How Is Entropy Calculated for a Density Matrix with Eigenvalues 0 and 1?

LagrangeEuler
Messages
711
Reaction score
22

Homework Statement


Calculate entropy for density matrix with eigenvalues ##0## and ##1##.



Homework Equations


##S=-\lambda_1 \ln \lambda_1-\lambda_2 \ln \lambda_2##
where ##\lambda_1## and ##\lambda_2## are eigenvalues of density matrix.


The Attempt at a Solution


How to calculate this when ##\ln 0## is not defined?
 
Physics news on Phys.org
LagrangeEuler said:

Homework Statement


Calculate entropy for density matrix with eigenvalues ##0## and ##1##.



Homework Equations


##S=-\lambda_1 \ln \lambda_1-\lambda_2 \ln \lambda_2##
where ##\lambda_1## and ##\lambda_2## are eigenvalues of density matrix.


The Attempt at a Solution


How to calculate this when ##\ln 0## is not defined?
I'm no expert, but...

What is the value of ##\displaystyle \lim_{\lambda_1\rightarrow 0}\left[\lambda_1\ln{\lambda_1}\right]##?

I think that is the only way to get a numerical answer here: make it approach what you want.
 
For a density matrix ρ with eigenvalues only 0 and 1, we have ρ = ρ^{2}. This is true only for pure states and thus we know the Von Neumann entropy must be zero. To calculate it numerically I would guess the approach Mandelbroth suggested is valid.
 
What is interpretation of that. For pure state entropy is zero. Why?
##-1\ln 1-\lim_{\lambda \rightarrow 0}\lambda \ln \lambda=-\lim_{\lambda \rightarrow 0}\lambda \ln \lambda ##
How to calculate this limit?
 
LagrangeEuler said:
What is interpretation of that. For pure state entropy is zero. Why?
##-1\ln 1-\lim_{\lambda \rightarrow 0}\lambda \ln \lambda=-\lim_{\lambda \rightarrow 0}\lambda \ln \lambda ##
How to calculate this limit?

To calculate the limit let t = 1/x. Then you have \lim_{t\to\infty}=\frac{log(1/t)}{t} = \frac{\infty}{\infty}. Then use L'Hospital's Rule and you will get the answer. Otherwise, you could just type it into wolfram alpha.
 
Tnx a lot! And physically why entropy of pure state is zero?
 
It is easy to see mathematically why the entropy of a pure state is zero. However, why its true physically seems a much harder question, one I'm not sure I know how to answer.
 
Entropy is in some sense a measure of our uncertainty about the state a system. If a system is in a mixed state, is our uncertainty big or small? What about when it is in a pure state?
 
Mute said:
Entropy is in some sense a measure of our uncertainty about the state a system. If a system is in a mixed state, is our uncertainty big or small? What about when it is in a pure state?

Pure state is minimum uncertainty so it makes sense Entropy would be zero.
 
Back
Top