11th Project Access Call – Awarded Projects

Projects from the following research areas:




Biochemistry, Bioinformatics and Life sciences (3)

Hydra – accelerating dehydration to probe water persistency

Project Title: Hydra – accelerating dehydration to probe water persistency
Project Leader:Walter Rocchia, Italian Institute of Technology, Italy
Resource Awarded: 67 932 330 core hours on FERMI


Sergio Decherchi, Italian Institute of Technology-Italy
Marco Jacopo Ferrarotti, Italian Institute of Technology-Italy
Roberto Gaspari, Italian Institute of Technology-Italy
Andrea Spitaleri, Italian Institute of Technology-Italy
Ivan Vialov, Italian Institute of Technology-Italy
Syeda Rehana Zia, Italian Institute of Technology-Italy

In this project we aim at the quantitative description of hydration and water kinetics in the orthosteric binding site of the adenosinic receptor. We propose a thorough investigation of the water network of a system for which we have a high-resolution crystal using microsecond-long plain molecular dynamics simulations. These simulations will allow us to provide a statistically robust description of the water motion in the binding cavity at all time scales, including events of water diffusion which happen on the multi-nanosecond to microsecond timespan. The large amount of data gathered will have a two-fold aim. First it will serve to rationalize the experimental kinetic affinity of a series of congeneric triazines for the GPCR target. Such affinity will be investigated in terms of stability of the key water molecules mediating the ligand-protein interaction. Secondly, the data obtained by plain MD will be used to validate new computational methods aiming at the rapid description and representation of hydration in protein sites. These accelerated protocols are conceived to offer a platform for the technological transfer of MD-based methods into the fast-paced industrial field of drug discovery.


Synergistic Effects of Ruthenium and Gold Anticancer Agents in the Allosteric Regulation of the Nucleosome Core Particle

Project Title: Synergistic Effects of Ruthenium and Gold Anticancer Agents in the Allosteric Regulation of the Nucleosome Core Particle
Project Leader:Ursula Rothlisberger, EPFL, Switzerland
Resource Awarded: 6 346 238 core hours on MareNostrum


Martin Bircher, EPFL-Switzerland
Esra Bozurt, EPFL-Switzerland
Elisa Liberatore, EPFL-Switzerland

Giulia Palermo, EPFL-Switzerland

Transition-metal compounds are effective anticancer agents, as they interfere with gene transcription leading to the modulation of primary tumors and metastasis. These compounds exert their action at the chromatin level, which tightly compacts the DNA in cells, allowing the packaging of the genome. The action of chromatin-binding agents has profound consequences on the replication and transcription machineries, thus resulting in a modulatory action of gene expression in cancer cells and in an important chemotherapeutic potential. Despite many years of research, the molecular targeting characteristics of these compounds have remained to a large extent elusive. Although the DNA is traditionally assumed to be the pharmacological target, recent evidences suggest that chromatin-associated proteins are crucially involved in the therapeutic effects. Here, we propose to study at atomistic level the mechanism of action of a promising class of Ru(II) and Au(I) based anticancer agents. The antimetastasis agents Rapta [Ru(II), Arene, PTA=1,3,5-triaza-7-phosphatricyclo ( decane], are highly promising drug candidates due to their low toxicity and selective activity against specific cancer cells types, while the Au(I) compound Auranofin – a clinically approved antiarthritic drug – has shown outstanding antiproliferative properties in both in vivo and in in vitro models. Unpublished high-resolution crystal structures indicate that Auranofin and Rapta-T (T = toluene) simultaneously bind at nucleosome core particles (NCP), which are the basic repeat units of chromatin. NCPs consist of chromosomal DNA (147 base pairs in length) wrapped around a protein histone core. Although Auranofin and Rapta-T bind at two distal sites of the NCP protein core, biochemical and quantitative ICP/MS experiments show a 3 fold increased binding of Auranofin when in combination with Rapta-T, suggesting a nontrivial synergistic effect of the two compounds. Here, we propose to run long time scale atomistic molecular dynamics (MD) simulations of the NCP as apo form and in complex with Auranofin and/or RAPTA-T in order to describe the mechanism of allosteric regulation. This computational approach will allow capturing the crucial conformational changes of the NCP, as induced by the in presence of these anticancer agents and provide a detailed understanding of the mechanism of allosteric regulation at the NCP level. Hybrid quantum mechanics/molecular mechanics (QM/MM) MD will be employed to in order to generate reliable in situ force fields for the protein-bound transition metal compounds. The outcome of these simulations will be used in collaboration with a team of experimental collaborators, directly inspiring biochemical essays, quantitative ICP/MS and calorimetry titration experiments, while also guiding the refinement of the crystal structures at level of the active site. Overall, this proposal aims to provide detailed insights on the molecular mechanisms underlying the cooperative binding of Auranofin and RAPTA-T, and its effects on the allosteric regulation of the NCP. This might open new routes for the induction of therapeutic effects. As an ultimate goal, we aim at deciphering how chromatin-binding agents could directly influence gene expression in cancer cells, thus opening new scenarios for cancer treatment.


Prot_Crowd_Diffuse_Meso – Combining Molecular Dynamics Simulations and Experiments to Assess Crowding and Diffusion of Membrane Proteins at the Mesoscale

Project Title: Prot_Crowd_Diffuse_Meso – Combining Molecular Dynamics Simulations and Experiments to Assess Crowding and Diffusion of Membrane Proteins at the Mesoscale
Project Leader:Mark Sansom, University of Oxford, United Kingdom
Resource Awarded: 11 100 000 core hours on MareNostrum


Matthieu Chavent, University of Oxford-United Kingdom
Anna Duncan, University of Oxford-United Kingdom
Antreas Kalli, University of Oxford-United Kingdom

With the recent improvements of both experimental and computational techniques it is now possible to gain insight into more complex biological systems. Nevertheless, at the frontier between these two fields lies a twilight zone called the mesoscale. This zone ranges from few hundreds of nanometres to micrometre in size and from microseconds to millisecond in time. Very recent results show that, in this area, proteins could adopt specific spatial organisation at the membrane impacting biological processes in eukaryotic cells. In parallel, recent work conducted by the present team and their collaborators has revealed the formation of “islands” of ~0.5 µm constituted by the aggregation of porins in bacterial outer membranes. This phenomenon has an essential role in the turnover of proteins at the surface of Gram negative bacteria . Despite increased efforts, both in computational and experimental methods, to cover this mesocale area, it is still very difficult to compare models and experimental observations, in part due to the limited scales of the simulations performed. Indeed, the complex dynamic interactions between lipids and proteins need to be better understood at a larger scale to take into account possible emerging phenomena which are not visible for small system sizes. That is why, in this project we propose to perform simulations of large enough time and length scales to make meaningful comparisons with experimental studies piloted in parallel by our colleagues. We will perform an exhaustive analysis of the dynamics and interactions of numerous membrane proteins ranging from simple transmembrane (TM) helices to larger radius integral membrane proteins, such as ion channels or outer membrane proteins. These systems will then be also compared with the dynamics of peripheral proteins to examine the respective influence of TM and juxta-membrane parts on the protein dynamics. The aforementioned proteins will be inserted in eukaryotic and bacterial membrane systems in order to compare the dynamics and aggregation of the proteins in two different types of organism, thus making our systems more biologically relevant. Simulating very large systems in order to reach time and length scales comparable to experiment is very challenging. Access to a Tier-0 machine will allow us to overcome the size and timescale issue by performing cutting edge simulations for time and length scales which are not routinely available (i.e. systems with a size of several hundreds of nanometres simulated for dozens of microseconds).



Chemical Sciences and Materials (4)

SHINE – Substrate-induced Polarization and Hybridization Effects in Supported Graphene Nanoribbons

Project Title: SHINE – Substrate-induced Polarization and Hybridization Effects in Supported Graphene Nanoribbons
Project Leader:Andrea Ferretti, CNR, Italy
Resource Awarded: 23 000 000 core hours on FERMI


Claudia Cardoso, CNR-Italy
Deborah Prezzi, CNR-Italy
Andrea Marini, CNR-Italy
Alice Ruini, University of Modena and Reggio Emilia-Italy

Graphene nanostructures have recently triggered a wealth of studies for their remarkable properties, which combine the unique electronic and mechanical features of graphene with the semiconducting behavior induced by quantum confinement. The possibility to produce graphene nanostructures with the needed atomic precision was recently demonstrated by exploiting on-surface synthesis techniques. On the theory side, most of the studies on excited-state properties rely so far on calculations for ideal isolated systems. However, the presence of a substrate is often required, e.g. for its catalytic role or to support further device applications, and has demonstrated to give rise to significant modifications of the intrinsic electronic and optoelectronic properties of the systems. The present project can be viewed as a follow-up of a previous PRACE project (EPIGRAPH- Pra06_1348), devoted to the study of A-GNRs, and their precursors, on Au (111) and Au (110) surfaces. Our main goal within the present project is the accurate first-principles investigation of edge- and quantum-confinement-effects in the excited-state properties of realistic graphene nanoribbons (GNRs), as altered by the coupling to a substrate. To this end, we will employ different levels of theory, from density-functional theory (DFT), for the study of ground-state properties, to many-body perturbation theory (MBPT) approaches, for an accurate investigation of excited-state properties. In close collaboration with our experimental partners, we have identified a number of relevant target systems to address, which comprise armchair and zigzag graphene nanoribbons on metallic substrates, such as gold and graphene. Our results will be directly compared with top-quality experimental data, resulting from scanning-tunneling spectroscopy (STS), angle-resolved photoemission spectroscopy (ARPES), and optical measurements (such as e.g. reflectance difference spectroscopy, RDS) of the above-mentioned systems.


MCTM@QMC – Static and dynamical correlation of Multi-Centre Transition-Metal complexes by Quantum Monte Carlo

Project Title: MCTM@QMC – Static and dynamical correlation of Multi-Centre Transition-Metal complexes by Quantum Monte Carlo
Project Leader:Leonardo Guidoni, University of L’Aquila, Italy
Resource Awarded: 22 984 932 core hours on FERMI


Matteo Capone, La Sapienza, Universit di Roma-Italy
Shibing Chu, La Sapienza, Universit di Roma-Italy
Sandro Sorella, SISSA-Italy
Daniele Bovi, University of L’Aquila-Italy
Daniele Narzi, University of L’Aquila-Italy
Daniele Ottaviani, University of L’Aquila-Italy
Fabio Pitari, University of L’Aquila-Italy

Multi-centre transition metal (TM) complexes have a fundamental role in different fields of Chemistry, Biology and Materials science. They are present in a large variety of catalysts in molecular form or as surfaces in bulk or nanostructured materials. In Biology they are present as catalytic and redox centres in many metalloenzymes and electron-transport proteins such as ferredoxin, hydrogenases, copper-based enzymes, and photosynthetic complexes. Due to the fundamental role of static and dynamical electron correlation in such systems, the accurate determination by first principles of their electronic and geometrical properties is still representing a challenge for Quantum Chemistry. Quantum Monte Carlo (QMC) methods are promising techniques for the study of the electronic structure of correlated molecular systems such as Multi-Centre-Transition-Metal Complexes. QMC algorithms are highly parallel in nature and thanks also to the relatively small memory requirements also for large systems, they have good performances and scalability for highly parallel computers. In the present proposal we plan to apply Quantum Monte Carlo and DFT+U techniques to characterize the electronic structure of Multi-Centre-Transition-Metal complexes which are fundamental in catalysis and biocatalysis. We will use the Extended Broken Symmetry method to reveal their electronic and structural properties with unprecedented accuracy. For the first time we will have the possibility to study these important systems without compromises in term of static and dynamical electron correlation. The project will shed new lights on the chemistry of these complexes as well as providing the scientific community with high-level reference calculations that might also be important for the benchmarking of more standard techniques. The applications to iron-sulfur clusters includes Fe2S2, which is found in ferredoxin proteina and the cubane Fe4S4 cluster, which is common in redox active metalloenzymes. As last subject we will study the Mn4CaO5 catalitic core of the photosystem II complex, which is the catalytic core of the water-splitting reaction in photosynthesis.


BIG-APPLE-NY — aB-initio InvestiGAtion of interParticle couPLing Effects in arrays of silicon NanocrYstals

Project Title: BIG-APPLE-NY — aB-initio InvestiGAtion of interParticle couPLing Effects in arrays of silicon NanocrYstals
Project Leader:Stefano Ossicini, Università degli Studi di Modena e Reggio Emilia, Italy
Resource Awarded: 45 000 000 core hours on FERMI


Matteo Bertocchi, Università degli Studi di Modena e Reggio Emilia-Italy
Ivan Marri, Università degli Studi di Modena e Reggio Emilia-Italy
Marco Govoni, University of Chicago- United States

In this proposal we aim at investigating effects induced by NCs interplay on the energetic structure, carrier multiplication (CM) dynamics and Auger recombination (AR) mechanisms of oxygen passivated silicon nanocrystals (Si-NCs). To this purpose, we use highly parallelized codes that combine density functional theory (DFT) and many body perturbation theory (MBPT) in the G0W0 approximation. Arrays of closely packed NCs have gained considerable interest between scientific community because they show new properties that are promising for photovoltaic (PV) and optoelectronic applications. NC-NC interaction can be exploited to alter carrier mobility and exciton recombination dynamics, to facilitate dissociation of photogenerated excitons and to generate systems that combine bulk-like transport properties and discrete excitonic optical behavior [1-3]. Moreover, NCs interplay seems to be able to promote a red-shift of the absorption energy threshold [4-16] and to be responsible for the occurrence of new two-site CM processes [17-19]. Effects induced by NC-NC interaction are not trivial to be isolated and investigated experimentally; theory can therefore play a fundamental role giving a direct access to the parameters that rule the optoelectronic properties in systems of strongly coupled NCs. Our scope is twofold, that is (i) identify the conditions that allow to exploit NC-NC interaction in order to increase the portion of solar spectrum that can be absorbed by an array of Si-NCs (by red-shifting the energy gap of the system) and (ii) identify the conditions that permit to maximize occurrence of non-dissipative CM effects (in particular two-site CM mechanisms that lead to the generation of Auger unaffected multiexciton configurations distributed among different NCs [20]), thus reducing effects induced by thermalization mechanisms in nanostructured solar cell devices. Realization of points (i) and (ii) will allow to enhance light harvesting and to decrease the occurrence of loss factors, thus improving the efficiency of nanostructured Si-based PV devices. Finally, we will investigate effects induced by wavefuntions delocalization on AR processes. In this project we will analyze three different systems obtained by placing two different O-terminated Si-NCs in the same simulation box. The considered systems are close to the experimental conditions where Si-NCs are often surrounded by oxygens. Due to the complexity of the treated systems (large number of electrons and high kinetic energy cutoff required to ensure a correct convergence) the planned calculations are highly computational demanding; for this motive points (i) and (ii) have been never addressed by theoretical modeling, despite they are of fundamental importance to discern how NCs interplay can be exploited to improve solar cell performances. The obtained results will be compared with the ones achieved in our previously approved PRACE projects [20-22], where isolated H- and O- terminated NCs were analyzed and where systems of interacting H-terminated Si-NCs were studied. Such comparison will allow to discern the role played by O- termination, NCs surface conformation, NCs size and obviously NC-NC separation on optoelectronic properties of closely packed Si-NCs.


BLUE ENERGY – Molecular simulation of aqueous electrolytes in nanoporous carbons: blue energy and water desalination

Project Title: BLUE ENERGY – Molecular simulation of aqueous electrolytes in nanoporous carbons: blue energy and water desalination
Project Leader:Mathieu Salanne, CNRS, France
Resource Awarded: 22 100 000 core hours on MareNostrum


Matthieu Haefele, CNRS-France
Lyderic Bocquet, Ecole Normale Superieure-France
Vojtech Kaiser, Ecole Normale Superieure-France
Benjamin Rotenberg, Universite Pierre et Marie Curie-France
Paul Madden, University of Oxford- United Kingdom

When fresh river water and salty sea water mix, a considerable amount of energy is dissipated. Several processes are currently examined for the exploitation of this yet untapped blue energy. Conversely, this explains why sea water desalination for drinkable water production requires so much energy. Since the proposal in 2009 of a new strategy to achieve both goals in a capacitor, using thermodynamic cycles based on the charge/discharge of electrodes under high/low salt concentration in the electrolyte, experimentalists and engineers tried to improve the process. A promising strategy is to use nanoporous carbon electrodes. However traditional models for the determination of the relevant quantities (electrical capacitance, amount of adsorbed salt and diffusion coefficients as a function of the electrolyte composition and electrode potential) cannot be used in the present case where interactions at the molecular level play an essential role. This is the bottleneck that we propose to overcome using molecular dynamics simulation. This will be achieved thanks, on the one hand, to the tools and methods developed in our previous work on supercapacitors based on ionic liquids or organic electrolytes in nanoporous carbon, and on the other hand, to our expertise on aqueous solutions under extreme confinement.



Earth System Sciences (2)

StratForce – Direct Numerical Simulations of forced stratified turbulence

Project Title: StratForce – Direct Numerical Simulations of forced stratified turbulence
Project Leader:Erik Lindborg, Royal Institute of Technology, Sweden
Resource Awarded: 35 000 000 core hours on FERMI


Geert Brethouwer, Royal Institute of Technology-Sweden
Andrea Maffioli, Royal Institute of Technology-Sweden
Peter Davidson, University Cambridge- United Kingdom
Pui-kuen Yeung, Georgia Institute of Technology- United States

Flows in the upper troposphere and stratosphere are invariably stably stratified due to an increase in temperature with height. The large scales involved in atmospheric flows means that these flows are also turbulent therefore the study of stratified turbulence is important to understand the atmosphere dynamics at intermediate scales up to 100km, where the Earth’s rotation is unimportant. Within this project, we propose to perform simulations of stratified turbulence that for the first time resolve all the physical scales involved, from the large scales to the smallest dissipative scales. The results will form a reference against which theories have to be tested. They will allow us to understand the dynamics of overturning motions in the atmosphere, which can inform turbulence parametrizations for weather prediction. We also wish to study the vertical structure of these stratified turbulent flows, which could allow meteorologist to know, for example, how many vertical layers to choose for their global circulation model and ultimately improve our ability of weather forecasting.


Study of liquid planetary materials at extreme conditions. (Fluid Silicates and the Magma Ocean)

Project Title: Study of liquid planetary materials at extreme conditions. (Fluid Silicates and the Magma Ocean)
Project Leader:Lars Stixrude, University College London, United Kingdom
Resource Awarded: 22 600 000 core hours on MareNostrum


Bing Xiao, University College- United Kingdom
Hang Cui, University College London- United Kingdom
Carlos Pinilla Castellanos, University College London- United Kingdom
Roberto Scipioni, University College London- United Kingdom

Deep melt may exist in the Earth today, and the magma ocean may have left signatures of its presence. However, these signals are still uninterpretable because of a lack of basic knowledge of the behavior of fluid silicates at extreme conditions: very little is known of the physics and chemistry of fluid silicates beyond the conditions of ongoing shallow magma genesis (< 3 GPa). We propose to solve this problem by performing new first principles quantum mechanical simulations in the range of pressure, temperature, composition relevant to the early Earth that have not yet been explored by experiment or theory. Simulations will include key homogeneous and heterogeneous systems of fluid silicates in liquid, vapor, supercritical, and solid forms, including simulations of pure phases, and phase coexistence. Wtthin our project MoltenEarth entitled Fluid Silicates at Extreme Conditions and the Magma Ocean’ we expect to change our views of magma ocean evolution and lead to new scenarios of Earth’s earliest evolution. What these scenarios might be is impossible to predict as they will be shaped by still unknown aspects of the physics and chemistry of silicate liquids at extreme conditions, which the MoltenEarth project aims to discover. Our work will be based on the use of state-of-the-art ab-initio methods for the simulations of systems containing hundreds and possibly thousands of atoms. The present project will allow us to predict quantities central to any attempt to understand magma ocean evolution: 1) The freezing interval of silicate liquids. Over what temperature range do silicate liquids freeze and what is the sequence of crystals that form? 2) The density of silicate liquids. How does the density of liquids depend on pressure and temperature and how does this compare with coexisting crystals? 3) The composition of silicate liquids. How are elements partitioned between coexisting liquid and crystal and how does this influence buoyancy? 4) The vaporization of silicate liquids. What is the nature of the first atmosphere that overlay the magma ocean after the giant impact? 5) The reaction of silicate liquids with core material. What is the nature of the bottom boundary of the magma ocean? What is the extent and nature of reaction? This project will almost certainly lead to the consideration of new evolutionary scenarios as the full richness of silicate crystallization is included, along with new ways of testing these against present-day observations. It will serve as the basis for evaluating what may be thought of as Earth’s initial condition: its state immediately following the moon-forming impact, including the depth of the magma ocean, and nature of its first atmosphere.



Engineering (2)

HAT- Homogeneous and Anisotropic Turbulence: Eulerian and Lagrangian statistics

Project Title: HAT- Homogeneous and Anisotropic Turbulence: Eulerian and Lagrangian statistics
Project Leader:Luca Biferale, University of Rome ‘Tor Vergata’, Italy
Resource Awarded: 27 000 000 core hours on FERMI


Fabio Bonaccorso, University of Rome ‘Tor Vergata’- Italy
Kartik Iyer, University of Rome ‘Tor Vergata’-Italy

Turbulence is a commonly occurring state of fluid motion both in nature and in engineering applications. Pollutant dispersion in the atmosphere, mixing inside a combustion chamber and weather patterns such as hurricanes, all involve turbulence. A characteristic feature of all turbulent flows is a wide range of scales in space and time dynamically active. On one side, in almost all cases, the large scales are directly effected by boundary conditions and other external forcing mechanisms, rendering them anisotropic (and non-homogeneous). On the other side, a recovery to small-scales isotropic statistics is believed to happen, far enough from boundaries. Very little is known about the relative importance of anisotropic with respect to isotropic fluctuations at changing the scale. We aim to advance the numerical state-of-the-art by performing a series of direct numerical simulations –at world record resolution– of randomly forced homogeneous anisotropic turbulence seeded with millions of particles, in order to access also Lagrangian properties. We will change both the degree of anisotropy and the typical scale of the injecting forcing mechanism, but keeping it homogeneous such as to have a precise control on the spatial statistics. We aim to produce the first data-set where leading isotropic and sub-leading anisotropic scaling properties can be measured at high Reynolds numbers. We aim to answer questions related to the universality properties of anisotropic turbulence for both Eulerian and Lagrangian domains. These are crucial problems also for improving turbulence modelling, where it is known that purely isotropic assumptions are failing.


Acronym – LES for FACTOR Full title – Large Eddy Simuation of combustor/turbine interactions

Project Title: Acronym – LES for FACTOR Full title – Large Eddy Simuation of combustor/turbine interactions
Project Leader:Laurent GICQUEL, TURBOMECA (SAFRAN group), France
Resource Awarded: 5 800 000 core hours on MareNostrum


Gabriel Staffelbach, CERFACS-France
Guillaume BONNEAU, TURBOMECA (SAFRAN group)-France

Charlie KOUPPER, TURBOMECA (SAFRAN group)-France

Gas Turbine (GT) engine is an essential part of any aeronautical industry. This unchallenged use for aircraft relies in the GT high power-to-weight ratio, which is the most effective of all engine concepts but also because of its flexibility, operability and its improvement potential. Over the last forty years, GT’s have evolved to yield ever-improved efficiencies, fuel consumption and significant reduction in emissions. To reach such levels, GT industry followed two paths: 1/ the use of lean burnt combustors for fuel or emission control and 2/ an increased turbine inlet temperature for improved work extraction. The former is today a limiting factor as the temperatures reached at the inlet of the turbine surpass the melting temperature of materials. This, plus the fact that this part of the engine is also the most aggressive: high temperature, high pressure with very large rotational velocities and strong mechanical as well as thermal stresses… makes new improvements subject to the limit of engineers to guaranty long term structural integrity of that component in such conditions. To improve our understanding of the turbine flow and the combustor/turbine interactions, the EU FACTOR project [3] invested in an experimental and numerical data basis generation. Two test facilities are to be produced: a reduced three sector combustor rig located at UNIFI (Florence, Italy) [4] and a real scale full annular combustor/turbine test rig operated by DLR (Göttingen, Germany) [5]. In parallel, high fidelity Computational Fluid Dynamics (CFD) tools relying on the Large Eddy Simulation (LES) technique [6,7] have been developed and successfully applied. The objective of this proposition is to invest further in this direction by providing state-of-art validations and predictions LES of combustor/turbine flows on the basis of the FACTOR project. To do so, fundamental steps dedicated to the LES sensitivity of the vane flow will need first be addressed. The grid resolution effect as well as the near wall turbulence modeling will be probed specifically in the context of the combustor/turbine Nozzle Guiding Vane (NGV) simulations (i.e. without the rotor of the turbine). For each modelisation (mesh resolution and wall treatment), the turbulence generated in the combustor will be analyzed to measure it intensity, scales, spectrum and shape. Then, the convection and distortion of the turbulent eddies in the vane passage will be gauged and related to important turbine characteristics: secondary flows, separation and heat transfer. To complement these actions, the effect of the vane clocking with respect to the fuel injection system will have also to be assessed relative to the grid resolution and turbulence modeling to indicate the degree of importance of such issues for industrial applications. Out of such computations to be complemented by experimental data, better understanding of the turbulence generation and transport in terms of intensity or time scale will be accessible thanks to such simulations. To finish, a fully integrated combustor/turbine LES of the Göttingen DLR test facility is foreseen to produce un-precedented high fidelity predictions to be gauged against experimental data on such systems where rotating parts are present.



Fundamental Constituents of Matter (3)

JOREK. Non-linear modeling of MHD instabilities and their control in tikamaks.

Project Title: JOREK. Non-linear modeling of MHD instabilities and their control in tikamaks.
Project Leader:Marina Becoulet, CEA/IRFM, France
Resource Awarded: 13 200 000 core hours on MareNostrum


François Orain, Max Planck Institute for Plasma Physics- Germany
Shimpei Futatani, Barcelona Supercomputing Center Spain- Spain
Guillaume Latu, CEA/IRFM-France
Chantal Passeron, CEA/IRFM-France
Eric Nardon, CEA/IRFM-France
Cristian Sommariva, CEA/IRFM-France
Guilhem Dif-Pradalier, IRFM/SCCP-France
Guido Huijsmans, ITER-France
Stanislas Pamela, CCFE- United Kingdom
Jane Pratt, University of Exeter- United Kingdom

The aim of the ITER project is the demonstration of the scientific feasibility of nuclear fusion as a future energy source. The present project is focusing on the important issues for the fusion reactor operation – MHD instabilities such as Edge Localised Modes (ELMs), plasma disruptions, Neoclassical Tearing Modes (NTMs) and some promising methods of their control in ITER. The non-linear MHD code JOREK will be used. We will benchmark our developments, validate against experiments, and carry out predictive simulations for ITER. JOREK code is involved in EUROfusion, JET1 and MST1 projects in 2015-2017. The present project continues study started in the previous PRACE project ( ra1904 in 2014 on CURIE, cpu time fully used resulting in many publications, including 2 PRL papers and invited and oral presentations on Conferences ). It is focusing on the 6 subjects: 1) ELMs physics where multi-cycles regimes, energy deposition, dynamics of filaments will be studied and compared with experiments. 2) ELM control by promising method using Resonant Magnetic Perturbations (RMPs) which were demonstrated to suppress /mitigate ELMs. 3) ELM control by pellets, which trigger ELM with high frequency reducing their size. 4) QH mode where saturated MHD instability Edge Harmonics Oscillations regulates the energy transport at the edge of the plasma thereby avoiding ELMs. Resistive wall and rotation will be included in modelling. 5) A disruption is an undesired plasma termination which arises from a loss of magnetic confinement provoked by a global instability, resulting in extremely large heat fluxes. Since disruption avoidance is not always achievable, disruption mitigation is essential for ITER and for future devices. Massive Gas Injection (MGI) is one of the main methods proposed to mitigate disruptions. In 2014 for the first time a thermal quench has been modelled with JOREK, different type of instabilities (ballooning and tearing) were detected depending on the parameters and in particular resistivity. This study should be continued for JET and predictions for ITER are still to be done. Initial work of runaway electrons implementation in JOREK will start (subject of the PhD thesis) 6) Modelling the suppression of tearing modes by electron cyclotron current drive (ECCD), including a quantitative model for the NTM threshold. Neoclassical tearing modes (NTMs) are phenomenon in high beta tokamak plasmas . They can be stabilized by replacing the missing bootstrap current inside the island with another non-inductively driven current. In particular, Electron Cyclotron Current Drive (ECCD) has been used successfully to suppress NTMs, and is also envisaged as the main NTM control method on ITER. Two models for the ECCD feedback on the plasma have been implemented in JOREK and validated in 2014. Large production runs are now required to study the effect of the ECCD on both classical and neoclassical tearing modes.


CONTQCDNf3 – The continuum limit of QCD with three dynamical quark flavours

Project Title: CONTQCDNf3 – The continuum limit of QCD with three dynamical quark flavours
Project Leader:Mauro Papinutto, “Sapienza” Universita di Roma, Italy
Resource Awarded: 74 000 000 core hours on FERMI


Martin Luescher, CERN- Switzerland
Mattia Dalla Brida, DESY- Germany
Stefan Schaefer, DESY- Germany
Rainer Sommer, DESY- Germany
Tomasz Korzec, Humboldt Universität zu Berlin- Germany
Gregorio Herdoiza, Universidad Autónoma de Madrid- Spain
Carlos Roberto Pena Ruano, Universidad Autónoma de Madrid- Spain
Stefan Sint, Trinity College Dublin – Ireland
Anastassios Vladikas, INFN- Italy
Leonardo Giusti, Universita di Milano Bicocca- Italy

Since the discovery of the Higgs boson, the Standard Model (SM) has been confirmed to accurately describe elementary particle physics and despite enormous effort, no sign for physics beyond the SM has been found. It is therefore extremely important to combine experiment and theory to perform precision tests that constrain the parameters of the SM (e.g., in the strongly interacting sector, the parameters of the Cabibbo Kobayashi Maskawa – CKM – matrix), eventually discovering inconsistencies between determinations from different physics processes – discrepancies that would point to the presence of new physics particles. This can be done for example by studying decays of hadrons, which are mediated through weak interactions of quarks, in the effective weak Hamiltonian formalism. In this formalism, the most difficult theoretical task amounts to compute non-perturbatively in Quantum Chromodynamics (the theory of strong interaction) matrix elements of multi-quark operators between hadron states. These matrix elements and, more in general, low-energy quantities in QCD can be computed by using the lattice regularisation, the only known definition which allows for first principle calculations through the evaluation of the functional integral with numerical Monte Carlo methods. In order to constrain as much as possible quantities like the CKM matrix elements extracted by combining experimental and lattice QCD quantities, the latter need to reach the same level of accuracy of the former. The goal of the present proposal is to perform precision continuum limit extrapolations (within 1% of accuracy) of phenomenologically relevant quatities, namely The QCD Λ-parameter, the strange and charm quark masses, the decay constant of the Ds meson and the matrix elements of the complete basis of Delta S = 2 four-quark operators which are relevant to study neutral Kaon mixing. In order to have complete control over systematic uncertainties we will simulate at quark masses small enough and volumes large enough to keep the chiral extrapolation and the finite volume effects under control. Small lattice spacings are needed to safely control the continuum limit extrapolation. In the last 10 years, there have been tremendous achievements for Monte Carlo algorithms. However, an important systematic effect underestimated until now is the entrapment of simulations in topological charge sectors of the field space in the vicinity of the continuum limit. An important element of the project consists in the novel treatment of this problem which will allow us to simulate lattices with three dynamical sea quarks (two degenerate light quarks, corresponding to pion masses in the range 260-310 MeV, and the strange one) down to lattice spacing of 0.04 fm, i.e. the finest lattice spacing ever simulated with dynamical quarks on large volumes. Reliable computation of physical observables at such small lattice spacings is achieved by using state of the art algorithms, suitably optimized for the computer architectures belonging to PRACE. We expect that, with most systematic effects carefully under control, the impact of our work will be that of providing benchmark estimates of important physical quantities which crucially depend on the correct approach to the continuum limit.


Amplification and propagation of intense tailored laser pulses in underdense plasmas

Project Title: Amplification and propagation of intense tailored laser pulses in underdense plasmas
Project Leader:Luis Silva, GoLP/IPFN – Instituto Superior Tecnico, Portugal
Resource Awarded: 69 000 000 core hours on FERMI


Eduardo Paulo Alves, GoLP/IPFN – Instituto Superior Tecnico-Portugal
Ricardo Fonseca, GoLP/IPFN – Instituto Superior Tecnico-Portugal
Luis Silva, GoLP/IPFN – Instituto Superior Tecnico-Portugal
Jorge Vieira, GoLP/IPFN – Instituto Superior Tecnico-Portugal
Marija Vranic, GoLP/IPFN – Instituto Superior Tecnico-Portugal
Raoul Trines, STFC Rutherford Appleton Laboratory- United Kingdom
Peter Norreys, University of Oxford- United Kingdom

This proposal leverages on the exploratory results previously obtained, closely linked with intense (intensities above 10^18 W/cm2) laser-plasma interactions, to explore how specially tailored pulses (with orbital angular momentum, or tailored intensity profiles e.g. sequences of trailing pulses) interact with plasmas and determine their potential impact on a wide range of scientific and technological applications. For instance, ultra-intense lasers can be used to excite relativistic shocks, providing a laboratory surrogate for the study of cosmic rays acceleration mechanisms. Intense lasers can also excite ultra-relativistic plasma waves that can accelerate electrons and positrons to very high energies in distances that are up to three orders of magnitude smaller than conventional accelerators. Moreover, the interaction between ultra-short and ultra-intense lasers with ultra-relativistic charged particle bunches can provide a test-bed in the laboratory to solve long standing mysteries associated with radiation reaction and non-linear QED effects. Despite the recent advances in laser technology, the generation of ultra-intense and ultra-short lasers with tailored profiles for many applications is still challenging, and the physics of the interaction of these exotic pulses still unexplored. It is therefore important to explore novel mechanisms capable to generate and amplify lasers beyond the intensity frontiers. We will explore very promising laser amplification regimes based on laser-plasma interactions. We will investigate Raman backscattering amplification, which uses a long 10 picosecond, low intensity pump laser pulse to amplify a shorter 10 femtosecond counter-propagating seed laser. Energy transfer from the pump to the seed occurs through an electron plasma wave. In addition, we will also study laser amplification via Brillouin backscattering, where the coupling between the seed and pump lasers is provided by an ion-acoustic plasma wave. The generation of ultra-high intensity lasers is also critical in the broad field of high energy density science, being particularly relevant to fast ignition. Fast ignition resorts to intense laser-matter interactions to achieve a fusion chain reaction capable to realise large amounts of energy. One of the outstanding challenges associated with fast ignition is to identify ideal conditions ensuring stable, instability free propagation until reaching the fast ignition fuel core. We will study these processes in order to find optimal regimes of energy transfer from specially designed intense laser pulses through a solid density and compressed plasma target where fusion reactions can occur. The focus of our research is thus to understand the physics of long laser propagation in plasmas driven by specially designed laser pulses. The relevant temporal and spatial scales associated with these scenarios are very disparate. Thus, full-scale, massively parallel kinetic particle-in-cell simulations are critical to address the challenges outlined in this proposal. This research proposal will take advantage of the outstanding numerical and visualization frameworks in our group. The unique computational infrastructures provided by PRACE and MareNostrum will be critical to explore some of the most exciting fundamental physics questions at the forefront of science identified in this proposal.



Mathematics and Computer Sciences (1)

EXA2CT: EXascale Algorithms and Advanced Computational Techniques

Project Title: EXA2CT: EXascale Algorithms and Advanced Computational Techniques
Project Leader:Tom Vander Aa, imec, Belgium
Resource Awarded: 5 000 000 core hours on FERMI, 5 000 000 core hours on MareNostrum


Imen Chakroun, imec-Belgium
Bram Reps, Universiteit Antwerpen-v
Omar Awile, Intel- France
Thomas Guillet, Intel- France
Chris Goodyer, NAG- United Kingdom

The EU-funded project EXA2CT (EU funding ref.no. 610741) brings together experts at the cutting edge of the development of numerical solvers, and HPC software architects for programming models and communication with the aim to develop parallel programming models and numerical solvers suitable for exascale computers. The proposed frameworks, programming techniques and numerical algorithms will be showcased on a number of open source proto-applications derived from real-world numerical HPC applications that are today developed in the European research community as well as in industry. EXA2CT is part of a wider European effort to resolve the challenges the exascale future still holds. The project has received in October 2014 a positive evaluation on the quality of the achieved results and the degree of collaboration between the involved partners on its first yearly review. The project is so far on time and the allocation of resources is coherent with the work performed.



Universe Sciences (3)

Magnetism, buoyancy, and near-surface shear in simulations of stellar convection

Project Title: Magnetism, buoyancy, and near-surface shear in simulations of stellar convection
Project Leader:Matthew Browning, University of Exeter,  United Kingdom
Resource Awarded: 19 600 000 core hours on FERMI, 9 000 000 core hours on MareNostrum


Laura Currie, University of Exeter- United Kingdom
Lucia Duarte, University of Exeter- United Kingdom
Lewis Ireland, University of Exeter- United Kingdom
Felix Sainsbury-Martinez, University of Exeter- United Kingdom
Maria Weber, University of Exeter- United Kingdom
Kyle Augustson, National Center for Atmospheric Research- United Kingdom
Mark Miesch, National Center for Atmospheric Research- United Kingdom
Nick Featherstone, University of Colorado- United Kingdom

Convection and magnetism play essential roles in the structure and evolution of stars and planets. All stars transport some of their energy by convective motions, as do many substellar objects (brown dwarfs and gaseous giant planets). Magnetism is likewise common to most of these objects, and the effects of those magnetic fields are far-reaching: for example, they play a crucial role in governing the rotational evolution of many stars, and likely impact the habitability of extrasolar planets. In most stars and planets, the magnetic fields are thought to be built by the convection itself, acting as a dynamo that can amplify seed magnetic fields and sustain them against Ohmic decay. But an understanding of how exactly this occurs has remained elusive. We still have no truly predictive theory for how the strength, geometry, or variability of stellar and planetary magnetism depends on basic parameters like mass, rotation rate, and age. Many useful insights into the dynamo process have come from numerical simulations. But the global simulations performed to date operate in parameter regimes far from those that prevail in stars, and lack (or understate) several key “ingredients” that likely play roles in generating the observed magnetism. Motivated by these puzzles, we aim here to build sophisticated 3-D models of the magnetism of low-mass stars and substellar objects. As Task A, we will carry out 3-D magnetohydrodynamic (MHD) simulations of convective dynamos in a global spherical geometry, modeling stars and substellar objects with unprecedented realism. As Task B, we will conduct more localized calculations of the near-surface convective layers, which involve faster, smaller-scale flows whose detailed properties cannot readily be explored in the global models.


Multi-scale simulations of Cosmic Reionization

Project Title: Multi-scale simulations of Cosmic Reionization
Project Leader:Ilian Iliev, University of Sussex,  United Kingdom
Resource Awarded: 14 000 000 core hours on MareNostrum


Romain Teyssier, University of Zurich- Switzerland
Anastasia Fialkov, Ecole normale superiore- France
Dominique Aubert, Observatoire Astronomique de Strasbourg- France
Pierre Ocvirk, Observatoire Astronomique de Strasbourg- France
Kyungjin Ahn, Chosun University- South Korea
REPUBLIC OF Karl Joakim Rosdahl, Leiden University- The Netherlands
Garrelt Mellema, Stockholm University- Sweden
Keri Dixon, University of Sussex- United Kingdom
Hannah Ross, University of Sussex- United Kingdom
Chaichalit Srisawat, University of Sussex- United Kingdom
David Sullivan, University of Sussex- United Kingdom
Peter Thomas, University of Sussex- United Kingdom
Jun-Hwan Choi, The University of Texas at Austin- United States
Paul Shapiro, The University of Texas at Austin- United States

The first billion years of cosmic evolution are one of the last largely uncharted territories in astrophysics. During this key period the cosmic web of structures we see today first started taking shape and the very first stars and galaxies formed. The radiation from these first galaxies started the process of cosmic reionization, which eventually ionized and heated the entire universe, in which state it remains today. This process had profound effects on the formation of cosmic structures and has left a lasting impression on them. This reionization process is inherently multi-scale. It is generally believed to be driven by stellar radiation from low-mass galaxies, which cluster on large scales and collectively create very large ionized patches whose eventual overlap completes the process. The star formation inside such galaxies is strongly affected by complex radiative and hydrodynamic feedback effects, including ionizing and non-ionizing UV radiation, shock waves, gas cooling and heating, stellar winds and enrichment by heavy elements. Understanding the nature of the first galaxies and how they affect the progress, properties and duration of the cosmic reionization requires detailed modelling of these complex interactions.The aim of this project is to combine a unique set of simulations of cosmic reionization covering the full range of relevant scales, from very small, sub-galactic scales, for studying the detailed physics of radiative feedback, all the way to very large cosmological volumes at which the direct observations will be done. These simulations will be bases on several state-of-the-art numerical tools, including Adaptive Mesh Refinement (AMR) techniques for achieving very large dynamic range in radiative hydrodynamics calculations (RAMSES-RT code), and a massively-parallel, highly numerically efficient radiative transfer method for accurate modelling at large scales (C2-Ray). We will complement the numerical simulations with semi-analytical galaxy formation modelling to explore the large parameter space available, to improve the treatment of reionizing sources in large-scale radiative transfer simulations as well as to derive detailed observational features of the first galaxies in different observational bands. The questions we will address are: 1) how do the radiative feedback from the First Stars hosted in cosmological minihaloes and dwarf galaxies affect the formation of early structures and subsequent star formation?; 2) how much does high-redshift galaxy formation differ from the one at the present day? What are the observational signatures of the first galaxies? 3) how important is the recently pointed out effect of local modulation of the star formation in minihaloes due to differential supersonic drift velocities between baryons and dark matter?; 4) how does the metal enrichment and the transition from Pop III (metal-free) to Pop II stars occur locally and how is this reflected in the metallicity distribution of the observed dwarf galaxies and globular clusters? and 5) How are these feedback effects imprinted on large-scale observational features?


PFMPD – Planetesimal Formation in Magnetized Protoplanetary Disks

Project Title: PFMPD – Planetesimal Formation in Magnetized Protoplanetary Disks
Project Leader:Chao-Chin Yang, Lund University, Sweden
Resource Awarded: 15 000 000 core hours on MareNostrum


Bertram Bitsch, Lund University-Sweden
Daniel Carrera, Lund University-Sweden
Karl Jansson, Lund University-Sweden
Anders Johansen, Lund University-Sweden
Michiel Lambrechts, Lund University-Sweden
Katrin Ros, Lund University-Sweden
Mordecai-Mark Mac Low, American Museum of Natural History- United States

As of now, more than 1900 planets have been detected outside of our own solar system. However, a comprehensive picture of how they were born remains lacking. The reason is that the process of planet formation involves complicated interaction between solid materials, gaseous medium, and magnetic fields. To study this process, numerical simulations on high-performance computing facilities are required. In this research project, we focus on studying how kilometer-sized celestial objects called planetesimals were formed in such a complex environment. We will use a public computer program called the Pencil Code to simulate a system of numerous pebbles and boulders moving inside a magnetized gas disk. We will observe how these solid bodies can be concentrated to form larger bodies by a mechanism called the streaming instability. To achieve this goal, we have implemented several state-of-the-art numerical techniques for the Pencil Code. These new implementations will help us produce a pilot study of planetesimal formation in a magnetized protoplanetary disk.