Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Liang Li Applicant's institution: ANL Applicant's division: NST Project Name: TM_LIO Project title: First-Principles Investigations of Transition-Metal-Substituted Li2IrO3 Cathode Material Associated funding: US Department of Energy, Office of Science, Basic Energy Science Other Systems: Internal: 200,000 core-hours, Carbon cluster at Center for Nanoscale Materials Science: lithium-ion batteries (LIBs) have seen substantial advances that enable the widespread use of portable electronic devices including mobile phones and laptop computers, however, considerably higher energy density and longer cycling stability are required for LIBs application in larger scale energy storage (e.g. grid electric energy storage). Various routes have been explored in the pursuit of improved energy density and enhanced cyclability, and the next-generation Li-rich cathode materials have the potential to meet the above-stated needs. Among these Li-rich materials, layered oxides have been studied extensively due to their high theoretical capacity. Recently, Li2IrO3, a layered Li-rich transition metal oxide, has demonstrated high reversible capacity due to its structural stability and involvement of oxygen redox in the electrochemical reactions.1,2 However, the use of Li iridate is substantially hindered by the high cost and large atomic mass of Ir. To enable the practical usage of this type of material, one feasible route is to replace the costly Ir with earth-abundant transition metal (TM) elements, while maintaining or ideally, improving the charge capacity. Another advantage of substituting heavy Ir with lighter elements would be the foreseeable increase of the specific energy. We herein propose a comprehensive computational study, performed in conjunction with experimental work by our collaborators, to investigate the electrochemical and thermodynamic properties of TM-substituted Li2IrO3. A series of TM concentration will be studied, which lead to different compositions, i.e. Li2TMyIr1-yO3 (y= 0.25, 0.50 and 0.75). It is known that oxygen redox plays a critical role in promoting the capacity of Li2IrO3 especially at high voltages, so in this study we will particularly focus on the reactivity of oxygen at various states of charge. The insights from this work are expected to be readily applied to other layered cathode materials, and provide practical guidance on designing low-cost and high-capacity layered cathode materials. Project description: The structure and energetics calculations of Li2-xTMyIr1-yO3 will be performed using the plane wave DFT code VASP, with supplied projector augmented wave (PAW) atomic potentials and Perdew-Becke-Ernzerhof (PBE) exchange-correlation functionals. On-site Coulomb interaction is accounted for by Hubbard U correction. Nonlocal dispersion corrections will also be included in the calculations to account for the van der Waals interaction in the layered materials. XANES spectra simulation will be carried out using the OCEAN code3 that utilizes the Bethe-Salpeter equation (BSE) formalism, built upon a DFT ground-state calculation. The following calculations are specifically proposed in this study, and detailed accounting of the computational time requested for each calculation is also presented: (a) The thermodynamically stable structure of Li2TMyIr1-yO3 will first be determined. Three different TM concentrations (y= 0.25, 0.50 and 0.75) will be investigated. A Li2IrO3 unit cell is used as the host material, which contains 8 Li atoms in total. In our previous study we have established a robust framework based on a structural enumeration technique, to identify the stable structures of layered materials with specific compositions.2 Enumeration of all possible atomic arrangements yield 8C2=28, 8C4=70 and 8C6=28 possible structures for y= 0.25, 0.50 and 0.75, respectively. Considering the structural symmetry, these numbers can approximately be reduced by half, so there are totally ~60 structures to investigate. Based on our previous calculations on Li2IrO3, each structural relaxation calculation normally takes 16 hours using 36 cores. Therefore, 60×16×36~35,000 core hours will be requested for determination of TM-substituted structures. (b) To monitor the structural variation and electronic properties of Li2TMyIr1-yO3 during Li extraction, the structures of Li2-xTMyIr1-yO3 (x=0.25, 0.5, 0.75, 1, 1.25, 1.5, 1.75) will also be examined exhaustively, and the DFT energies obtained in this process will be used to determine the most stable configuration at various Li concentration. Bader charge analysis of transition metals and oxygen atoms will also be performed to understand the charge compensation mechanism. After structural enumeration, the electrostatic energies of Li2-xTMyIr1-yO3 will be calculated and 30 structures with the lowest electrostatic energies will be selected for further DFT calculations. Considering 3 different TM concentrations (y= 0.25, 0.50, 0.75) and 7 intermediated delithiated compositions, 3×7×30=630 calculations are needed. Therefore, all together we expect to need 630×16×36~363,000 core hours. (c) The XANES spectra of stable Li2-xTMyIr1-yO3 phases will be simulated. The XANES simulation is built upon ground-state Kohn-Sham eigenstates with a large number of empty bands being considered. The treatment of the BSE formalism is computationally demanding and typically requires 48 hours on 128 cores for the systems under study. The number of stable phases is unknown unless the DFT calculations described in the prior section are completed, however, a total number of 5 would be a reasonable estimation for each y value, with the fully lithiated Li2TMyIr1-yO3 being included. We therefore request 3×5×48×128~92,000 core hours for XANES calculations. (d) To gain further insights into oxygen reactivity in Li-rich materials and to test the robustness of the computational scheme described above, we plan to extend this project to investigate other Li‑rich layerd transition metal oxides with the same stoichiometry as Li2IrO3, such as Li2RuO3. The energetics of Li2-xRuO3 at various states of charge will be computed following the procedure presented in our previous work.2 Similar as Li2-xTMyIr1-yO3, the complete delithiation process of Li2RuO3 involves 7 intermediate compositions. 30 structures will be evaluated for each intermediate composition, with each calculation taking 16 hours on 36 cores. Four representative compositions of Li2-xRuO3 (x=0. 0.5, 1, 1.5) will be selected for XANES simulation, the computational load of which is similar as Li2-xTMyIr1-yO3. Therefore, our investigation of Li2-xRuO3 requires all together 7×30×16×36+4×36×128~140,000 core hours. The total time requested for this project is approximately 35,000+363,000+92,000+140,000~630,000 core hours. Industry partnership: Project URL: Requested allocation: 630000 Q1: 157500 Q2: 157500 Q3: 157500 Q4: 157500 Justification: The determination of the lowest-energy structure requires an exhaustive search through the vast configurational spaces. The structural relaxation calculations were benchmarked using 4 cores of Blues node. On Blues, if two nodes (32 cores) were used, then as many as 87% core hours are consumed, compared with that using 4 cores. So using 32 cores on Blues is quite efficient. We have not performed any scaling test on Bebop using the bdwall partition, but based on the performance of Blues, our request of using 36 cores (1 node) on Bebop should be efficient. Another demanding part in this project is the X-ray Absorption Spectra calculation. These calculations are performed using OCEAN package, which utilizes a GW and Bethe-Salpeter equation (BSE) formalism and thus is computationally expensive. We optimize the performance of our code by dividing the calculations into two steps: first in the DFT calculations will be performed using two KNL nodes (128 cores), then in the following core-hole screening stage, the calculations will be performed using bdwall partition using only 1 node (36 cores). Storage requirements: The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System