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: Pierre Darancet Applicant's institution: ANL Applicant's division: NST Project Name: Tas2_stacking Project title: Understanding the stacking dependence of the electronic structure of TaS2 Associated funding: BES Other Systems: Carbon cluster at the Center for Nanoscale Materials: 100,000 hours Science: Using density functional theory (DFT) calculations, we aim at understanding the conditions of emergence of the “Mott” insulating behavior of the layered transition-metal dichalcogenide (TMDC) 1T-TaS2. Among TMDCs, 1T-TaS2 is known for its so-called “Star of David” CDW in which an in-plane, periodic lattice distortion involving 13 tantalum atoms transforms the high symmetry metal into an odd-electron-number insulator, suggesting a Mott transition [1]. Interestingly, density functional theory studies have consistently found that the “Mott” insulating phase is destroyed by out-of-plane interactions [2,3], and that the CDW phase is, in fact, a highly anisotropic, out-of-plane metal --a phenomena we explained by the weak effective electron-electron interactions in the CDW state [3]. In the past two years, new experimental findings [4,5,6] have subsequently challenged the DFT view by observing a ground state Mott insulating phase that can be tr ansformed into a long-lived one-dimensional metallic state by light [4], and voltage pulses [5,6,7,8]. In this project, we will attempt to explain these behaviors by computing the electronic structure of different out-of-plane models of the CDW. Specifically, we aim at understanding the effect of the experimentally-determined and theoretically-ignored 13-layer periodicity of the CDW, never accounted for in previous modeling efforts. As a 13-layer-thick CDW involves a unit cell of 507 atoms with ~13^13 possible packings of the CDW centers, we have reduced the total configuration space by computing all the energy-favorable packings for bilayers (5 symmetry-independent configurations), and trilayers (85 configurations) of TaS2 as well as the corresponding Wannier Hamiltonian with the CNM cluster. Extrapolating these energies to the 13-layer unit cell, we have identified ~10 promising, energy-allowed configurations for the 13-layer-thick CDW, and aim in this proposal at computin g their total energies, electronic structure, and electronic Wannier functions. These calculations will serve three purposes: (1) they will provide a direct characterization of the likely experimental structure and determine whether TaS2 is a Mott or an Anderson insulator; (2) they will validate/invalidate our trilayer-based model for the computation of the structural properties; (3) through the computation of the low-energy electronic Wannier function, they will provide an effective packing-dependent Hamiltonian which we will use to predict the electronic structure of all possible packings. References: 1. J. A. Wilson and A. D. Yoffe, Adv. Phys. 18, 193 (1969), and B. Sipos, A. F. Kusmartseva, A. Akrap, H. Berger, L. Forro, and E. Tutis, Nat. Mater. 7, 960 (2008); 2. Y. Ge and A. Y. Liu, Phys. Rev. B 82, 155133 (2010); A. Y. Liu, Phys. Rev. B 79, 220515 (2009) 3. P. Darancet, A. J. Millis, and C. A. Marianetti, Phys. Rev. B 90, 045134 (2014) 4. Y. Yu et al., Nature Nanotechnology 10, 270–276 (2015) 5. A. Tsen et al., PNAS 112 (49) 15054-15059 (2015) 6. I. Vaskivskyi et al., Science Advances, E1500168 (2015) 7. D. Cho et al., Nature Communications 7, 10453 (2015) 8. L. Ma et al., Nature Communications 7, 10956 (2016) Project description: As suggested above, the proposed work involves 10-15 density functional theory structural relaxation of a periodic, 507-atom unit cell. These calculations will involve the VASP DFT package (planewave, periodic boundary conditions), with PAW pseudopotentials and a 4x4x1 kpoint sampling using the GGA(PBE)+U functional. Following our previous results [3], we will neglect spin-orbit coupling to ease the memory requirements as it has shown little effect in the low-symmetry CDW structure. Despite this approximation, the 2873 electrons of the system require the use of large memory nodes and the 64 GB nodes on the Blues cluster are expected to be ideal for such calculations. We have built the full tentative structural models based on 85 DFT-optimized trilayer geometries to minimize the number of ionic displacements necessary to complete the energy minimization. We have made preliminary calculations using the same Fourier grid (on a trilayer + 10 layers of va cuum supercell), and, based on the 64GB gen4 nodes of Carbon, we estimate that each structural relaxation will take up to 40,000 cpu hours on Blues. The total requested time is then 400,000 core-hours. Dr. Pierre Darancet will be responsible for carrying out the simulations. Industry partnership: Project URL: Requested allocation: 400000 Q1: 200000 Q2: 200000 Q3: 0 Q4: 0 Justification: 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