16th Project Access Call – Awarded Projects

Results of the 16th Call for Proposals for Project Access.

Projects from the following research areas:

Biochemistry, Bioinformatics and Life sciences (7)

 Elucidating chemo-mechanical coupling in myosin biomolecular motor with atomic resolution

Project Title: Elucidating chemo-mechanical coupling in myosin biomolecular motor with atomic resolution
Project Leader: Marco Cecchini   
Resource Awarded: 16.9 million core hours on Joliot Curie – KNL

Details

Team Members:
Immaculada Martinez-Rovira, ALBA synchrotron, SPAINRachel Delorme, CNRS, FRANCETim Schneider, CNRS, FRANCEConsuelo Guardiola, CNRS, FRANCE

Abstract:
The development of artificial molecular machines is an exciting avenue in nanotechnology. Chemists have successfully synthesized prototypical motors, but none of them even approaches the complexity or the efficiency of their biological versions, which have been fine-tuned by Evolution. Here, we focus on the biomolecular motor myosin and aim at a mechanistic understanding of the recovery stroke, the functional step that couples ATP hydrolysis to the re-priming of the motor. Recent X-ray crystallography and MD simulations of myosin VI by us have indicated the existence of a novel intermediate along the recovery stroke that is consistent with a ratchet-like mechanism. Intrigued by these observations, we set out to provide a molecular understanding of chemomechanical transduction in myosin using the string method with swarms of trajectories. The energetics of the transition will be explored by bias-exchange umbrella sampling along the optimized string. This analysis will provide a mechanistic understanding of the recovery stroke with unprecedented resolution and unveil how myosin motors can operate in a Brownian, fluctuating environment. The analysis on myosin VI will be complemented by the string optimization of the recovery stroke of the prototypical Dictyostelium discoideum myosin II, to compare processive versus non-processive molecular motors. The elucidation of the principles underlying myosin’s function will be crucial for the development of bio-inspired synthetic molecular machines..

 New approaches in radiotherapy: hadron minibeam radiation therapy

Project Title: New approaches in radiotherapy: hadron minibeam radiation therapy
Project Leader: Yolanda Prezado
Resource Awarded: 16 million core hours on MareNostrum

Details

Team Members: Immaculada Martinez-Rovira, ALBA synchrotron, SPAINRachel Delorme, CNRS, FRANCETim Schneider, CNRS, FRANCEConsuelo Guardiola, CNRS, FRANCE

Abstract:
Radiotherapy (RT) is one of the most frequently used methods for cancer treatment (above 50% of patients will receive RT). Despite remarkable advancements, the dose tolerances of normal tissues continue being the main limitation in RT. Finding novel approaches that allow increasing normal tissue resistance is of utmost importance. This would make it possible to escalate tumour dose, resulting in an improvement in cure rate. With this aim, we have proposed a new approach, called hadron minibeam radiation therapy (HADRONMBRT), which combines the prominent advantages of charged particles for RT and the remarkable tissue preservation provided by the use of submillimetric field sizes and a spatial fractionation of the dose, as in minibeam radiation therapy (MBRT). The main objectives of this project are to optimise the minibeam generation, to develop calculation tools allowing a biological-optimisation and to compare treatment plans. HADRONMBRT may open the door to an efficient treatment of very radioresistant tumours, like gliomas. In addition, it can specially benefit paediatric oncology (brain and central nervous system). To be able to carry out all this research an intense work in simulation is needed. The small field sizes used require important calculation resources. Only large clusters, like the ones in PRACE would allow us to perform these calculations.

 Mechanochemical coupling in the rotary cycle of F1FO-ATPase

Project Title: Mechanochemical coupling in the rotary cycle of F1FO-ATPase
Project Leader: Jacek Czub
Resource Awarded: 40 million core hours on SuperMUC

Details

Team Members: Milosz Wieczor, Gdansk University of Technology, POLAND

Abstract:
ATP synthase is — to use a phrase due to Boyer — ”a splendid molecular machine” that plays a central role in cellular bioenergetics, but due to its complex structure and multiple inhibition mechanisms has also emerged as an attractive target for the development of novel therapeutics. In recent years, experimental data shed new light on certain aspects of the rotary catalysis, revealing an unexpectedly modified stepping pattern in the mammalian enzyme, as well as the allowing to precisely determine the relative arrangement of individual subunits. The the current project is thus aiming to build on these new insights to advance our understanding of the mechanism of action of ATP synthase, with a special emphasis on the mechanochemical coupling between the rotor and catalytic subunits. In particular, we intend to use state-of-the-art molecular simulation techniques to examine the energy conversion mechanism in the catalytic portion of ATP synthase and the intersubunit torque transmission. Additionally, we plan to refine the assembled molecular model of the full membrane-embedded enzyme based on experimental data to elucidate the intricate relationship between structural and functional properties of this remarkable nanomotor. Ultimately, we wish to reconcile the available spectroscopic and structural data and provide a consistent thermodynamic, kinetic and mechanistic description of the synthetic cycle of mammalian ATP synthase.

Conformational dynamics, assembly and inhibition of GPCR oligomers

Project Title: Conformational dynamics, assembly and inhibition of GPCR oligomers
Project Leader:  Nicolas Floquet
Resource Awarded: 20 million core hours on Joliot Curie – KNL

Details

Team Members: Pedro Victor RENAULT DE BARROS, Centre National de la Recherche Scientifique (CNRS) UMR5247 Institut des Biomolécules Max Mousseron, FRANCEBartholomé DELORT, Institut des Biomolécules Max Mousseron, FRANCEMaxime LOUET, Institut des Biomolécules Max Mousseron, FRANCEIrina Moreira, Center for Neuroscience and Cell Biology (CNC), PORTUGAL

Abstract:
G-Protein Coupled Receptors (GPCRs) together form one of the largest family of human proteins, involved in nearly all physiological / pathological processes. Recently, we identified the possibility for these receptors to form tetramers in which each central receptor is coupled to its cognate G-Protein. In this proposal, we will use different modelling tools to study the conformational dynamics of these GPCR oligomers, to better understand how they assemble at the membrane surface and how to inhibit the resulting protein : protein interfaces. The starting model will be that we validated recently and comprising two Ghrelin (GHSR-1a) and two Dopamin D2 receptors, coupled respectively to their cognate Gq and Gi hetero-trimeric G-Proteins. This model will first be refined by All-Atoms and Coarse-Grained molecular dynamics performed at the µs time-scale. The resulting trajectories will permit both to compute the collective motions in these large assemblies and highlight some hot-spots at the protein : protein interfaces. The expected results will drive the synthesis and testing of new peptides targeting these interfaces. The need for Tiers-0 is fully justified by the size of the studied systems (nearly 500 000 atoms) and by the multiplicity of the calculations that will be required to achieve our different goals. These calculations will permit to shade light on these complex assemblies and how they function at the molecular scale.

In silico drug trials in the beating ischaemic human heart

Project Title: In silico drug trials in the beating ischaemic human heart
Project Leader: Blanca Rodriguez
Resource Awarded: 54.6 million core hours on MareNostrum

Details

Team Members: Ruth Aris, Barcelona Supercomputing Center, SPAINMarina Lopez, Barcelona Supercomputing Center, SPAINAlfonso Santiago, Barcelona Supercomputing Center, SPAINMariano Vazquez, Barcelona Supercomputing Center, SPAINOliver Britton, The University of Oxford, UNITED KINGDOMAlfonso Bueno-Orovio, The University of Oxford, UNITED KINGDOMJulia Camps, The University of Oxford, UNITED KINGDOMFrancesc Levrero, The University of Oxford, UNITED KINGDOMAurore Lyon, The University of Oxford, UNITED KINGDOMFrancesca Margara, The University of Oxford, UNITED KINGDOMPeter Marinov, The University of Oxford, UNITED KINGDOMHector Martinez, The University of Oxford, UNITED KINGDOMAdam McCarthy, The University of Oxford, UNITED KINGDOMAna Minchole, The University of Oxford, UNITED KINGDOMElisa Passini, The University of Oxford, UNITED KINGDOMJakub Tomek, The University of Oxford, UNITED KINGDOMCristian Trovato, The University of Oxford, UNITED KINGDOMXin Zhou, The University of Oxford, UNITED KINGDOM

Abstract:
Cardiovascular disease stands as the major cause of death worldwide, primarily dominated by ischaemic heart disease. Ischemic heart disease results in complex electro-mechanical abnormalities in the human heart, which may lead to sudden cardiac death. Limitations in experimental and clinical techniques hamper their investigation, and there is an urgent need for novel approaches to yield effective improvements in diagnosis and treatment to decrease the mortality of such deadly condition. In this application, a systematic HPC investigation of human heart electro-mechanics will be conducted to unravel key factors determining abnormalities caused by ischaemic disease and pharmacological therapy. Clinical imaging and electrophysiological datasets will be integrated to construct patient-specific anatomically-based electro-mechanical models of ischaemic human hearts, requiring tens of millions of nodes due to numerical constraints. The large disparity in scales of the cardiac electro-mechanics problem further exacerbates convergence requirements, and the need for highly efficient and multi-physics solvers. To meet such needs, we will exploit large HPC platforms (MareNostrum IV) using Alya, a multi-physics solver in the PRACE Benchmark Suite, specifically designed to tackle large-scale coupled problems, and with almost linear scalability up to a hundred thousand cores. Following intensive validation of the personalised electro-mechanical models against clinical electrophysiological recordings and image-based deformation maps, systematic HPC simulation studies will be conducted to evaluate the safety and efficacy of anti-arrhythmic therapy building on available datasets from our pharmaceutical industry collaborators. We expect the results will identify the most effective pharmacological treatments for each ischaemic heart disease scenario. The project will demonstrate the power of HPC for in silico human clinical trials. In collaboration with our clinical and industrial partners we will investigate their translation for precision medicine and drug development pipelines.

LDbud – Lipid Droplet Biogenesis

Project Title: LDbud – Lipid Droplet Biogenesis
Project Leader: Stefano Vanni
Resource Awarded: 68 million core hours on Piz Daint

Details

Team Members: Pablo Campomanes, University of Fribourg, SWITZERLANDValeria Zoni, University of Fribourg, SWITZERLAND

Abstract:
Cellular membranes play a huge role in a myriad of key cellular processes as diverse and fascinating as vesicle trafficking, receptor functioning or synaptic transmission. Yet, due to intrinsic limitations of biophysical and biochemical methods when dealing with a membrane-mimetic environment, a high-resolution knowledge of the basic physicochemical principles that govern these processes is still limited. In this context, in silico modeling is a well-established alternative strategy, since computational-based studies offer the possibility to investigate the behavior of matter directly at the atomistic level under highly controlled conditions. In this project, I propose to use coarse-grained (CG) molecular dynamics (MD) simulations to investigate a poorly characterized cellular membrane, that of intracellular lipid droplets (LDs). From a biophysical perspective, the membrane of LDs is remarkably fascinating since it has a unique composition, consisting of a hydrophobic core of neutral lipids surrounded by a phospholipid monolayer. I will specifically focus on LD biogenesis, and namely on the budding of nascent LDs from the bilayer membrane of the endoplasmic reticulum (ER). To do so, we will use recently developed chemical-specific CG parameters explicitly optimized to study oil-water interfaces in combination with new “computational assays” developed to address large-scale membrane remodeling phenomena using MD simulations.

Molecular view on assembly and disassembly of the signaling complex consisting of the human β2-adrenergic receptor and its cognate G protein

Project Title: Molecular view on assembly and disassembly of the signaling complex consisting of the human β2-adrenergic receptor and its cognate G protein
Project Leader: Kristyna  Pluhackova
Resource Awarded: 15 million core hours on MareNostrum

Details

Team Members:

Abstract:
The ability of a cell to react to extracellular stimuli is enabled by a complex protein machinery attached to the cell membrane. The core of this machinery consists of transmembrane proteins termed G protein-coupled receptors (GPCRs) and their associated G proteins found in the cell interior. Due to their important roles in numerous crucial signaling pathways, GPCRs are targeted by one third of all current drugs. This project aims at deciphering the pathways and the underlying free-energy landscapes of the binding and unbinding processes between a G protein and a prominent G protein coupled receptor, β2 adrenergic receptor, at different stages of the activation cycle. The results are expected to provide new important details on the mechanism of signal transmission from the cell exterior to the cell interior. Using fully atomistic bias-exchange metadynamics molecular dynamics simulations we will address here the assembly, the structure, and the stability of the signaling complex in the pre-activation state of the signaling cycle and the disassembly of the complex in the post-activation state. The simulations will deliver molecular determinants of the G protein-exerted effects on the receptor and an atomistic view on the (un)binding pathways. Complementary multifunctional atomic force microscopy imaging will be performed to validate the simulations by obtaining the binding force profiles of the G protein to the receptor along the dissociation/association pathways. This collaborative effort – combining atomistic molecular dynamics simulations and atomic force microscopy – aims to provide first information at atomistic resolution on the assembly and disassembly of the G protein-coupled receptor/G protein signaling complex. The results are expected to play important roles (i) in elucidation of the GPCR/G protein selectivity and (ii) in development of drugs altering the transmembrane signaling by influencing the GPCR/G protein complex formation and stability.

top

Chemical Sciences and Materials (10)

First Principles Simulations of Proton-Irradiated Ices

Project Title: First Principles Simulations of Proton-Irradiated Ices
Project Leader: Daniel  Muñoz-Santiburcio
Resource Awarded: 15 million core hours on MareNostrum

Details

Team Members:

Abstract:
Understanding radiation effects on different materials is of paramount importance for many scientific and technological fields like those related to nuclear energy, space industry, laser- and ion-based materials processing and therapeutic applications. The particular case of radiation effects on pure and mixed water ice is very important in Astrochemistry and Prebiotic Chemistry, being water ice present in cosmic dust grains and on the surface of many bodies in the Solar System like asteroids and several satellites (e.g. Europa, Callisto and Ganymede). Despite previous extensive experimental investigations, the mechanism by which radiation damage on solids occurs is still a matter of debate. Clearly, an accurate theoretical description from first-principles simulations is needed in order to rationalize the experimental results. However, performing such simulations is a challenging task because of the non-adiabatic nature of the interaction radiation-solid, and few of such studies exist to date. In the present project, the process of proton irradiation on pure and mixed water ice at space conditions is studied by means of state-of-the-art Molecular Dynamics simulations coupled to Real-Time Time-Dependent Density Functional Theory, thereby adequately describing in real-time the non-adiabatic process by which the proton interacts with the ice matrix, promoting electronic and structural changes and chemical reactions of possible prebiotic interest. In subsequent Molecular Dynamics simulations in the adiabatic regime (Born-Oppenheimer MD) together with the Metadynamics technique, the corresponding reaction networks will be fully explored, characterizing the mechanism and energetics of the reactions triggered by the proton irradiation process.

 XchangePH – Exact Exchange effects on anharmonic vibrational spectra

Project Title: XchangePH – Exact Exchange effects on anharmonic vibrational spectra
Project Leader: Matteo Calandra
Resource Awarded: 17 million core hours on Joliot Curie – KNL

Details

Team Members:

Abstract:
The goal of this project is to go beyond state of the art in the calculation of non-perturbative anharmonic spectra and of the electron-phonon coupling by using advanced approximations for the exchange correlation potential including various degrees of Hartree-Fock exchange. In several previous works, we demonstrated that the self-consistent harmonic approximation (SSCHA) developed by us the allows for accurate description of anharmonic phonon spectra in the non-perturbative limit in superconductors, ferroelectrics and charge density wave systems. We found, however, that SSCHA calculations based on total energies and force constants obtained via density functional theory with LDA/GGA functionals, substantially underestimate the ferroelectrics and charge density wave critical temperatures. This is due to the error introduced by the approximate exchange correlation functional on total energies, forces and volumes. In this proposal, we plan to apply the SSCHA using hybrid functionals as total energy and force engines. We will apply the method to bulk and single layer SnTe ferroelectrics, to the calculation of the ultralow mobility of SnSe and to the determination of TCDW in single layer NbSe2 and TiSe2. In several of these systems it is known that the inclusion of exact exchange substantially improves the description of the electronic structure and structural properties. Beside tackling anharmonicity in these systems, we will also extend the finite difference approach to the calculation of the electron-phonon interaction using hybrid functionals, a method that we pioneered during these years. In particular, we will study the effects of exact exchange on the electron phonon coupling of metallic dichalcogenides such as single layer NbSe2 and TiSe2. Our proposal will lead to a new paradigm and benchmark in the calculation of anharmonic phonon spectra of solids and in the description of the electron-phonon interaction due to the better inclusion of exchange effects.

Towards an atlas of pockets on the kinome

Project Title: Towards an atlas of pockets on the kinome
Project Leader: Sergio Decherchi
Resource Awarded:  16 million core hours on Marconi – Broadwell

Details

Team Members: Pietro Ballone, Istituto Italiano di Tecnologia, ITALYClaudio Peri, Istituto Italiano di Tecnologia, ITALYFrancesca Spyrakis, Università di Torino, ITALY

Abstract:
Protein kinases represent about 30% of all the targets of pharmaceutical interest. Kinases are involved in a number of severe diseases such as cancer, several forms of immune deficiencies, hypertension, diabetes and inflammation. Developing proper strategies to inhibit or solicit their activity, therefore, is of paramount importance for health care. The obvious strategy in inhibiting a kinase is represented by targeting the orthosteric ATP binding site, which is conserved across all the kinase group. Several drugs conceived in this way proved effective, a notable example being Dasatinib. However, it is important to consider that other proteins bind ATP. Hence, this chemotherapeutics drug has important side effects, since it is only partially selective towards its target, i.e., tyrosine kinase, whose ATP pocket is similar to those of many other proteins. Additionally the emergence of mutations in ATP binding sites often gives resistance and limits the drug usage against cancer. For those reasons, it is crucial to devise alternative therapeutic strategies where now the target pocket is no longer the orthosteric ATP site but other allosteric pockets possibly not conserved across the kinome. Our project aims at identifying different allosteric pockets together with regular emerging patterns on kinases by exploiting massive molecular dynamics (MD) simulations on the whole kinome. Molecular dynamics runs on the microsecond time scale are powerful tools to observe and diagnose the motions that represent the signature of cryptic pockets. A new method (Poketron) recently developed and published by our group will ease the post processing of long trajectories, identifying unknown pockets in kinases, and providing information on pockets communication or “cross-talk” that represent an important aspect of allosteric communication. Together with Pocketron, Machine Learning tools will be used to detect regularities. The successful completion of our plan will represent the first extensive mapping of relevant pockets on such a vast family of enzymes, and will provide a valuable tool for many other researchers/studies.

 PEROMAT – Electronic and charge transfer properties of perovskites

Project Title: PEROMAT – Electronic and charge transfer properties of perovskites
Project Leader: Kari Laasonen
Resource Awarded:  10 million core hours on Joliot Curie – KNL

Details

Team Members:

Abstract:
Perovskite semiconductors have recently attracted much attention in various technological domains, including optoelectronics and photovoltaics. In particular, solar cells based on a perovskite light-absorber have recommended themselves as an inexpensive and flexible multifunctional alternative to traditional silicon-based technologies. However, their vast implementation is slowed down by the instability issues. Modification of chemical composition, suggested to improve the stability, unavoidably affects their electronic and charge transfer properties, often resulting in non-optimal characteristics. Despite a large number of case studies, a general scheme for controllable tuning of perovskite properties is still lacking. Thus, it is essential to rationalize the mechanisms of perovskite electronic structure modulation. To approach this goal, current project will systematically investigate the impact of composition and crystal structure on the electronic and charge transfer properties of perovskites, applying the tools of computational materials science. A particular attention will be drawn to the synergetic effects of the components, fine-tuning of spin-orbit coupling, impact of organic cation species on the carrier transport and trapping characteristics, as well as excitonic effects. The outcome of this work will elucidate the atomistic-level design principles of perovskites with pre-defined electronic structure and provide a practical guidance on the targeted optimization of perovskite compositions for the experimental community.

DynamcIs – Role of protein, lipid and water dynamics in the mechanism and function of mitochondrial complex I (cI)

Project Title: NDynamcIs – Role of protein, lipid and water dynamics in the mechanism and function of mitochondrial complex I (cI)
Project Leader: Vivek Sharma  
Resource Awarded:  15 million core hours on MareNostrum

Details

Team Members: Noora Aho, University of Helsinki, FINLANDOuti Haapanen, University of Helsinki, FINLANDAapo Malkamäki, University of Helsinki, FINLANDValerio Chiarini, University of Helsinki, FINLAND

Abstract:
Life on earth is supported by a constant supply of energy in the form of ATP, which is synthesized in the mitochondria of the cell by a process called oxidative phosphorylation (OX-PHOS). In OX-PHOS, three large membrane-bound enzymes (respiratory complexes I, III and IV) couple the electron transfer reactions to the pumping of protons across the membrane. The proton electrochemical gradient established across the membrane drives the synthesis of ATP via ATP synthase. The first enzyme in the electron transport chain of mitochondria is complex I, which accepts electrons from the food-stuff (NADH) and catalyzes two-electron reduction of ubiquinone. This reaction is coupled to proton pumping against a membrane gradient of ~200 mV. The contribution of complex I to the total ATP synthesis is enormous, ca. 40 %, thereby, making it one of the key enzymes in biological energy conversion. The complex I from human mitochondria has been found to be associated with a large number of pathologies. In particular, it is known that point mutations and assembly problems in complex I cause various neurodegenerative and mitochondrial disorders, for which treatment options are rather limited. The three-dimensional structure of bacterial complex I, which is half the size of mitochondrial version, has been available for some time. However, one of the biggest breakthroughs in the mitochondrial sciences occurred this year (August 2017), when the first electron microscopy structure of entire human mitochondrial complex I was solved at a resolution of ~3.7 Å. This massive L-shaped ~1 MDa complex comprises 45 different subunits and its intricate architecture poses numerous questions; how the long range coupling (ca. 200 Å or 20 nm) between electron transfer and proton pumping is achieved, what roles do 31 accessory subunits play in the coupling and regulation of the enzyme, and how malfunction of enzyme is responsible for mitochondrial disorders. By utilizing our earlier extensive experience in the modelling and simulation of bacterial complex I, our major goal is to perform fully atomistic classical molecular dynamics simulations of entire complex I from human mitochondrion in physiological environment. The simulations will shed light on the most fundamental aspects such as dynamic electron tunnelling in complex I, long-range coupling between electron and proton transfer, and on the roles water and lipids in enzyme function. Together with our experimental collaborations, this very timely and multidisciplinary project has a breakthrough potential in opening novel routes of drug discovery against mitochondrial dysfunction.

 SHP2-ReM-MD – The regulatory mechanism of the SHP-2 protein: a Molecular Dynamics investigation.

Project Title: SHP2-ReM-MD – The regulatory mechanism of the SHP-2 protein: a Molecular Dynamics investigation
Project Leader: Gianfranco Bocchinfuso
Resource Awarded: 15 million core hours on Marconi – Broadwell

Details

Team Members: Paolo Calligari, Università di Roma Tor Vergata, ITALYAntonio Palleschi, Università di Roma Tor Vergata, ITALYLorenzo Stella, Università di Roma Tor Vergata, ITALY

Abstract:
Mutations of the gene coding for the protein SHP-2 are involved in several forms of leukemia and other cancers. In addition, they cause different developmental disorders. Our aim is to fully understand the effects of these mutations on the structure and function of the protein, and to design new molecules to contrast their consequences. SHP-2 plays a critical role in the regulation of a number of important signaling pathways in the cell and it is the first identified phosphatase for which amino acid substitutions increasing the activity were associated with the development of cancers. The SHP-2 structure includes two N-terminal Src homology 2 domains (N-SH2 and CSH2) followed by a protein tyrosine phosphatase (PTP) domain, containing the catalytic site, and a C-terminal tail. In the basal state, SHP-2 is auto-inhibited, as a loop of the N-SH2 domain is inserted into the PTP active site (“closed” conformation). SHP-2 activation is mediated by interactions between its SH2 domains with partners containing short amino acid motifs comprising a phosphortyrosine residue. This event is coupled with a rearrangement of the domains, whose final effect is a greater accessibility of the active site on the PTP domain (“open” conformation). Many pathological “gain of function” single amino acid substitutions act by favoring the “open” state. Notwithstanding the biological relevance of this process, the structure of the active state is unknown and its characterization could contribute to clarify the mechanism of action of different pathological mutations. In Work-Package 1, we will investigate the transition between the “closed” and “open” states for the wild type protein and four different pathological mutants, by molecular dynamics computer simulations. To this end, different approaches will be applied to improve the efficacy of the simulations, including Replica Exchange Molecular Dynamics simulations and calculation of the binding energy between PTP and N-SH2 domains. Different lines of evidence show that an efficient inhibition of the interaction of SHP2 with its molecular partners could be a good strategy to reduce the pathological effects of SHP-2 mutants. In Work-Package 2, we plan to design an inhibitor of these interactions by starting from a phosphopeptide (PP) able to bind the N-SH2 domain. The binding free energies of different PPs will be estimated, starting with Umbrella Sampling (US) simulations to evaluate the Potential of Mean Force (PMF) along the coordinate joining the centers of masses of PPs and N-SH2. Of note, these studies will be carried out synergistically with the experimental studies of a project funded by the Italian foundation AIRC (Italian Association for Cancer Research), thus allowing quantitative comparisons between computational and experimental data.

OPTEL2D Opto-electronic properties of 2D Transition Metal Dichalcogenides with DFT and post-DFT simulations

Project Title: OPTEL2D Opto-electronic properties of 2D Transition Metal Dichalcogenides with DFT and post-DFT simulations
Project Leader: Maurizia Palummo
Resource Awarded: 49.5 million core hours on Marconi – KNL

Details

Team Members: Claudio Attaccalite, CNRS, FRANCEConor Hogan, CNR, ITALYDaniele Varsano, CNR, ITALYGiancarlo Cicero, Politecnico di Torino, ITALYElena Cannuccia, University of Rome “Tor Vergata”, ITALYAndrea Pianetti, University of Rome “Tor Vergata”, ITALYStella Prete, University of Rome “Tor Vergata”, ITALY

Abstract:
Two-dimensional (2D) and layered materials possess novel combinations of electronic and optical properties and thus present a unique opportunity in condensed matter research and semiconductor devices. If experimental techniques to grow 2D materials on large areas, continue advancing, the new properties of these materials may enable a paradigm shift in semiconductor-based technologies leading to flexible and ultrathin next-generation electronic and opto-electronic devices. First-principles methods are playing an important role in the scientific development of this research area. Calculations can nowadays synergically complement experiments and greatly help the microscopic interpretation of physical processes, as well predict novel properties and new 2D materials. Generally speaking, 2D materials possess physical properties that are very different from conventional bulk materials, including a high sensitivity to defects and impurities, molecular functionalization, weak dielectric screening and strong light-matter interaction. In particular, the reduced dimensionality results in striking many-body effects such as gigantic transport gap renormalization and strongly bound but spatially delocalized excitons. From a theoretical/computational point of view this means that it is important to overcome the mean-field density functional theory (DFT) framework and to use post-DFT approaches, based on many-body perturbation theory (namely GW and BSE methods) in order to obtain an accurate knowledge of their electronic and optical excitations. Although graphene is still the king of the 2D realm, the absence of an intrinsic gap is problematic for several applications and the scientific attention over the world is now focusing on novel 2D materials with sizable band-gap such as the family of transition metal dichalcogenides. They have strong light-matter interaction and mobilities similar to bulk semiconductors, coupled to a robust air stability. While multi-layers have a very negligible light emission, monolayers emit light due to their direct gap. Moreover it has been shown that TMD/TMD or TMD/graphene hetero-junctions can achieve ultrahigh power densities in ultra-thin PV cells [see PI CV]. In this regard it is timely to understand how to improve the Quantum Yield (QY) of ML which, at the moment, is too low to be used in real light-emitting devices and also to study how to increase the photovoltaic efficiency in TMD heterostructures. The OPTEL2D project aims at achieving a systematic and coherent study by means of DFT and post-DFT simulations of several TMD materials. The main objectives are: i) to unveil the role of many-body effects of pristine group-VI Hexagonal-TMD with the main focus on the calculation of radiative lifetimes as well as on the analysis of fine-structure of the exciton series (dark and bright) by means of non-linear optical spectra calculations with a non-perturbative inclusion of spin-orbit interaction. ii) to investigate how many-body effects modify the opto-electronic properties of orthorombic ZT-phase of TMD and in particular to focus on the possible presence of excitonic instabilities iii) to reveal the role of atomic and molecular doping in the modulation of the opto-electronic properties of TMD monolayers iv) to engineering the opto-electronic properties of TMD heterostructures by mixing monolayers of different atomic structural phase.

 NANOMOTION – Monolayer protected gold nanoparticles, on the move: Toward the design of nanoparticles with programmed recognition abilities

Project Title: NANOMOTION – Monolayer protected gold nanoparticles, on the move: Toward the design of nanoparticles with programmed recognition abilities
Project Leader: Marco De Vivo
Resource Awarded: 36 million core hours on Marconi – KNL

Details

Team Members: Sebastian Franco, Italian Institute of Technology, ITALYJissy Kuriappan‎, Italian Institute of Technology, ITALYLaura Riccardi, Italian Institute of Technology, ITALY

Abstract:
The self-assembly of a monolayer of ligands on the surface of noble-metal nanoparticles provides a unique pathway to the realization of ordered and complex molecular structures, with innovative applications emerging in fields such as nanomedicine, chemosensing, and even catalysis (nanozymes). The molecules that form the coating monolayer are the main contributor to a nanoparticle’s functionality. Through a combined computational/experimental study, we have recently demonstrated that coating ligands form transient, protein-like binding pockets in functionalized gold nanoparticles (AuNPs). These transient pockets are responsible for the recognition of small organic molecules [Riccardi et al. Chem – Cell Press 2017]. Building upon our recent study, this project aims at progressing toward the rational design of tailored coating groups to form functionalized nanoparticles with programmed recognition ability. Therefore, we ask to access PRACE resources to characterize, with unprecedented details, the fundamental molecular mechanisms for the formation of binding pockets in the monolayer of AuNPs and, importantly, understand in depth how their formation is coupled with the presence of small analytes in solution. Moreover, we will study the underling mechanisms of nanoparticle aggregation, using a set of AuNPs bearing different coating monolayers. Results will be integrated with new experiments, towards the computational design of intelligent nanodevices with recognition abilities for small molecules such as drugs, metabolites and small molecular markers for cancer – a major advance for nanotechnology.

Acoustic surface plasmon in vicinal Au surfaces from first principle electron energy loss spectra simulations

Project Title: Acoustic surface plasmon in vicinal Au surfaces from first principle electron energy loss spectra simulations
Project Leader: Nathalie Vast
Resource Awarded: 15 million core hours on Juwels

Details

Team Members: Michèle RAYNAUD, CEA -DRF – IRAMIS, FRANCENathalie VAST, CEA -DRF – IRAMIS, FRANCEOleksandr MOTORNYI, Ecole Polytechnique, FRANCEAndrea DAL CORSO, International School for Advanced Studies (SISSA-ISAS), ITALY

Abstract:
In the past few years, acoustic surface plasmons (ASP) observed in Be (0001), Au (111) and Au (788) surfaces attracted much attention due to their unusual properties – they are able to propagate along the surface without distortion and with a much lower velocity compared to the light speed. Although their origin and properties are not yet fully understood. Untill now, first principle calculations have been performed on Au, Ag and Cu (111) surfaces and showed that indeed ASP can be predicted from ab initio time dependent density functional theory (TDDFT). However, simulations of the vicinal gold surfaces like (332) and especially (788) using traditional TDDFT methods will have extremely high computational cost. In this project we aim to perform a thorough first principle TDDFT study of acoustic surface plasmons in vicinal Au surfaces (332) and (788) using a novel approach to TDDFT – the Liouville-Lanczos recursion. In this context we plan obtain EELS spectra for these surfaces for various values of wavevector in order to identify and quantify ASP, study conditions for their excitations and compare the ASP dispersion obtained with previous experiments to the ASP dispersion predicted by TDDFT.

 DEFMAT – Defect Chemistry in Electrochemical Materials

Project Title: DEFMAT – Defect Chemistry in Electrochemical Materials
Project Leader: Ulrich Aschauer
Resource Awarded: 33 million core hours on SuperMUC

Details

Team Members:

Abstract:
Defects are always present in real materials and affect or enable functionality. This project aims to determine by density functional theory (DFT) based electronic structure calculations, the role defects such as bulk point defects and surface point and extended defects have on the functional properties. We will investigate methodological aspects such as the application of a novel self-consistent on-the-fly procedure to determine Hubbard U to account for strongly correlated transition metal d orbitals using the DFT+U method and apply it to defect-induced magnetic or ferroelectric functionality in strained perovskite thin films. We will however also investigate Li ion battery materials under operating conditions, compute experimentally accessible spectroscopic data and study the effect of surface reconstruction and dissolution on the surface chemistry of photocatalytic materials. These calculations are very relevant to experiment and all this work is carried out in close collaboration with experiment. Obtaining computational results on a timescale compatible with these experiments can however be challenging, especially since the materials we investigate are computationally expensive. For this reason, a successful collaboration with experiment requires access to large computational resources under the PRACE Tier-0 umbrella.

top

Earth System Sciences (3)

HiResNTCP: High-resolution near-term climate predictions

Project Title: HiResNTCP: High-resolution near-term climate predictions
Project Leader: Francisco Doblas-Reyes
Resource Awarded: 33 million core hours on Marenostrum

Details

Team Members:

Abstract:
The future evolution of climate in the near term, from one year to a decade or so ahead, is of significant importance to our changing society. Near-term climate prediction (NTCP) is concerned with the climate evolution in those time scales, where climate is impacted by internal low frequency variability, in addition to changes due to anthropogenic or natural forcing. NTCP, by contrast with long-term climate projections, aims to produce a skilful and reliable prediction of the future evolution of climate considering both external forcing and internal climate variability, as well as their interaction. To achieve this, NTCP systems start from estimates of the observed climate system as the model’s initial condition. The ability of current climate models to make skilful NTCP is usually tested by performing re-forecasts, ensembles of predictions that start from initial states in the past. The sample of re-forecasts can be compared with the observed evolution of the climate system over the same period, a validation that is essential if users are to have confidence in the forecasts. One of the key open science questions around NTCP is motivated by recent studies establishing that the typical atmospheric and oceanic resolutions used for global climate experimentation are a serious limiting factor to correctly reproduce both climate mean state and variability. Seasonal climate prediction experiments performed with PRACE resources have shown the positive impact of using unprecedented resolution increases in global climate models upon forecast quality and climate reproducibility. However, the impact of equivalent increases in resolution has not yet been tested in NTCP because the technical and scientific challenges have been too important until now. Taking advantage of the lessons being learned from multi-model coordinated global climate experiments at high resolution, the HiResNTCP proposal aims at investigating the potential improvement of NTCP in terms of model climate, including the reduction of the drift problem, and forecast quality associated with an unprecedented increase in global climate model resolution. The experiment proposed consists of a continuous run that will generate initial conditions for a set of 10-member ensemble re-forecasts produced over the period 1960-2018. Such an experiment would not be possible without appropriate access to tier-0 computing resources and the associated support offered by PRACE. A total of 33 Mcore-hours are requested in this proposal. An ambitious set of analyses will be performed. The project will be carried out by members of the Earth Sciences Department of the Barcelona Supercomputing Center with the EC-Earth global climate model. The data produced will be made publicly accessible via the Earth System Grid Federation data node hosted by the Department..

ReSuMPTiOn – Revealing SubMesoscale Processes and Turbulence in the Ocean

Project Title: ReSuMPTiOn – Revealing SubMesoscale Processes and Turbulence in the Ocean
Project Leader: Laurent Brodeau
Resource Awarded:  40 million core hours on MareNostrum

Details

Team Members:
Aurélie Albert, CNRS – IGE, FRANCEBernard Barnier, CNRS – IGE, FRANCEJulien Le Sommer, CNRS – IGE, FRANCEJean-Marc Molines, CNRS – IGE, FRANCEThierry Penduff, CNRS – IGE, FRANCEJacques Verron, CNRS – IGE, FRANCEAurélien Ponte, IFREMER, FRANCEJérôme Chanut, Mercator Océan, FRANCEFlorent Lyard, Observatoire Midi-Pyrénées, FRANCEBrian Arbic, University of Michigan, UNITED STATES

Abstract:
ReSuMPTiOn aims to determine the contribution of kilometric-scale oceanic flow features, known as submesoscale turbulence, in several aspects of the global oceanic circulation known for their importance in regulating the Earth’s climate. To achieve this, realistic numerical simulations of the full 3D oceanic flow in the North Atlantic basin, at ultra-high horizontal and vertical resolution, will be designed and performed using the NEMO ocean model. This type of ultrahigh resolution ocean flow simulations (~1 km horizontal grid resolution on 300 vertical levels, with the inclusion of tidal motions), has only been made achievable recently thanks to the advance in HPC capabilities of modern supercomputers. Besides providing the project team and the ocean/climate community with a pioneering dataset for studying the interactions between fine-scale features and the larger-scale components of oceanic flow, the proposed set of simulations will also be used for preparing upcoming satellite missions, including the SWOT altimetry mission. The SWOT mission will indeed provide observations of the surface ocean dynamics at an unprecedented small-scale resolution. The present project will allow to develop a system ready to prepare, assimilate, interpret, and further expand (e.g. below the surface) the information that will be obtained from the SWOT mission. This new generation of high spatial resolution altimetric satellite observations from the SWOT mission, combined to the set of proposed ground-breaking ocean simulations, is expected to lead to major breakthroughs in our understanding of the roles played by fine-scale ocean turbulence (and tidal motions) in influencing the large-scale ocean circulation, and thereby, the climate system. The simulations of this project will also be used as a testbed for designing the next generation of models for the Copernicus Marine Environment Monitoring Service.

NEWA-ProRun: New European Wind Atlas Production Run

Project Title: NEWA-ProRun: New European Wind Atlas Production Run
Project Leader: Andrea Hahmann
Resource Awarded: 56.7  million core hours on MareNostrum

Details

Team Members:
Björn Witha, Carl von Ossietzky Universität Oldenburg, GERMANYMartin Doerenkaemper, Fraunhofer IWES, GERMANYArnau Folch, Barcelona Supercomputing Center, SPAINJorge Navarro, CIEMAT, SPAINJesús Fidel González Rouco, Universidad Complutense de Madrid, SPAINTija Sile, University of Latvia, LATVIAStefan Söderberg, WeatherTech Scandinavia AB, SWEDEN

Abstract:
The New European Wind Atlas (www.neweuropeanwindatlas.eu) Project is funded under the European Commission’s 7th Framework Programme ERA-Net Plus that comprises nine funding agencies from eight EU Member States. The project aims to create a new and detailed European wind resource atlas. The NEWA project is developing a new reference methodology for wind resource assessment and wind turbine site suitability based on a mesoscale-to-microscale model-chain. This new method will produce a more reliable wind characterization than current models, leading to a significant reduction of uncertainties on wind energy production and wind conditions that affect the design of wind turbines. The NEWA-ProRun (Production Run) project is the most ambitious Wind Atlas mesoscale model run made to date on a worldwide level. This Wind Atlas will have a very high horizontal resolution that spans over climatological timescales. The mesoscale simulation will be performed with a mesoscale model along with a novel strategy that optimizes its parallelization. Also, we will use a state of the art model setup that has been optimized for creating wind atlases. The NEWA-ProRun also includes a unique uncertainty map based on a multiphysics ensemble of mesoscale model simulations. Therefore, the requested allocation of computational resources includes: (1) an unprecedented long simulation, extending over 30 years that covers all of Europe at a grid spacing of 3 km x 3 km, and (2) a series of shorter ensemble simulations to estimate the uncertainty of the mesoscale runs.

top

Engineering (8)

TRAAD – TRansonic Airfoil AeroDynamics

Project Title: TRAAD – TRansonic Airfoil AeroDynamics
Project Leader: Prof Neil Sandham
Resource Awarded: 35  million core hours on Hazel Hen

Details

Team Members:
Florian Hammer, University of Southampton, UNITED KINGDOMSatya Jammy, University of Southampton, UNITED KINGDOMJae Wook Kim, University of Southampton, UNITED KINGDOMDavid Lusher, University of Southampton, UNITED KINGDOMMarkus Zauner, University of Southampton, UNITED KINGDOM

Abstract:
As computers increase in performance, numerical simulation of transonic flow over wings from first principles, without relying on turbulence modelling, is becoming feasible at Reynolds numbers high enough to capture the essential flow physics. In this project we will simulate transonic flow over a laminar-flow wing, which is designed to maintain a slightly favourable pressure gradient to deliver laminar flow up to the shock interaction on the upper surface. This is representative of next-generation aircraft wing profiles, which aim to reduce both friction and wave drag. Direct numerical simulations (DNS) will be run at a Reynolds number of half a million, which is high enough, based on preliminary simulations, to capture shock-wave/boundary layer interactions leading to transition as well as the appearance of incipient buffet, an unsteady phenomenon that limits the flight envelope of commercial aircraft. We will use the DNS to validate large eddy simulations (LES) and then apply LES in much wider computational domains. From the simulations will be able to elaborate: (i) multi-time scale phenomena from Kolmogorov, via Kelvin-Helmholtz, up to buffet, (ii) the nature of buffet, in particular two-dimensional versus three-dimensional cell behaviour, and (iii) near-field noise sources. The simulations will provide much-needed databases to study shock-wave/boundary-layer interactions for a complete lifting surface. They are also an essential preparation for simulations at Reynolds numbers of the order of one to ten million, as we look forward to exascale machines.

 FWING – Future smart Wing design

Project Title: FWING – Future smart Wing design
Project Leader: Marianna  Braza
Resource Awarded: 35  million core hours on Hazel Hen

Details

Team Members: JAN-BERNHARD VOS, CFS Engineering, SWITZERLANDAlain Dervieux, Sophia Antipolis – Méditerranée, FRANCEYannick Hoarau, University of Strasbourg, FRANCEMichalis Fragiadakis, National Technical University of Athens, GREECEAlistair Revell, The University of Manchester, UNITED KINGDOM

Abstract:
The project FWING – “Future smart Wing design” aims at providing an innovative design for the wings of future aircrafts. Able to be both deformed and submitted to optimal vibrations by means of electroactive smart actuators (placed under the “skin” of the lifting surface) these new designs have been under the scope of our team’s research over the past years. The CFD-SM (Computational Fluid Dynamics – Structural Mechanics) strong coupling approaches will take into account the properties of the smart materials already used in experimental studies in which tests were performed aiming to define a range of optimal deformations and trailing-edge vibrations of the high-lift flap of an Airbus wing (A3xx type). The present proposal aims at creating a numerical environment for testing morphing concepts, aiming at the optimal smart actuating behaviour in order to achieve smart wing design through turbulence control. To this end, hybrid moprhing concepts utilizing the effect of multiple types of smart actuators acting at different time scales (Shape Memory Alloys for low-frequency/high deformation – new generation of piezo-actuators for higher-frequency/low deformation actuation) are of essence. The goal, a simultaneous increase of the aerodynamic performance in all phases of the flight (low subsonic for take-off and landing and transonic regimes in cruise), a significant increase of lift, decrease of drag and noise reduction. Part of the morphing concepts studied are bio-inspired from large span hunting birds among others. An overview of our approaches in collaboration with Airbus can be found in the film dedicated by the CNRS Journal to our studies in collaboration with the LAPLACE Laboratory. *The present computational project is part of the SMS “Smart Morphing and Sensing” (www.smartwing.org) coordinated by Dr. Marianna Braza, project leader of the present PRACE proposal. In this context, advanced and innovative electroactive morphing prototypes are created and the design will be greatly enabled by the High-Fidelity CFD-SM methods involved in the present PRACE – FWING proposal. The numerical code NSMB –Navier Stokes Multi Block in which the Project leader’s group has developed advanced turbulence modelling approaches for high-Reynolds number subsonic and transonic flows, has been proven to be an excellent tool able to accompany the needs of the team’s projects. Efficiently coupled with a structural solver at our disposal, it will provide the capacity to perform simulations involving the modelling and the behaviour laws describing the electroactive materials.

MOST-SEA The mechanics of sediment transport under sea waves

Project Title: MOST-SEA The mechanics of sediment transport under sea waves
Project Leader: Giovanna Vittori
Resource Awarded: 30  million core hours on Marconi – KNL

Details

Team Members: Marco Mazzuoli, University of Genoa, ITALY

Abstract:
In the engineering practice, the initiation of sediment motion at the bottom of sea waves and the pick-up rate of the sediment from the bottom is computed in terms of the sediment characteristics and of averaged flow quantities. Many important effects, such as the interaction of the sediment grains with the turbulent structures which form in the boundary layer at the sea bottom and the effects of the unsteadiness of the flow are not accurately described by the currently used models. Moreover, existing models often provide poor predictions when compared to the measurements. In the present project we plan to perform four direct numerical simulations of the oscillatory boundary layer that forms over a movable bed made of 30 layers of sediments. The results, obtained for two different values of the Reynolds number and for three values of the diameter of the spheres, will allow us to investigate in detail the mechanics of the sediments and their interaction with the flow. The final aim of the investigation is to provide indications to develop a physically-based method to predict sediment transport at the bottom of sea waves.

DILPART

Project Title: DILPART
Project Leader: Francesco Picano
Resource Awarded: 32  million core hours on Marconi – KNL

Details

Team Members: Federico Dalla Barba, University of Padova, ITALYLuca Brandt, Royal Institute of Technology, SWEDENPedro Costa, Royal Institute of Technology, SWEDEN

Abstract:
Turbulent suspensions of solid particles in viscous fluids are often found in several contexts, such as biological and environmental flows, and industrial processing. Despite this, our current understanding and prediction ability are limited by experimental difficulties while fully-resolved simulations are only possible in simple configurations. Different simulation models have been proposed to predict multiphase flows, depending on the typical particle length scales, and the global volume or mass fractions. For suspensions of small particles in dilute to semi-dilute conditions, the so-called 1-way and 2-way coupling point-particle approximation is often applied. In this case, each particle is treated as a material point, subjected to a drag force determined by the local fluid and particle velocity. This approximation has been widely used to study particle transport and turbulence modulation. However, results from the point-particle model are debated and not always reliable, with several quantitative discrepancies between numerical and experimental data. This is observed in basic quantities like settling velocities, clustering intensity, particle accelerations, turbulence modulation, and even drag in turbulent channel flows. This raises the following scientific questions: How accurate are point-particle approaches? And how can we improve these models, anyway required for large-scale applications? The development of high-fidelity numerical techniques, such as Immersed Boundary Methods (IBM), together with the continuous increase in computing power, make a first-principles approach of the problem feasible under certain conditions. Particle-resolved simulations, where the local flow around each particle is simulated, have been recently proved to be crucial e.g. for understanding the dynamics of turbulent dense suspensions. The main aim of this project is to answer the scientific questions above by simulating, for the first time, turbulent flows laden with small particles (smaller than flow length scales) in very dilute (1-way), dilute (2-way) and semi-dilute (2-way/4-way) conditions, resolving the flow at the particle-fluid interface. These simulations can then be directly compared to the data of an equivalent point-particle simulation, forming the basis of future modeling efforts. Hence the simulations will provide a state-of-the-art dataset that will be an important milestone for the development of new models that can accurately predict particulate flow, without resolving the particle-fluid interface.

ROHITU – Roughness in highly turbulent Taylor-Couette and Rayleigh-Bénard flows

Project Title: ROHITU – Roughness in highly turbulent Taylor-Couette and Rayleigh-Bénard flows.
Project Leader: Detlef Lohse
Resource Awarded: 15  million core hours on MareNostrum

Details

Team Members: Vamsi Arza, University of Twente, NETHERLANDSMartin Assen, University of Twente, NETHERLANDSPieter Berghout, University of Twente, NETHERLANDSAlexander Blass, University of Twente, NETHERLANDSChong Shen Ng, University of Twente, NETHERLANDSRichard Stevens, University of Twente, NETHERLANDSRoberto Verzicco, University of Twente, NETHERLANDSXiaojue Zhu, University of Twente, NETHERLANDS

Abstract:
Many wall bounded flows in nature and technology are affected by surface roughness of the wall. In some cases this has adverse effects, e.g. drag increase leading to higher fuel costs, in others it is beneficial like for mixing enhancement or transfer properties. Computationally, it is notoriously difficult to simulate these flows, because of the large scale separation of highly turbulent flows and the special handling of irregular boundaries. From a physics perspective, many questions are still unanswered, one of the most urgent ones being the effects of roughness topology on local and global flow features. In this proposal, we aim at investigating three specific points which will contribute to a better understanding of the relationship between roughness topology and the flow dynamics. First, we plan to investigate the effect of three surface characterization parameters, the mean surface height in the transitionally rough regime, the mean streamwise gradient of the rough surface and the fractal dimension of the rough surface. By a superposition of different rough elements with various length scales, we can systematically study the effect of the fractal dimension on the fluid flow. Second, we plan to investigate the transition from the smooth wall flow towards the transitionally rough regime. Third, we plan to investigate the effect of roughness on the large scale structures. The studies will be carried out in two paradigmatic and complementary systems in turbulence research, Taylor-Couette (TC) and Rayleigh-Bénard (RB) flow. In particular we wish to study the effect of wall roughness on the flow structures in the vicinity of the wall, higher order statistics, energy budgets and large-scale flow features (e.g. plume ejection, Taylor rolls and the large-scale circulation). We aim at gaining detailed insight into the mechanisms underlying a change in drag (TC) or enhanced heat transfer (RB), distinctive to turbulent flows over rough surfaces. Direct Numerical Simulations (DNS) are carried out with the “AFiD” code (www.afid.eu). Wall roughness is implemented by means of the Immersed Boundary Method (IBM).

CPCROR – Characterization of the pressure fluctuation developing over an airfoil and a 3D CROR blade for the prediction of trailing edge noise

Project Title: CPCROR – Characterization of the pressure fluctuation developing over an airfoil and a 3D CROR blade for the prediction of trailing edge noise
Project Leader: Bodart Julien
Resource Awarded: 16.5  million core hours on MareNostrum

Details

Team Members: Thibault Bridel Bethomeu, ISAE-SUPAERO, FRANCEGonzalo Saez-Mischlich, ISAE-SUPAERO, FRANCE

Abstract:
This proposal addresses the fundamental characterization of the pressure fluctuation in turbulent boundary layers developing over an airfoil and a rotating CROR blade. It aims at extracting relevant quantities to feed reduced order models used to predict the so-called trailing edge noise, known to significantly contribute to the total noise emitted by a CROR engine. This project is part of a broader effort, which brings together a large number of partners, including industrial partners (AIRBUS), aeroacoustic modeling experts (LMFA, Université de Lyon) and wind-tunnel measurement experts (DNW-TWG windtunnel operated by NLR & DLR). This research plan is funded through two dedicated CLEANSKY-2 projects, i) SCONE that addresses the numerical simulation and modeling part and ii) CRORTET in which large experimental facilities are in use. In this project, we specifically target three geometries, which will provide an unprecedented database to the project members but also to the aeroacoustic community, with open access to the results obtained with the widely studied Valeo CD airfoil at industrially relevant flow regimes. This work leverages a first set of preliminary studies aiming at evaluating the optimal grid resolutions, domain sizes, and relevant flow regimes in order to fully capture the statistical quantities associated with the pressure field.

VIVALdI-HPC of Vortex Induced VibrAtions for flow controL and energy harvestIng

Project Title: VIVALdI-HPC of Vortex Induced VibrAtions for flow controL and energy harvestIng.
Project Leader: Oriol  Lehmkuhl
Resource Awarded: 27  million core hours on SuperMUC

Details

Team Members: Juan Carlos Cajas, Barcelona Supercomputing Center (BSC-CNS), SPAINGuillaume Houzeaux, Barcelona Supercomputing Center (BSC-CNS), SPAINDaniel Pastrana, Barcelona Supercomputing Center (BSC-CNS), SPAINIvette Rodriguez, Universitat Politècnica de Catalunya, SPAIN

Abstract:
Geophysical flows from wind to oceanic currents represent a clean source of energy widely available. Structures based on vortex-induced vibrations are one of the mechanisms for harvesting part of this energy in the range of frequencies where flow-induced vibrations originates a strong coupling between the oscillating body and the fluctuating wake. In the present project high fidelity simulations of a cylindrical body oscillating in a free-stream from sub-critical to super-critical Reynolds numbers will be carried out for first time by means of wall-resolved LES. The project aims at shedding light in the interactions of the cylinder with the fluid and specially with the boundary layer, but also the characterisation of the wake topology. The project focus on study the influence of the Reynolds number and the range of frequencies in which vortex induced vibrations reinforce the aerodynamic forces on the cylinder. In addition to this, for the cases corresponding to the critical and supercritical Reynolds numbers were the wake width is reduced as a consequence of the instabilities in the boundary layer, localised roughness or a trip wire is added to the cylinder surface in the location where boundary layer is forced to separate from the cylinder so as to passively control flow separation and transition to turbulence to widen the wake. The main idea behind the use of these passive devices is to maximise the ratio of the amplitude of oscillations to the cylinder diameter (A/D). This is the first time this kind of simulations are performed at this level of modelisation, being a step forward in the understanding of the physics of fluid-structure interaction in the range of industrial applications.

Large Eddy Simulations of Compressor Turbulence

Project Title: Large Eddy Simulations of Compressor Turbulence
Project Leader: Luca di Mare
Resource Awarded: 30  million core hours on Marconi – KNL

Details

Team Members:

Abstract:
The project is aimed at conducting Large Eddy Simulations (LES) of turbulent flow in gas turbine compressor passages. The geometry to be simulated represents the stator passages in a research compressor rig in the Whittle laboratory at the University of Cambridge. Computations will be performed on sectors of the annulus comprising between one and four passages. The domain sizes will vary between 170 and 700 million cells and will require between 110 and 220 nodes on the MareNostrum IV cluster. The computations are intended to study the computational performance of the proprietary multi-block structured solver, AU3X, and to characterise the interaction between passage and wall turbulence and periodic flow phenomena, for example, wakes generated by upstream blade rows. The data generated by this study will also be used to build better steady-state computational fluid dynamics models for turbulent flows in turbomachinery compressors. The project is part of a long-term numerical and experimental investigation into turbulent flow in gas turbine compressor passages that has been carried out at the University of Oxford, Imperial College London and the University of Cambridge.

top

Fundamental Constituents of Matter (8)

Interdisciplinary simulations of Fermi superfluids far from equilibrium

Project Title: Interdisciplinary simulations of Fermi superfluids far from equilibrium
Project Leader: Gabriel Wlazłowski
Resource Awarded: 87.2 million core hours on Piz Daint

Details

Team Members: Marco Antonelli, Polish Academy of Sciences, POLANDBrynmor Haskell, Polish Academy of Sciences, POLANDVadym Khomenko, Polish Academy of Sciences, POLANDMaciej Marchwiany, University of Warsaw, POLANDJanina Grineviciute, Warsaw University of Technology, POLANDKonrad Kobuszewski, Warsaw University of Technology, POLANDPiotr Magierski, Warsaw University of Technology, POLANDJanusz Oleniacz, Warsaw University of Technology, POLANDBuğra Tüzemen, Warsaw University of Technology, POLANDNicolas Schunck, Lawrence Livermore National Laboratory, UNITED STATESIonel Stetcu, Los Alamos National Laboratory, UNITED STATESKenneth Roche, Pacific Northwest National Laboratory, UNITED STATESAurel Bulgac, University of Washington, UNITED STATESShi Jin, University of Washington, UNITED STATESKazuyuki Sekizawa, University of Washington, UNITED

Abstract:
Superfluidity is a generic feature of various quantum systems at low temperatures. It has been experimentally confirmed in many condensed matter systems, in 3He and 4He liquids, in nuclear systems including nuclei and neutron stars, in both fermionic and bosonic cold atoms in traps, and it is also predicted to show up in dense quark matter. Once the physical system undergoes phase transition and becomes superfluid its dynamical properties change dramatically. For example, any rotational motion can be sustained by quantum vortices only, which can spin forever since the superfluids have zero viscosity. The superfluid properties manifest differently in various systems. In heavy nuclei, the smallest superfluid systems in nature, superfluidity works as a lubricant and allows them to fission into smaller fragments. In the case of the opposite process – fusion of two nuclei – it is predicted that it hinders the formation of a heavy nucleus. In old neutron stars, superfluid interior is responsible for observable effect known as neutron star glitch – a sudden decrease of rotation period of the star. Finally, in rotating superfluids quantized vortices form regular lattices, which if excited, become “tangled” like spaghetti, giving rise to a chaotic motion known as quantum turbulence. Although, all of these effects may appear as rather distinct, there exists a common theoretical framework, providing accurate description of these phenomena – a fully microscopic method based on Time-Dependent Density Functional Theory, extended to superfluid systems. Within this project, we aim at ambitious goal of solving fundamental problems of non-equilibrium quantum many-body dynamics in superfluid Fermi systems related to: 1) quantum turbulence in spin-polarized unitary Fermi gas, 2) induced fission and quasi-fission of atomic nuclei and 3) superfluid dynamics in the neutron star crust. All these phenomena are of great scientific importance and their description based on microscopic framework pose a long standing challenge. It is partly due to the computational cost – a fully microscopic description requires to solve about a million of 3D, complex, coupled, nonlinear partial differential equations that has become possible to handle only recently with top-tier supercomputers. The unified framework will be applied by a team consisting of experts in condensed matter, nuclear physics, astrophysics and HPC. The European top-tier supercomputers will allow to make a qualitative leap in the description and understanding of these phenomena leading possibly to breakthroughs in a number of topics of great interest related to ultracold atomic gases, low-energy nuclear physics and physics of neutron stars. Moreover, the studies will provide an important feedback concerning the range of applicability of methods based on density functionals in the context of superfluid fermionic systems.

EFluctQCD – Electric charge fluctuations in strongly interacting matter

Project Title: EFluctQCD – Electric charge fluctuations in strongly interacting matter
Project Leader: Christian Schmidt
Resource Awarded: 55 million core hours on Marconi – KNL

Details

Team Members: Sayantan Sharma, Johannes Gutenberg Universität Mainz, GERMANYLorenzo Dini, Universitaet Bielefeld, GERMANYFrithjof Karsch, Universitaet Bielefeld, GERMANYEdwin Laermann, Universitaet Bielefeld, GERMANYHauke Sandmeyer, Universitaet Bielefeld, GERMANYSwagato Mukherjee, Brookhaven National Laboratory, UNITED STATESPatrick Steinbrecher, Brookhaven National Laboratory, UNITED STATES

Abstract:
A central goal of the physics program pursued with ultra-relativistic heavy ion beams at hadron colliders such as the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory (BNL) in the USA and the Large Hadron Collider at CERN, Geneva, is to map out the phase diagram of strong interaction matter. Higher order moments of conserved charge fluctuations provide a sensitive tool for detecting the boundary between the low and high temperature phases of strongly interacting matter in heavy ion experiments as well as in numerical calculations performed in the framework of lattice QCD. We propose here to study higher order moments of electric charge fluctuations at vanishing and non-vanishing values of the baryon chemical potential. This extends our earlier studies of fluctuations of net baryon-number fluctuations. However, studies of electric charge fluctuations are far more demanding as they are dominated by contributions arising from the light pion sector of the theory. Typical problems of lattice QCD calculations arising from finite volume and finite cut-off effects are much more pronounced in this case. The calculations proposed here shall be performed on lattices of size 643×16. This will extend existing calculations on this size lattices and, in combination with existing results on smaller lattices we will be able to extract continuum extrapolated results in a wide range of temperatures.

MIMOSA HPC Matter Irradiating by beaMs with Orbital angular momentum: Simulations and Applications, HPC

Project Title: MIMOSA HPC Matter Irradiating by beaMs with Orbital angular momentum: Simulations and Applications, HPC
Project Leader: Rachel Nuter 
Resource Awarded: 15.4 million core hours on Joliot Curie – KNL

Details

Team Members: Stefan Skupin, Université de Lyon, FRANCEMichael GRECH, Universite Pierre et Marie Curie, FRANCERomain Beuton, University of Bordeaux, FRANCEDavid Blackman, University of Bordeaux, FRANCEBenoit Chimier, University of Bordeaux, FRANCEGuillaume Duchateau, University of Bordeaux, FRANCEPedro Gonzalez de Alaiza Martinez, University of Bordeaux, FRANCEPhilip Korneev, National Research Nuclear university, RUSSIAN FEDERATIONIllia Thiele, Chalmers University, SWEDEN

Abstract:
The goal of the MIMOSA project is to study the interaction of intense structured laser beams with matter in order to achieve an active control of such processes as acceleration of charged particles, generation of TeraHertz (THz), controlled plasma jet formation for astrophysics-related studies and laser-induced material modification. We will perform extensive numerical experiments on laser-matter interaction with intense laser beams carrying Orbital Angular Momentum (OAM), develop theoretical models and propose, at the end of the project, a set of relevant experiments. The originality of this project consists in studying material modifications induced by OAM laser pulses, generation of electron vortices and localized magnetic fields, optical control of the divergence of energetic proton beams, generation of directed electromagnetic pulses in THz domain. The intense OAM laser pulses (illustrated in Figure 1) are not characterized with a cylindrical symetry; their modelisation is only accessible with 3D codes. These challenging studies require 3D numerical simulations modeling the interaction between matter and structured laser beams. These simulations are time and memory consuming. They can only be launched with modern and highly parallelized codes. We have developped in CELIA and in ILM such numerical tools, that we routinely used in Tier 1 in their 2D geometry. This MIMOSA HPC project will enable us to routinely launch 3D simulations in Tier0 system and to achieve our challenging studies.

Eulerian and Lagrangian Plasma Simulations of Kinetic Turbulence ELPS

Project Title: Eulerian and Lagrangian Plasma Simulations of Kinetic Turbulence ELPS
Project Leader: Francesco Califano
Resource Awarded: 60 million core hours on Marconi – KNL

Details

Team Members: Matteo Faganello, Aix-Marseille University, FRANCEDimitri Laveder, Observatoire de la Côte d, FRANCECarlo Cavazzoni, CINECA, ITALYSilvio Sergio Cerri, Princeton University, UNITED STATESMatthew Kunz, Princeton University, UNITED STATES

Abstract:
Turbulence and magnetic reconnection in weakly collisional plasmas are frontier problems in plasma (astro)physics research and present formidable challenges in computational physics. These intimately connected physical processes are currently the targets of numerous theoretical efforts and satellite space missions. From a physical point of view, the great complexity and difficulty in the study of turbulence in weakly collisional plasmas stems from the coupling of a wide range of disparate phenomena at different scales and frequencies, ranging from the large-scale “fluid” dynamics to small-scale kinetic processes, that give rise to strongly nonlinear, multi-scale, multi-physics dynamics. To model and investigate these processes, the six-dimensional phase-space evolution of the particle distribution function must be self-consistently integrated in time alongside the Maxwell equations that govern the evolution of the electromagnetic fields. The solution of this system of equations necessitates an enormous investment of computational resources (CPU time, memory) and a suite of sophisticated algorithms and diagnostics designed to elucidate the dynamics within a very non-intuitive high-dimensional space. In this project, we propose a new synergistic approach that leverages the individual advantages of two different but complementary state-of-the-art numerical codes to span four decades in wavenumber of the energy spectrum. Our scientific objectives will be i) to quantify how energy is redistributed into waves, coherent structures, (kinetic) instabilities and particle acceleration; ii) to highlight the reconnection of magnetic fields as the main driver of sub-ion scales turbulent fluctuations and iii) to study the back reaction on the system due to the loss of local thermodynamic equilibrium. The requested PRACE allocation will enable a groundbreaking, first-principles study of kinetic turbulence in weakly collisional, magnetized plasmas, such as those found in the solar wind and in other hot, dilute astrophysical environments. We emphasize that our proposed study is likely to challenge the current paradigm of magneto-kinetic turbulence, which posits a local scale-by-scale phase-space cascade of free energy, and will impact our understanding of turbulence, magnetic self-organization, and plasma microphysics in a wide variety of systems.

SOLeCORE — Self-Organisation from SOL to CORE in Magnetised Turbulent Plasmas

Project Title: SOLeCORE — Self-Organisation from SOL to CORE in Magnetised Turbulent Plasmas
Project Leader: Guilhem Dif-Pradalier
Resource Awarded: 40 million core hours on MareNostrum

Details

Team Members: Elisabetta Caschera, Commissariat à l, FRANCEVirginie Grandgirard, Commissariat à l, FRANCEGuillaume Latu, Commissariat à l, FRANCEChantal Passeron, Commissariat à l, FRANCEYanick Sarazin, Commissariat à l, FRANCE

Abstract:
With the development of large machines like the ITER tokamak, controlled fusion makes a huge step forward towards mastering the energy of the stars for a civil usage. The steady international progress regarding the achieved fusion performance relies on our ability to understand and predict the confinement properties of the plasma. Turbulent transport is the key player in this matter. The advent of High Performance Computing has allowed to start addressing this issue–one of the few “great unsolved problems” in physics and mathematics–by means of first-principle simulations. State-of-the-art models use the 5-dimensional gyrokinetic description, as required by the low collisionality of such hot and dilute plasmas. This international and strongly competitive activity has already led to critical breakthroughs, especially regarding turbulence self-organisation and regulation via large scale flows. The central challenge is now to provide a unified view of turbulence properties in the experimentally relevant flux-driven regime, and when multiple scales and disparate regions of the plasma are self-consistently modelled. Especially, the outer region is long known to play a critical role as it stores a significant fraction of the overall kinetic energy content in experimental discharges. Yet, it is hardly addressed in numerical simulations because of numerical and physics bottlenecks. The present 3 year project addresses groundbreaking issues when edge and core plasma interplay. It builds upon our long standing effort to develop the unique gyrokinetic code GYSELA, which is now mature enough and contains the necessary ingredients to address this problem as a whole. The backbone of this proposal is the interplay between edge and core turbulence, with the final goal to make reliable predictions regarding heat confinement time in tokamak plasmas. Understanding turbulence self-organisation from micro to meso- and even macro-scales when multiple instability mechanisms coexist, and when the confined plasma core and the highly fluctuating edge–including the so-called scrape-off layer with open magnetic field lines–interact dynamically, is the target. Three frontier problems will be addressed: (i) understanding and predicting the interaction between edge and core turbulences, with special emphasis on the dynamics of large scale flows and comparison to experiments; (ii) investigating transient dynamics, i.e. cold and hot bursts, so as to probe quantitatively and qualitatively nonlocal interactions, and to assess the resilience of the edge to perturbations; (iii) elucidating the possible role of the edge in the puzzling “isotope effect”, corresponding to the observed degradation of confinement from Deuterium to Hydrogen plasmas, while common scaling laws predict an opposite trend. The PI has strong expertise in this field, will be fully committed to this activity in the next three years, as part of his implication in several European projects which share the same ultimate goal. We thus believe our present proposal to be both of great general interest and extremely timely.

HFlavLat – Precision Heavy Flavour Physics from Lattice QCD

Project Title: HFlavLat – Precision Heavy Flavour Physics from Lattice QCD
Project Leader: Carlos Pena
Resource Awarded: 22.9 million core hours on MareNostrum

Details

Team Members: José Ángel Romero, Consejo Superior de Investigaciones Científicas, SPAINAndrea Bussone, Universidad Autónoma de Madrid, SPAINGregorio Herdoiza, Universidad Autónoma de Madrid, SPAINJavier Ugarrio, Universidad Autónoma de Madrid, SPAINDavid Preti, Università di Torino, ITALY

Abstract:
The aim of this project is to improve state-of-the-art results for the leptonic and semi-leptonic decay amplitudes of charmed mesons from Lattice QCD, in view of the new generation of experimental results coming from dedicated facilities. Specifically, we aim at decreasing the uncertainty in the determination of |V_{cd}| and |V_{cs}|, and to do first studies of leptonic decays of charmonium states. Furthermore, a first step towards direct computations in B physics using relativistic b quarks on the lattice will be made – a crucial task in view of the existing tensions in the B sector of the Standard Model and the expected availability of more precise experimental results in the next few years. Accurate control of several relevant systematic uncertainties will be achieved by using existing N_f=2+1 ensembles at very fine lattice spacings, and an action with optimal scaling properties for the determination of observables.

The chiral limit in (2+1)-flavor QCD

Project Title: The chiral limit in (2+1)-flavor QCD
Project Leader: Dr. Olaf Kaczmarek
Resource Awarded: 75 million core hours on Piz Daint

Details

Team Members: Heng-Tong Ding, Central China Normal University, CHINASheng-Tai Li, Central China Normal University, CHINAJishnu Goswami, University of Bielefeld, GERMANYFrithjof Karsch, University of Bielefeld, GERMANYAnirban Lahiri, University of Bielefeld, GERMANYLukas Mazur, University of Bielefeld, GERMANYHauke Sandmeyer, University of Bielefeld, GERMANY

Abstract:

The goal of this proposed project is to understand the nature of the chiral transition in (2+1)-flavor QCD and to determine the chiral transition line in the QCD phase diagram at small to moderate values for the light quark chemical potential. By utilizing Highly Improved Staggered Quarks (HISQ) on Nt=12 lattices the project improves on earlier studies through considerably reduced lattice cut-off effects, notably taste violations. In combination with previous results on Nt=6 and Nt=8 lattices it allows for reliable continuum as well as chiral extrapolations of the results. The project aims at establishing scaling behavior much closer to the continuum limit than before, with consequences for improving the understanding of the nature of the chiral transi-tion. By analyzing the scaling behavior of the chiral susceptibility in more detail, in particular by disentangling and subtracting the UV divergence of the connected part, this will allow to consolidate the universality class of the chiral transition and its critical light quark mass. These calculations would further constrain and reduce the errors of the non-universal scaling parameters, which are needed in the determination of the curvature of the chiral transition line. The chiral transition line shall, furthermore, be set in relation to the freeze-out curve to facilitate the interpretation of present and future experimental results on fluctuations and correlations of conserved quantum numbers, in particular in view of existence and location of the critical end point in the QCD phase diagram. This critical endpoint, if it exists, is likely to be connected to the (2+1)-flavor chiral phase transition at zero chemical potential. In fact, in the chiral limit the critical end point will be a tri-critical point at which the line of second order chiral transitions turns into a line of first order transitions.

Large-scale SUSY phenomenology with GAMBIT

Project Title: Large-scale SUSY phenomenology with GAMBIT
Project Leader:  Pat Scott
Resource Awarded: 42 million core hours on Marconi – KNL

Details

Team Members: Peter Athron, Monash University, AUSTRALIACsaba Balazs, Monash University, AUSTRALIAAndrew Fowlie, Monash University, AUSTRALIAPaul Jackson, The University of Adelaide, AUSTRALIAMartin White, The University of Adelaide, AUSTRALIAJonathan Cornell, McGill University, CANADAMarcin Chrząszcz, CERN, SWITZERLANDNicola Serra, Universität Zürich, SWITZERLANDSebastian Wild, Deutsches Elektronen-Synchrotron (DESY), GERMANYFlorian Bernlochner, Karlsruhe Institute of Technology (KIT), GERMANYFelix Kahlhoefer, RWTH Aachen University, GERMANYRoberto Ruiz de Austri, University of Valencia, SPAINFarvah Mahmoudi, Université Lyon 1, FRANCEJulia Harz, Université Pierre et Marie Curie, FRANCESuraj Krishnamurthy, The University of Amsterdam, NETHERLANDSChristoph Weniger, The University of Amsterdam, NETHERLANDSTorsten Bringmann, University of Oslo, NORWAYTomas Gonzalo, University of Oslo, NORWAYAnders Kvellestad, University of Oslo, NORWAYAre Raklev, University of Oslo, NORWAYJan Conrad, Stockholm University, SWEDENJoakim Edsjö, Stockholm University, SWEDENSanjay Bloor, Imperial College London, UNITED KINGDOMBenjamin Farmer, Imperial College London, UNITED KINGDOMSebastian Hoof, Imperial College London, UNITED KINGDOMJames McKay, Imperial College London, UNITED KINGDOMRoberto Trotta, Imperial College London, UNITED KINGDOMAaron Vincent, Imperial College London, UNITED KINGDOMAndy Buckley, University of Glasgow, UNITED KINGDOMGregory Martinez, University of California Los Angeles, UNITED STATESChristopher Rogan, University of Kansas, UNITED STATES

Abstract:

The Global and Modular Beyond-the-Standard Model Inference Tool (GAMBIT) is a project aimed at producing the most rigorous analyses and comparisons possible of theories for particle physics theories Beyond the Standard Model. It achieves this by combining the latest experimental results from dark matter searches, high-energy collider experiments such as the LHC, flavour physics, cosmology and neutrino physics. It then compares these results to the most accurate theoretical predictions of cross-sections, particle masses, scattering and decay rates, cosmic ray fluxes and neutrino oscillations using cutting-edge statistical methods, in order to produce the most up-to-date and complete picture of the search for dark matter and new physics possible. The GAMBIT codebase has been developed over a period of five years by a team of 30 experimentalists, theorists, statisticians and computer scientists, working in very close collaboration. It draws on the expertise of members of nearly all of the leading particle and astroparticle experiments around the world, as well as many of the leading pieces of software in the field. To date, GAMBIT has led to three landmark physics papers [1-3]. Two of these [2,3] have focused on supersymmetry, arguably the most promising theoretical framework for explaining dark matter and predicting the existence of other new particles. Due to computational constraints however, the most extensive of these analyses was able to explore just 7 of the 25 most interesting parameters of this framework. We are currently carrying out work on a 9-parameter version on a Tier 1 facility. The power of the PRACE Tier 0 infrastructure will allow us to expand our investigations to 11, 13 and 15-parameter versions, moving us closer to the ultimate goal of eventually exploring all 25 parameters. References: [1] GAMBIT Collaboration: P. Athron, et al. Status of the scalar singlet dark matter model, EPJC in press [arXiv:1705.07931] [2] GAMBIT Collaboration: P. Athron, et al. Global fits of GUT-scale SUSY models with GAMBIT, EPJC in press [arXiv:1705.07935] [3] GAMBIT Collaboration: P. Athron, et al. A global fit of the MSSM with GAMBIT, EPJC in press [arXiv:1705.07917].

top

Mathematics and Computer Sciences (2)

MultiSoot-EL – Particle tracking for soot size distribution predictions in LES of gas turbines

Project Title: MultiSoot-EL – Particle tracking for soot size distribution predictions in LES of gas turbines
Project Leader: Eleonore Riber 
Resource Awarded: 15 million core hours on Joliot Curie – KNL

Details

Team Members: Bénédicte Cuenot, CERFACS, FRANCELucien Gallen, CERFACS, FRANCEOlivier Vermorel, CERFACS, FRANCE

Abstract:
Expected stringent particulate matter (PM) emission legislation for gas turbine combustors is currently motivating considerable efforts to better understand, model and predict soot formation. To design the next generation of combustion chambers, numerical simulation has become an essential tool, especially in terms of pollutant emissions prediction. Considering that the chemistry of soot particles strongly depends on their size, the present contribution proposes to use a semi-deterministic Lagrangian approach to predict the soot particles size distribution. The Lagrangian approach is applied to a gaseous non-premixed burner computed with Large Eddy Simulation (LES) for three operating points. The gaseous chemistry is described with an Analytically Reduced Chemistry to guarantee a good prediction of both flame structure and gaseous soot precursors. Results are validated against measurements and compared in terms of accuracy and CPU cost with an Eulerian semi-empirical model. To the authors knowledge, it is the first time that such an approach is used for soot, as only Eulerian formulations or stochastic Lagrangian approaches are reported in the literature up to now to predict soot size distribution.

ALISIOS – Advanced controL of wInd farm uSing computatIOnal fluid dynamicS

Project Title: ALISIOS – Advanced controL of wInd farm uSing computatIOnal fluid dynamicS
Project Leader: Paolo Schito
Resource Awarded: 30 million core hours on Marconi – KNL

Details

Team Members: Bibiana Garcia-Hevia, Centro National de Energias Renovables, SPAINSugoi Gomez-Iradi, Centro National de Energias Renovables, SPAINLuca Bernini, Politecnico di Milano, ITALYAlberto Zasso, Politecnico di Milano, ITALYBart Doekemeijer, Delft University of Technology, NETHERLANDSJan-Willem van Wingerden, Delft University of Technology, NETHERLANDS

Abstract:
Current practice in wind farm operation is that every turbine has its own controller that optimizes its own performance in terms of energy capture. This way of operating wind farms means that each wind turbine operates based only on the available information on its own measurements. Wind turbines are connected because of the formation of wakes, that may lead to losses in power generation for the downwind turbines: Barthelmie reports losses in power generation of up to 23% in the offshore Lillgrund wind farm due to wake effects. This gets the wind farm to operate in a non-optimum way, since wind turbines are not operating as players of a major system. The major reasons for this non-optimum approach of wind farms operation is the fact that turbines were developed as single machines and only as a second step were arranged in a wind farm. There is also a lack of knowledge and tools which can model the dynamics of the flow inside the wind farm, how wind turbines modify this flow, and how the wind turbines are affected by the perturbed flow. In addition, this lack of tools deals to also a lack of advanced control solutions, because there is not any available tool which can help on developing and testing virtually advanced control concepts for wind farms. Within the H2020 CL-WINDCON project, new innovative advanced open and closed loop wind farm control algorithms will be developed, to enable to treat the entire wind farm as an integrated optimization problem. The wind farm control algorithms will be validated using experimental data (full-scale and wind tunnel) and using so-called high fidelity numerical simulations that are proposed within this project. High-fidelity simulations will be performed using the Simulator for Onshore/Offshore Wind Farm Applications (SOWFA) open-source tool, which is based on the OpenFOAM framework. With SOWFA it is possible to model the characteristics of the atmospheric boundary layer for the reproduction of the natural wind features. Furthermore, it is possible to reproduce the wake effects of the wind turbines and to evaluate the performance of the wind turbines. This framework couples the aerodynamics of the rotor with the structural dynamics of the wind turbine and the control strategy. These high-fidelity simulation tools can be used to analyse and validate the current method of control inside wind farms. This will allow to pin-point the weak points, and to hypothesize potential gains using coordinated control methods. Secondly, high-fidelity simulations play a crucial role in the development and validation of surrogate wind farm models, with trade off modelling accuracy with computational tractability for real-time optimization. Finally, controllers synthesized using such surrogate models can be tested and the benefits of such control algorithms can be hypothesized for virtual and existing wind sites.

top

Universe Sciences (6)

SPOTSIM – Spot-forming convection simulations

Project Title: SPOTSIM – Spot-forming convection simulations
Project Leader: Petri Käpylä
Resource Awarded: 28.5 million core hours on MareNostrum

Details

Team Members: Maarit Käpylä, Max-Planck-Institut for Solar System Research, GERMANYNishant Singh, Max-Planck-Institut for Solar System Research, GERMANYAxel Brandenburg, NORDITA, SWEDEN

Abstract:
The solar magnetic field is generated by dynamo processes operating in its interior. The large-scale magnetic field is manifested by dark sunspots, where strong magnetic fields inhibit convective heat transport and lower the temperature, at the solar surface. The current paradigm of solar dynamo and spot formation assumes that strong magnetic flux tubes are generated in the tachocline, a layer of strong shear at the interface between the convection zone and the radiative interior, whence they rise without appreciably interacting with the highly turbulent convection on the way to the surface to form sunspots. This paradigm has been recently challenged by numerical simulations which indicate that strong distributed magnetic fields can be generated also in the bulk of the convection zone and that solar-like magnetic cycles can be obtained without tachoclines. However, a mechanism to form magnetic field concentrations is required for sunspots to form in a distributed dynamo. A promising candidate is the Negative Effective Magnetic Pressure Instability (NEMPI) which occurs in a stratified fluid under the influence of large-scale magnetic fields. The instability has been demonstrated to exist in idealised simulations but yet in turbulent convection. Previous convection simulations used to study magnetic structures have relied on either uniform imposed large-scale fields or boundary conditions to feed the system with magnetic flux. Such models produce spot-like structures but they leave the question of their formation process untouched. In the present project we use 28 million CPU hours to study the formation of magnetic structures with turbulent convection simulations where a more realistic large-scale magnetic field configuration is either imposed or generated self-consistently by the interaction of highly stratified turbulence and the overall rotation and shear in the system. We use high resolution simulations that can enable the first self-consistent models of spot-forming dynamos.

Energy Transfer across Boundary Layers in the Earth’s Magnetosphere

Project Title: Energy Transfer across Boundary Layers in the Earth’s Magnetosphere
Project Leader: Takuma Nakamura
Resource Awarded: 20 million core hours on Marconi – KNL

Details

Team Members: Philippe Bourdin, Austrian Academy of Sciences, AUSTRIARumi Nakamura, Austrian Academy of Sciences, AUSTRIAWilliam Daughton, Los Alamos National Laboratory, UNITED STATES

Abstract:
In this project, a series of large-scale fully kinetic plasma simulations will be performed to understand realistic energy transfer physics in collisionless space plasmas. Space such as between planets, stars and even galaxies is almost commonly filled with plasma with its density small enough to neglect particle collisions. In such a collisionless system, the boundary layer between regions with different plasma properties plays a central role in transferring energy and controlling the dynamics of the system itself. In a representative collisionless system, the Earth’s magnetosphere, the energy input from the solar wind is transferred and changes its properties through different physical processes at various boundary layers, which eventually leads to the global dynamics of the magnetosphere and various energetic space weather phenomena. Although a number of theoretical, numerical and experimental studies have been performed to understand the boundary layer physics and related energy transfer processes in the magnetosphere, quantitative aspects of the transfer processes are still poorly understood. This is mainly because the realistic transfer processes basically involve a broad range of temporal and spatial scales from the electron kinetic to magneto-hydrodynamic (MHD) scales, which were difficult to be handled by previous research tools. Thus, the main goal of this project is to quantify the realistic energy transfer processes covering all necessary scales by effectively combining state-of-the-art fully kinetic simulations which cover a broad range of scales and high-resolution in-situ spacecraft observations which cover necessary scales to provide realistic parameters to the simulations. To this end, we will systematically perform large-scale fully kinetic simulations using the high-performance VPIC code under realistic conditions obtained from the recently launched high-resolution MMS (Magnetospheric Multiscale) spacecraft. This project is timely because (i) the proposed systematic simulations covering full electron-to-MHD scales are feasible only by the combination of high-performance VPIC code and high-performance processors on the Tier-0 system (MareNostrum), and (ii) providing realistic initial conditions to the simulations from real observations resolving the electron-scales are feasible only by the MMS spacecraft.

Relics of cosmic reionization and dark matter free-streaming in the Lyman-alpha forest

Project Title: Relics of cosmic reionization and dark matter free-streaming in the Lyman-alpha forest
Project Leader: Ewald Puchwein
Resource Awarded: 20 million core hours on Joliot Curie – KNL

Details

Team Members: Laura Keating, University of Toronto, CANADAJonathan Chardin, Université de Strasbourg, FRANCEMatteo Viel, SISSA, ITALYMartin Haehnelt, University of Cambridge, UNITED KINGDOMGirish Kulkarni, University of Cambridge, UNITED KINGDOMDebora Sijacki, University of Cambridge, UNITED KINGDOMAvery Meiksin, University of Edinburgh, UNITED KINGDOMJames Bolton, University of Nottingham, UNITED KINGDOMGeorge Becker, University of California, Riverside, UNITED STATESVid Iršič, University of Washington, UNITED STATES

Abstract:
The intergalactic medium (IGM) is the rarefied material which spans the vast distances between galaxies in the Universe. In the early Universe this material is hot and ionized. Then the IGM cools due to the cosmic expansion and becomes neutral. It remains neutral until the first stars and galaxies form. Their ionizing UV emission reionizes the Universe. Understanding this cosmic reionization process in detail is not only interesting in its own right, but also highly relevant for galaxy formation and cosmology. In particular, reionization probes the ionizing photon output of galaxies and suppresses the formation of stars in low mass halos where photoheating can unbind the gas component. Cosmological constraints are also affected. For example, the forest of Lyman-α absorption lines visible in the spectra of high-redshift quasars that arises from neutral hydrogen along the line of sight is one of the most sensitive probes of the small scale matter distribution and hence of the properties of dark matter. Unfortunately, the signatures of free streaming of dark matter in the Lyman-α forest are degenerate with the thermal and ionization state of the IGM shortly after cosmic reionization. To make further progress here an improved modelling of the IGM both during and after cosmic reionization is needed. We, hence, propose an ambitious simulation programme designed to achieve exactly that. We will use a combination of large, very high resolution cosmological hydrodynamical simulations and post-processing radiative transfer simulations to better understand the legacy of the patchy reionization process in the z ≳ 5 Lyman-α forest. This will give us new insights into the nature of the sources of ionizing radiation and the exact timing of cosmic reionization. At the same time it  will enable us to use the best available Lyman-α forest data at these redshifts to more accurately constrain the small-scale matter distribution by breaking degeneracies with the thermal and ionization state of the IGM. This will result in much improved and more robust constraints on the free streaming of dark matter. Our simulations will model a range of dark matter models that have been introduced to alleviate small-scale problems of cold dark matter, such as warm, fuzzy and mixed dark matter. Their detailed signatures in the Lyman-α forest will be predicted and tested against the most recent observational data.

CIMS-EHres – Convection In Massive Stars, going to Extremely-High resolution simulations of stellar interiors

Project Title: CIMS-EHres – Convection In Massive Stars, going to Extremely-High resolution simulations of stellar interiors
Project Leader: Cyril Georgy
Resource Awarded: 40 million core hours on MareNostrum

Details

Team Members: Raphael Hirschi, Keele University, UNITED KINGDOMLaura Scott, Keele University, UNITED KINGDOMW. David Arnett, University of Arizona, UNITED STATESCasey Meakin, University of Arizona, UNITED STATESAndrea Cristini, University of Oklahoma, UNITED STATES

Abstract:
Convection is as of today a major weakness in the classical, one-dimension modelling of stars. To correctly interpret the huge amount of data that will come out of current and future facilities such as the Gaia mission, the Ligo/Virgo interferometers, the E-ELT and the James Webb Space Telescope, it is crucial to correct this situation by developing next generation stellar evolution codes. The aim of this project is to perform extremely high resolution 3-dimension hydrodynamics simulations of convection during two different nuclear burning stages: carbon burning and neon burning, characterised by different physical conditions, such as sound speed, bulk Richardson number, etc. These simulations, reaching Reynolds numbers as high as 10,000, will firmly be in the turbulent regime, and will provide a precious insight on the physics of the convective boundaries. Crucial information about the position and shape of the boundary, as well as about the mixing processes occurring through this boundary will be obtained. In the mean term, these simulations will allow us to develop new algorithms to treat convection in classical 1d stellar evolution codes, capturing accurately the average behaviour of the complex 3-dimension simulations. This is a mandatory step to push stellar evolution in a new, predictive era.

MENHIRS – Mainstay Ensemble of Nyx HI Reionization Simulations

Project Title: MENHIRS – Mainstay Ensemble of Nyx HI Reionization Simulations
Project Leader: Jose Onorbe
Resource Awarded: 18 million core hours on Juwels

Details

Team Members:

Abstract:
The latest measurements of CMB electron scattering optical depth reported by the Planck satellite significantly reduces the allowed range of HI reionization models, pointing towards a later ending and/or less extended phase transition than previously assumed. During HI reionization the intergalactic medium (IGM) is photoheated to ∼ 104 K, and owing to long cooling and dynamical times of the diffuse gas, comparable to the Hubble time, its vestiges persist in the IGM to much later times. Existing and upcoming observations of the Lyman-α (Ly-α) forest at z ∼ 5 − 6 can be used to detect these extant thermal signatures of HI reionization by comparing them with accurate hydrodynamical models of HI reionization. This can provide new independent constraints not only on the timing of HI reionization but also on the sources that reionized the universe. We propose to run a grid of 27 state-of-the-art cosmological hydrodynamical simulations with extremely high dynamical range (Lbox = 80 h−1Mpc and 40963 resolution elements) in order to explore the physical parameter space characterized by the onset of reionization, its duration and the heat injected into the IGM. The suggested dynamical range of the simulations is required to attain the necessary precision in the Ly-α forest predictions. The number of simulations is motivated by the requirement to guarantee a good sampling of parameter space and to create accurate emulator of the Ly-α statistics. For the proposed work, we will use the Nyx code, a massive parallel cosmological N-body + hydrodynamical code. We will employ a well tested, new approximate method already implemented in the code by the PI to model inhomogeneous reionization. The method allows to explore the full physical parameter space and provide accurate fits to high-z Ly-α observations, at a cost dramatically less than using full radiative transfer simulations. To achieve these goals we request 24,000,000 core hours on the MareNostrum-4 system based on the tests done in our PRACE Preparatory Access Project at MareNostrum, Curie, Hazel Hen and Marconi. We also offer to the TAC the possibility of moving this allocation to other PRACE systems (Hazel Hen, Curie or Marconi) if they decide is more suitable. The PI has secured funding from the University of Edinburgh to continue research on this topic for at least the next two years..

Shining a light through the dark ages

Project Title: Shining a light through the dark ages
Project Leader: Pierre Ocvirk
Resource Awarded: 68 million core hours on Piz Daint

Details

Team Members: Romain Teyssier, Universitaet Zurich, SWITZERLANDDominique Aubert, Universite de Strasbourg, FRANCEJoseph Lewis, Universite de Strasbourg, FRANCE

Abstract:
The Epoch of Reionization (EoR) takes place during the first billion years of the Universe. It begins as the first stars form, and sees the early cold neutral Universe turn into a warm, ionised, UV-transparent state. Observing this transition, and the sources powering it, offer a unique opportunity to understand structure formation in the prime youth of the Universe, and the community has massively invested towards this goal. With SKA starting construction, and multiple 21cm experiments already collecting data and producing results (LOFAR), the need for accurate predictions of the 21 cm signal is stronger than ever. Yet, the conflicting requirements of very large volumes (100s of Mpc) and very high mass resolution make this a very challenging task, and models have to resort to sub-grid recipes for a number of processes. This project aims at improving our understanding of the interaction between the cold pre-reionization intergalactic medium (IGM) and the ionizing fronts sweeping the Universe during the EoR, by performing cutting-edge, fully coupled radiation-hydrodynamics simulations of the end of the dark ages with our code RAMSES-CUDATON on Piz Daint. We will simulate small 1 h-1 Mpc3 volumes irradiated by UV ionisation fronts, at very high resolution (12 Msun mass resolution and 60 pc physical, requiring up to 20483 grids), to study the propagation of the front and the photo-evaporation of the IGM and mini-haloes during the onset of the EoR. Such massive simulations are only feasible on Piz Daint, with our code RAMSES-CUDATON, optimised for GPU usage. The latter has already been deployed successfully on up to 16384 hybrid nodes, with very good parallel performance, putting the team in an ideal position to exploit the power of Piz Daint to achieve the scientific goals of the project. The final product, in the form of clumping factors and photon consumption rates, will be published and picked up by the community to improve the description of the recombining Hydrogen gas in large scale models of the EoR.

top

Share: Share on LinkedInTweet about this on TwitterShare on FacebookShare on Google+Email this to someone