12_variation_perturbation_matrixhq

# To 2 2 2 2 2 2 2 2 derivative

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 2 H1,1S 2, 2 H 2, 2 S1,1 H1,1 H 2, 2 H12, 2 4 2 S1,1S 2, 2 S1, 2 S1,1S 2, 2 S12, 2 2 boils down to values for matrix elements , and , matrix elements ˆ H i , j f i Hf j d Si , j f i f j d When S is not the Identity Matrix (Basis Functions are Not Orthonormal) E has eigenvalues on diagonal, zero off diagonal C is the matrix of each column of eigenvectors H is the Hamiltonian matrix S is the overlap matrix for the basis functions (H-ES)C=0 right multiply each term by C, rearrange HC=ESC left multiply each side by C-1S-1, rearrange C-1S-1HC=C-1S-1ESC E is diagonal C-1S-1HC=EC-1S-1SC matrix times its inverse is identity matrix C-1S-1HC=E It takes both the eigenvector coefficients and the basis overlap elements to diagonalize the Hamiltonian If Basis Functions are Orthogonal (S=I): it takes only c-1Hc=E to diagonalize the Hamiltonian 17 10/7/2013 2x2 Matrix Example - Particle-in-a-Box with a Trial Wavefunction Symmetric about center at /2 and goes to zero at walls ( 0, ). Can only model the symmetric states, i.e. all of the odd quantum number solutions for particle in a box f n ( x) x n ( a x) n 1 use only two terms and f1 ( x) x(1 x) x x boils down to values for matrix elements 1 2 d 2 ˆ H1,1 f1* Hf1d x x 2 x x2 2 0 2m dx 1 2 dx x x2 0 2m m 2 2 2 31 2 0 2 0 2 1 1 m 2 3 2 x 1 0 2 dx x 1 0 2 2 d 1 2 x dx 2x3 x 4 2m dx 2 x3 2 x 4 x5 2 1 1 1 3 m 4 5 0 m 3 2 5 1 2 x 3 x 4 dx 1 2 d 2 2 ˆ H 2, 2 f 2* Hf1d x 2 2 x 3 x 4 x 2 x3 x 4 2 0 2m dx 2 2 2 12 x 12 x 2 dx x 2 2...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online