Approximate Methods and Linear Algebra

In terms of essentials we only have a few more topics. Specifically we have four goals  

To describe and illustrate the simplest form of first-order perturbation theory 

To explain the secular equation which results from the LCAO approximation 

While this may seem to be intense ad hoc mathematics, it represents what took this author at least 5 years to locate and assemble into the necessary tools to function using the linear combination of atomic orbitals and a sense of how to approach seemingly impossible problems in quantom chemistry. We offer it here as a shortcut for interested undergraduates and auxiliary mathematics as an abbreviated course in linear algebra, which can be useful in a number of areas of chemistry. We will mention that this chapter is usually the end of a nine week (six credit, two semester) course and example 2 is almost always treated and tested. 

Now that we have some better understanding of where the H atom orbitals come from, the next topic should be the electronic structure of molecules and ways to treat problems for which we are unable to solve the Schrodinger equation exactly. Recall the difficulty of solving the Schrodinger equation for just one electron in the H atom. Then perhaps you may faint when you consider the notion of how one might treat the electronic structure of benzene with 12 atoms and 42 electrons Well, there is no known exact solution for even the He atom with only two electrons so do not faint but continue to wonder about how we are going to treat the multielectron case for molecules. There are two main methods the variation method and perturbation theory. In this chapter, we will emphasize the variation method, which is the most powerful mathematical approach, and give a few key examples. However, we will first mention the basic approach of perturbation theory but without much elaboration since it is the weaker of the two methods. 

---

Direct and iterative methods. Recall that the final results of the difference approximation of boundary-value problems associated with elliptic equations from Chapter 4 were various systems of linear algebraic equations (difference or grid equations). The sizes of the appropriate matrices are extra large and equal the total number N of the grid nodes. For... 

This allows us to represent partial differential equations as found in the balance equations using the collocation method. Equation (11.47) is a solution to a partial differential equation represented by a system of linear algebraic equations, formed by the interpolation coefficients, oij, and the operated radial functions. The interpolation coefficients are solved for using matrix inversion techniques to approximately satisfy the partial differential equation... 

A linear algebraic system of rate equations for the fast species results, which can be solved a priori. Hence a strongly reduced (in the number of species to be treated) system is obtained. This concept originates from astrophysical applications and from Laser physics. It is in some instances also referred to as collisional-radiative approximation , for the fast species, lumped species concept , bundle-n method or intrinsic low dimensional manifold (ILDM) method in the literature. We refer to [9,12,13] for further references on this. 

The calculation of temperatures and equilibrium compositions of gas mixtures involves simultaneous solution of linear (material balance) and nonlinear (equilibrium) algebraic equations. Therefore, it is necessary to resort to various approximate procedures classified by Carter and Altman (Cl) as (1) trial and error methods (2) iterative methods (3) graphical methods and use of published tables and (4) punched-card or machine methods. Numerical solutions involve a four-step sequence described by Penner (P4). 

In this paper, an inverse problem for galvanic corrosion in two-dimensional Laplace s equation was studied. The considered problem deals with experimental measurements on electric potential, where due to lack of data, numerical integration is impossible. The problem is reduced to the determination of unknown complex coefficients of approximating functions, which are related to the known potential and unknown current density. By employing continuity of those functions along subdomain interfaces and using condition equations for known data leads to over-determined system of linear algebraic equations which are subjected to experimental errors. Reconstruction of current density is unique. The reconstruction contains one free additive parameter which does not affect current density. The method is useful in situations where limited data on electric potential are provided. 

The boundary conditions are applied in the finite element method in a different way than in the finite difference method, and then the linear algebra problem is solved to give the approximation of the solution. The solution is known at the grid points, which are the points between elements, and a form of the solution is known in between, either linear or quadratic in position as described here. (FEMLAB has available even higher order approximations.) The result is still an approximation to the solution of the differential equation, and the mesh must be refined and the procedure repeated until no further changes are noted in the approximation. 

The determination of the concentration of the components (tT ) at the collocation points can be obtained by solving the set of non linear algebraic equations [5(N + 1)] resulting from insertion of equations (B.41-B.45) into equation (B.40) and excluding the surface where the concentrations are known. These equations are solved numerically by an IMSL (International Mathematical and Statistical Library) subroutine called ZSPOW based on a variation of Newton s method which uses a finite difference approximation to the Jacobian and takes precautions to avoid large step sizes or increasing residuals. 

The linear variation method is the most commonly used method to find approximate molecular wave functions, and matrix algebra gives the most computationally efficient method to solve the equations of the linear variation method. If the functions /i, in the linear variation function [Pg.228]

---