Unitary matrix of a hermitian form matrix

Click For Summary

Homework Help Overview

The discussion revolves around finding a unitary matrix U for a given Hermitian matrix H, such that U*HU is diagonal. The original poster presents the matrix H and their attempts to derive the unitary matrix, including eigenvalues and eigenvectors.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to find eigenvalues and corresponding eigenvectors, then normalize them to form the unitary matrix U. They express confusion regarding the validity of their matrix and question the transformation of an eigenvector to achieve a real number in the unitary matrix.

Discussion Status

Participants are exploring the properties of the eigenvectors, particularly focusing on the orthogonality of the vectors associated with a repeated eigenvalue. Guidance has been provided regarding the construction of an orthogonal basis from the eigenvectors.

Contextual Notes

The discussion highlights the challenge of ensuring orthogonality among eigenvectors, especially in the context of repeated eigenvalues, which may affect the formation of the unitary matrix.

Locoism
Messages
77
Reaction score
0

Homework Statement



Given the matrix H=
\begin{array}{cc}
4 & 2+2i & 1-i \\
2-2i & 6 & -2i \\
1+i & 2i & 3 \\
\end{array}

Find a unitary matrix U such that U*HU is diagonal
(U* is the conjugate transpose of U, and U* = U-1)

The Attempt at a Solution



I find the eigenvalues
λ1 = 9
λ2 = 2
λ3 = 2
and the corresponding eigenvectors
v1 = [ 1-i , -2i , 1 ]
v2 = [ -1+i , 0 , 2 ]
v3 = [ -1-i , 1 , 0 ]
Normalizing them and using them as column vectors for U I get
\begin{array}{cc}
(1-i)/\sqrt{7} & (-1+i)/\sqrt{6} & (-1-i)/\sqrt{3} \\
-2i/\sqrt{7} & 0 & 1/\sqrt{3} \\
1/\sqrt{7} & 2/\sqrt{6} & 0 \\
\end{array}

but this doesn't work. I try U*U and I try U*HU but neither come out right.
What am I missing?

*edit* I've just noticed that U isn't in hermitian form because of v1. Is there a way of transforming v1 so I get a real number for u1,1? Like multiplying it my the conjugate of 1-i? I'm sure it has something to do with the repeated eigenvalue, but I don't know what to do
 
Last edited:
Physics news on Phys.org
Hi Locoism! :smile:

Your v2 and v3 form a basis for the eigenspace of eigenvalue 2.
However, they are not orthogonal.
 
Ok but then do I construct an orthogonal basis from v2 and v3 (using gram-schmidt)?
 
Yep.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 16 ·
Replies
16
Views
3K
Replies
2
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 12 ·
Replies
12
Views
4K