Square root of Dirac Delta function

Click For Summary
The discussion revolves around the challenge of dealing with the square root of the Dirac Delta function, \sqrt{\delta(x)}, in the context of calculating wave functions for quantum mechanics problems. The initial approach involves differentiating \sqrt{\delta(x)} and attempting to integrate it, but these methods yield unsatisfactory results. Alternative definitions of the Dirac Delta function are explored, yet they also fail to resolve the issue. An attempt to square the wave function to eliminate the square root does not provide a solution either. The conversation highlights the complexity of manipulating the Dirac Delta function and seeks further insights or corrections from participants.
ismaili
Messages
150
Reaction score
0

Homework Statement



I wonder how to deal with the square root of Dirac Delta function, \sqrt{\delta(x)}. Actually, this comes from a problem which asking readers to calculate the wave function of a free particle and of a harmonic oscillator at time t, provided that the wave function at time t=0 is given, i.e. \psi(x,t=0) = \sqrt{\delta(x-a)}.

The way to obtain \psi(x,t) in this problem is by the integral with the propagators.
However, I have no idea how to deal with \sqrt{\delta(x)}.

Homework Equations



\sqrt{\delta(x)}=?



The Attempt at a Solution



I tried to differentiate it,
\frac{d}{dx}\sqrt{\delta(x)} = \frac{1}{2\sqrt{\delta(x)}}\delta&#039;(x)<br /> = -\frac{1}{2\sqrt{\delta(x)}}\frac{\delta(x)}{x} = -\frac{1}{2}\frac{\sqrt{\delta(x)}}{x}
\Rightarrow <br /> \int \sqrt{\delta(x)}f(x)dx = -2\int xf(x)d(\sqrt{\delta(x)}) <br />
But this is still no good...

I also tried the other definition of Dirac Delta function,
\delta(x)\sim\lim_{\epsilon\rightarrow 0}e^{-x^2/\epsilon}
\Rightarrow \sqrt{\delta(x)} = \lim_{\epsilon\rightarrow 0}e^{-x^2/2\epsilon}<br /> =\lim_{\epsilon&#039;\rightarrow 0}e^{-x^2/\epsilon&#039;} = \delta(x) ??

still no good...

I also tried to calculate \psi(x,t)^2 to try to get rid of the square root,
but it seems doesn't help.

Is there anyone who has any ideas about \sqrt{\delta(x)} ?
Any help will be appreciated, thanks.

 
Physics news on Phys.org
ismaili said:

Homework Statement



I wonder how to deal with the square root of Dirac Delta function, \sqrt{\delta(x)}. Actually, this comes from a problem which asking readers to calculate the wave function of a free particle and of a harmonic oscillator at time t, provided that the wave function at time t=0 is given, i.e. \psi(x,t=0) = \sqrt{\delta(x-a)}.

The way to obtain \psi(x,t) in this problem is by the integral with the propagators.
However, I have no idea how to deal with \sqrt{\delta(x)}.

Homework Equations



\sqrt{\delta(x)}=?



The Attempt at a Solution



I tried to differentiate it,
\frac{d}{dx}\sqrt{\delta(x)} = \frac{1}{2\sqrt{\delta(x)}}\delta&#039;(x)<br /> = -\frac{1}{2\sqrt{\delta(x)}}\frac{\delta(x)}{x} = -\frac{1}{2}\frac{\sqrt{\delta(x)}}{x}
\Rightarrow <br /> \int \sqrt{\delta(x)}f(x)dx = -2\int xf(x)d(\sqrt{\delta(x)}) <br />
But this is still no good...

I also tried the other definition of Dirac Delta function,
\delta(x)\sim\lim_{\epsilon\rightarrow 0}e^{-x^2/\epsilon}
\Rightarrow \sqrt{\delta(x)} = \lim_{\epsilon\rightarrow 0}e^{-x^2/2\epsilon}<br /> =\lim_{\epsilon&#039;\rightarrow 0}e^{-x^2/\epsilon&#039;} = \delta(x) ??

still no good...

I also tried to calculate \psi(x,t)^2 to try to get rid of the square root,
but it seems doesn't help.

Is there anyone who has any ideas about \sqrt{\delta(x)} ?
Any help will be appreciated, thanks.

I have an idea, but I'm not sure if it's correct.
Consider
f(0) = \int \delta(x)f(x)dx = \int \sqrt{\delta(x)}\sqrt{\delta(x)} f(x)dx<br /> = \int \sqrt{\delta(x)}\sqrt{\delta(x)} dg(x)
where
\frac{dg}{dx} = f
,then integration by part gives,
f(0) = -2\int g(x)\frac{\delta&#039;(x)}{2\sqrt{\delta(x)}} dx <br /> = \int g(x) \frac{\sqrt{\delta(x)}}{x} dx

Now, let g(x) = xh(x), we have,
f(0) = \int h(x)\sqrt{\delta(x)} dx
,i.e.

<br /> \int h(x)\sqrt{\delta(x)} dx = \left[h(x) + xh&#039;(x)\right]_{x=0}<br />

I'm not sure if I were correct.
Any help would be appreciated!

-----------

Gosh, I found one mistake just now,
I calculated the integration by part wrong,
I omitted one factor of \sqrt{\delta(x)}
So the final formula is incorrect.
 
Last edited:
First, I tried to show that ##f_n## converges uniformly on ##[0,2\pi]##, which is true since ##f_n \rightarrow 0## for ##n \rightarrow \infty## and ##\sigma_n=\mathrm{sup}\left| \frac{\sin\left(\frac{n^2}{n+\frac 15}x\right)}{n^{x^2-3x+3}} \right| \leq \frac{1}{|n^{x^2-3x+3}|} \leq \frac{1}{n^{\frac 34}}\rightarrow 0##. I can't use neither Leibnitz's test nor Abel's test. For Dirichlet's test I would need to show, that ##\sin\left(\frac{n^2}{n+\frac 15}x \right)## has partialy bounded sums...