Bulletin of the American Physical Society
APS March Meeting 2017
Volume 62, Number 4
Monday–Friday, March 13–17, 2017; New Orleans, Louisiana
Session E7: FirstPrinciples Modeling of ExcitedState Phenomena I: Methodological AdvancesFocus

Hide Abstracts 
Sponsoring Units: DCOMP DCP DMP Chair: Volker Blum, Duke University Room: 266 
Tuesday, March 14, 2017 8:00AM  8:12AM 
E7.00001: Koopmanscompliant functionals as spectral theories: molecules, solids, and liquids Nicola Marzari, Ngoc Linh Nguyen, Nicola Colonna, Andrea Ferretti Koopmanscompliant functionals\footnote{I. Dabo, M. Cococcioni, and N. Marzari, arXiv:0901.2637 (2009); I. Dabo {\it et al.}~Phys.~Rev.~B 82 115121 (2010).} enforce a generalized criterion of piecewise linearity with respect to the fractional removal or addition of an electron (i.e. a charged excitation) from any orbital in local or semilocal totalenergy DFT functionals. By doing so they lead to beyondDFT orbitaldensity dependent functionals that are able to deliver spectroscopic properties. We'll present an overview of the current status when applied to ionization potentials, electron affinities, photoemission spectra, and band gaps of molecules, solids, and liquids, with results that are comparable or slightly superior to manybody perturbation theory, but with functionals that rely only on the physics of the PBE generalizedgradient approximation. [Preview Abstract] 
Tuesday, March 14, 2017 8:12AM  8:24AM 
E7.00002: Screening and linear response in Koopmanscompliant functionals Nicola Colonna, Ngoc Linh Nguyen, Andrea Ferretti, Nicola Marzari The need to describe relaxation effects in the fractional removal or addition of an electron requires screening the orbitaldependent corrections of Koopmanscompliant functionals. Here, we present a general method to incorporate orbitalbyorbital screening based on linearresponse theory. We illustrate the importance of such generalization when dealing with challenging system containing orbitals with very different chemical character, such as transitionmetal complexes. Results for ionization potentials, when compared with manybody perturbation theory (MBPT) and experiments, show a remarkably good performance, comparable to the most accurate MBPT approach (G0W0@PBE0). [Preview Abstract] 
Tuesday, March 14, 2017 8:24AM  8:36AM 
E7.00003: Ultrafast phase transitions in advanced materials: review of some experiments and a new theoretical approach Roland Allen, Ayman AbdullahSmoot, Michelle Gohlke, David Lujan, James Sharp, Ross Tagaras This talk will review some experimental studies of advanced materials responding to fast intense laser pulses, including lightinduced superconductivity in cuprates [1]. A new method will be introduced for treating ultrafast phase transitions, such as those involving superconductivity, magnetism, charge density waves, and spin density waves. This method is made possible by the fact that the densityfunctionalbased technique emphasized here (and also standard densityfunctional approaches and other firstprinciples techniques, as long as they include nuclear motion) can yield a true electronic temperature [2]. Illustrative results will be presented for a simple model, with the electronic temperature immediately after the laser pulse calculated as a function of the fluence. \\ 1. D. Fausti, R. I. Tobey, N. Dean, S. Kaiser, A. Dienst, M. C. Hoffmann, S. Pyon, T. Takayama, H. Takagi, and A. Cavalleri, “LightInduced Superconductivity in a StripeOrdered Cuprate”, Science 331, 189 (2011). \\ 2. Zhibin Lin and Roland E. Allen, “Ultrafast equilibration of excited electrons in dynamical simulations”,J. Phys. Condens. Matter 21, 485503 (2009). [Preview Abstract] 
Tuesday, March 14, 2017 8:36AM  8:48AM 
E7.00004: A nonadiabatic exchangecorrelation potential for stronglycorrelated materials: local impurity approximation and beyond Volodymyr Turkowski, Shree Ram Acharya, Talat S. Rahman We formulate a nonadiabatic TimeDependent DensityFunctional Theory (TDDFT) for materials with strong local (onsite) electronelectron interactions. In this approach, the TDDFT exchangecorrelation (XC) potential is derived from the expression for the single electron localinspace selfenergy obtained from the manybody Dynamical MeanField Theory (DMFT) solution for an effective Hubbard model (ShamSchluter equation). We attest to the validity of the formalism through good agreement of our TDDFT results with the nonequilibrium DMFT solution for the ultrafast excited state charge dynamics of the Hubbard model for a system perturbed by a short laser pulse. To include the effects of spatial nonlocality in the XC potential, we propose a generalization of the formalism by including a momentumdependent correction to the DMFT electron selfenergy (the dynamical vertex approximation). We apply the approach to analyze the ultrafast breakdown of the insulating phase in VO2 and show that the TDDFT results are also in a good agreement with available experimental data. The developed approach can be used to study the ultrafast response of complex strongly correlated materials, a task that current manybody approaches fail to address fully because of their inherent computational demands. [Preview Abstract] 
Tuesday, March 14, 2017 8:48AM  9:00AM 
E7.00005: Quasiparticle spectra obtained through stochastic manybody methods Vojtech Vlcek, Roi Baer, Eran Rabani, Daniel Neuhauser We present the linearly scaling stochastic approach to manybody perturbation theory and to calculations of quasiparticle energies in $G_0W_0$ approximation and beyond. Our approach overcomes the steep scaling of conventional deterministic schemes. Further, it allows a simple incorporation of higher order interactions (vertex corrections). Exemplifying on covalently bonded systems (nanocrystals and polymer chains), we show practical calculations of quasiparticle spectra, and selfenergies for large systems with thousands of electrons. The linear scaling is fundamental nature of our approach, which does not rely on a particular character of the electronic structure (e.g., there is no need for sparsity of the density matrices). The scaling prefactor is small so the stochastic $G_0W_0$ method is thus a method of choice for all systems from few tens to thousands  and in the near horizon hundreds of thousands  of electrons. [Preview Abstract] 
Tuesday, March 14, 2017 9:00AM  9:12AM 
E7.00006: Temperature dependence of the oneelectron Green's function within the cumulant formalism J. J. Kas, J. J. Rehr Recently there has been renewed interest in the cumulant expansion for the oneelectron Green's function due to its success in explaining inelastic losses and manybody effects beyond the GW approximation in xray photoemission and absorption spectra. For example, the approach has shown great promise in reproducing the observed multipleplasmon satellites in the XPS of metals and semiconductors,\footnote{J. Zhou et al., J. Chem. Phys. 143, 184109 (2015).} as well as the satellite structure in chargetransfer systems such as CeO2.\footnote{J.J. Kas, J.J. Rehr and J.B. Curtis, Phys. Rev. B 94, 035156 (2016).} Here we investigate the role of temperature on these satellite features using a finite temperature cumulant expansion for the retarded oneelectron Green's function. The cumulant is related to the retarded GW selfenergy, which is determined from the Matsubara Green's function and response function within the RPA. We apply the method to the uniform electron gas over a wide range of temperatures, and we discuss the implications of these results on measurements of XPS and Compton scattering, which can be used as a thermometer for warm dense matter. [Preview Abstract] 
Tuesday, March 14, 2017 9:12AM  9:24AM 
E7.00007: A ManyBody Formalism of $\Delta$SCF Approach for Simulating XRay Spectra from FirstPrinciples YUFENG LIANG, John Vinson, Sri Pemmaraju, Walter Drisdell, Eric Shirley, David Prendegast Accurately reproducing Xray spectral fingerprints for materials characterization relies heavily on how to correctly model the manyelectron response to the generation of an Xray core hole. In this talk, we present a novel firstprinciples theory for simulating Xray spectra that is based on manyelectron wavefunctions. The proposed theory go beyond the electronhole correlations within the BetheSaltpeter Equation and consider higherorder vertex corrections up to the level of MahanNozi\'eresDe Dominicis (MND) theory. An efficient algorithm is invented to incorporate these manyelectron processes by using linear algebra rather than iterating over all Feynman diag [Preview Abstract] 
Tuesday, March 14, 2017 9:24AM  9:36AM 
E7.00008: The Hubbard dimer: exact dynamic exchangecorrelation kernel, single and double excitations of a strongly correlated problem Jaime Ferrer, Diego Carrascal, Neepa Maitra, Kieron Burke The Hubbard dimer is an exactly solvable model of a strongly correlated problem [1]. We develop here its exact frequencydependent exchangecorrelation kernel. Armed with this, we analyse the behaviour of the single and double excitations of the model as they evolve from the weak correlation regime deep into the stronglycorrelated MottHubbard regime. [1] D. J. Carrascal, J. Ferrer, J. C. Smith and K. Burke, The Hubbard dimer: a density functional case study of a manybody problem, Journal of Physics: Condensed Matter 27, 393001 (2015). [Preview Abstract] 
Tuesday, March 14, 2017 9:36AM  9:48AM 
E7.00009: Linkedcluster formulation of screened electronhole interaction from explicitlycorrelated geminal functions without using unoccupied states Arindam Chakraborty, Michael Bayne The accurate determination of the electronhole interaction kernel remains a significant challenge for precise calculations of optical properties in the GW+BSE formalism. The inclusion of unoccupied states has long been recognized as the leading computational bottleneck that limits the application of this approach for large finite systems. In this work, we present an alternative derivation that avoids using unoccupied states to construct the electronhole interaction kernel. The central idea of our approach is to use explicitly correlated geminal functions for treating electronelectron correlation for both ground and excited state wave functions. We demonstrate with diagrammatic techniques that the frequencydependent electronhole kernel can be expressed in terms of connected closedloop diagrams. We show that the cancelation of disconnected diagrams is a consequence of the linkedcluster theorem in realspace representation and the resulting renormalized operators are equivalent to infiniteorder summations of particlehole diagrams. The derived electronhole interaction kernel was used to calculate excitation energies in atoms, molecules, clusters and quantum dots and the results for these systems were compared with CIS, TDHF, TDDFT, EOMCCSD, and GW+BSE calculations. [Preview Abstract] 
Tuesday, March 14, 2017 9:48AM  10:24AM 
E7.00010: Connecting Interface Structure to Energy Level Alignment at Aqueous Semiconductor Interfaces Invited Speaker: Mark Hybertsen Understanding structurefunction relationships at aqueous semiconductor interfaces presents fundamental challenges, including the discovery of the key interface structure motifs themselves. Important examples include the alignment of electrochemical redox levels with the semiconductor band edges and the identification of catalytic active sites. We have developed a multistep approach, initially demonstrated for GaN, ZnO and their alloys, motivated by measured high efficiency for photocatalytic water oxidation. The interface structure is simulated using ab initio molecular dynamics (AIMD).\footnote{N. Kharche, et al., \textit{Phys. Chem. Chem. Phys.} 16, 12057 (2014)} The calculated, average interface dipole is combined with the GW approach from manybody perturbation theory to calculate the energy level alignment between the semiconductor band edges and the centroid of the occupied 1b1 energy level of water and thus, the electrochemical levels.\footnote{N. Kharche, et al., \textit{Phys. Rev. Lett.} 113, 176802 (2014)} Cluster models are used to study reaction pathways.\footnote{M. Z. Ertem, et al., \textit{ACS Catal.} 5, 2317 (2015)} The emergent interface motif is the full (GaN) or partial (ZnO) dissociated interface water layer. \newline \newline Here I will focus on the aqueous interfaces to the stable TiO$_{\mathrm{2}}$ anatase (101) and rutile (110) facets. The AIMD calculations reveal interface water dissociation and reassociation processes through distinct pathways: one direct at the interface and the other via a spectator water molecule from the hydration layer. Comparisons between the two interfaces shows that the energy landscape for these pathways depends on the local hydrogen bonding patterns and the interplay with the interface template. Combined results from different initial conditions and AIMD temperatures demonstrate a partially dissociated interface water layer in both cases. Specifically for rutile, structure and the GWbased analysis of the interface energy level alignment agree with experiment. Finally, hole localization at different interface structure motifs will be discussed. \newline \newline Work performed in collaboration with J. Lyons, N. Kharche, M. Ertem and J. Muckerman, done in part at the CFN, which is a U.S. DOE Office of Science Facility, at BNL under Contract No. DESC0012704 and with resources from NERSC under Contract No. DEAC0205CH11231. [Preview Abstract] 
Tuesday, March 14, 2017 10:24AM  10:36AM 
E7.00011: Reduced order polarizability method for large scale GW calculations Minjung Kim, Subhasish Mandal, Eric Mikida, Kavitha Chandrasekar, Eric Bohm, Nikhil Jain, Laxmikant Kale, Glenn Martyna, Sohrab IsmailBeigi The GW method is an important tool for accurate calculation, from first principles, of excited electronic systems. However, the GW method has not been routinely applied to large scale materials physics or chemistry problems due to its heavy computational load and large memory requirements. The most computationally intense part of GW calculation is the calculation of the polarizability matrix: for standard ``sumoverstates'' approaches, it scales as N$^4$ where N is the number of electrons in the system. As part of our team's effort towards developing massively parallel GW software that can be readily applied to largescale systems, we have implemented a realspace algorithm which greatly reduces the number of fast Fouriertransform to build polarizability matrix (in a plane wave basis). Using this realspace representation of the polarizability matrix, we are then able to develop two types of cubicscaling polarizability methods that use interpolation or Gaussian quadrature to simplify the treatment of energy dependencies. We will describe the methods and their accuracies and efficiencies when applied to crystalline materials. [Preview Abstract] 
Tuesday, March 14, 2017 10:36AM  10:48AM 
E7.00012: Constructing the GW selfenergy of a point defect from the perfect crystal and the near neighborhood of the defect Dmitry Skachkov, Mark van Schilfgaarde, Walter Lambrecht The fullpotential linearized muffintin orbital method allows for a real space representation of the GW or quasiparticle selfconsistent (\textit{QS})GW selfenergy $\Sigma_{\mathrm{R,L;R\mbox{'}+T,L\mbox{'}}}$. This can be used to construct the selfenergy matrix for a point defect system in a large supercell from that of the perfect crystal in the primitive cell and the selfenergy of the defect site and its near neighborhood, obtained selfconsistently in a smaller supercell. At the interface between both regions we can average the two types of $\Sigma _{\mathrm{R,L;R\mbox{'}+T,L\mbox{'}}}$ matrix blocks. The result relies on the limited range of the selfenergy matrix in real space. It means that we can calculate the quasiparticle energy levels of the defect system at essentially the cost of a DFT calculation and a few \textit{QS}GW calculations for relatively small systems. The approach presently focuses on quasiparticle energy levels of band structures of the defect system rather than total energies. We will present test results for As$_{\mathrm{Ga\thinspace }}$in GaAs, Zn$_{\mathrm{Ge}}$ in ZnGeN$_{\mathrm{2}}$, N$_{\mathrm{O}}$, V$_{\mathrm{O}}$, V$_{\mathrm{Zn}}$, and N$_{\mathrm{O}}$V$_{\mathrm{Zn}}$ in ZnO. [Preview Abstract] 
Tuesday, March 14, 2017 10:48AM  11:00AM 
E7.00013: Spinwave excitations and electronmagnon scattering from manybody perturbation theory Christoph Friedrich, Mathias C. T. D. M\"uller, Stefan Bl\"ugel We study the spin excitations and the electronmagnon scattering in bulk Fe, Co, and Ni within the framework of manybody perturbation theory as implemented in the fullpotential linearized augmentedplanewave method. Starting from the \textit{GW} approximation we obtain a BetheSalpeter equation for the magnetic susceptibility treating singleparticle Stoner excitations and magnons on the same footing. Due to approximations used in the numerical scheme, the acoustic magnon dispersion exhibits a small but finite gap at $\Gamma$. We analyze this violation of the Goldstone theorem and present an approach that implements the magnetic susceptibility using a renormalized Green function instead of the noninteracting one, leading to a substantial improvement of the Goldstonemode condition [1]. Finally, we employ the solution of the BetheSalpeter equation to construct a selfenergy that describes the scattering of electrons and magnons. The resulting renormalized band structures exhibit strong lifetime effects close to the Fermi energy. We also see kinks in the electronic bands, which we attribute to electron scattering with spatially extended spin waves. \\ {}[1] M\"uller et al., Phys. Rev. B \textbf{94}, 064433 (2016). [Preview Abstract] 
Follow Us 
Engage
Become an APS Member 
My APS
Renew Membership 
Information for 
About APSThe American Physical Society (APS) is a nonprofit membership organization working to advance the knowledge of physics. 
© 2020 American Physical Society
 All rights reserved  Terms of Use
 Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 207403844
(301) 2093200
Editorial Office
1 Research Road, Ridge, NY 119612701
(631) 5914000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 200452001
(202) 6628700