DECI 10th Call

Find below the results of DECI-10 (Distributed European Computing Initiative) call.

Projects from the following research areas:

 

 

Astro Sciences (6)

CONVDYN13

Project Title: Convection-driven dynamos in rapidly rotating late-type stars
Project Leader: Petri Käpylä, University of Helsinki, Finland
Resource Awarded: 5 419 008 core hours on RZG – Hydra
Details

Collaborators:
Axel Brandenburg – AlbaNova University Center, Nordita, Sweden
Elizabeth Cole – University of Helsinki, Finland
Maarit Mantere – University of Helsinki, Finland
Jörn Warnecke – AlbaNova University Center, Nordita, Sweden
Abstract:
The magnetic field of the Sun and other similar stars is thought to be generated by a dynamo in the outer layers of the star where energy is transported by fluid motions. In the Sun the global magnetic field is manifested by sunspots which are strong concentrations of magnetic field. The amount of sunspots varies cyclically with a period of roughly 11 years. In the course of a cycle spots first appear about 40 degrees away from the equator. As the cycle progresses, spots appear closer and closer to the equator. Such behavior has also been reproduced by numerical simulations recently.

The Sun is now 5 billion years old and its rotation period is roughly one month. In its youth, the solar rotation was much faster than what it is now. The reason for the deceleration are the magnetic fields: the Sun has lost much of its angular momentum via the solar wind. In the early days also the global magnetic field of the Sun was very different from what it is today: observations of young solar-like stars suggest that magnetic fields are much more non-axisymmetric and that the spots appear at higher latitudes and are much larger than in the present-day Sun.

We study the dynamos in rapidly rotating stars by numerical simulations that cover rotation rates well above the solar values, probing the Sun during its first billion years.

Dissipative_Phenomena

Project Title: Dissipative phenomena on Local Group dwarf galaxies
Project Leader: Dr. Stefano Pasetto, University College London, London, UK
Resource Awarded: 686 400 core hours on EPCC – Archer and EPCC – Blue Joule
Details

Collaborators:
Prof. Dr. Emeritus Cesare Chiosi – University of Padova, Physics & Astronomy, Padova, Italy
Prof. Yutaka Fujita – Osaka University, Earth and Space Science, Osaka, Japan
Prof. Dr. Eva Grebel – University of Heidelberg, Germany
Dr. Daisuke Kawata – University College London, London, UK
Abstract:
Dwarf galaxies are the most common type of galaxies in the universe and dominate in number the population of galaxies around our Galaxy, i.e. the Local Group (LG). They result from a complex interplay between dark-matter-driven hierarchical structure formation and feedback-controlled conversion of the baryonic matter into stars. Within the cosmological paradigm, we know that dwarf galaxies play a central role in interacting with a major galaxy. This results in a large number of bodies penetrating or navigating throughout the halo of a larger galaxy such as e.g., is the case for our galaxy, the Milky Way (MW) and its orbiting satellites. In their orbits these satellites are then subject to environmental effects such as tidal forces and ram pressure, and instabilities processes that rule their star formation history. The purpose is this study is to investigate the interplay between environment and internal processes in the star formation history of dwarf galaxies neighbouring a hosting system. In particular our goal is to explore the environmental effects on the formation and evolution of dwarf galaxies in relation with the properties of the hosting environment.

GalChem

Project Title: Galactic Chemodynamics in the Era of Gaia
Project Leader: Brad Gibson, University of Central Lancashire, Preston, UK
Resource Awarded: 5 940 000 core hours on SURFSARA – Cartesius
Details

Collaborators:
Dr. Christopher Few – University of Exeter, School of Physics, Exeter, UK
Dr. Daisuke Kawata – University College London, London, UK
Dr. Chiaki Kobayashi – University of Herdfortshire, Centre for Astrophysics Research, Hatfield, UK
Abstract:
We propose to run two families of simulations of Milky Way-type systems, as a pathway to the goal of more realistic Milky Way simulations; the codes employed are our particle-based GCD+ and our new chemistry-enhanced, grid-based code, RAMSES-CH. One family is based upon “zoom” style cosmological simulations, for which we will run 8 simulations with different merger histories and environments, and explore how these characteristics impact upon the formation and evolution of each galaxy’s sub-components, including their stellar and gaseous discs, bulge, and stellar and hot coronal halo. The other family is a controlled simulation of the Galactic disc; using high resolution simulations. We will study the formation processes of the detailed structures in the Milky Way-type galaxies such as spiral arms and their effects on gas and stellar dynamics and star formation. Our codes are unique in terms of their self-consistent treatment of chemical evolution in non-linear dynamical simulations, which allow them to predict the distribution of elements in stars and gas as a function of position, kinematics, and age, allowing their comparison directly with observations. The associated chemical information will allow to self-consistently generate mock observational data, using our stellar population synthesis package, linking directly to the anticipated returns from future missions and experiments, such as ESA’s Gaia and the Gaia-ESO Survey, and directly compare with the observed kinematical and chemical properties of stars in the Milky Way. The mock observational data and simulated outputs will be made publicly available.

Galsim

Project Title: Simulations of galaxies, feedback and the intergalactic medium
Project Leader: Dr. Ilian Iliev, University of Sussex, Physics and Astronomy, Sussex, UK
Resource Awarded: 10 692 000 core hours on CSC – Sisu
Details

Collaborators:
Ms. Charlotte Clarke – University of Sussex, Physics and Astronomy, Sussex, UK
Dr. Keri Dixon – University of Sussex, Physics and Astronomy, Sussex, UK
Dr. Aurel Schneider – University of Sussex, Physics and Astronomy, Sussex, UK
Prof. Peter Thomas – University of Sussex, Physics and Astronomy, Sussex, UK
Abstract:
The understanding of the formation and clustering properties of galaxies is one of the key problems in modern cosmology. The advent of very large, sensitive, high-resolution radio interferometer arrays like the Pan-European Low Frequency Array (LOFAR), ALMA, Square Kilometer Array (SKA) and its pathfinders (e.g., ASCAP in Australia), as well as ground and space-based facilities like Hershel, the Dark Energy Survey (DES) and the Euclid satellite is expected to revolutionize our knowledge of how galaxies form and evolve and of their relations to the Cosmic Web of structures. The correct interpretation of such observations requires detailed numerical simulations. The main goal of this project is to produce detailed, simulated sky data based on a suite of large simulations, mainly geared towards radio interferometer surveys like LOFAR, ALMA, SKA and its pathfinders. Additionally, it consists of an important step towards the prediction/simulation of the Euclid data, which will be a major challenge for computational cosmology in the next decade. We are members of the Euclid Simulations Working Group and within it have been identified as one of the very few groups capable of carrying out the trillion-particle (1012) large-scale, N-body simulations needed for proper modelling for that mission. Due to the complex nature of the formation of galaxies, we will combine extremely large N-body simulations of dark matter with semi-analytic galaxy modelling. As a sub-project we will also perform detailed radiative hydrodynamic simulations of the effect of radiative and supernova feedback on the cold gas fraction in galaxies using the Adaptive Mesh Refinement (AMR) code Enzo. The results of these simulations will be used to inform and improve our semi-analytical models of the neutral hydrogen and molecular hydrogen fractions in galaxies. In this way we will able to track the enormous range of scale covered by new generation surveys and produce a complete picture of the formation of galaxies and its components within the dark matter dominated cosmic web. The detailed galactic structure and properties depends also on the nature of dark matter, which currently is one of the major unknowns of physics. Cosmic structure formation is potentially capable of tightly constraining the parameter space of possible dark matter candidates. We will run simulations of different dark matter models, measuring quantities like the halo abundance, halo profiles and subhalo distributions, which are key measures to extract the nature of dark matter. Applying the semi-analytical modelling will then give us the possibility to make concrete predictions for upcoming galaxy surveys, which will lead to better constraints on the mass of the dark matter particle.

HYDRAD

Project Title: Hydrodynamic stability of rotating flows in accretion disks
Project Leader: Dr. Bjoern Hof, Max Planck Institute for Dynamics and Self-Organization, Complex Dynamics and Turbulence, Germany
Resource Awarded: 6 063 998 core hours on RZG – Hydra and VSB-TUO – Anselm
Details

Collaborators:
Prof. Marc Avila – FAU Erlangen-Nürnberg, Institute of Fluid Mechanics, Germany
Dr. Markus Rampp – RZG – Computing Centre of the Max-Planck-Society, Garching, Germany
Liang Shi – Max Planck Institute for Dynamics and Self-Organization, Complex Dynamics and Turbulence, Germany
Abstract:
The origin of turbulence in accretion discs has been debated for many decades. In order for matter to accrete, angular momentum has to be transported outward and as molecular viscosity is too small a turbulent viscosity has been put forward to explain observable accretion rates. The velocity profiles of discs (Keplerien profiles) are centrifugally stable and therefore a different instability mechanism is required for turbulence to arise. While in hot discs turbulence can arise through the magnetorotational instability, cooler discs lack sufficient ionization and it is unclear how turbulence sets in. In analogy to pipe and other shear flows it has often been argued that turbulence in disc flows could also be triggered by perturbations of finite amplitude. State-of-the-art laboratory experiments by different groups yield contradictory results and call for a numerical investigation. Here we perform numerical simulations to clarify if such a subcritical instability mechanism may be responsible for turbulence in Keplerian flows at high Reynolds numbers.

PLANETESIM-2

Project Title: Towards an initial mass function of planetesimals
Project Leader: Dr. Anders Johansen, Lund University, Astronomy and Theoretical Physics, Lund, Sweden
Resource Awarded: 7 500 000 core hours on FZJ – JuRoPA
Details

Abstract:
The planets of the solar system and the exoplanets – planets that orbit stars other than the sun – are a fascinating research area. Fuelled by new detection methods that find more and more planets around other stars, satellite missions to other planets, moons and asteroids in our solar system, and the ever growing power of supercomputers, planet research is in rapid development and enjoys lots of interest from the broad public. The field is closely tied with the perhaps most fundamental question of all: how common is life in the universe? – a question pursued by astronomers, geologists, biologists, physicists and chemists in concert. Planets form in discs of gas and dust orbiting young stars as dust and ice particles collide and grow to ever larger bodies. An important stage in the planet formation process is the formation of km-scale planetesimals. Planetesimals are building blocks of both terrestrial planets like the Earth and of the solid cores of gas giant planets such as Jupiter. A fundamental problem is that cm-sized pebbles do not stick when they collide. Supercomputer simulations performed by members of our group have identified a surprising phenomenon that allows growth from pebbles to planetesimals: pebble-sized particles concentrate in dense filaments that protect them from gas drag, in a process related to why bicycle riders and migrating geese travel in groups. The densities of pebbles gets so high that gravity takes over and leads to gravitational collapse to form planetesimals. The aim of this research project is to use high-resolution computer simulations to understand the birth sizes of planetesimals. The asteroid belt between Mars and Jupiter and the Kuiper belt beyond Neptune are examples of planetesimal belts left over from the planet formation process. The largest asteroids and Kuiper belt objects have sizes that are similar to the largest planetesimals that form in the computer simulations, but an important feature of both these populations is that the size distribution of the planetesimals show a break around 50 km in radius. This has been dubbed the missing intermediate-sized planetesimals problem. Previously we have in our computer simulations only been able to form the largest planetesimals (with radii of 150-1500 km) from overdense filaments of pebbles. Small planetesimals form from small-scale particle overdensities and hence it requires very high resolution simulations to model their formation. The computational resources granted by PRACE in May 2012 has allowed us to run for the first time 2563 simulations of planetesimal formation on some of Europe’s most powerful supercomputers. This work is making good progress and has already produced exciting preliminary results (see detailed description at the end of the proposal). Here we ask for an additional 2,500,000 CPU hours to push to the crucial 5123 resolution that will allow us to determine the initial mass function of planetesimals down to 30 km in radius and compare critically to the observed properties of the Kuiper belt and the asteroid belt.
top

Bio Sciences (7)

APOP20X3

Project Title: Rational Design of new antagonists of XIAP & Survivin receptors involved in apoptosis.
Project Leader: Dr. Fernando Blanco, Trinity College Dublin, Dublin, Ireland
Resource Awarded: 1 564 984 core hours on PSNC – Cane
Details

Abstract:
Apoptosis is the process of programmed cell death (PCD) that may occur in multicellular organisms. Biochemical events lead to characteristic cell changes (morphology) and death. These changes include blebbing, cell shrinkage, nuclear fragmentation, chromatin condensation, and chromosomal DNA fragmentation. Briefly, the diseases in which the implication of apoptosis mechanisms has been described fall into two groups: diseases in which there is an increase in cell survival and diseases in which there is an increase in cell death. There are two signaling pathways that initiate the apoptotic suicide program; the intrinsic and extrinsic cellular mechanisms, which are not completely independent of each other (Figure 1). However, regardless of the mechanism by which this process of death occurs, both converge to some activation of intracellular proteins called caspases, involved both in the initiation and execution of apoptosis. The intrinsic mechanism takes place in the mitochondria, the metabolic bioenergetic centre of eukaryotic cells, and is regulated by the family of Bcl-2 proteins and involves, in addition to other proteins, the IAP family (Inhibitors of Apoptosis Proteins). The starting point of this research project is the search of new modulators of particular targets involved in apoptosis. Specifically the research will focus on the families of XIAP and Survivin inhibitors, as well as the regulating family of Bcl-2 proteins. The use of computational methods, like molecular dynamics (AMBER software), will provide the theoretical support to the rational design of new potential candidates. The analysis of protein-ligand interaction energies by using MMPBSA protocol implemented in AMBER software will be employed to establish a ranking of the best pairs ligand/pose to be used for the synthesis/purchase of series of compounds to evaluate biologically by experimental collaborator groups.

fplb

Project Title: Interaction of a homologous series of fluorescent probes with different lipid bilayers
Project Leader: Dr. Luís Loura, Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Resource Awarded: 4 216 934 core hours on WCSS – Supernova
Details

Collaborators:
Dr. Hugo Filipe – Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Dr. Maria Joao Moreno Silvestre – Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Abstract:
The ability of molecules to cross tight endothelia such as the blood-brain barrier (BBB) is a major drawback for the discovery of new drugs to treat Central Nervous Systems (CNS) disorders, with important social and economical implications. Following this subject, the kinetic and thermodynamic characterization of the interaction of amphiphiles with lipid bilayers is important to predict the interaction of amphiphilic drugs with biological membranes, a property that determines their pharmacokinetics and bioavailability. According to Overton’s rule, a positive linear dependence of the permeability coefficient through a cell monolayer with the partition coefficient is predicted, assuming the diffusion through the lipid bilayer as the rate limiting step. For amphiphilic drugs, this diffusion step corresponds to their translocation across the bilayer. Therefore, the determination of the translocation rate constant and the partition coefficient of a drug into a lipid bilayer may be used to determine its rate of passive permeation. Compositional asymmetry across the two bilayer leaflets is a striking feature in many eukaryotic plasma membranes. Sphingolipids are almost exclusively distributed in the exoplasmic leaflet, whereas the cytoplasmic leaflet contains mainly glycerophospholipids; cholesterol (Chol) being equilibrated between both. However, the experimental preparation of stable asymmetric membranes is very difficult, which makes theoretical work very important in this kind of systems. Molecular Dynamics (MD) is a powerful tool to study the interaction of amphiphiles with lipid bilayers as it gives atomistic details that often cannot be obtained experimentally. Additionally, the rates of translocation of phospholipids (in the order of hours to days), or cholesterol (in the order of seconds) are much longer than a typical MD simulation time scale, maintaining the lipid asymmetry stable during the sampling process. To gain predictive power for the permeation of amphiphilic drugs, the parameters obtained must be understood in terms of the interactions established. This may be achieved performing equilibrium MD and Umbrella Sampling to calculate the Potential of Mean Force (PMF). The PMF gives the free energy profiles of the amphiphiles at different depths in the bilayer allowing the calculation of the partition coefficient and the rate of translocation. Here we propose to calculate the PMF for the interaction of a homologous series fatty amines (NBD-Cn, n=4 to 16) in lipid bilayers with different compositions. In order to keep simplicity we mimic the cellular external monolayer with palmitoylsphingomyelin (SpM):Chol(6:4) and the inner monolayer with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC):Chol(1:1). Currently, we are also proceeding with the experimental characterization of the interaction of NBD-Cn with bilayers of different lipid compositions, POPC (liquid-disordered state), POPC:Chol and SpM:Chol (liquid-ordered state). The comparison between simulated and experimental results, in the case of symmetric membranes, allows extrapolation for experimentally inaccessible amphiphiles. Moreover, the atomistic detail provided by the MD simulations will give very important information regarding the effect of the asymmetry on the kinetic processes compared to the behavior in the symmetric bilayers.

HIV1-GSL

Project Title: Functional roles of glycosphingolipids in modulating entry of human immunodeficiency virus type 1 (HIV-1) into host cell
Project Leader: Ilpo Vattulainen, Tampere University of Technology, Finland
Resource Awarded: 7 000 000 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Clifford Lingwood – Hospital for Sick Children, Toronto, Canada
Abstract:
Acquired immunodeficiency syndrome (AIDS) is a sexually transmitted disease, caused by the human immunodeficiency virus (HIV). The virus severely damages the immune system leaving the individual susceptible to life-threatening infections and cancers. For productive infection, HIV penetrates the plasma membrane to deliver its genomic RNA into the target cells.The entry of HIV-1 requires the sequential interaction of the viral exterior envelope glycoprotein gp120, with its primary receptor CD4 and a chemokine co-receptor on the host cell surface. These interactions initiate a fusion of the viral and cellular membranes, resulting in entry of the viral nucleocapsid into the host cell.

There are significant evidences that the lipid composition of the target membrane can influence the efficiency of virus-cell fusion. Lipid rafts primarily consisting of glycosphingolipids (GSLs), sphingomyelin and cholesterol have been implicated for the infectious route of HIV-1 entry. GSLs may enhance the fusion process by serving as direct binding sites for the envelope glycoprotein gp120 on the cell surface; they may promote the clustering of primary and co-receptors and regulate their lateral mobility in the target cell membrane; finally they may also trigger the conformational changes in viral spike required for membrane fusion. Even for non-CD4 expressing cells, GSLs provide potential alternative means of viral host-cell entry. In addition to the role of GSLs as facilitators of HIV infection, some of them can act as a natural resistance factor for HIV prevention.

HIV has proven to the one of the most difficult pathogen to overcome. Currently, there are no effective vaccines that can provide specific and long lasting immunity to the virus. Despite huge efforts to develop potential therapeutics, the spread of HIV continues to increase worldwide. Targeting and manipulating host cell lipids is recently emerging as a promising approach to prevent HIV infection, based on the role of raft-specific-lipids in HIV biology. However, the exact mechanism by which GSLs support viral fusion remains to be determined.

In this study we will perform atomistic molecular dynamics simulations and free energy calculations to understand the complex biology of GSL-mediated host cell entry of HIV-1. The research will be carried out in close collaboration with experimental partners, the focus being to uncover the atomistic details of GSL-gp120 interactions which is crucial to initiate viral entry, and the objective being to unlock the mechanism(s) by which GSLs promote or prevent HIV-1 infection. The project will not only promote our current state-of-the-art understanding about this epidemic disease, but it will also likely provide clues for its prevention.

INPHARMA

Project Title: Integrating molecular dynamics simulations and NMR-based re-scoring for protein-ligand docking
Project Leader: Dr. Teresa Carlomagno, European Molecular Biology Laboratory (EMBL), Structural and Computational Unit, Heidelberg, Germany
Resource Awarded: 750 000 core hours on EPCC – HeCToR XE6
Details

Collaborators:
Dr. Lars Skjaerven – European Molecular Biology Laboratory (EMBL), Structural and Computational Unit, Heidelberg, Germany
Abstract:
Protein-ligand docking is an important computational tool in structure based drug design. It allows for rapid prediction of the binding orientation of small molecules to a target protein. The presence of a reliable protein-ligand model facilitates systematic optimization of lead compounds to high affinity drug candidates. However, current docking protocols suffer from a low success rate due to deficiencies in the ranking of the ensemble of docking orientations (referred to as docking modes). Moreover, the inability to properly account for protein flexibility inflicts additional challenge to obtain a reliable model for arbitrary protein ligand pairs. While moderate protein flexibility (i.e. side-chain rotations) can be modeled on-the-fly during the docking experiment, more extensive backbone and domain rearrangements are generally not accessible during conventional docking protocols. In this respect, molecular dynamics (MD) simulations have an acknowledged capability to model protein flexibility in a realistic solvated environment, but also in modeling ligand-induced effects. While being computationally exhaustive, the key benefit of employing MD is the fully flexible model – i.e. all atoms are allowed to move in accordance to inter-atomic interactions governed by detailed empiric force fields. A step-by-step propagator is applied to provide a time evolution of the individual atom motions allowing for an explicit evaluation of protein dynamics. The conformational diversity obtained from an MD simulation can be used in a subsequent docking experiment, referred to as ensemble docking. This approach has shown remarkable results in sampling the correct binding mode, which would otherwise not be sampled by docking to a single static structure. However, as for conventional docking, the major barrier remains a robust scoring function. We have recently developed a new NMR-based technique that facilitates re-scoring, based solely on experimental data, of binding modes generated by protein-ligand docking. The approach, entitled INPHARMA (Interligand NOEs for PHARmacophore Mapping), has shown promising results in correctly discriminating between docking modes of a number of ligands for protein kinase A (PKA) (Orts et al, 2008; and unpublished data). The INPHARMA approach exploits NOEs occurring between two ligands binding in a fast exchange to the same target protein making it attractive for fragment-based drug discovery. These interligand INPHARMA-NOEs results from a transfer of magnetization mediated by the receptor protein. INPHARMA-NOEs can be theoretically estimated for pairs of protein-ligand complexes generated by computational docking and subsequently compared to the experimental NOEs providing means to discriminate between docking conformers. In this project, we aim at combining a fully flexible model of the target protein by MD simulations with INPHARMA-guided rescoring of docking modes to overcome docking deficiencies in scenarios where structural data is limited. Specifically, when (i) only an apo structure is available, or (ii) when the structural data derives from a low quality homology model. While we have so far successfully benchmarked our INPHARMA-based rescoring protocol for several docking scenarios to PKA, including an induced-fit scenario with domain rearrangements upon ligand binding (manuscript in preparation), we now extend our research to new target proteins including G-protein coupled receptors (GPCRs) and cyclin-dependent kinase 2 (CDK2) — both being of major pharmaceutical interest. While experimental data is currently being collected for sets of ligands in our lab, we are in the need of extensive computational power to employ MD simulations of the relevant targets for subsequent INPHARMA-guided ensemble docking. The computational projects proposed here, is a vital component in the proposed protocol, which should be applicable to a wide range of systems.

NANODROPS

Project Title: Interaction of hydrophobic molecules: triglycerides and nanoparticles with lipid biomembranes
Project Leader: Himanshu Khandelia, University of Southern Denmark, Odense, Denmark
Resource Awarded: 10 141 200 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Luca Monticelli – INSERM, Paris, France
Julian Shillcock – EPFL, Lausanne, Switzerland
Abstract:
The goals of this project are to:
Understand the Mechanism of formation of lipid droplets (LD) in lipid membranes
Understand the interaction of nanoparticles with lipid membranes.
The common thread in the two sub-projects is the overarching goal of understanding the interaction and aggregation behavior of hydrophobic nano-sized particles in lipid membranes.
Cylindrically shaped lipids self-assemble into flat membranes, which are 4 nm thick, and comprise of a hydrophobic core shielded from water by hydrophilic lipid headgroups. The interaction of external agents with lipids therefore depends on the agents’ size, shape and amphipathic nature. Hydrophobic molecules such as triglyceride (TGLs), and fullerene nanoparticles partition readily into lipid membranes, but not being cylindrical, do not align with the hydrophobic lipids tails. Although computational investigations of the interactions of amphipathic and small hydrophobic molecules with membranes is commonplace, the interaction of large hydrophobic molecules or aggregates in membranes is relatively unchartered territory.
The aggregation of TGLs in membranes drives LD formation in the endoplasmic reticulum (ER) of cells. LD formation is a key step in many cellular processes including cholesterol transport, and is also implicated in several diseased states. The mechanism of formation of LDs, and their extrusion mechanism into the cytoplasm remain unknown and heavily disputed. We have previously shown (see references) that TGLs can spontaneously aggregate in flat lipid bilayers as stable discs up to 14 nm in diameter, and this is the form of a nascent lipid droplet. In this project, we will simulate the complete formation and extrusion of the lipid droplet from the lipid bilayer, and will investigate the role and localization of specific proteins or curvature-inducing lipids in the extrusion mechanism.
As industrial and research applications of nanoparticles increase in scale and complexity, it is important to predict the interactions of nanoparticles with biological materials such as phospholipid membranes and vesicles. In particular, fullerenes, which are all-carbon cage molecules with exciting applications in drug delivery and photovoltaic devices, can easily partition into the bilayer, and probably mediate their toxicity via the cell membrane. We will study the interaction of fullerenes and other nanoparticles with lipid membranes, as a function of size, focusing on transport through the membrane and on aggregation behavior. We will also explore how the large-scale fluctuations of membranes can lead to sorting and separation of nanoparticles based on their size, shape and surface properties. Also, how nanoparticles modify membrane fluctuations to generate new steady states that break the unperturbed membrane’s translational symmetry.

PTACRB-2

Project Title: Identifying Proton Pathways in a Multi-Drug-Resistance Membrane Transporter by EVB-MD Simulations
Project Leader: Dr. Jose Faraldo-Gomez, Max Planck Institute of Biophysics, Theoretical Molecular Biophysics Group, Frankfurt, Germany
Resource Awarded: 2 950 000 core hours on CYFRONET – Zeus BigMem and ICHEC – Fionn-thin
Details

Collaborators:
Dr. Claudio Anselmi – Max Planck Institute of Biophysics, Theoretical Molecular Biophysics Group, Frankfurt, Germany
Prof. K. Martin Pos – University of Frankfurt, Institute of Biochemistry, Germany
Prof. Gregory Voth – University of Chicago, Department of Chemistry, USA
Abstract:
The emergence of multi-drug resistance in pathogenic bacteria is a threat to human health on a global scale. Resistance is in part conferred by an array of proteins in the bacterial membranes whose function is to remove a variety of toxic compounds from the cell interior – for example, man-made antibiotics. Novel strategies to inhibit these drug-efflux systems are therefore of great interest from a biomedical standpoint. To be able to do so, however, much remains to be understood about their molecular mechanisms – and in particular about how drug-transport is energized.
Multi-drug resistance transporters have been identified in several membrane protein subfamilies. For example, the main drug-efflux pump in Escherichia coli, termed AcrB, is a member of the so-called RND family. Like other RND pumps, AcrB is a trimeric protein, and functions through an intricate conformational mechanism that entails remote allosteric effects within each protomer as well as cross-talk between them. Importantly, to power this complex mechanism AcrB harvests the energy stored in the membrane in the form of an electrochemical gradient of H+. That is, AcrB facilitates the downhill passage of H+ across the membrane, through a structural mechanism that is tightly coupled to its pumping activity.
Previous large-scale molecular dynamics simulations in our group, based on a novel high-resolution crystal structure of AcrB, have revealed a number of water channels within its transmembrane domain. Importantly, these channels are redefined in each of the states adopted by the transporter during its conformational cycle, and are consistent with the concept of alternating-access, which is key for tight conformational coupling. However, whether or not these channels are actual proton pathways remain to be demonstrated.
We now plan to elucidate this question, by examining the energetics of H+ passage along each of the water channels identified thus far. To do so, we will calculate the potential of mean force associated with H+ movement, via a systematic series of umbrella-sampling molecular dynamics simulations. To be able to simulate the dynamics of the permeating H+, we will use the most up-to-date implementation of the Multi-State Empirical Valence Bond method, in combination with the CHARMM molecular-mechanics force field.
In sum, the proposed calculations will employ state-of-the-art methods to reveal a level of mechanistic insight not yet attained for any H+-driven membrane transporter, and in particular the multi-drug resistance efflux pump AcrB. These investigations will be carried out in parallel with novel experimental work, both at the functional and structural levels.

TransMem

Project Title: Translocation of Biomolecules Across Cell Membranes
Project Leader: Dr. Mikael Lund, Lund University, Department of Theoretical Chemistry, Lund, Sweden
Resource Awarded: 5 200 115 core hours on SURFSARA – Cartesius
Details

Collaborators:
Dr. Pavel Jungwirth – Academy of Sciences of the Czech Republic, Institute of Organic Chemistry and Biochemistry, Prague, Czech Republic
Anil Kurut – Lund University, Department of Theoretical Chemistry, Lund, Sweden
Abstract:
Cell penetrating peptides have drawn considerable attention recently, thanks to their ability to cross phospholipid membranes and deliver various molecular cargos, such as nucleic acids, proteins, quantum dots, and various drugs, inside the cells. A crucial structural requirements for an effective penetration of a peptide through the phospholipid bilayer is the presence of multiple guanidinium cations, which are present in the side chains of arginines. Confocal microscopy and flow cytometry techniques have demonstrated that oligoarginines containing six or more amino acids internalize into the membrane more efficiently than equally long lysine oligomers. However, the molecular mechanism of penetration into and translocation across the phospholiúpid bilayer of these peptides is poorly understood. Among the possible mechanistic explanations inverse micelle formation, electroporation, endocytosis, and anion mediated energy-independent diffusion through the membrane could be considered. Nevertheless, none of these mechanisms is able to fully rationalize the difference in membrane permeability of arginine containing peptides and those containing the other cationic amino acids, i.e., lysines. Within the present proposal we aim at clarifying the molecular mechanisms of membrane permeabilities of different cationic peptides of varying length and composition. We will explore the ability of guanidinium cationic groups present in arginine containing peptides (but not of ammonium groups in oligolysines) to pair via like-charge ion pairing and consequences thereof on membrane permeation. Homo- ion pairing of guanidinium cations is one of the driving forces for oligoarginine aggregation in water. and at the lipid bilayers, as shown by our previous all-atom molecular dynamics simulations. By a combination of all-atom and coarse-grained molecular simulations we will ask and aim to answer the following question: Does the higher charge density of membrane adsorbed oligoarginine aggregates, in comparison to single oligoarginines or oligolysines, play a decisive role in the ability to penetrate across the cellular membrane? Succesfully answering this question by means of extensive simulations will allow us to suggest a plausible and experimentally testable molecular mechanism of translocation of cell penetrating peptides across the cellular membrane. This will have direct consequences for devising new strategies of drug delivery into cells.
top

Earth Sciences (6)

CELESTE

Project Title: Cosmo-Eulag semi-operational testing
Project Leader: Dr. Zbigniew Piotrowski, Institute of Meteorology and Water Management, National Research Institute, Warsaw, Poland
Resource Awarded: 5 500 000 core hours on CSCS – Rosa
Details

Abstract:
It is proposed to perform routine testing of the next generation dynamical core for European high resolution (1 km and better) regional weather prediction. The focus is to provide robust and efficient solver for atmospheric flows over complicated topography of Alps that also poses basic conservative properties. To date, the new EULAG dynamical core of COSMO Consortium for Small Scale Modelling weather forecasting framework has been extensively tested in a range of idealized and semi-realistic experiments. Currently, COSMO runs a project CELO aiming at operationalization of the EULAG dynamical core. The weather forecasting framework of COSMO with EULAG dynamical core will be run daily in close-to-operational configuration to test its robustness and forecast quality. This aims at acceleration of the process of operationalization of the new dynamical core and expedited provision of next generation of weather prediction framework. On the scientific side, the proposal bases on exceptionally powerful and well established EULAG model developed at the National Center of Atmospheric Research in Boulder, CO, USA and the successful COSMO operational weather prediction model.

HyVaMPI

Project Title: Hybrid-Vlasov modelling of plasma instabilities
Project Leader: Dimitry Pokhotelov, Finnish Meteorological Institute, Helsinki, Finland
Resource Awarded: 3 500 000 core hours on UIO – Abel
Details

Collaborators:
Sanni Hoilijoki – Finnish Meteorological Institute, Helsinki, Finland
Ilja Honkonen – Finnish Meteorological Institute, Helsinki, Finland
Yann Kempf – Finnish Meteorological Institute, Helsinki, Finland
Minna Palmroth – Finnish Meteorological Institute, Helsinki, Finland
Sebastian von Alfthan – Finnish Meteorological Institute, Helsinki, Finland
Abstract:
Space environment is a natural laboratory to study the fourth state of matter, the ionised gas known as plasma. Numerical modelling of space plasma is an innovative research field of great scientific interest and of great importance for the society. Plasma dynamics in the Earth’s space environment driven by solar activity creates space weather causing harmful effects on spacecraft and space-based technological solutions for positioning and communications. To facilitate the future numerical forecast of space weather it is necessary to model plasma dynamics on global scales comparable to the Earth’s magnetic environment but also on small scales of plasma particle kinetics. Most innovative approach to space weather simulations is based on the idea of hybrid modelling implemented in the novel plasma code developed at the Finnish Meteorological Institute. The newly developed hybrid code, the Vlasiator, aims to reproduce plasma dynamics self-consistently in a coupled system of solar wind and the Earth’s magnetic field while also allowing to model local processes known as plasma instabilities. The new code, developed in a framework of the Starting Grant from the European Research Council, requires extensive parallel computations and has already demonstrated great scaling performance on large supercomputers including European Tier-0 systems. For the physical verification and further developments of the code it is essential to run a series of simulations replicating various plasma phenomena known to exist in the Earth’s space environment and to compare the code’s predictions to known spacecraft observations and theoretical predictions. Numerical modelling is vital for understanding the fundamental properties of plasma instabilities especially the nonlinear development and instability saturation mechanisms. The most important plasma instabilities playing key role in energy/mass transfer from the solar wind throughout the Earth’s magnetic environment will be addressed in this project.

MOTUS

Project Title: A High-Resolution Modelling Study of the Turkish Straits System Utilizing HPC
Project Leader: Gianmaria Sannino, ENEA, Rome, Italy
Resource Awarded: 2 625 000 core hours on ICM – Boreasz
Details

Collaborators:
Rolf Isele-Holder – RWTH Aachen University, Mechanical Engineering, Aachen, Germany
Prof Emin Ozsoy – Middle East Technical University, Institute of Marine Sciences, Mersin, Turkey
Dr Adil Sozer – Middle East Technical University, Institute of Marine Sciences, Mersin, Turkey
Abstract:
The requirements for hydrodynamic and coupled modeling in the Turkish Straits System (TSS) are reviewed, methods based on advanced modeling techniques addressing strong topographic control and non-linear hydrodynamics in a coastal sea rich with processes are proposed. The scientific questions on the role of the TSS in coupling the adjacent basins of the Mediterranean and Black Seas with highly contrasting properties, in a region of high climatic variability and materials transport depending critically on the cycle of water can only be answered by model predictions of the processes that determine the integral properties of the coupled sub-systems. This can only be achieved if the entire TSS is modeled as a finely resolved integral system that accounts for the high contrasts in seawater properties, steep topography, hydraulic controls, fine and meso-scale turbulence, nonlinear and non-hydrostatic effects, thermodynamic states and an active free-surface in the fullest extent, based on well represented fluid dynamical principles. The extreme environment that needs to be represented as a whole and with the full details of its highly contrasting properties creates a high demand on computational resources which can only be met by supercomputing. Techniques are proposed to evaluate the impact of vertical and horizontal resolution and curvilinear coordinate transformations on predicted properties of the TSS

SPAITAC

Project Title: Seasonal Prediction of the Arctic Ice and Tropical Atlantic Cyclones
Project Leader: Prof. Francisco Doblas-Reyes, Institut Català de Cienciès del Clima, Climate Forecasting Unit (CFU), Barcelona, Spain
Resource Awarded: 5 625 000 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Dr. Louis-Philippe Caron – Stockolm University, Department of Meteorology (MISU), Stockholm, Sweden
Dr. Virginie Guemas – Institut Català de Cienciès del Clima, Climate Forecasting Unit (CFU), Barcelona, Spain
Dr. Colin Jones – Swedish Meteorological and Hydrological Institute (SMHI), Rossby Centre, Norrköping, Sweden
Dr. Klaus Wyser – Swedish Meteorological and Hydrological Institute (SMHI), Rossby Centre, Norrköping, Sweden
Abstract:
The Arctic Ocean has experienced a sharp decline in sea ice extent and thickness in the recent decades, which threatens to leave the Arctic ice-free during the summer months in a foreseeable future. Insights of pan-Arctic sea-ice cover a few months in advance would be of key interest for the marine accessibility of the Arctic Seas and for the Arctic population whose livehoods depend on fishing and hunting. In parallel, the tropical Atlantic is currently experiencing a remarkably high level of hurricane activity: the last three hurricane seasons have produced at least 17 tropical cyclones, making it the most active 3-year period since the beginning of the hurricane record in 1851. The recent passage of Sandy over New York serves as a powerful reminder to the incredible destructive power these storms can wreck on exposed populations. If both Arctic sea ice and Atlantic hurricanes show a clear trend over the recent past, they both also display relatively large internal natural variability which can induce changes from one year to another of the same order of magnitude as the changes associated with the long-term trend. Previous studies suggest that such observed changes are predictable a few weeks to a few months ahead. Accordingly, we plan to evaluate the ability of the most recent version of a state-of-the-art Earth system model, EC-Earth, which participated in the CMIP5 (Fifth Climate Model Intercomparison Project) project at the basis of the IPCC (Intergovernmental Panel on Climate change) AR5 (Fifth Assessment Report), to predict these two important features of the climate system at the seasonal scale. In CMIP5, EC-Earth was run using a resolution of 1.125° in the atmosphere and about 1° in the ocean with enhanced resolution in the tropics (ORCA1 configuration). A new version, with significantly higher resolution, is now available and will become the standard version for the upcoming years. This new version can be run with either 1) a resolution of ~0.7° in the atmosphere and 1° in the ocean, or 2) a resolution of ~0.35° in the atmosphere and 0.25° in the ocean. In the second configuration, the enhanced atmospheric resolution significantly improves the realism of climate simulations, in particular, of the hurricane activity. Similarly, the high ocean resolution improves the representation of oceanic fronts, eddies and convection among other oceanic features, and potentially the representation of the Artic sea ice shrinking rate. In two series of retrospective seasonal forecasts over the last two decades using these configurations, we will assess the skill of EC-Earth in predicting tropical Atlantic cyclones and Arctic sea ice, and the benefits derived from increased resolution. This project represents a key step in the development of an operational high-resolution seasonal forecast system. It will contribute to the European-funded SPECS (Seasonal-to-decadal climate Prediction for the improvement of European Climate Services) project and to the PICA-ICE (Previsión Interanual de la Cubierta de hielo marino del Árctico y su Impacto en el Clima de Europa) project funded by the Spanish Ministry of Economy and Competitiveness.

WIND-FORECAST

Project Title: Development Of A Multi-Scale Atmospheric Flow Simulation Tool For Short-To-Medium Term Wind Power Forecasts
Project Leader: Dr. Gokhan Kirkil, Kadir Has University, Faculty of Engineering and Natural Sciences, Istanbul, Turkey
Resource Awarded: 140 000 core hours on CYFRONET – Zeus
Details

Collaborators:
Dr. Yasemin Ezber – Istanbul Technical University, Eurasia Institute of Earth Sciences, Istanbul, Turkey
Abstract:
In this project, we aim at development of a multi-scale multi-physics atmospheric flow simulation tool for short-to-medium term wind power forecasts at wind farms. Wind power forecasts are very valuable for daily and day-ahead operations of power generation and utility companies. Numerical simulations of atmospheric flows from meso-scale down to turbine scale might help these companies in forecasting power from wind turbines over a range of time scales. However, there are few challenges for development of such an operational tool. First, there are hardly any single numerical simulation tool that can simulate the range of scales exist in this problem altogether. Therefore, integration of few models, each one developed for simulation of certain range of scales, is needed in order to accurately handle this problem. Second, each of these tools and their aggregate simulation capability should be validated against existing measurement data which require interrogating and analysing 4D simulation data (i.e. mean wind speed data). In this project, we propose a work plan for step-by-step development of such an operational forecasting tool as well as investigating its validation and uncertainty quantification using state-of-the-art statistical and measurement techniques.

WISER

Project Title: A Weather Climate Change Impact Study at Extreme Resolution
Project Leader: Dr. Alan Gadian, University of Leeds, School of Earth and Environment, UK
Resource Awarded: 6 000 000 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Dr. Ralph Burton – University of Leeds, School of Earth and Environment, UK
Mr. James Groves – University of Leeds, School of Earth and Environment, UK
Abstract:
A NCAS (NERC’s National Centre for Atmospheric Science) group led by Dr Alan Gadian plans to run a future weather / climate change project to examine the changes in Weather patterns in Western Europe Climate is now a weather problem. Climate models typical have insufficient resolution to resolve and simulate regional weather processes such as rainfall and convection. There are large errors in the prediction of subtropical rainfall patterns, largely miss the Indian Monsoon, and cannot resolve blocking anticyclones which are very important in European weather patterns. This project will embed a nested simulation using the NCAR WRF (Weather Research and Forecasting) model within climate reanalyses and predictions from ECMWF and NCAR (US National Center for Atmospheric Research). The nested WRF simulations will cover +/- 65 deg of latitude and will be nested down to 4 km resolution over western Europe, capturing almost all important weather scales. A series of simulations will be completed 1. Validation runs, driven by Era Interim data will include 1970-present day reanalysis and calibration. 2. Predictive runs will include present day to 2080, and driven by the the US CAM model Runs are planned in 10 year blocks, each with several completed simultaneously on 5000- 10000 cores. Each 10 year block will use ~300 hours on 5000-10000 cores on the IBM Blue Gene/Q. The work is part of an ongoing collaboration between NCAR’s Nested Regional Climate Modelling group and NCAS, with active support of the Hartree Centre. The project aims to provide new insight into the technology of regional climate modelling (especially from the validation “hindcast” runs) and provide statistical predictions of the European regional impacts of climate change. Data will be provide to JRC, Natural Hazards, European Flood Forecasting system.
top

Engineering (8)

DNSTF

Project Title: Direct numerical simulation of finite size fibres in turbulent flow
Project Leader: Prof. Gustav Amberg, KTH, Department of Mechanics, Stockholm, Sweden
Resource Awarded: 8 437 500 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Dr. Minh Do-Quang – KTH, Department of Mechanics, Stockholm, Sweden
Abstract:
The goal of this project is to adapt a numerical model of the dynamical behaviour of finite-size fibres in high Reynolds number, turbulent flow. The chosen approach is to start from direct numerical simulations for both turbulent flow and to resolve the flow around individual fibres. The turbulent flow is modelled by an entropy lattice Boltzmann method and the interaction between fibres and carrier fluid is modelled through an external boundary force method (EBF). Direct contact and lubrication force models for fibre-fibre interactions and fibre-wall interaction are taken into account to allow for a full four- way interaction. This model will allow us to study the influence of wall effects and interaction effects on the turbulent flow. A code (named SLILAB) has been developed at Mechanics department, Royal institute of Technology, Sweden. The new version of SLILAB is now based on the Palabos library environment and offers full coupling between fluid flow and solid particle motion. This code has been tested for a year in some HPC centers in Sweden. Its performance has a high parallel efficiency. Due to the limitation of CPU hours we can get from Swedish National Infrastructure for Computing (SNIC) and the capacity of the computer we can access, the simulation we have done up to now is for a small turbulent flow in a small channel size (friction Reynolds number=180 in a box 2cm height and 3cm width). To get a better knowledge about the complicated phenomena at a scale that can be used to study the dynamical behaviour of fibres in an application such as a the head-box in a paper-making machine, or in a groove of a refiner, etc, we need access to a large scale simulation resources as the ones provided by PRACE.

ERPP

Project Title: Exact Regularized Point Particle Method for the momentum coupling in particle-laden turbulent flows
Project Leader: Dr Paolo Gualtieri, Universita di Roma la Sapienza, Dipartimento di Meccanica e Aeronautica, Italy
Resource Awarded: 5 910 000 core hours on CYFRONET – Zeus BigMem
Details

Collaborators:
Dr Francesco Battista – Universita di Roma la Sapienza, Dipartimento di Meccanica e Aeronautica, Italy
Abstract:
Turbulent multiphase flows laden with small solid particles or liquid droplets are involved in many areas of applied science. Turbulent transport and collisions of water droplets was found to accelerate the process of rain formation. In applied technology, the dynamics of small droplets is crucial in designing the injection system of internal combustion engines. In fact, after the atomization of the fuel stream, the droplets evolve in a turbulent surrounding flow where preferential segregation occurs. Such phenomena impacts on the evaporation, mixing and combustion of the fuel leading to an overall loss of heat conversion into mechanical work or to the formation of unburned gasses and soot. The relevant physical aspect consists in the particles finite inertia that prevents them from following the fluid trajectories leading to `preferential accumulation` phenomena. The effect of turbulent transport on particle dynamics has been studied extensively in many flow configurations. Much less is known about the effect of disperse phase on the carrier flow. In the so-called two-way coupling regime the particles volume fraction is still very small to neglect inter particles/droplets collisions and hydrodynamic interactions but the mass loading on the fluid is of order one due to large density ratios. In such conditions the momentum coupling between the phases is relevant and must be considered in numerical simulations. In this proposal we present a new exact methodology which is able to capture the momentum exchange between a carrier turbulent flow and hundred thousands of small inertial particles. In fact, the local distortion of the carrier flow can be captured only resolving the boundary of each particle on the computational grid (Resolved Particle Simulation). However this approach is feasible only for a small number of large particles i.e. for particles whose diameter is larger than the smallest fluid length-scale. When hundred thousands of small particles are injected in the flow such methods cannot be pursued. In the limit of small particles diameters the disperse phase can be considered as material points i.e. as point source/sink of momentum for the carrier fluid. In a nutshell the coupling between the two phases is achieved by describing the disturbance flow produced by the disperse phase on the carrier fluid in terms of an exact regularized unsteady Stokes solution. This approach achieves an highly computational efficiency. In fact, the disturbance flow produced by the particles/droplets is strongly compact in space and localized around their actual position meaning that at each time step only few fluid grid points perceive such disturbance. Computational efficiency is mandatory since the dynamics of the disperse phase and the correspondent alteration of the turbulent flow depends on many physical parameters such as the mass load, the particles/droplets size, the density ratio and the particles/droplets relaxation timescale. Hence, at least a four-dimensional parameter space must be explored with numerical simulations demanding for efficient momentum coupling algorithms as we propose here.

GREENLIGNITE

Project Title: Clean Energy from Lignite Gasification
Project Leader: Dr. Ismail Tuncok, Solaris Engineering and Consulting, R&D Applications, Ankara, Turkey
Resource Awarded: 2 240 000 core hours
Details

Collaborators:
Dr. Aytekin Gel – ALPEMI Consulting, Phoenix, USA
Abstract:
Major consumers of the world’s energy resources today are in dire need of secure and affordable energy solutions. While oil reserves are dwindling, coal reserves, are still abundant globally. Hence, coal is the major source of electrical power today but coal burning has been also a major source of pollution due to various challenges, which need to be overcome before coal can be used as a sustainable source of energy. Initiatives such as clean coal technologies can pave the way to environmental sustainability, and create a new and improved image for coal-based energy while the technology is evolving for energy production from renewables. Coal gasification is considered one of the critical technologies for clean energy in the near future as it can provide an opportunity to meet the ever-increasing energy demands of the both developed and developing countries across the world in an environmentally sustainable manner. Despite its potential, this technology has been met with resistance mainly due to the uncertainties in the economic viability and the environmental impact of the process. To help address these issues, high-resolution computer simulations are being proposed to provide information to industrial stakeholders to address these uncertainties. The overall performance of coal gasifier is directly linked to a few key operating parameters. Coal feed rate, reaction kinetics, solids and syngas re-circulation rates are key factors in the daily operation of the gasifier. The production of polluting species in the exit syngas, low energy content of the syngas, locally high temperatures, and un-reacted coal are some of the problems linked to these factors, which are quite difficult to understand and quantify without the aid of high resolution simulations for large scale operation. The proposed project intends to employ a validated computational fluid dynamics models for reacting multiphase flows in fluidized beds and gasifiers to investigate several of these important design parameters and their interactions specifically for Turkish lignite and biomass based feedstock. Confidential

HIGHERFLY

Project Title: Immersed methods for insect flight aerodynamics
Project Leader: Dr Fehmi Cirak, The University of Cambridge, Dept of Engineering, Cambridge, UK
Resource Awarded: 5 000 000 core hours on EPCC – Archer and EPCC – HeCToR XE6
Details

Collaborators:
Dr Jakub Sistek – Institute of Mathematics of the AS CR, Constructive Methods of Mathmatical Analysis, Prague, Czech Republic
Abstract:
The aim of this project is to perform high-fidelity fluid-structure interaction simulations of insect and bio-inspired micro-air-vehicle flight. The wing is modelled as a thin compliant shell structure and the viscous, incompressible fluid flow is approximated with a direct numerical approach. The flow equations are discretized using a fixed block-structured grid of finite elements in order to ascertain algorithmic simplicity and scalability. On the other hand, the structure equations are discretized with a moving unstructured mesh of finite elements. The discretized fluid and structure equations are coupled with a partitioned approach. The coupling is strong in the sense that within each time step the fluid and structure equations are iterated until the interface equilibrium conditions are satisfied. The systems of equations to be solved have up to several billions of unknowns and are very computing intensive. As it is necessary to simulate the flapping of the wing over several stroke cycles, it is necessary to solve the system of equations up to several thousand times. The obtained computational results will be validated with experimental data gathered with a recently completed flapper in the Department of Zoology in Cambridge.

JOSEFINA

Project Title: Jet nOiSe intEraction with Fuselage in INstalled configurAtion
Project Leader: Dr. Sergey Karabasov, Queen Mary University of London, School of Engineering and Materials Science, London, UK
Resource Awarded: 6 250 000 core hours on PDC – Lindgren
Details

Collaborators:
Prof. Victor Kopiev – Central Aerohydrodynamic Institute (TsAGI), Moscow, Russia
Dr. Philip Ridley – NAG Ltd, Oxford, UK
Abstract:
Control of cabin noise is an important problem that has to be considered by both engine and aircraft manufacturers for optimising the sound absorption properties of acoustic insulation in the cabin against the weight penalty. Boundary layer and engine noise, e.g., jet and fan noise, are the typical main components of the external noise sources that can transmit to the cabin. For propulsive jets in high-altitude cruise conditions, the broadbandshock- associated jet noise can be very important. This is typically the case for high bypass ratios engines (BPR~7-8) and conventional engine installations under the aircraft wing. Due to the increased jet proximity to the fuselage wall, the importance of the jet noise component for cabin noise transmission is likely to further increase for ultra-high bypass ratio (UHBR) engines (target BPR values ~ 15-18). Notably, in accordance with the international noise regulation laws, the UHBR engines are expected to be in service by 2020 in order to reach the target level in reduction of community noise that scales with a high power of jet exit velocity. This calls for a fundamental study of new strategies for cabin noise reduction. The growing concern of industry for the effect of jet noise transmitted to the cabin pushed major aircraft manufacturers, Boeing and Airbus, to perform expensive full-scale flight tests. On the other hand, high-resolution computational modelling can be used to study the effect of jet with accounting for dual-stream jet installation effects due to the presence of nacelle, pylon and wing on fuselage wall in flight conditions. This project seeks to perform several high-resolution calculations with the goal to obtain acoustic pressure loading on fuselage surface for a range of jet/cruise conditions based on a representative installed jet nozzle geometry. For unsteady calculations, the CABARET Monotonocally Integrated Large Eddy Simulations (MILES) method will be used that combines low dissipation and shock capturing (Karabasov and Goloviznin 2009, Faransov et al 2012). CABARET scheme is low-dispersive and nondissipative, uses a low dissipative shock-capturing flux correction, and is robust for nonuniform grids. It is as fast and as simple for implementation for relatively complex geometries through the use of implicit (unstructured hexahedral) block structure as conventional shockcapturing finite-volume schemes and affordable for supercomputing with a good scalability up to a few thousands of computational cores (Ridley and Karabasov, CUG 2010). In comparison with the conventional 2nd order schemes, the dispersion and dissipation errors of CABARET scheme are a few orders of magnitude lower. Recently, a numerical investigation of the effect of jet noise on acoustic loading of the fuselage in flight conditions has been conducted in Central Aerohydrodynamic Institute (TsAGI), Russia by Kopiev and co-workers (Faranosov et al, Russian-Chinese Conference on Aviation Science and Technology 2012). In this project, with the help of computing power provided, we are going to extend the results of the latter work to a higher resolution and more realistic jet configurations.

LargeRB2013

Project Title: Very high-resolution computation of turbulent Rayleigh-Bénard convection in large aspect ratio cells
Project Leader: Prof. Janet D. Scheel, Occidental College, Department of Physics, Los Angeles, USA
Resource Awarded: 6 817 501 core hours on EPCC – Blue Joule
Details

Collaborators:
Prof.Dr. Jörg Schumacher – Technische Universität Ilmenau, Institute for Thermodynamics and Fluid Mechanics, Ilmenau, Germany
Prof. Ladislav – Charles University, Faculty of Mathematics and Physics, Prague, Czech Republic
Prof. Katepalli R. Sreenivasan – New York University, Physics and Courant Institute, New York, USA
Dr. Pavel Urban – The Academy of Sciences of the Czech Republic, Institute of Scientific Instruments of the ASCR, v.v.i., Brno, Czech Republic
Abstract:
Many turbulent flows in nature and technology are driven by sustained temperature differences. Applications range from cooling devices of chips to convection in the Earth and the Sun. Turbulent Rayleigh-Bénard convection (RBC) is the paradigm for all these convective phenomena because it can be studied in a controlled manner, but it still has enough complexity to contain the key features of convective turbulence in the examples just mentioned. RBC in cylindrical cells has been studied intensely over the last few years in several laboratory experiments, mostly in slender cells of aspect ratio smaller than or equal to unity in order to reach the largest possible Rayleigh numbers. The key question in RBC is the mechanism of turbulent transport of heat and momentum. Since the fluid motion is driven by a constant temperature difference between the top and bottom plates, thin boundary layers of temperature and velocity will form on these walls as well as on the side walls of the cell. A deeper understanding of the global transport mechanisms is possible only if we understand the dynamical coupling between the boundary layers and the rest of the flow in the bulk of the cell. This interaction has been the center of research but no experiment to date has adequately resolved the boundary layers at high Rayleigh numbers. Numerical simulations have attempted to do this recently but the resolution needed to obtain the smallest flow scales has not yet been employed. It is crucial to resolve all the dynamically important scales to represent the flow faithfully. Indeed, recent superfine resolution in isothermal turbulence has led to some enlightening results. That experience suggests that extending superfine simulations to RBC in aspect ratios larger than unity allows us to better disentangle the central physics of turbulent heat transport in many applications. Specifically, we propose to make a significant step forward with direct numerical simulations (DNS) of RBC (a) by resolving scales never accessed before, and (b) by simulating the flow using an aspect ratio significantly larger than unity. Our efforts are based on the Nek5000 software which has been developed for solving flow equations with spectral accuracy. The DNS will provide necessary new insights into the global circulation of the convective flow and into the details of the increasingly intermittent boundary layers that are coupled to the circulation. The superfine data will also help to resolve differences of the small-scale temperature statistics in the bulk compared to passive scalar turbulence. The DNS of the underlying flow equations directly resolve the turbulence from the largest vortices on the order of the system size down to the dissipative Kolmogorov scale and finer. Thus, they provide the full three-dimensional structure of the flow fully resolved in space and time. Our proposed study will provide new insights of the turbulent transport processes which have remained vague until now. As such, the proposed research will be transformational in the domain of fluid dynamics, with potential relevance for natural and technological applications.

MoDSS

Project Title: Molecular Dynamics Simulations of Superspreaders
Project Leader: Prof. Ahmed E. Ismail, RWTH Aachen University, Mechanical Engineering, Aachen, Germany
Resource Awarded: 577 500 core hours on RZG – Hydra
Details

Collaborators:
Rolf Isele-Holder – RWTH Aachen University, Mechanical Engineering, Aachen, Germany
Abstract:
Trisiloxane surfactants are widely used in various products such as paints, inks, and herbicides because of their capacity to enable “superspreading,” the greatly enhanced spreading of aqueous solutions on hydrophobic surfaces. As these surfactants are toxic and chemically unstable, alternative superspreading agents are eagerly desired, which is why there have been numerous studies in the last 20 years in academia and industry with the aim of uncovering the mechanisms that facilitate the ultra-rapid spreading. Since experimental setups are unable to resolve fully the processes occurring on the molecular scale, the driving mechanisms of superspreading have not yet been completely identified, leading to the existence of competing theories. Trisiloxane surfactants consist of a hydrophobic octamethyltrisiloxane head group and a hydrophilic polyethylene glycol tail, whose chain lengths can be varied. It has been shown in experiments that the spreading rates of aqueous trisiloxane surfactant solutions depend on the lengths of this hydrophilic chain, the hydrophobicity of the examined substrate, and the concentration of trisiloxane surfactants in the solution. The dependence of the spreading rates is non-trivial in a sense that maximum spreading rates occur for intermediate values for any of the three quantities just mentioned, which suggests that the process of superspreading results from competing mechanisms, which are not even understood rudimentary. We will perform large-scale Molecular Dynamics simulations of trisiloxane surfactantladen water droplets on various surfaces. These simulations will clarify open questions on the fundamental behavior of trisiloxane surfactants in solutions and the various interfaces involved in the process and in this way contribute to the understanding of superspreading and the development of alternative environmentally friendly superspreading agents.

waveclim

Project Title: Nearshore wave climate analysis for the West coast of Ireland and effects induced on the wave resource by arrays of wave energy converters
Project Leader: Prof. Frederic Dias, University College Dublin, Dublin, Ireland
Resource Awarded: 2 295 000 core hours on CSCS – Rosa
Details

Abstract:
The aim of this project is a detailed assessment of the wave climate on the west coast of Ireland and a study of the effects induced on the wave resource by arrays of Wave Energy Convertors (WECs). The west coast of Ireland features one of the largest concentrations of wave energy in the world and thus, it is a likely target for large scale deployment of WECs. Following newly developed guidelines for wave energy resource assessment we will perform a 10 years climate analysis for coastal areas while fully accounting for wave directionality. Recent investigations have revealed a considerable potential for exploitation of the nearshore regions. Indeed, in the nearshore, the directional spread of the wave energy resource is narrower than in the offshore. Furthermore, the proximity to land will likely facilitate both the installation and maintenance of the devices. However, most wave climate studies for the west coast of Ireland have focused on the offshore areas and hence there is a real need for an assessment of the wave resource close to the shoreline. Due to limitations in spatial resolution, the existing operational wave forecasting models at various meteorological centres around the world (in particular WAM at the European Centre for Medium-Range Weather Forecast ECMWF and WAVEWATCH at U.S. National Oceanic and Atmospheric Administration NOAA) do not resolve sufficiently accurately coastal areas (depths smaller than 50 m). Nonetheless, the quality of these models for the offshore areas is high. One of the main reasons for the high quality is the assimilation of satellite data (altimeter and SAR). With these in mind, we will construct local wave models with high spatial resolution, driven by high-quality boundary input (consisting in wave directional spectra) and wind forcing from ECMWF. Together with the issue of survivability in harsh sea conditions, the accessibility of the devices for maintenance are key in selecting a viable technology. Performing maintenance at sea is notoriously expensive and risky as evidenced by both the offshore oil industry and, more recently, the offshore wind industry. Indeed, deployment and maintenance activities require sufficiently long, contiguous time intervals where the sea state is sufficiently calm. In the case of WECs however, two opposing criteria would need to be balanced: the sites targeted for the deployment of WEC farms are at the same time characterized by energetic sea states, where such calm weather-windows have small duration. Finally, arrays of WECs will be included inside the operational code to study their effects on the wave resource.
top

Materials Science (10)

AIDMP

Project Title: Ab-initio design of mutant proteins
Project Leader: Prof. Fernando Nogueira, Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Resource Awarded: 1 200 000 core hours on FZJ – JuRoPA
Details

Collaborators:
Dr. Bruce Milne – Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Dr. Micael Oliveira – Universidade de Coimbra, Center for Computational Physics, Coimbra, Portugal
Abstract:
In this project, we will study what determines the visible light absorption characteristics of channelrhodopsin proteins. With this study we will determine the key environmental factors leading to colour shifts and devise a model to predict the changes in the environment (like mutations of one amino acid) required to obtain a pre-determined colour shift. With the help of this model we expect to guide biochemists in synthesising novel (mutant) forms of this protein that are able to react to light of different colours. We will use a real-space real-time implementation of Time-Dependent Density-Functional Theory that has already proved to be quite accurate for biochromophores.

DIVI

Project Title: DIVIDE ET IMPERA
Project Leader: Dr. Jan-Willem Handgraaf, Culgi BV, Leiden, The Netherlands
Resource Awarded: 3 500 000 core hours on ICHEC – Fionn-thin, ICHEC – Stokes and SURFSARA – Cartesius
Details

Collaborators:
Dr. Paul Becherer – Culgi BV, Leiden, The Netherlands
Dr. Monica Belacu – Culgi BV, Leiden, The Netherlands
Prof. Maurizio Fermeglia – University of Trieste, Engineering & Architecture, Trieste, Italy
Dr. Ruben Gracia – Culgi BV, Leiden, The Netherlands
Prof. Doros Theodorou – National Technical University of Athens, School of Chemical Engineering, Athens, Greece
Dr. Elena Tocci – ITM-CNR, , Rende, Italy
Abstract:
Culgi is a technical consulting company in the area of chemical informatics and modelling in materials and life sciences. One of its main activities is the development, distribution and support of the Chemistry Unified Language Interface (CULGI). As a result of our vast experience with industrial and academic partners, in various field as chemicals, pharmaceuticals, aerospace, petroleum, we started to develop a novel protocol to automatically fragment and parameterize any chemical component for further use in simple, rapid particle based simulations. The DECI-10 programme, by the extensive computational resources offered, will enable the creation of a representative database of molecular fragments with known physico-chemical properties and interaction parameters calculated using Monte Carlo techniques, quantum mechanics, thermodynamics information and optimisation techniques.
The key to successful design in any field of computational chemistry is property prediction from a thermodynamic analysis of soft interfaces, based on molecular structures. This implies a number of elements: • generation of such structure in silico (structure implies the need for a spatial theoretical method, as opposed to the typical bulk methods in engineering) • calibration (hence one needs to decompose molecules into fragments, to access a experimental databases, to calibrate such fragments) • validation using realistic industrial case studies • rapidity (costing of less than one hour) and meaningful for an end-user in an industrial context (fragments will have to interact through soft potentials, to reduce the computational cost) • easy-to-use by industrial end-users (for example, as an extension of existing engineering thermodynamics methods currently used in an industrial setting). Within these elements as targets, we aim for a mind set change with respect to the acceptance and use of rational modelling tools in computational chemistry.
The automated fragmentation-parameterization (AFP) of a reduced ultra-structure model will consist of scripted software modules, that allows for atomic reading of a certain relevant chemical (e.g. lipid, peptide, protein, polysaccharide), that will then be converted into a suitable coarse-grained model for the mesoscopic modelling. AFP consists of three steps: (i) assignments of so-called coarse-grained ‘beads’ to groups of atoms (automated fragmentation), (ii) calculation of intra-molecular bead interactions and (iii) calculation of intermolecular bead interactions. The AFP project will rely on proprietary background methods from Culgi B.V., such as a quantum-thermodynamic approach to bead interactions and the fragmentation algorithm. Additional software will be used (after necessary modifications included): NWChem and COSMO RS packages to calculate the interaction parameters between the fragments. To decide on the best manner of fragmentation and parameterization we plan to work on a large database of very different molecules and to perform the calculations at multiple description levels. We will start with a set of chemical substances (such as polymers, solvents, surfactants) for which thermophysical data is available (binary mixtures information from the NIST database). Once the protocol is validated, it can be compared with currently available coarse-grained models for proteins or it can be applied to many other types of chemicals (drugs, agrochemicals, crude oils, electronic materials, etc.)

EXC-XMCD

Project Title: The excitonic effects on X-ray circular and linear dichroism at 2p edges
Project Leader: Dr. Dominik Legut, Vysoka Skola Banska – Technical University of Ostrava, Nanotechnology Centre, Ostrava, Czech Republic
Resource Awarded: 2 000 000 core hours on PSNC – Chimera
Details

Collaborators:
Dr. Robert Laskowski – Vienna University of Technology, Institute of Materials Chemistry, Vienna, Austria
Abstract:
The understanding of the X-ray magnetic circular/linear dichroism (XMCD/XMLD) in metals and magnetic insulating materials has advanced tremendously in last years [Stoehr and Siegmann 2006, Antonov 2007, 2008]. The basic advantage of the XMCD is the ability to estimate the orbital and spin moments based on the sum rules in ferromagnetic materials (Thole 1992, Carra 1993). There are not any such sum rules in XMLD, however this technique (using effect that is quadratic in magnetization) has the advantage over XMCD to detect antiferromagnetic materials (v.d. Laan 1998). With the help of the first- principles (ab initio) calculations the origins of both effects, XMCD and XMLD, were revealed to large extent in the case of transition metals in recent decades (Kunes 2003). However, for oxide systems and/or early d-metals the structures of the XMCD/XMLD spectra are much more complex, often reasonably well modeled by `multiplet theory` codes, i.e. codes which are not parameter free, but not in a such good agreement as if the spectra were modeled by pure quantum mechanics codes, e.g. using density functional theory (DFT) parameter free technique. Following deficiencies may originate from the improper description of many-body effects, particularly from quasiparticle electron-hole: missing satellite peaks at high(low) energy side of 3d transition atoms in pure metals and insulating compounds, incorrect branching ratio of the X-ray absorption spectra (XAS) at 2p edges (L 3/L2), and fine-structure features (oscillations) of XMCD/XMLD in the magnetic insulating materials. To some extent these experimentally observed features can be resembled using multiplet calculations that strongly depend on number of empirical parameters. Here, we would like to identify for the first time the role of the excitonic effects on the XAS, XMCD, XMLD of magnetic solids with different electron screening (magnetic metals and insulators) with ferro-(XMCD) and antiferro-(XMLD) ordering. We will widen the existing knowledge of XAS of nonmagnetic solids (Laskowski 2012) by use of developed Bethe-Salpeter (BSE) equations (Onida 2002) and by further investigations of selected magnetic metallic (3d transition metals) and insulating or half-metallic materials, e.g. NiO, CrO2, respectively. We will continue in further development of the code and hence we intend to determine the magnitude of the quadruple contributions with respect to the dipolar contributions of the simulated spectra using BSE technique, as found in some materials (Rueff 2004). Our approach is at the atomistic level using first-principles (quantum mechanics) calculations and therefore no empirical parameters for such calculations are required. The only inputs needed are atomic number of constituents and some structural information, e.g. body-centered cubic structure in the case of Fe. However, computational demand to perform BSE calculations is about 3 orders of magnitude higher than for the independent single particle approximation for the same structure and material.

InterDef

Project Title: Interface defects
Project Leader: Dr. Matthew Watkins, University College London, London, UK
Resource Awarded: 3 739 250 core hours on SURFSARA – Cartesius
Details

Collaborators:
Dr. Francisco López Gejo – Centro Física Materiales, Chemical Physics Complex Materials, San Sebastian – Donostia, Spain
Prof. Alexander Shluger – University College London, London, UK
Abstract:
This project will focus on computational modelling of mechanisms underpinning formation and performance of resistive random access memory (ReRAM). In general, resistive memory technologies rely on switching between high and low resistance states of materials by the application of electrical stimuli. They compete with other concepts in the fast-growing market of non-volatile memory devices. Dynamic random access memories (DRAM) and Flash currently represent the dominant memory technologies. DRAMs are fast and show almost unlimited endurance (cycle times) for write and read operations. However, these volatile memory devices immediately lose the stored information if the supply voltage is cut-off. Flash memories are non-volatile and their technology is characterized by a compact structure. However, Flash has write times in the 10 μs range, write voltages of >12 V and a high write energy. ReRAMs represent a highly promising non-volatile memory technology because of their very high endurance, ultrafast switching (

LipoSim

Project Title: Large scale simulations of liposomes as drug carriers
Project Leader: Prof. Leif Eriksson, University of Gothenburg, Department of Chemistry and Molecular Biology , Göteborg, Sweden
Resource Awarded: 8 750 000 core hours on PDC – Lindgren
Details

Abstract:
Massively parallel computations will be employed in order to simulate drug delivery from a drug loaded liposomes into a cellular plasma membrane. The use of PRACE facilities enables us to understand at a detailed atomic level, how different lipid mixtures, drug concentrations and sizes of the liposome or micelle based carriers influence the ability of these to actually bring the drug molecules to their intended targets. We focus in this landmark study on different compounds aimed for cancer therapy, with the aim to thereby have established a system to be used in the study of a wide range of other compounds in the future, in order to better optimize conditions for their usage in connection with lipid vesicles for drug delivery. Deeper insight into these processes will also allow researchers to design new molecules that better dissolve into and transfer between liposome and cell, or that are able to diffuse out of the liposome as response to small variations in the local environment.

In order to be able to model the complex interactions involved in the fusion of a lipid carrier into a membrane, the computations will be carried out using more than 1,000 computer nodes in parallel, with each system corresponding to approximately 10,0 00,000 atoms. Another advantage with the current model set – up is also that it can be used to explore membrane budding, vesicle formation and related phenomena, of importance to understand cell signalling processes.

The work is carried out in a collaboration between researchers at the universities of Stockholm and Gothenburg.

MEGAREACT

Project Title: Metal catalysed gasification reactions
Project Leader: Prof. Kim Bolton, University of Boras, School of Engineering, Borås, Sweden
Resource Awarded: 750 000 core hours on UIO – Abel
Details

Collaborators:
Abas Mohsenzadeh Syouki – University of Boras, School of Engineering, Borås, Sweden
Abstract:
Gasification, the conversion of carbonaceous material to a gaseous product with an employable heating value, is one of the most important and effective production methods of energy carriers employed toward sustainable development. Making use of quantum mechanics simulation tools, mechanisms of reactions during the metal catalysed gasification process are investigated. In particular, we obtain reactant and transition state energies and frequencies that are used to obtain the reaction barriers and Arrhenius pre-exponentials of the elementary reactions on a transition metal catalyst surface. These data will subsequently be used in kinetic modelling of the entire reaction. The first reaction that will be studied is the water gas shift (WGS) reaction, which is important in almost all gasification reactions.

Novel_Anticoagulants

Project Title: Discovery of Novel Anticoagulants
Project Leader: Dr. Horacio Pérez-Sánchez, University of Murcia, Parallel Computer Architecture Group, Murcia, Spain
Resource Awarded: 480 000 core hours on CINECA – PLX
Details

Abstract:
The challenge of an aging society with anticoagulant requirements of increasing complexity encourages the search for new anticoagulant molecules. Heparin is widely used as activator of antithrombin, but incurs side effects. This research project is motivated by the potentially numerous applications of a technique that has been recently developed by the PI and collaborators and implemented in the GPU program BINDSURF, and from the promising results that he has obtained. Applying state-of-the-art computational drug discovery methods he has discovered a compound that, working as heparin cofactor, binds with nanomolar affinity to antithrombin causing partial activation, and which paves the way for the discovery of novel anticoagulants based on its molecular structure. At the same time, little is known about the mechanism by which this compound activates antithrombin. Unfortunately, this situation is common for many current drug discovery methods. Understanding clearly the biological and functional consequences of protein-ligand interactions and implementing them in a drug design strategy would dramatically accelerate the discovery and design of compounds with required functional properties. The main aim of my proposal is to apply an improved integrated computational-experimental strategy for the discovery of bioactive compounds with desired functional properties. The range of applicability of this methodology will cover any drug discovery campaign. As proof of concept I will focus on its application to the discovery of novel anticoagulants. The results of this research are expected to improve insights into the discovery process and design of bioactive compounds with desired functional properties (i.e., agonists or antagonists). Finally, the research should result in novel compounds with unknown scaffolds (no patent issues). These results can form the basis of an attractive new generation of drug discovery approaches.

RODCS

Project Title: Reactivity on doped ceria surfaces: ab initio optimisation of catalytic activity
Project Leader: Prof. Graeme W. Watson, Trinity College Dublin, Dublin, Ireland
Resource Awarded: 3 7450 000 core hours on CSCS – Rosa
Details

Abstract:
In this environmentally conscious age the search for new and improved catalysts for pollution abatement is inexorable. Ceria (CeO2) represents a material which has long been associated with pollution control via the catalytic degradation of exhaust gases, most notably in the three-way catalyst (TWC) for automotive emissions. In recent times, research has been directed towards methods of improving the catalytic ability of ceria, typically through the incorporation of different metal ions into the ceria lattice and onto the structure of the surfaces, a process known as doping. The use of divalent noble metal ions (Cu, Pd, Pt) is one area where research has been particularly focussed, mainly due to their use in TWCs. However, trivalent metal ions (Y, La, Bi) have also shown promise for enhancing ceria’s catalytic activity. Despite the strong promise, a number of questions remain regarding the manner of how the catalytic activity is improved by these dopants. The aim of this study is to use computational modelling to fully investigate the role these divalent and trivalent ions play in key catalytic reactions. Initially, this will be achieved by ascertaining the effect the dopants have on the oxygen storage capacity (OSC) of both bulk ceria and at its surfaces, i.e. their ability to release and subsequently take up oxygen ions. Following this, the effect of the dopants on the absorption of the emission gases CO, NO and NO2 will be considered, as well as the subsequent oxidation of CO and reduction of NOx (x = 1,2). This project will thus provide insight into the effect of these dopants at the nanoscopic level, allowing the active sites of CeO2 to be elucidated and unlocking the cause of the reported increase in OSC over pure CeO2. Subsequently, this may provide sufficient understanding to allow the selection of novel doping strategies to optimise catalytic performance. This approach will therefore allow the rational chemical design of next generation catalysts for pollution abatement.

SCosPtS

Project Title: Surface Chemistry on stepped platinum surfaces
Project Leader: Prof.Dr. Marc Koper, Leiden University, Catalysis and Surface Chemistry, Leiden, The Netherlands
Resource Awarded: 2 280 000 core hours on CSC – Sisu
Details

Collaborators:
Dipl. Phys. Manuel Kolb – Leiden University, Catalysis and Surface Chemistry, Leiden, The Netherlands
M.Sc. Hongjiao Li – Leiden University, Catalysis and Surface Chemistry, Leiden, The Netherlands
Abstract:
Understanding the influence of steps and kinks on reactivity is of great importance in catalysis. Most catalytically important systems are composed of nanoparticles which consist mostly of surfaces with a varying defect density. Due to computation costs most DFT simulations however only focus on either simple high symmetry facets or on very short terraces. Increasing availability of HPC resources makes it possible to simulate systems with longer terraces. This allows for clear seperation between the effects of the step area and the previously studied highly symmetric terraces.
Our project consists of two parts. In one we will model the complete pathway for the dissociation of dimethyl- ether (DME) for a regularly stepped platinum surface (Pt(411)) and compare the energy barriers along the pathway for the step area and the terrace area. This in conjunction with earlier results for the flat terrace will give us the possibility to directly model the influence of steps on the reactivity of the surface. We will compare these simulations directly to recent experimental results obtained in our electro-chemistry labs, which have shown the importance of considering (100) terraces vs (100) steps.
The second part of the project will follow a ultra-high vacuum project in our group in which we try to model the electrochemical conditions on a stepped platinum surface in UHV. This will be accomplished by covering the surface in a mono-layer thin water film and then introducing hydrogen and oxygen to the system. Our simulations will try to mirror this system by using earlier results for the water coverage and running ab-initio molecular dynamics simulations after the addition of hydrogen or oxygen into the system.
The main aim here is to study the non-linear effect that steps and defects have on the water structure on an important catalytic surface, i.e. platinum.

TheoMoMuLaM

Project Title: Theoretical Modelling of Multifunctional Layered Materials
Project Leader: Prof. Philippe Ghosez, University of Liege, Physics, Liege, Belgium
Resource Awarded: 2 835 000 core hours on UIO – Abel
Details

Collaborators:
Dr. Eric Bousquet – University of Liege, Physics, Liege, Belgium
Abstract:
The research project TheoMoMuLaM goal is to explore from first-principles new exotic properties at layer interfaces of oxides and fluorides. The group will use density functional theory calculations and effective Hamiltonian techniques to design new optimised materials.
Oxide-based materials have gained a lot of attention in the last two decades due to their promising properties for technological applications. More interesting is the recent discovery of the apparition of totally new exotic phenomena at surfaces and interfaces of oxides. Two spectacular examples are the observation of a twodimensional electron gas (2DEGs) at the interface between LaAlO3 and SrTiO3 and the improper ferroelectricity property in PbTiO3/SrTiO3 superlattice. The most fascinating result from these two examples is that it is possible to find a new property at the interface between two materials that is totally absent in the initial parent compounds: a metallic 2DEG at the interface of two good insulators LaAlO3 and SrTiO3 or the improper ferroelectric behaviour in PbTiO3/SrTiO3 superlattices that cannot exist in the bulk parent compounds. Two of the actual main challenges are to use these interface effects in order to find optimised magnetoelectric multiferroic materials for spintronic applications or enhanced thermoelectric materials for new renewable source of energy.
The TheoMoMuLaM projects aims at using atomistic fist-principles simulations in order to understand the microscopic origin of these useful properties in bulk materials and in layered materials. The artificially layered systems, such as superlattices, are the target systems since they have a high potential to exhibit enhanced and optimised magnetoelectric and thermoelectric properties at the interface and thus even if the parent materials doesn’t even show such properties.

TransMem

Project Title: Translocation of Biomolecules Across Cell Membranes
Project Leader: Dr. Mikael Lund, Lund University, Department of Theoretical Chemistry, Lund, Sweden
Resource Awarded: 5 200 115 core hours on SURFSARA – Cartesius
Details

Collaborators:
Dr. Pavel Jungwirth – Academy of Sciences of the Czech Republic, Institute of Organic Chemistry and Biochemistry, Prague, Czech Republic
Anil Kurut – Lund University, Department of Theoretical Chemistry, Lund, Sweden
Abstract:
Cell penetrating peptides have drawn considerable attention recently, thanks to their ability to cross phospholipid membranes and deliver various molecular cargos, such as nucleic acids, proteins, quantum dots, and various drugs, inside the cells. A crucial structural requirements for an effective penetration of a peptide through the phospholipid bilayer is the presence of multiple guanidinium cations, which are present in the side chains of arginines. Confocal microscopy and flow cytometry techniques have demonstrated that oligoarginines containing six or more amino acids internalize into the membrane more efficiently than equally long lysine oligomers. However, the molecular mechanism of penetration into and translocation across the phospholiúpid bilayer of these peptides is poorly understood. Among the possible mechanistic explanations inverse micelle formation, electroporation, endocytosis, and anion mediated energy-independent diffusion through the membrane could be considered. Nevertheless, none of these mechanisms is able to fully rationalize the difference in membrane permeability of arginine containing peptides and those containing the other cationic amino acids, i.e., lysines. Within the present proposal we aim at clarifying the molecular mechanisms of membrane permeabilities of different cationic peptides of varying length and composition. We will explore the ability of guanidinium cationic groups present in arginine containing peptides (but not of ammonium groups in oligolysines) to pair via like-charge ion pairing and consequences thereof on membrane permeation. Homo- ion pairing of guanidinium cations is one of the driving forces for oligoarginine aggregation in water. and at the lipid bilayers, as shown by our previous all-atom molecular dynamics simulations. By a combination of all-atom and coarse-grained molecular simulations we will ask and aim to answer the following question: Does the higher charge density of membrane adsorbed oligoarginine aggregates, in comparison to single oligoarginines or oligolysines, play a decisive role in the ability to penetrate across the cellular membrane? Succesfully answering this question by means of extensive simulations will allow us to suggest a plausible and experimentally testable molecular mechanism of translocation of cell penetrating peptides across the cellular membrane. This will have direct consequences for devising new strategies of drug delivery into cells.

top

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