Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: John Tse Project Name: SnI4 Division: XSD Project title: Modeling of the structure and vibrational dynamics of materials at High Pressure Associated funding: DOE and NSERC (Canada) Other Systems: Science: Electronic structure calculations are to be used to assist the assignments of vibrational density of states (VDOS) obtained from high pressure nuclear resonant inelastic x-ray scattering and diffraction experiments. Several phase transition systems will be studied with the combination of simulations and x-ray experiments done primarily at APS. There are two objectives for this renewal request. [1] Re-examination of the Sn vibrational density in the revitrified face center cubic phase from amorphous SnI4 at 60GPa. [2] Interpretation of the Eu and Fe atomic vibrational density of states in superconducting EuFe2As2 determined from nuclear inelastic scattering experiments. ][1[ SnI4 This is a continuation of the previous project. Using previous CPU allocation, we have performed First Principles Molecular Dynamics calculations and correctly reproduced the high pressure transformation sequence, viz. an ambient crystalline phase to an intermediate crystal structure prior to amorphization and subsequent recrystallization to a face-center cubic (FCC) phase at ∼64 GPa. The calculated Sn vibrational density of states (VDOS) were in good agreement with experiment available at the time below 60 GPa. The results were published in J. Chem. Phys., 143, 164508 (2015). However, recent experiments extended beyond 60 GPa have found that the Sn VDOS in the FCC phase differ significantly from theoretical predictions. The reason is not obvious. We propose to investigate the dynamics of the high pressure phase in more detail in order reveal the cause of this discrepancy. [2]EuFe2As2 Recently the Eu and Fe VDOS of EuFe2As2 under pressure have been measured with nuclear resonance inelastic scattering. It was found that upon compression the a-axis increases with a concomitant decrease in c-axis with increasing pressure. Furthermore, an anomalous softening of the lattice vibrations were observed in the Eu and Fe VDOS. The compression of the c-axis indicates the shortening of the interlayer As-As contact. This will lead to the change in the electronic state of the Fe atoms and this may be the cause of pressure-induced superconductivity in the tetragonal phase. The goal here is to investigate and uncover the link between the phonon softening with changes in the chemical bonding and the appearance of superconductivity using First-principles calculations. Project description: Static and dynamic First-principles electronic structure calculations will be performed to characterize the structure and vibrations of both the SnI4 and EuFe2As2 systems. The VASP code will be used primarily for geometry optimization. In previous work molecular dynamics were also performed with the VASP code. From our experience, on a small system consisted of 32SnI4 molecules in a cubic box(160 atoms) using only one k-point in the simulation, the scaling of VASP was quite reasonable up to 64 CPUS. In the new study, we wish to study to investigate system size effect to the vibrational frequencies. For this purpose, we wish to use the update Qbox code implemented with the ultrasoft pseduopotential (QBall). QBall/Qbox have been demonstrated to scale well on large computer cluster,s particularly on the IBM Blue/gene. Our first goal is to swtup QBall on the cluster and test the scalabiity on increasing the system size incrementally. In the primitive cell of SnI4 there are 4 molecules (20 atoms), we believe a model consisted of 3x3x4 or 4x4x4 replicate of the unit cell should be feasible. Since the Sn and I atoms in the high pressure FCC phase is positionally disordered, we shall employed the special quasi-random structure method (SQS) to generate the model. This is different from our previous study in which the structure obtained from the compression of the low pressure structure with molecular dynamics calculations was used. We expect the SQS will give a better representation of the disorder solid. For the EuFe2As2 calculations, a proper account of the van der Waals interaction between the u-Fe-As layers is essential. We will use a non-published version of Qbox from the author F. Gygi that included the non-local van der Waals functional. First-principles molecular dynamics calculations will be performed in the pressure range from 0 - 15 GPa. Will investigate the change of chemical bonding using Bader's quantum theory of atoms-in-molecule e analysis of the charge density. The Fe and Eu VDOS as a function of pressure will be computed and compare with experiments. Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 491520 Q1: 122880 Q2: 122880 Q3: 122880 Q4: 122880 Justification: Storage requirements: The MD trajectory files are quite large. From previous experience for each run, the total size of all output files are close to 1 Gb. We request 1 TB of disk storage Thank You, The LCRC Accounts System