Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Zhenpeng Yao Project Name: CEES-II-Wolverton Division: Computing, Environment, and Life Sciences Project title: DFT Computation of Hybrid and Li-ion Battery Materials Associated funding: CEES-II EFRC Other Systems: Science: The expanded use of renewable but intermittent energy sources coupled with increasing demand for electric transportation vehicles, greatly enhanced the need for advanced energy storage technologies with high energy density and long cycle life. Lithium ion battery (LIB) has been the most prominent electrochemical energy storage technology over the past decades and enabled the wireless evolution of portable electronic devices. Therefore, the development of new high energy electrode materials for rechargeable lithium batteries is becoming more essential. We are interested in several different high energy density lithium ion battery systems including the traditional conversion materials and the novel hybrid Li-ion/Li-O2 materials. The conversion-type materials (e.g., Co3O4 and NiO) can offer significantly larger capacities because of their abilities to make use of all possible valence states of the metal cations, but also it may pose significant kinetic challenges (e.g. , reversibility and voltage hysteresis). At the same time, the hybrid materials (like Li5FeO4) can realize high capacities by releasing and re-accommodating lithium and oxygen reversibly during charging and discharging, which also suffer from cyclability problems. Density functional theory (DFT) has been facilitating lithium-ion battery design and improvement by enabling predictions of operation voltage, phase diagram, lithium-ion transportation and so on for various electrode materials. Using LCRC, we will be able to provide new mechanisms and clarify discrepancies in previous experimental/computational studies on traditional conversion materials and the novel hybrid Li-ion/Li-O2 materials. Project description: Using computational resources from LCRC, we succeeded to predict new candidates for the hybrid Li-ion/Li-O2 materials by the high throughput DFT method. We will continue this project to identify the mechanisms behind the poor cyclability of current hybrid material (Li5FeO4). We will focus on studying the phase evolution during the charge and discharge of current hybrid material and look into the thermodynamic/kinetic obstacles. With computational support from LCRC, we have developed an accurate, predictive computational approach for the metastable lithiation process of a conversion material: Co3O4 for lithium-ion batteries. Based on this method coupled with in-situ transmission electron microscopy (TEM) observations, we have unraveled the detailed lithiation reaction pathway of Co3O4 by systematically exploring the energetics of stable and metastable LixCo3O4 structures using first-principle calculations. Based on what we have learned from Co3O4, we plan to apply our method to other conversion material systems in order to obtain a more general understanding of the detailed mechanism for conversion materials during charging and discharging. Our study could help future experiments to overcome the current limitations of the conversion-type electrode materials and promote the development of more advanced LIBs. Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 2165760 Q1: 541440 Q2: 541440 Q3: 541440 Q4: 541440 Justification: To identify the kinetic obstacles during the charge and discharge of current hybrid material, defect formation energies and ionic migration barriers would be evaluated for all the intermediate phases (~5). Considering the number of defect types (~6), charge states (~3), geometrically different configurations (~4), and an averaged relaxation computation time of 96 cores for 24, defect related calculations would consume 5*6*3*4*96*24 = ~829,440 CPU hours. Ionic migration barrier calculations are required for all the intermediate phases (~5) considering different types of ions (~3), geometrically different pathways (~4), and an averaged nudged elastic band theory computation time of 96 cores for 72, 5*3*4*96*72 = ~414,720 CPU hours are needed. The phase prediction method used for studying the conversion reactions is effective yet computational resource consuming. For each material system we planned to work on, a series of stoichiometries (~20) along the lithiation/delithiation process will be considered. For each of these stoichiometries, there could be up to ten thousands of geometrical configurations. To save computational time, we rank all the configurations with their electrostatic energies and select lowest ones (~4) to be relaxed in DFT. The relaxation for each of them usually needs 96 cores for 24 hours on average. We will choose several typical conversion materials (~5) to study which will spend: 5*20*4*96*24 = ~921,600 CPU hours in total. To summarize, we request a total amount of 2,165,760 CPU hours in the next fiscal year (~541,440/Qtr.). Storage requirements: Thank You, The LCRC Accounts System