Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Phay Ho Project Name: XRayCluster Division: CSE Project title: Intense x-ray interactions of heterogeneous nanoclusters Associated funding: DOE Other Systems: Science: The x-ray free-electron laser (XFEL) pulses, currently available at the Linac Coherent Light Source (LCLS) and Spring-8 Angstrom Compact Free Electron Laser (SACLA) facilities provide researchers with unprecedented tools to resolve the dynamics of atoms and electrons with atomic resolution. A recent mode of operation involves sequences of two pulses with different photon energies and controllable femtosecond delay between pulses enabling pump-probe experiments to investigate matter under extreme conditions and create real-time “movies” to track the ultrafast processes as they evolve. However, radiation damage induced by high intensity x-ray radiation in these crystals is unavoidable. In particular, x-ray photoionization, inner-shell transitions (Auger and fluorescence) and secondary ionization can take place with high probability within the pulse duration, and these processes can damage the structure of the molecules and potentially limit the usefulness of the se intense x-ray pulses for molecular structural determination. Our goal is to investigate the fundamental mechanisms of the damage in clusters of varying size, with an ultimate aim of understanding phenomena from molecular to the nanometer scale. Project description: In order to obtain an atomistic view of the dynamical x-ray damage processes and the subsequent structural distortion on the target system throughout the x-ray pulse, we use a combined Monte-Carlo/Molecular-dynamics (MC/MD) computational model. At first, the rates of all inner-shell transitions and the cross sections of photoionization of each subshell will be obtained with Hartree-Fock-Slater calculation. During the x-ray pulse, the occurrences of the photo-ionization and inner-shell decay processes and the site of their occurrences are treated by Monte-Carlo type methods. Subsequently, the dynamics of the photoelectrons, Auger electrons and the atoms/ions in the systems are tracked using molecular dynamic method. The advantage of this MC/MD model is that it allows us to compute the time-dependent coherent x-ray diffraction pattern of the target system by tracking the position of the delocalized electrons and the configuration (electronic configurat ion and charge state) and positions of atoms/ions. Recently, we have modified our MC/MD code to enable calculation of atomic transition rates on-the-fly. This development is essential for clusters containing heavy elements, which have large number of electronic configurations, and it avoids storing of all the transition rates in each compute node that requires tens or hundreds on GB go memory. For this coming year, we plan to investigate the effects of x-ray energy, pulse duration, pulse fluence and particle size (1 to 25 nm) on the complex mechanism of radiation damage with nanoscale doped rare-gas core-shell clusters as our target systems. Our choice of the target systems is motivated by the idea that a sacrificial layer (tamper) is added on the sample to prevent the rapid loss of electrons in the sample during the XFEL, resulting in improvement of the quality of sample's structural information recorded in the scattering patterns. In particular Helium dopant will be investigated for their ability to protect an embedded Xenon cluster. For Xe/He clusters, various particle sizes (core size and dopant thickness) will be examined. Among other things, we will monitor the fluorescence spectrum, x-ray diffraction pattern, charge-state distribution, energy spectra, lattice dynamics, and dynamics of photoelectrons, Auger and secondary electrons in order to gain a more complete dynamical picture of the interaction of intense x-ray pulses with nanoscale materials. The results will aide the development for larger systems, approaching the experimental sizes of 100 nm. Our code includes a modified LAMMPS code to perform MD calculation. LAMMPS is known to have good scalability with MPI/OpenMP parallelization on high-performance computing machines. Our code has been used to investigate the ionization dynamics of homogeneous system on Mira and show excellent scaling efficiency up to 8192 nodes (65536 cores). Since Monte-Carlo sampling is used, each calculation is repeated for many realizations until a numerical convergence is reached for the physical quantity of interest. Thus, many simulations, each uses 8-32 compute nodes, will be bundled as a 128-node job. For this year, we plan to submit 40 jobs, each will require 6 hours and 128 nodes. This corresponds to a total of 409600 core-hours. In addition, we will need about 2400 core-hours for development. Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 412000 Q1: 103000 Q2: 103000 Q3: 103000 Q4: 103000 Justification: Storage requirements: 1 TB Thank You, The LCRC Accounts System