## LM Suite

## LM Suite

#### Overview:

This suite offers a range of first principles programs that can be used to model different nanoscale structures. The common basis of these codes is the use of a linear muffin tin orbital basis (LMTO) set to describe the electronic structure of the different materials. These orbitals are expanded in terms of the angular momentum index around each atom and can be considered crystal extensions of atomic orbitals.

#### LM Programs available:

**LMTO-ASA**

In typical LMTO calculations under the atomic sphere approximation (ASA), crystals are divided up into regions (muffin tins) around each atom where the LMTO basis is expanded and interstitial or free space regions outside of the muffin tins. One of the main challenges in this approach is insuring that the wavefunctions and their derivatives are continuous at the muffin tin boundaries. LMTO-ASA calculations depend on the fact that the system is close-packed and can be described as a collection of slightly overlapping atomic spheres. Since LMTO approaches rely on localized orbitals, it is essential that the muffin tin regions describe most if not all of the space in the system.

For materials where there is a lot of free space or voids, LMTO-ASA approaches can run into problems. In these situations, it is essential to pad the system with empty spheres (muffin tins with Z=0) to accurately describe the wavefunction throughout the system. Routines are available which can automatically generate empty spheres to fill the space within a system completely.

The LMTO-ASA approach produces a compact basis set that leads to efficient calculations with speeds that rival those found in empirical tight-binding approaches.

**Full Potential LMTO**

The constraints associated with the ASA approximation have led to research into techniques that move beyond the muffin tin boundaries. This has led to different full potential approaches that allow systems with large regions of open space to be calculated with confidence. The LM Suite provides a full potential approach based on Hankel functions which can be used for a wide range of systems. However, while computational accuracy is increased, the computational expense of the calculation also increases.

**GF LMTO Approach**

A program based on a Green's function description of materials is also available in the LM Suite. The program, lmgf, calculates the Green's function for a system in the LMTO basis and uses this information to determine a range of properties including the density of states, band structure, and magnetic moment.

**Principle Layer Green Function LMTO Approach**

LMPG is based on a Green's function description of materials and is ideally suited for examining the properties of multilayer systems or nanoscale devices between contacts. In this code, the system is divided up into three regions, two contacts and a device region. The two contact regions are taken to extend to infinity in the positive and negative z directions, respectively. The device region is divided up into a series of layers where only nearest neighbor interactions between layers are considered. Green's function approaches are a natural choice for transport calculations since the information on the contacts can be incorporated into the Hamiltonian for the device region through an additional self energy term. Scattering from impurities, phonons, and other entities can also be included directly in the formalism through self energy contributions. The lmpg code has been used to examine transport in devices ranging from magnetic tunnel junctions to atomic point contacts.

#### Applications:

- Electronic Structure
- Magnetic properties of materials (exchange coupling, magnetoresistance, spintronics)
- Electronic transport
- Impurities in solids
- Optical Properties of materials

#### Developers:

- M. van Schilfgaarde (Arizona State University)
- A. T. Paxton (Queen's University, U. K.)
- J. Klepeis (Lawrence Livermore National Laboratory)
- M. Methfessel (IHP, Germany)
- T. Kotani (Osaka University, Japan)
- D. A. Stewart (Cornell University)
- S. Faleev (Sandia National Laboratories)
- A. Chantis (Arizona State University)

#### Getting Started:

- LMTO Documentation: LMTO-ASA basic package (version 6.15)
- Running LM Suite on the Nanolab Cluster
*Contact Derek Stewart, stewart (at) cnf.cornell.edu, for more information*

#### Relevant Books and Research Articles:

- Methfessel, M., van Schilfgaarde, M., and Casali, R. A. "A full-potential LMTO method based on smooth Hankel functions", in Electronic Structure and Physical Properties of Solids: The Uses of the LMTO Method, Lecture Notes in Physics, 535, edited by H. Dreysse, Springer-Verlag, Berlin (2000).
- Kurdnovsky, J., Drchal, V., Sob, M., Weinberger, P. Electronic Structure of Disordered Alloys, Surfaces, and Interfaces, Kluwer Academic Press, (1996).
- M. W. Finnis, A. T. Paxton, M. Methfessel, and M. van Schilfgaarde, "Crystal Structures of Zirconia from First Principles and Self-Consistent Tight-Binding",
*Physical Review Letters*,**81**, 5149 (1998) - M. van Schilfgaarde and V. P. Antropov, "First Principles Exchange Interactions in Fe, Ni, and Co",
*Journal of Applied Physics*,**85**, 4827 (1999) - Belashchenko, K. D., Tsymbal, E. Y., van Schilfgaarde, M., Stewart, D. A., Oleinik, I. I., and Jaswal, S.,"Effect of Interface Bonding on Spin-Dependent Tunneling from oxidized Co Surfaces",
*Physical Review B*,**69**, 174408, (2004). - Stewart, D. A., and van Schilfgaarde, M., "Digitally doped magnetic resonant tunneling devices: High tunneling magnetoresistance systems",
*Journal of Applied Physics*,**93**, 7355, (2003). - Faleev, S., Leonard, F., Stewart, D. A., van Schilfgaarde, M., "Ab-initio TB-LMTO method for non-equilibrium electron transport in nanosystems",
*Physical Review B*,**71**, 195422, (2005). - Faleev, S. V., van Schilfgaarde, M., Kotani, T., "All-electron self-consistent GW approximation: Application to Si, MnO, and NiO",
*Physical Review Letters*,**93**, 126406, (2004).

#### Questions, Comments...

Please contact:

Derek Stewart

stewart (at) cnf.cornell.edu

Cornell Nanoscale Facility