ESS-ILL user meeting - topical session on atomic-scale simulations in neutron scattering

20 Jan 2021, 13:00 → 21 Jan 2021, 18:30 Europe/Stockholm

Andrew McCluskey (Diamond Light Source), Danielle Adonis (European Spallation Source ERIC), Gregory Tucker (ESS), Miguel Gonzalez (Institut Laue Langevin (ILL)), Peter Fouquet (Institut Laue-Langevin), Thomas Holm Rod (European Spallation Source ERIC), Wojciech Potrzebowski (European Spallation Source ERIC)

During the last decades, computer simulations have become an essential tool in many fields, including neutron scattering. The increasing complexity of the systems studied at neutron facilities limits the validity of standard analytical approaches in many cases, calling for a symbiotic approach where neutron data (often complemented by additional experimental techniques) and modelling are used together to further understand such complex systems. This approach facilitated by the advances in hardware and software seen during recent years, has resulted in computer simulations becoming part of the standard toolbox for many experimental scientists, without requiring extensive training or huge resources.

The goal of this workshop is to showcase some representative examples of such uses, highlighting in particular recent work by young researchers showing how atomistic simulations (classical MD, spin dynamics, DFT, Monte Carlo, etc.) can help in understanding and interpreting neutron scattering data, as well as enhance the value of experimental data by leading to a better comprehension.

This virtual meeting will be held on Wednesday 20th and Thursday 21st January 2021, 13:00 - 18:30.

The list of confirmed speakers can be found in the timetable, but additional contributions (especially from young scientists) are welcomed and will be selected among the abstracts submitted before December 18th.

Wednesday, 20 January

13:00 → 13:15 Welcome

Welcome.pdf

13:15 → 15:20 Biology

Chair person: Wojciech Potrzebowski

13:15 Unraveling water dynamics and restructuring upon α-synuclein fibril formation by neutron spectroscopy and molecular dynamics simulations

Protein amyloid fiber formation is the pathological hallmark in various neurodegenerative diseases such as Parkinson’s or Alzheimer’s. The physico-chemical origin of protein fibrillation, as well as the role that hydration-water might play remain elusive. We combined neutron spectroscopy and molecular dynamics simulations on hydrated powders of α-synuclein and tau to investigate both structural and dynamical properties of the protein-hydration water system. Hydration-water dynamics is enhanced in the fiber state of both α-synuclein and tau, as shown by increased water mean-square displacements and broader quasi-elastic neutron scattering spectra. Molecular dynamics simulations of hydrated α-synuclein powders evidence a compact monomeric state and a more extended fiber state, in which interaction between the NAC segment and the N-terminal and C-terminal regions is reduced. As a consequence, water around N-terminal and C-terminal is no more constrained by the hydrophobic residues in the NAC segment, resulting in increased dynamics and entropy. The increase in water dynamics upon fiber formation is larger for tau than for α-synuclein. Since the latter contains a much smaller fraction of disordered residues in the fiber state than the former, we suggest that residual fiber disorder correlates with hydration water dynamics. The entropic driving force that increased water dynamics presents for fiber formation is suggested to be maximized in amyloids with an extensive fuzzy coat such as tau.

Speaker: Dr Kevin Pounot (ILL)

Kevin Pounot.pdf

13:40 Short-time diffusive dynamics of proteins in a naturally crowded environment

We investigate the effect of crowding on the short-time tracer diffusion of model proteins on the nanometer length and (sub-)nanosecond timescales both experimentally and with the aid of computer simulations [1]. Experimentally, we dissolve polyclonal immunoglobulin (Ig) antibody proteins of natural isotopic abundance as tracers in perdeuterated Escherichia coli cell lysate as crowder to mimic a biological environment and at the same time optimize the sensitivity for the tracers using incoherent quasi-elastic neutron scattering. We subsequently compare the ensemble-averaged dynamics of Ig in lysate to the dynamics of Ig in pure water as a function of concentration. In both cases, remarkably, the diffusion of Ig only depends on the total macromolecular volume fraction φ in the sample, within the experimental accuracy. To shed light on how polydispersity affects the short-time self-diffusion of proteins in crowded environments, we perform computer simulations which have proven to provide accurate information on the diffusion of proteins in crowed environments [2,3] and to interpret and rationalize neutron scattering experiments [4,5]. Because of the short times scales involve in our experiments, diffusion is mainly modulated by hydrodynamic interactions (HI) mediated by the aqueous media and therefore an accurate description of the HI most be included in the simulation scheme. For these reason, we perform simulations based on Stokesian Dynamics [6] in which HI are considered explicitly and short-time properties can be calculated. In our approach, the lysate polydisperse system is modeled using hard spheres. We demonstrate an intricate dependency between the diffusion of a tracer on the crowder composition, specifically on the ensemble effective radius, Reff = (<Ri^3>)^1/3. For tracers with a radius close to Reff, the tracer diffusion is similar to that of a monodisperse system, whereas deviations are observed for significantly different tracer radii. Notably, the hydrodynamic radius of Ig is close to the lysate effective radius, which explains the surprising insensitivity to the polydispersity observed in the experiments. The simulation results further show that polydispersity slows down larger macromolecules more effectively than smaller ones even at nanosecond timescales. This has obvious implications for the functioning of the cellular machinery. Our simulations also confirms the predictive power, on the nanosecond timescale, of coarse-grained molecular dynamics simulations.

[1] Grimaldo M., Lopez H., et al. (2019) J. Phys. Chem. Lett. 10, 1709–1715
[2] McGuffee, S. R.; Elcock, A. H. (2010) PLoS Comput. Biol. 6, e1000694.
[3] Ando, T.; Skolnick, J. (2010) Proc. Natl. Acad. Sci. USA 107, 18457–18462.
[4] Bucciarelli, S. et al. (2016) Science Advances 2, e1601432.
[5] Wang, G. et al. (2018) J. Phys. Chem. B 122, 2867–2880
[6] Brady, J. F.; Bossis, G. (1988) Annu. Rev. Fluid Mech. 20, 111–157

Speaker: Hender Lopez-Silva (TU Dublin)

Hender Lopez Silva.pdf

14:05 Sucrose and trehalose as agents for stabilization of proteins: Insights from atomistic molecular dynamics simulations

Finding harmless stabilizing compounds for proteins and nucleic acids is an important task in modern food and pharmaceutical industries. In this work sucrose and trehalose were investigated as stabilizing agents for myoglobin and the Aβ(1-42) peptide. Atomistic molecular dynamics (MD) simulations were carried out for several systems of myoglobin/Aβ(1-42) in both water and aqueous solutions of sucrose or trehalose. Analysis of resulting MD trajectories and computed self-intermediate scattering functions showed that trehalose slowed down the dynamics of myoglobin and Aβ(1-42) more than sucrose. The difference was particularly large in the case of Aβ(1-42). It was also observed that the rotational motions of the disaccharide molecules slowed down by the presence of myoglobin/Aβ(1-42), and also in this case the effect was largest for Aβ(1-42). Perhaps slightly unexpected from the observed dynamics, it was found that the trehalose molecules exhibited less direct interactions with myoglobin/Aβ(1-42) than sucrose, but caused a stronger reduction of the water dynamics in the hydration shell of myoglobin/Aβ(1-42), and thereby also slowed down and stabilized myoglobin/Aβ(1-42) more efficiently. This mechanism can explain why trehalose generally is a better stabilizing agent than sucrose and other types of sugar.

Speaker: Dr Inna Ermilova (Chalmers University of Technology)

Inna Ermilova.pdf

14:30 Structure and Dynamics of Huntingtin. A Segmental Labelling Approach

Structure and Dynamics of Huntingtin. A Segmental Labelling Approach
Xamuel L. Lund [1], Amin Sagar [1], Frank Gabel [2], Anne Martel [3], Pau Bernadó [1]
[1] Centre de Biochimie Structurale (CBS), INSERM, CNRS, Université de Montpellier, 34090 Montpellier, France
[2] Université de Grenoble Alpes, CEA, CNRS, IBS, 38044 Grenoble, France
[3] Institut Laue-Langevin (ILL), 38044 Grenoble, France

Huntington’s Disease (HD) is a neurodegenerative pathology caused by a mutation in the huntingtin gene (1). When the number of consecutive CAG trinucleotides in the first exon (exon-1) of this gene, encoding the intrinsically disordered N-terminal region of the protein Huntingtin (Htt), is increased beyond 35 (pathological threshold), HD symptoms are manifested and the severity and age of onset are correlated with the length of the expansion (1,2). The exon-1 of Htt is a low complexity region that contains the N-terminal 17 residues (N17), a poly-glutamine tract and a proline rich region. We will attempt to elucidate the structural differences between non-pathological and pathological Htt constructs using SANS measurements of Htt exon-1 deuterated in a residue-specific manner (glutamines and prolines) in combination with atomistic ensembles. In order to prioritize the different deuteration patterns and the optimal deuteration ratio of the buffer, the neutron scattering profiles of Htt ensembles in the eight different deuteration schemes and six H2O/D2O ratios were computed and a realistic noise-level was added (3). Using the ensemble optimization method tool (EOM), the ability of each of the samples to yield relevant structural information was evaluated (4). The chosen constructs will then be produced by a Cell-free protein expression system using the appropriate mixture of deuterated amino acids and measured at beamline D22 of ILL. We expect to derive accurate models of the Htt Exon-1 and shed light to the structural bases of the pathological threshold in HD. The proposed strategy paves the way to the characterization of low complexity regions in proteins, which remain elusive for the other structural biology techniques.
1. Saudou F, Humbert S. The Biology of Huntingtin. Neuron. 2016 Mar 2;89(5):910–26.
2. Zuccato C, Valenza M, Cattaneo E. Molecular Mechanisms and Potential Therapeutical Targets in Huntington’s Disease. Physiol Rev. 2010 Jul 1;90(3):905–81.
3. Svergun DI, Richard S, Koch MHJ, Sayers Z, Kuprin S, Zaccai G. Protein hydration in solution: Experimental observation by x-ray and neutron scattering. Proc Natl Acad Sci. 1998 Mar 3;95(5):2267.
4. Estaña A, Sibille N, Delaforge E, Vaisset M, Cortés J, Bernadó P. Realistic Ensemble Models of Intrinsically Disordered Proteins Using a Structure-Encoding Coil Database. Structure. 2019 Feb 5;27(2):381-391.e2.

Speaker: Mr Xamuel Loft Lund (Centre de Biochimie Structurale (CBS), INSERM, CNRS, Université de Montpellier, 34090 Montpellier, France)

Xamuel_Lund.pdf

14:55 Solution Structure of the Gloeobacter violaceus Ligand-Gated Ion Channel Probed by Small Angle Neutron Scattering

Pentameric ligand-gated ion channels transform chemical signals into electrical ones, a process during which they undergo subtle conformational cycling resulting in rapid, reversible gating of an intrinsic transmembrane pore. The pH-gated bacterial channel GLIC has proved a valuable model system for this receptor family, in part due to its accessibility to X-ray crystallography - which has provided structures of closed, open, and intermediate conformations. However, such models raise critical questions as to whether the X-ray structures represent actual functional states, and if the cycle includes additional conformations such as highly contracted or expanded states. Small-angle neutron scattering (SANS) offers an approach to characterizing solution-phase structure, increasingly applicable to biomolecular problems on this scale. Here we characterized GLIC by SANS, in combination with inline size-exclusion chromatography (SEC), match-out deuterated detergents, and computational modeling of multiple putative structural models. SEC-SANS under resting conditions (pH 7) enabled the characterization of non-aggegated species which fit predictions from resting-state models, particularly following unrestrained molecular dynamics simulation of X-ray structures. Curves collected under activating conditions (pH 3) best fit intermediate models along the open-closed transition sampled in simulation, whereas they diverged from expanded-pore models. In addition to supporting less-expanded models of the open state, this work demonstrates that SEC-SANS, in combination with molecular modeling and simulations, can distinguish solution-phase functional states of an ion channel. As neutron source brilliance increases and inline SEC-SANS set-ups become increasingly accessible, such methods will offer valuable tools for the elucidation of receptor conformational cycling and pharmaceutical development.

Speaker: Marie Lycksell (Stockholm University)

Marie Lycksell.pdf

15:20 → 16:00 Coffee Break

16:00 → 18:30 Nanomaterials & Liquids: 2

16:00 An investigation of 3He in C60: A DFT, theoretical and INS study

In the past two decades, advances in the organic chemistry of fullerenes has enabled chemists to synthesise fully closed fullerene cages with atoms/molecules trapped inside [1-4], allowing the study of interesting effects that arise due to the confinement. In this report, a methodology to determine the INS spectrum of a single atom trapped inside a nano-cavity in a near-spherical symmetry by using a combination of theoretical and numerical methods is presented, along with experimental INS results of the 3He@C60 system in order to see if the atom inside behaves like a particle in a box or a harmonic oscillator as predicted by DFT simulations of the potential shape inside the cage. Previous experiments allowed the assignment of the first transition of 3He inside the cage, and based on those results, a theoretical energy diagram and INS spectrum of this system based on
the particle in a box system, is calculated and compared to experimental results obtained on IN1-LAGRANGE which show new features.
[1] K. Komatsu, M. Murata and Y. Murata,Science, 2005,307, 238–240
[2] Y. Morinaka, F. Tanabe, M. Murata, Y. Murata and K. Komatsu,Chem. Commun.,2010,46, 4532–4534
[3] K. Kurotobi and Y. Murata,Science, 2011,333, 613–616.
[4] A. Krachmalnicoff, et al, Nature Chemistry, 2016,8, 953–957

Speaker: Mohamed Aouane (U. Southampton and Institut Laue-Langevin)

Aouane_20_01_2021.pdf

16:25 Neutron scattering and MD simulations applied to geo-inspired nanotubes

Imogolite is an aluminosilicate clay mineral, composed of single-walled nanotubes with stoichiometry of 3(OH)Al2O3SiOH, and having a nanometric inner diameter. Thanks to their one-dimensional inner cavity, they are a model system for studying the dynamics of nanoconfined water. We will present, by means of molecular dynamics (MD) simulations, a study of the evolution of the structure and the dynamics of both the nanotube and water adsorbed inside it, as a function of temperature. Simulations will be presented together with the results of elastic and quasielastic neutron scattering experiments, obtained on IN13 and IN16B.

Speaker: Arianna D’Angelo (Laboratoire de Physique des Solides, CNRS-Université Paris Saclay and Institut Laue-Langevin)

ILL-ESS_Dangelo_A.pdf

16:50 Water dynamics in carbon nitride – a joint QENS, DFT and AIMD study.

Understanding the nanoscale dynamics of water molecules incorporated in layered and porous solids is key in designing next-generation materials for water filtration devices and fuel cells and requires the combination of experimental measurements and MD simulations. Cold neutrons, allowing investigations of molecular motions on a pico- to nano-seconds time regime, offer critical complementarity to computational modelling studies, where experimental parameters are used to feedback into model inputs, providing an effective methodology to understand properties of materials, as well as guiding advanced designs. We have previously demonstrated that crystalline carbon nitrides with a layered polytriazine imide (PTI) structure spontaneously absorbs ~9wt% of water from the atmosphere [1]. Here we present a comprehensive study combining QENS measurements with DFT and ab initio molecular dynamics calculations to describe and quantitative determine water transport throughout graphene-like materials.
Our data indicate that water travels as single-file through nanometer-sized voids ordered within the PTI layers. The process shows a permeance rate similar to subnanometres carbon nanotubes and graphene oxide assemblies. The transport mechanism involves sequence of orientation reversals directed by hydrogen bonding interactions with -N= and -NH groups decorating the ring voids within the carbon nitride nanosheets [2]. The latter resembles the selective transfer through Aquaporin channels in biological systems where water diffusion is facilitated by hydrogen bonding interactions between transported molecules and protein. The results suggest that these PTI-layered nanoporous materials represent excellent candidates for inclusion in membrane systems for next-generation energy and nanofiltration device applications.

References
[1] Suter T. et al. “Formation of an ion-free crystalline carbon nitride and its reversible intercalation with ionic species and molecular water” Chem. Sci., 2019, 10, 2519-2528.
[2] Foglia F. et al., “Aquaporin-like water transport in nanoporous crystalline layered carbon nitride” Sci. Adv., 2020, 6, 39.

Speaker: Karolina Lisowska (Department of Chemistry, University College London)

ESS-ILL (1).pdf

17:15 THE STRUCTURE OF WATER IN CALCIUM-SILICATE-HYDRATES STUDIED BY NEUTRON DIFFRACTION WITH ISOTOPIC SUBSTITUTION.

Water is an essential component in hydraulic binders, exerting control on the hydration reactions and on the long-term behavior of cementitious structures acting as a vector of ionic transportation. Here, a protocol based on water sorption isotherms, X-ray diffraction and thermogravimetric analysis was designed to obtain C-S-H with controlled distribution of interlayer, adsorbed and bulk pore water. Neutron diffraction with isotopic substitution was then used to obtain information about the local ordering of water of these samples. The hydrogen isotope first-order difference method provided a composite radial distribution function centered on the water hydrogens. We also applied molecular dynamics simulations of C-S-H models with varying water contents, which showed a deviation of the H-bond characteristics of the adsorbed water as compared with bulk water. These experimental and computational results provide for the first time a detailed structural description of water and the local environment of hydrogen in C-S-H.

Speaker: Mrs Zhanar Zhakiyeva (ILL)

Zhakiyeva.pdf

17:40 Neutron reflectivity and MD simulation study of ionic liquids and deep eutectic solvents at a solid electrode

Ionic liquids (ILs) have been described as molten salts with a melting point below 100 °C, consisting of a large asymmetric organic cations and organic or inorganic anions. Since the ionic liquids are essentially ionic conductors, their utilization as novel electrolytes for electrochemical devices, such as batteries, electric double layer separators, dye-sensitized solar cells and fuel cells has been the subject of intense studies. To gain an impression of the structures and reactions that occur in electrochemical systems, one consider the interface between a metal and an electrolyte solution. The study of the structure and dynamics of ionic liquids on solid materials is of great importance for understanding electrochemical processes occurring at the interface. A two- and three-electrode electrochemical cells, were developed to carry out neutron reflectivity (NR) measurements. Experimental neutron reflectivity data were fit with the model obtained by Molecular Dynamics (MD) simulations, widely used computational technique that can give valuable information about the composition, structure and dynamics of the layers formed at the liquid/solid interfaces. In-house software was developed and used to calculate neutron reflectivity directly from the MD trajectory. The experimental reflectivity data can be fit very well using a model, obtained from MD simulations. It has been shown in this work that joined use of NR and MDs is an approach, which provides new insights in the structure of complex interfaces.

Speaker: Nebojsa Zec (Helmholtz Zentrum Geesthacht)

ZEC.pdf

Thursday, 21 January

13:00 → 15:05 Soft matter (incl. biology)

13:00 Enhancing refinement with quantum mechanics in neutron protein crystallography

Speaker: Octav Caldararu (LU)

Octav Caldararu.pdf

13:25 New on-line DFT approach to analyse neutron diffraction crystallography and SANS data

In the recent years, software packages implementing the Density Functional Theory (DFT) formalism have made considerable advances. In particular, the advent of modern supercomputing techniques made possible routine quantum-mechanical simulations of systems that were difficult to treat only few years ago. This opens up possibilities of new interdisciplinary collaborations that can benefit from the high level of detail offered by DFT.

In this presentation, we will illustrate the main ideas behind a scientific workflow which employs DFT to complement the information obtained from SANS experiments and neutron crystallography of biological systems of many thousand atoms, of a scale up to thousands of residues (up to more than 20 thousands atoms). We will briefly show some examples of such workflow by applying the recently developed BigDFT code to structural data of enzymes, among which the recent neutron crystallography data of SARS-CoV-2 main protease.

Speakers: Luigi Genovese (CEA-Grenoble), Viviana Cristiglio (Institut Laue-Langevin)

BigDFT_Cristiglio_Genovese.pdf

13:50 Diffusive-like motions in a solvent free myoglobin-polymer hybrid revealed by neutron scattering and MD simulations.

The interaction between proteins and hydration water stabilizes protein structure and promotes functional dynamics, with hydration water enabling protein flexibility. However, some engineered solvent-free protein-polymer hybrids have been shown to preserve protein structure, function and dynamics [1,2]. Here we have studied the dynamics of myoglobin conjugated with a polymer surfactant corona. We performed elastic, quasi-elastic and inelastic neutron scattering at ILL (Grenoble) and MLZ (Garching) to probe dynamics of the protein-polymer hybrid on the ns-ps timescale. In addition, we combined the measurements with MD simulations to gain molecular insights into the molecular motions and verify the relevance of the model used to treat scattering data. After validating the simulations by computing the neutron scattering functions directly from the trajectories, we analyzed hydrogen bond dynamics, mean-square displacements and displacement densities. Both polymer and protein scattering data, measured between 200 and 300 K, have been interpreted using a two-well model, which highlighted coupled motions above the dynamical transition. Further analysis of the MD simulations indicated diffusive-like motions in the polymer and confined motions in the protein [3]. Based on these results, we postulate that liquid-like polymer dynamics plasticize the conjugated protein in a qualitatively similar way as do hydration-water translational motions.

(1) Perriman, A. W.; Brogan, A. P. S.; Cölfen, H.; Tsoureas, N.; Owen, G. R.; Mann, S. Reversible Dioxygen Binding in Solvent-Free Liquid Myoglobin. Nat. Chem. 2010, 2 (8), 622–626.
(2) Gallat, F.-X.; Brogan, A. P. S.; Fichou, Y.; McGrath, N.; Moulin, M.; Härtlein, M.; Combet, J.; Wuttke, J.; Mann, S.; Zaccai, G.; Jackson, C. J.; Perriman, A. W.; Weik, M. A Polymer Surfactant Corona Dynamically Replaces Water in Solvent-Free Protein Liquids and Ensures Macromolecular Flexibility and Activity. J. Am. Chem. Soc. 2012, 134 (32), 13168–13171.
(3) Schirò G, Fichou Y. , Brogan A.P.S., Sessions R., Lohstroh W., Zamponi M., Schneider G.J., Gallat F.X., Paciaroni A., Tobias D.J., Perriman A., Weik M. Diffusive-like motions in a solvent free protein-polymer hybrid, under review.

Speaker: Yann Fichou (Institut de chimie et biologie des membranes et nano-objets (CBMN))

Yann Fichou.pdf

14:15 LiquidLib: An MPI/OpenMP Parallelized Toolbox for Analyzing Molecular Dynamics Simulations with Applications to Neutron Scattering Experiments

Neutron scattering represents a collection of powerful techniques useful for studying the microscopic sturcture and dynamics of materials at the atomic level. However, the interpretation of experimental data is not straightforward and usually requires the incorperation of computer simulations. LiquidLib (http://z-laboratory.github.io/LiquidLib/) is a post-processing package for analyzing the trajectory of molecular dynamics (MD) simulations with applications to neutron scattering experiments. LiquidLib is able to analyze the trajectories generated from popular pacakges such as LAMMPS, GROMACS, VASP, etc. and compute various statistical quantities including the pair distriubution function, structure factor, mean squared displacement, self and collective intermediate scattering functions, self and collective van Hove correlation functions, etc. Since its initial released in 2017, LiquidLib has gained more than two hundered active users from more than twenty countries. With the latest support of MPI/OpenMP parallelization, LiquidLib is able to provide a multifold speedup of the computations and handle large trajectories acrossing multiple processors with or without shared memory.

Speaker: Yanqin Zhai (University of Illinois at Urbana-Champaign)

Yanqin_Zhai.pdf

14:40 Using molecular modeling and neutron scattering experiments to investigate PNIPAM microgels

Poly(N-isopropylacrylamide), commonly known as PNIPAM, is a synthetic polymer mostly exploited for its thermo-responsive nature, but it is also of interest because of its affinity with proteins in terms of energy landscape and amphiphilic chemical composition. Depending on the specific synthesis protocol, different molecular architecture can be originated, such as linear chains or microgels, i.e. polymer network particles with a colloidal size. In this talk I will discuss two studies on the behavior of this synthetic soft material where atomistic molecular dynamics simulations are combined to elastic incoherent neutron scattering experiments.
In the first part I will discuss the behavior of concentrated PNIPAM microgel suspensions across the volume phase transition temperature, where particles go from a swollen state at low temperature to a collapsed state at high temperature. I will show that the volume phase transition in PNIPAM-based systems can be detected at different time- and length-scales as well as in overcrowded conditions [1].
In the second part I will focus on the low temperature behavior of PNIPAM microgels, providing evidence of the occurrence of a dynamical transition akin to that observed in proteins [2,3]. This study is based on a nanoscale model of a microgel network in water, which quantitatively reproduces neutron scattering experiments. By correlating the information extracted from the analysis of the polymer relaxations times, water self-diffusion coefficients and hydrogen bonding interactions I will show that water-polymer coupling plays a driving role in the phenomenon. I will also report the observation of a low temperature dynamical transition in PNIPAM linear chains, which suggest a wide generality of the phenomenon, independently on the macromolecular architecture [4].

References
[1] Zanatta M. et al. “Atomic scale investigation of the volume phase transition in concentrated PNIPAM Microgels” J. Chem. Phys., 2020, 152, 204904.
[2] Zanatta M. et al. “Evidence of a low-temperature dynamical transition in concentrated microgels” Science advances 2018, 4, eaat5895.
[3] Tavagnacco L. et al. “Water-polymer coupling induces a dynamical transition in microgels” The journal of physical chemistry letters, 2019, 10, 870-876.
[4] Tavagnacco L. et al. “Protein-like dynamical transition of hydrated polymer chains” submitted arXiv:2007.11860.

Speaker: Letizia Tavagnacco (CNR-ISC and Department of Physics, Sapienza University of Rome)

Tavagnacco.pdf

15:05 → 16:00 Coffee Break

16:00 → 18:05 Hard condensed matter

16:00 A combined computational and experimental approach to studying structure-property relationships in complex oxide ion conductors

Solid oxide ion conductors are a remarkable class of compounds with many technological applications including fuel cell electrolytes, membranes, and oxygen sensors. Knowledge of the conduction pathways in these materials is crucial for development of new conductors with improved properties. Quasi-elastic neutron scattering experiments provide a unique opportunity to observe oxygen dynamics on the microscopic scale and, when combined with ab initio molecular dynamics simulations, can lead to atomic-level insight into the diffusion processes in complex conductors. We demonstrate this on two excellent oxide ion conductors: La2Mo2O9 and Bi0.913V0.087O1.587. In both cases we were able to elucidate intricate, temperature-dependent migration pathways that had not been previously identified and link them to specific structural motifs in each compound. Extracting this level of information relies on the combined use of experimental and computational studies and is difficult to replicate using the more conventional experimental methods alone.

Speaker: Chloe Fuller (U. Durham)

Chloe Fuller.pdf

16:25 Lattice dynamics and thermal conductivity in complex metallic alloys with atomic simulations in neutron scattering

Complex metallic alloys (CMAs) are long range ordered materials, characterized by large cells, comprising several hundreds of atoms and cluster building blocks. A key property of CMAs is the low lattice thermal conductivity (1.3 W/m.K), which suggests a potential application for CMAs for thermoelectricity.
Engineering lattice thermal conductivity commonly implies controlling the heat flow carried by waves of atomic vibrations called phonons. It is thus fundamental to quantify the heat transmitted and the distance travelled by a single phonon. However, this requires the knowledge of its lifetime, the determination of which is still at the limits of instrumental and numerical capabilities.
Here, we present quantitative measurement and calculation of phonon lifetimes in a single crystal of the clathrate-I phase Ba7.81Ge40.67Au5.33 and o-Al13Co4 which is an approximant of the quasicrystal, decagonal phase AlNiCo CMAs renowned for its puzzling glass-like thermal conductivity. Surprisingly, we find acoustic phonons with long lifetimes travelling over distances from a hundred to tens of nanometers and which are found to dominate the thermal transport. Considering only the three low-energy acoustic phonons, and the observed energy dependence of their lifetime leads to a calculated thermal conductivity in very good agreement with the experimental one. Atomistic simulations show that the finite phonon lifetime is an anharmonic effect, due to structural disorder, explaining the weak temperature of the phonon lifetime. This findings prove that structural complexity is at the origin of the low thermal conductivity in these systems, leading to a drastic reduction of the phase space for long-living heat carrier. Our results provide a novel picture of thermal transport in the systems and underline the state of the art for the simulations to reproduce the observed phonons lifetime and thermal conductivity.

Speaker: Pierre-Francois Lory (Automotive Cell Company, Joint Venture SAFT-PSA)

P-F_Lory.pdf

16:50 Reproducing Anharmonicity in INS Measurements with DFT

Thermal expansion (positive or negative) is the result of phonon anharmonicity. These phonons can be probed using inelastic neutron scattering (INS). By measuring the generalized density of states (GDOS), a dynamical fingerprint of polycrystalline samples is created, which is often compared to DFT calculations. However, finite difference phonon calculations are based on the assumption of independent harmonic phonon modes. For calculating thermal properties, the quasi-harmonic approximation (QHA) is often used to introduce temperature affects into calculations. How valid an approximation this is, depends on each compound. One test of the QHA is to calculate thermal expansion and compare to measurements. However it is possible that these can match fortuitously.
With GDOS measurements at only two different temperatures, one can extract a somewhat crude but very useful plot of anharmonicity vs phonon energy. This can be compared with the implicit anharmonicity calculated with the QHA (see attachment). Comparing the calculated with the measured anharmonicity directly provides a great additional way to support or reject the use of the QHA in your calculations.

References:
1. S. d’Ambrumenil, M. Zbiri, S. J. Hibble, A. M. Chippindale, D. S. Keeble, C. Wright, and N. H. Rees. Phys. Rev. B, 100(17):174302, 2019.
2. S. d’Ambrumenil. Negative Thermal Expansion in Transition-Metal Cyanides. PhD thesis, University of Reading, Institut Laue-Langevin, 2020.

Speaker: Stella d’Ambrumenil (DFT)

17:15 Tunable bonding in the incipient metal thermoelectric GeTe

The group IV-VI chalcogenides have important thermoelectric applications. GeTe has emerged as a promising non-toxic candidate, especially when the high-temperature cubic phase is suppressed to room temperature. However, even the mechanism of phase transition is disputed, as is the presence of disorder. Here we apply ab-initio MD, synchrotron X-ray and dynamic neutron pair distribution function (PDF) analysis to GeTe. We show that previous reports of disorder and symmetry breaking are entirely due to highly damped and anharmonic phonons. As predicted by metavalent bonding theories, this arises due to a softening in local bonding on heating, which strengthens long-range <100>c correlations. This picture is consistent with reported changes in resistivity and dielectric constant, and shown to be ubiquitously present in other polarisable binary chalcogenides and hR6 structured elements. Our results unify the results of local probes and spectroscopy as applied to binary chalcogenides, and should inspire a re-examination of other highly anharmonic energy materials such as hybrid perovskites and ’rattling’ thermoelectrics.

Speaker: Dr Simon Kimber (Universite Bourgogne Franche-Comte)

17:40 Persistent homology for magnetism

Persistent homology (PH) is a relatively new method from algebraic topology that can be used to find features in discrete datasets. The PH algorithms are distributed in a number of popular Python packages, making it easy to start with data analysis. We have shown that this method is useful in Monte Carlo simulations of Heisenberg spins and can reveal the phase diagram. The barcode (a concept in PH) visualizes at which length scales the data has interesting features. The method also has applicability for experimental data, such as post processing the neutron count rate.

Speaker: Mr Bart Olsthoorn ( NORDITA - Nordic Institute for Theoretical Physics)

Olsthoorn.pdf

18:05 → 18:30 Closeout

UserMeeting_conclusion.pdf