrophysical Transients

Allocation: Barbora CPU 26000, Barbora VIZ 40, Karolina VIZ 40

Abstract: There is overwhelming evidence that new power sources and more complex geometries are strictly necessary for explaining the observed light curves of certain astrophysical transients, e.g. supernovae in the presence of surrounding medium. Numerical models are challenging due the need of performing radiation hydrodynamic simulations over many orders of magnitudes of both time and space, which typically implies an elevated computational cost. In this context, we have developed a new tool that solves the multi-dimensional radiation hydrodynamic equations using the moving-mesh technique. We have shown that this approach is ideal for simulating astrophysical transients even if their geometry is not spherically symmetric over ten orders of magnitude spatially and temporarily. Thus, we propose the use this new capability for conducting high-resolution studies for synthesizing observational signatures of theoretical models of astrophysical explosions interacting with their surrounding medium with complex geometries, e.g. supernovae, stellar mergers, and tidal disruption events, I.e. when a star is destroyed due to the strong tidal pull of a super-massive blac hole. Our results would provide a large set of synthetic light curves from a systematic numerical study of multidimensional astrophysical transients making use of self-consistent radiation hydrodynamics sampling appropriate domain size and durations.

 

Researcher: Libor Veis

Project: Large-scale DMRG calculations of novel polycyclic aromatic hydrocarbons

Allocation: Barbora VIZ 40, Karolina CPU 27000, Karolina GPU 2000, Karolina VIZ 40

Abstract: In this project, we will apply our massively parallel implementation of the density matrix renormalization group (DMRG) method [1], which is a method of choice for strongly correlated systems requiring very large active spaces, on the electronic structure problem of novel polycyclic aromatic hydrocarbons (PAHs) recently prepared via the ultra-high vacuum (UHV) on-surface chemical synthesis [2] and characterized experimentally by means of the scanning probe microscopy with the unprecedented sub-molecular resolution [3]. In particular, we will continue our previous project and study different super-ring annulene-within-annulene (AWA) structures exhibiting peculiar global aromatic character, which contradicts the well-known Huckel’s [4n + 2] rule [4]. Our state-of-the-art calculations are expected to provide a thorough theoretical insight into the structure-properties relationship and help to explain the unusual electronic properties of these PAHs.

 

Researcher: Andrzej Kadzielawa

Project: Symmetric Special Quasirandom Structure Generator

Allocation: Barbora CPU 13000, Barbora VIZ 40, Karolina CPU 12400, Karolina VIZ 40

Abstract: With the introduction of the novel fusion reactor vessels, there is demand for the materials applicable that are respecting the so-called fail-safe scenarios. Currently, the best performing materials are the alloys of Tungsten. The problem of using such solid solutions is their decomposition at 1000°C into Tungsten- and Chromium-rich grains. As the understanding of the WCr system is limited, it is crucial to comprehend the thermodynamical behavior of the W-Cr system and its stability. For this purpose, we devised a method of predicting the (in)stability of the alloy, based on the composition via the high-throughput calculations of the ab-initio supercell systems with electronic, phononic, and configuration entropy taken into account. The crucial point of this approach is a resources-conserving generation of the so-called Special Quasirandom Structures, allowing for the incorporation of the low-but-still-of-essence symmetries of the colored lattices.

 

Researcher: Michal Cifra

Project: SubTHz spectra of functional vibration modes of proteins

Allocation: Barbora VIZ 40, Karolina CPU 12000, Karolina GPU 12000, Karolina VIZ 40

Abstract: Understanding the interaction of electromagnetic field with the biological matter at the molecular level is crucial both for bionanotechnological applications as well as for setting safety limits for undesired electromagnetic exposure of complex biological systems, including humans. In the current computational project, we aim to explore how the electromagnetic field interacts with tubulin – a protein that serves several important roles in cells, including cell division. Our focus is on frequencies of electromagnetic field which match the frequencies of biologically functional protein vibrations. The frequencies of these collective protein vibration modes lie typically in subTHz - THz frequency range depending on the size and other properties of the protein molecule. The goal of the project is to employ molecular dynamics simulations and modal analyses to calculate the spectra of tubulin vibration modes that are associated with the tubulin conformational transitions. Such knowledge will guide experiments and provides the first step for the development of electromagnetic control of tubulin-based nanomachines.

 

Researcher: Martin Friak

Project: AI meets magnetism of materials

Allocation: Barbora CPU 28600, Barbora GPU 1000, Barbora VIZ 40, Karolina CPU 30000,  Karolina GPU 4000,  Karolina VIZ 40

Abstract: Magnetic properties of atoms in magnetic crystals very sensitively depend on atoms around them. These structure-property relations represent one of the most difficult problems in materials science. Their deeper understading would open ways to future design of new magnets. Therefore, computationally demanding quantum-mechanical calculations are used to analyze them. Modern AI-tools, e.g., neural networks, combined with structural descriptors are computationally much less deman-ding. But neural networks must be first trained on tens of thousands of training cases. We plan to use quantummechanical calculations to build a sufficiently big database of local magnetic moments in different crystals, train a neural network and apply it when designing new magnetic materials.

 

Researcher: Petr Kulhanek

Project: Predicting DNA Mutability by Thermodynamic Characterization of Mismatched Base Pairs

Allocation: Barbora VIZ 40, Karolina CPU 6640, Karolina VIZ 40

Abstract: Incorrect base-pairing (mismatches) in double-stranded DNA can result in inherited genetic diseases, cancer, and aging. Thus, organisms developed several mechanisms to remove such errors to keep the genetic information intact. The mismatch repair pathway (MMR) increases the fidelity of DNA replication by several orders. Errors are detected by the MutS enzyme, which seeks for the property of dsDNA altered due to a mismatch. Previous studies have found that base-pair opening is more important than overall DNA bending in mismatch discrimination from regular Watson-Crick base pairs. In this study, we want to improve the description of the base-pair opening by employing new geometrical parameters (collective variables) to better describe the thermodynamics of mismatched base-pair in syn conformations. Moreover, the thermodynamic stabilities will be determined in various sequence contexts of DNA. Such information is essential for understanding the appearance of mutation coldspot and hotspot sequences with perspective for the early diagnosis of inherited diseases.

 

Researcher: Martin Hurta

Project: Evolutionary Design of Movement Disorders Classifier

Allocation: Barbora VIZ 40, Karolina CPU 2000, Karolina VIZ 40

Abstract: The evolutionary design (ED) takes inspiration from biology to develop new methods of computer programming that deliver solutions to unusual and challenging problems. Our research focuses on applying ED to real-world problems, particularly signal processing connected to the Parkinson's disease (PD) movement abnormalities classification problem. PD's symptoms treatment can cause motor abnormalities called levodopa-induced dyskinesia (LID), which worsen the patients' quality of life. Recently, the LID-classifier model, developed using evolutionary design technique (particularly cartesian genetic programming), has been introduced. LIDclassifier aims to help clinicians adjust a patient's medication dosage and find a tolerable balance between the benefits and side effects. In this project, we plan to further extend this work by including hardware-oriented requirements in the automated process of the classifier design. The automated design will be performed using cartesian genetic programming with the coevolution of adaptive size fitness predictors (CGPcoASFP). Found hardware-oriented LID-classifier can be then implemented directly in a home-wearable device and thus help clinicians monitor Parkinson's patients during their everyday activities directly at home. According to this case study, we plan to introduce a methodology for automated design of hardwareoriented implementation of classifiers of other movement disorders accompanying neurological conditions.

 

Researcher: Jan NAVRÁTIL

Project: Role of lattice defects, N-dopant, and single-atom Pt and Pd co-catalysts in TiO2(001) anatase in water splitting for green hydrogen production

Allocation: Barbora VIZ 40, Karolina CPU 25000, Karolina VIZ 40

Abstract: Hydrogen is an attractive energy carrier due to its high energy-to-mass ratio that can replace fossil fuel consumption, meeting global energy demands. When produced from renewable energy, it is considered as green hydrogen. To this end, photocatalytic water splitting has attracted increasing attention, and titanium dioxide (TiO₂) became the most prominent photocatalyst used due to its low cost, chemical stability, earth abundance and nontoxicity. However, TiO₂ wide bandgap reduces its potential for the absorption of visible light.

The possibility to enhance the visible light photoactivity due to the vacancy defects and nitrogen-dopant of TiO₂(001) anatase will be investigated by using computational methods based on density functional theory. The effect of single-atom Pt and Pd co-catalysts on the properties of TiO₂ and water-splitting potential-energy profiles will be studied in detail. Computational results will shed new light on the construction of efficient catalytic water-splitting systems.

 

Researcher: Radek Halfar

Project: Development and evaluation of algorithms for personalized medicine

Allocation: Barbora GPU 200, Barbora VIZ 40, Karolina CPU 868, Karolina GPU 400, Karolina VIZ 40

Abstract: This project focuses on researching new algorithms for protein behaviour and its interaction with ligands. These algorithms are an essential part of developing new substances (drug candidates) for specific groups of patients. For this purpose, modern methods in artificial intelligence and optimization algorithms are used.

 

Researcher: Pablo Nieves

Project: New approaches to calculate magnetoelasticity efficiently

Allocation: Barbora CPU 50600, Barbora VIZ 40, Karolina CPU 17578, Karolina GPU 15581, Karolina VIZ 40

Abstract: Magnetostriction is a physical phenomenon in which the process of magnetization induces a change in shape or dimension of a magnetic material. Nowadays, in many technical applications such as electric transformers, motor shielding, and magnetic recording, magnetic materials with extremely low magnetostriction are required. By contrast, materials with large magnetostriction are needed for many applications in electromagnetic microdevices as actuators and sensors. In this proposal, we aim to test new approaches to compute magnetostriction within the framework our developed software MAELAS.

 

Researcher: Jan Kuriplach

Project: Structure, magnetism and defects in inverse Heusler alloys Mn2FeZ (Z = Si, Al, Sn)

Allocation: Barbora VIZ 40, Karolina CPU 3000, Karolina VIZ 40

Abstract: The purpose of the proposed project is to investigate computationally the interplay between point defects and structural and magnetic properties of selected inverse Heusler alloys with the composition Mn2FeZ (Z = Si, Al, Sn). The results of ab initio calculations and simulations will be compared and confronted with experimental data, which enables to check the adequacy of theoretical approaches used and to assist the proper interpretation of measurements. Studying Heusler alloys will bring a deeper understanding of their fundamental physics and will pave and ease the way to their possible applications.

 

Researcher: Dominika Maslarova

Project: Tuning of the acceleration length in laser-based electron accelerators

Allocation: Barbora VIZ 40, Karolina CPU 8672, Karolina VIZ 40

Abstract: Laser-based electron accelerators represent a promising concept of the next-generation accelerators. Their main advantage is a remarkably short acceleration length, caused by a significant acceleration gradient (up to ~100 GV/m), about thousand-fold times higher than in the conventional radiofrequency accelerators (up to ~50 MV/m). Consequently, such accelerators introduce a more compact and cheaper option, leading to better accessibility to electron accelerators in research, medical and industrial facilities. In one of the most popular methods, called laser wakefield acceleration, electrons are injected into a plasma wave (wakefield), generated and dragged by a few-tens-of-fs, ultra-intense (≳10¹⁸ W/cm²) laser pulse (driver) in optically transparent plasmas. The plasma wave consists of several periods, each containing an acceleration and deceleration phase. The injection of an electron beam into the wakefield represents a crucial step of the entire acceleration process. The injection methods are primarily focused on the injection of an electron beam to the first wakefield period. However, this limits the acceleration length because electrons trapped in a plasma wave travel slightly faster than the wave. Therefore, they eventually outrun the wave, stop being accelerated and start to be decelerated. In our recent research, indications of a possible adjustment of the period of injection were observed in the simulations. It has been previously theoretically analyzed, that if the injection is triggered into later wakefield periods, electrons can be gradually accelerated in each of the periods, sliding forward from one period to another, until they reach the first one. However, the control of the injection into later wakefield periods itself has not been analyzed before. The discovery could, therefore,  immensely prolong the whole acceleration length. Also, higher energies can be achieved, which is one of the main goals of the accelerator community. In case the new simulations verify this preliminary outcome, the new method will be experimentally tested.

 

Researcher: Tadeas Kalvoda

Project: Retrieving data. Wait a few seconds and try to cut or copy again.

Allocation: Barbora CPU 21000, Barbora VIZ 40, Karolina VIZ 40

Abstract: The interactions of transition metal ions with water molecules in aqueous solution, resulting in formation of solvation shells, plays an important role in various aspects (desired or undesired) of biochemical reactions involving these ions. Therefore, a precise description of solvation effects of such ions is desirable, potentially improving the modelling of thermodynamic properties of metal-protein reactions, as well as drug design. Hydration of small, doubly charged “hard” ions is particularly problematic to describe because of their strong electronic polarization effects to the solvent molecules, which are difficult to capture using standard force field molecular dynamics. Ab initio molecular dynamics treats (valence) electrons explicitly which allows it to include such effects.

 

Researcher: Martin Kejík

Project: Computational studies of the synthesis and catalytic properties of octakis(trimethylstannyl)spherosilicate-based porous hybrid single-site metallosilicates

Allocation: Barbora CPU 26000, Barbora VIZ 40,  Karolina CPU 20000, Karolina VIZ 40

Abstract: Heterogeneous (solid) catalysts are the unsung heroes of the industrial chemical synthesis as over 80 % of all industrial syntheses rely on them. Catalysts in general are compounds that are not consumed themselves, but their presence steers chemical reactions towards desired products. The most desirable form of a catalyst is a solid as it can be conveniently separated from liquid/gaseous “production mass”. Unfortunately, solid catalysts are hard to develop and even harder to understand how they work. The current shift away from fossil resources has created much pressure on the development of novel and more efficient heterogeneous catalysts. In the past century the humanity has mostly relied on decomposition/breaking processes to produce valuable chemicals – coal and oil (complex molecules) were broken into intermediate-sized valuable molecules. In the current century the humanity must master the opposite process – take small molecules from renewable resources (water, carbon dioxide, methane …) and join them in a controlled fashion to produce valuable compounds. As myriads of various ways to combine the small molecules exist, the catalysts are the key to making this future possible. Our experimental research is aiming to develop catalysts for the conversion of ethanol (a renewable resource) into butadiene (a precursor to plastics; traditionally obtained from fossil resources). In order to gain more insight and to steer our experimental endeavors we hope to employ the methods of computational quantum chemistry.

 

Researcher: Urszula Wdowik

Project: Low-dimensional magnetism and vibrational dynamics of quantum magnets with chain-like crystal structure

Allocation: Barbora VIZ 40, Karolina GPU 3500, Karolina VIZ 40

Abstract: Organometallic compounds formed by ladder-like infinite chains with μ3-bridging chloro-ligands and exhibiting low-dimensional magnetism have received significant experimental and theoretical interest not only because of their unconventional magnetic properties resulting from the interplay between quantum fluctuations and geometrical frustration, but also due to their potential application in molecular spintronics and quantum technologies. The molecular-based magnets can exhibit superior properties compared with their inorganic analogues in terms of chemical stability and tunability. On the other hand, their complex crystallographic and magnetic structure as well as dynamics of their lattices pose a real challenge for experimentalists. This project addresses ab initio identification of the structural features, magnetic interactions, electronic and vibrational properties of Cu(tn)Cl2 (tn=1,3-diaminopropane=C3H10N2), Cu(en)Cl2 (en=ethylendiamine=C2H8N2),Cu(en)SO4, and Cu(en)2CrO4 organometallic compounds. Results of the proposed theoretical investigations will serve for comparative analysis with experimental data provided by the experimental group lead by A. Orendáčová, UPJŠ, Košice, Slovakia within bilateral Czech-Slovak cooperation.

 

Researcher: Martin Surkovsky

Project: Deterministic Road Traffic Simulator – II. phase

Allocation: Barbora CPU 1000, Barbora VIZ 40, Karolina CPU 700, Karolina VIZ 40

Abstract: A usage of historical data observed from traffic combined with high performance computing resources for traffic simulation can optimize traffic flow in a macro scale level. A set of algorithms used for route planning has been developed in ADAS laboratory, including in-house traffic simulator and a scalable, distributed system for serving vehicle routing requests which is able to handle tens of thousands individual cars.

The aim of this effort is to reduce number of non-deterministic events in the traffic simulator to increase the reproducibility of results. In this way we will be able to fine-tune the setting of routing algorithms without need of doing several experiments with a particular setting to get statistically significant results.

 

Researcher: Jakub Sebesta

Project: Mechanical properties and stability of Nb, Al based HEAs

Allocation: Barbora CPU 34600, Barbora GPU 5000, Barbora VIZ 40, DGX-2  600, Karolina GPU 4300, Karolina VIZ 40

Abstract: An interesting part of materials that bring unusual mechanical properties are multielement alloys called High Entropy Alloys (HEA) based on a number of principal metallic elements. The name itself refers to the significant contribution of the entropic term to the stability of alloys, given by a large number of constituents. Although this type of alloy has been studied for a quite long time, there are still unexplored areas due to possible complex composition. An interesting group of these compounds are already derivatives from the very first one, Cantor's alloys based on 3d metals with Nb admixture. It has been observed that Nb doping can lead to improved mechanical properties. Recently, we devoted our effort of understanding the magnetic interaction of Cantor's alloys (PRM 2019, PRB2021) where we modify them with 3d and sp metal substitution. Therefore we would like to know how these derivatives change the mechanical and especially magneto-elastic properties and ground state magnetic behavior of selected derivatives (e.g. Nb and Al) of the Cantors alloys.

 

Researcher: Damien Lucien Michael Gagnier

Project: Angular momentum transport during common-envelope self-regulated spiral-in phase

Allocation: Barbora VIZ 40, Karolina CPU 36000, Karolina VIZ 40

Abstract: Common-envelope evolution is the name given to a short-lived period in the life of a binary system during which two stars orbit and interact inside a single common-envelope. Depending on the complex interplay between the binary and the shared envelope structure, the common-envelope can ultimately be expelled, or, alternatively, the two stars can merge into a single object. Common-envelope evolution is expected to be at the origin of a variety of objects and transients such as peculiar stars, planetary nebulae, red transient events, and the emission of gravitational waves, for example. However, while common-envelope evolution is arguably the most crucial major process in binary evolution, it is also the least-well-constrained and more generally one of the most important unsolved challenges in star evolution. In this project, we will perform 3D-hydrodynamic simulations of common-envelope evolution on a spherical grid, focusing on the quasi-steady self-regulated spiral-in phase, that is after the initial and rapid plunge-in of the secondary star inside the primary’s envelope. In particular, we will investigate the orbital evolution of the binary by examining the various torques exerted on the system, as well as the shared envelope dynamics by characterising the associated transport processes.

 

Researcher: Martin Kolisko

Project: Phylogenomic analyses of Excavates

Allocation: Barbora VIZ 40, Karolina CPU 8344, Karolina VIZ 40

Abstract: A well resolved phylogenetic tree of eukaryotes is essential for our understanding of origin and evolution of the eukaryotic cell and its diverse structures and organelles. One of the most persisting and difficult problems to solve, is the phylogenetic position of different excavate groups and their relationships to each other. This is mostly due to their very diverse nature at the DNA and protein sequence level, which causes the true phylogenetic signal in data to be drown-out by random noise or erroneous signal. The state-of-the-art method for resolving ancient evolutionary relationships is phylogenomics where hundreds of genes get combined into one dataset. Here we plan to use such dataset enriched for novel, previously unstudied, excavate taxa, combined with cutting edge phylogenomic methods to amplify the true evolutionary signal and reduce the amount of noisy or erroneous data.

 

Researcher: Martin Klajmon

Project: Reliable Diffusivities of Amorphous Materials from Polarizable Molecular Dynamics Simulations

Allocation: Barbora VIZ 40, Karolina CPU 23500, Karolina FAT 1000, Karolina VIZ 40

Abstract: Transport properties, such as the diffusion coefficient (D), are important characteristics of materials. These properties and also their temperature evolution may serve as the basis for evaluating other physicochemical properties, such as phase transitions (e.g., the glass transition between liquid and amorphous glass-like solid state of a material, or recrystallization molecular crystals, such as pharmaceuticals). Instead of labor-intensive experiments, the transport properties can be calculated using molecular dynamics (MD) simulations. Here, a user faces two essential questions: which computational technique within MD and which force field (FF) to use to achieve the best results possible? To address the former question, we will aim at developing and testing a suitable technique to calculate as-reliable-as-possible diffusion coefficients of selected industrially, environmentally and pharmaceutically important compounds (e.g., ionic liquids and pharmaceuticals ingredients), with a special focus on their amorphous state at lower temparatures, where the currently recommended technique to obtain D from MD trajectories tends to fail. Once identified, this technique will allow us to address the second question whether new polarizable FFs hold their improved accuracy in predicting D not only at room and higher temperatures (as recently demonstrated) but also at temperatures corresponding to the amorphous solids or subcooled liquids.

 

Hlavní řešitel: Ondrej Marsalek

Project: An efficient path to MP2-based machine learning potentials in the condensed phase

Allocation: Barbora VIZ 40, Karolina CPU 22800, Karolina VIZ 40

Abstract: Today, just a few years after being noticed in the broader scientific community in physics and chemistry, machine learning interaction potentials are widely employed in modern molecular and condensed phase physics. They are easy to use and calculate physical properties of atomistic systems much more efficiently than the underlying electronic structure methods. However, the construction of a machine learning potential that reliably calculates energies and forces for a system in a wide range of different states is anything but trivial and can be associated with large computational costs as well as substantial effort on the part of the researcher. In this project, we propose a shortcut to a new machine learning potential for high-level electronic structure calculation methods by taking advantage of existing training sets for lower-level methods. This will eventually enable us to perform molecular dynamics in extensive simulations of large molecular systems driven by interactions that correspond to a high-quality electronic structure method, something that was not possible up to this point due to prohibitive computational cost.

 

Researcher: Michael Bakker

Project: Investigations into 31P NMR Chemical Shifts in Phosphorylated Intrinsically Disordered Proteins through a MD/ADMA/DFT Approach and Machine Learning

Allocation: Barbora GPU 8000, Barbora VIZ 40, Karolina CPU 10000, Karolina GPU 4000, Karolina VIZ 40

Abstract: The aim of this project is the proper and complete computation of ³¹P NMR Chemical Shifts for specific intrinsically disordered proteins (e.g. hTH1, RNAP II CTD, etc.). These IDPs are of particular importance due to their contributions and implications with neurogenetic diseases (e.g. Azlheimer‘s or Parkinson‘s). Additionally, better understanding of intrinsically disordered proteins has great value for advanced mutation screening and protein functionality. Unfortunately, traditional techniques such as x-ray crystallography have difficulties describing the flexible nature inherent to these chaotic macromolecules. Computational methods, such as the ones proposed in this submittance can be utilized to alleviate these problems. Snapshots derived from molecular dynamics trajectories can be employed successfully to create an effective ergodic conformational ensemble. This ensemble better reflects the flexible nature of IDPs than singular PDB structures. This project‘s aim is to properly evaluate the implementation of MD, the necessity of an optimization, the selection of an effective and rapid basis set, and the possible inclusion of modern machine learning tools such as cluster analysis. Further applications of this research will be employed on later molecules and inevitably used to develop effective computational tools in the future.

 

Researcher: Jan Kočí

Project: Archaic introgression in humans: automated finding of optimal admixture graphs and cross-validation on different types of data

Allocation: Barbora VIZ 40, Karolina CPU 3900, Karolina VIZ 40

Abstract: Ancestors of modern humans have a history of interbreeding with other hominins, most notably the Neanderthals and the Denisovans. However, many aspects of this history are still unclear and hotly debated. One thing is clear: the history of admixture with archaic hominins was complex and involved more gene flows besides the well-established ones (Neanderthal to non-Africans and Denisovan to Papuans and Australians).

Most publications to date only considered a few alternative models of this history incorporating population divergence and admixture events, often crafted manually and fitted to one chosen type of genetic data. During our investigations using a new automated tool for finding well-fitting admixture graphs, we have found many models with different topologies which have fits to the data statistically indistinguishable from those of published models. This undermines confidence in the published models and the conclusions built on them.

In this project we will test several dozens of the best candidate graphs of archaic introgression history obtained from a large batch of automated searches of the model space. Thus a set of models fitting f-statistics well will be generated. These models will then be fitted to an independent type of genetic data, site patterns (using the Legofit package). We hope this approach will allow us to find models fitting both data types well. To our knowledge, this meticulous approach has not been applied in archaeogenetic research before.

 

Researcher: Jan Zemen

Project: Modeling Magneto-Optical Spectra in Antiperovskite Nitrides with Ferrimagnetic Phase

Allocation: Barbora CPU 25800, Barbora VIZ 40, Karolina VIZ 40

Abstract: The aim of the project is to calculate Magneto-optical (MO) spectra in antiperovskite nitrides with formula unit Mn_4-xNi_xN where x=0, 0.5, 1, and compare them to measured MO spectra to gain insight into the magnetic structure of the ordered alloy below and above room temperature. The magnetic structure of Mn-based antiperovskite nitrides was first studied by neutron diffraction in bulk samples in 1970s. A range of collinear and non-collinear magnetic phases have been detected which host useful effects even in the bulk form, e.g., the invar effect, barocaloric effect. In recent years, several groups have grown these materials in thin-film form aiming for applications in non-volatile memory devices, sensors, and actuators. We have simulated and measured the Anomalous Hall Effect (AHE) and MO spectra in Mn₃NiN films recently. The agreement of both sets of data suggests that the non-collinear antiferromagnetic (AFM) phase transforms into a collinear ferrimagnetic (FIM) phase above the Neel temperature in compressively strained Mn₃NiN film on BaTiO₃ substrate. Here we propose to explore related FIM phases predicted in Mn_4-xNi_xN this year. We will focus on phases where the symmetry of the magnetic structure allows for non-vanishing Berry curvature, i.e., there is an intrinsic contribution to anomalous Hall conductivity and to MO Kerr effect (MOKE). Understanding the dependence of AHE and (MOKE) on chemical composition will futher increase the potential of this class of materials for applications in spintronic and optoelectronics.

 

Researcher: Miroslav Kolos

Project: Interactions Modeling of Nitridated Zero-Valent Iron Nanoparticles with Chlorinated Hydrocarbons

Allocation: Barbora VIZ 40, Karolina CPU 13281, Karolina VIZ 40

Abstract: Chlorinated hydrocarbons, such as trichloroethylene (TCE), are highly toxic pollutants of groundwater, spreading worldwide. Remediation of groundwater using conventional technologies is ineffective and expensive from the long-term perspective. Therefore, modern remediation techniques are widely used based on the usage of nanoscale zero-valent iron nanoparticles (nZVI) and their modifications. Pure nZVI nanoparticles have strong dechlorination capability towards TCE, proven in experimental and theoretical studies. However, their reactivity decreases quickly due to the oxidation process in a water environment. The possibility of stabilizing the nZVI is nitriding, which has been used for decades to improve the corrosion resistance of iron and steel materials. Products of such nitriding process of nZVI are stable in the water environment and preserved dechlorination capability towards TCE. Few experimental studies were conducted on nitrided nZVI for remediation of contaminated groundwater, but still little is known about detailed mechanisms of the dechlorination process. Experimental studies do not observe the fundamental interactions on the nanoscale level, so it is necessary to model these reactions. To improve the practical use of such modified nZVI, a better understanding of interaction and reaction mechanisms is needed. Ideally, the determination of the fundamental surface reaction mechanisms will enable the design of optimal nitrided nZVI particles for field-scale applications.

 

Researcher: Matus Dubecky

Project: Many-body physics of 2D materials III.

Allocation: Barbora VIZ 40, Karolina CPU 82031, Karolina VIZ 40

Abstract: MXenes are layered two-dimensional (2D) materials based on transition metal carbides/nitrides important for their advantageous properties, like, e.g., stability or tunable band gap. Accurate modeling od their properties, like band gaps, or exciton binding energies, requires non-empirical many-body methods to reach benchmark accuracy. This project focuses on application of reference stochastic quantum many-body ab initio methods, like variation and fixed-node diffusion quantum Monte Carlo, to promising and experimentally available MXene systems. In addition to the accurate estimate of materials properties like stability, band gaps, and exciton binding energies, of practical importance, new insights into MXene many-body physics are envisaged to emerge within the project. The obtained results will ultimately serve as a guide toward selection of a cheaper mean-field models for scalable larger-scale models of these promising materials.

 

Researcher: Christian Sippl

Project: Harvesting seismic waveform data for microseismicity with deep learning approaches

Allocation: Barbora VIZ 40, Karolina CPU 2250, Karolina GPU 6000, Karolina VIZ 40

Abstract: The strongest and most devastating earthquakes occur along subduction zones, thus a detailed understanding of the processes involved in the buildup of future large subduction earthquakes can have high social and economic significance. Recent studies have shown that detailed observations of many thousands of small earthquakes (“microseismicity”) on the interplate contact as well as in the downgoing plate can yield critical insights into the current state of the subduction system.

We plan to conduct a comparative study between four subduction regions by harvesting large amounts of available raw seismic data for microseismicity using an automated and deep learning based approach. For this, we will combine existing and recently published algorithms for seismic arrival time picking and phase association into a single automated workflow. The retrieved large catalogs of small earthquakes (expected to contain hundreds of thousands of events) will form the base for several lines of further analysis, including seismic tomography, statistical seismology and combined inversion with GPS data.

All of these lines of research are aimed at characterizing the different plate margins, and at understanding what parameters (e.g. age and temperature of the incoming plate, subduction angle) may govern differences in the observed characteristics between different subduction zones.

 

Researcher: Jakub Planer

Project: Tunable Charge Injection Layers for Organic Semiconductors

Allocation: Barbora VIZ 40, Karolina CPU 7100, Karolina VIZ 40

Abstract: In recent time, light weight, flexibility, low-cost and tunability of organic semiconductors (OS) have drawn the attention of a broad scientific community and semiconductor industry, making them usable in many applications such as active-matrix organic light-emitting diodes (OLEDs) in smartphones, organic field-effect transistors (OFETs), solar cells, memories, photoswitches and sensors. In such

 

Researcher: Jan Hečko

Project: SOLPS-ITER simulations of discharges with seeding in the COMPASS tokamak

Allocation: Barbora CPU 4000, Barbora VIZ 40, Karolina VIZ 40

Abstract: The management of the outbound plasma heat flux is one of the most important challenges that need to be solved for the successful operation of future thermonuclear reactors. One approach to mitigate the heat flux deposited onto the plasma-facing divertor tiles in a tokamak is to utilize the detached mode of divertor operation, during which a “cushion” of neutral particles is created by injecting neutral gas between the ionized plasma and the material surface. This mode has been recently achieved in the COMPASS tokamak both in the standard and the reversed configuration of the magnetic field. A radically different behaviour was observed in each of the aforementioned configurations, mainly in the spatial distribution of visible light radiation sources in the plasma. The goal of this project is to investigate the differences and attempt to explain the underlying physics, which requires the coupling of the experimental data with simulations of the scrape-off layer (SOL) plasma using the SOLPS-ITER code.