How to factorize the hydrogen atom Hamiltonian?

Click For Summary

Discussion Overview

The discussion revolves around the factorization of the hydrogen atom Hamiltonian, specifically exploring how to express the Hamiltonian in a different form that incorporates the position operator. Participants are examining the mathematical identities and operator relationships involved in this process.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant presents the hydrogen atom Hamiltonian and proposes a factorization involving constants β and γ, asking for clarification on how to derive this form.
  • Another participant suggests working backwards from the proposed factorization using the operator identity provided, detailing the steps to verify the factorization.
  • A later reply acknowledges an error in the calculation of a commutator and presents a revised equation relating the constants β and γ to the Hamiltonian.
  • One participant points out that the sum calculated in the previous post is multiplied by 1/(2m) in the Hamiltonian, indicating a potential oversight in the earlier analysis.
  • Another participant challenges the previous calculations, suggesting an alternative interpretation of the commutator and presenting a different expression for the sum involved in the factorization.

Areas of Agreement / Disagreement

Participants express differing views on the calculations and interpretations of the commutators and the resulting expressions. There is no consensus on the correct approach or final form of the factorization, as multiple competing views remain.

Contextual Notes

Some calculations depend on specific assumptions about the operators and their commutation relations, which may not be universally agreed upon. The discussion reflects ongoing uncertainties and refinements in the mathematical treatment of the Hamiltonian.

yamata1
Messages
61
Reaction score
1
Hello,
The hydrogen atom Hamiltonian is
$$H=\frac{p^2}{2m} -\frac{e^2}{r}\tag{1}$$
with e the elementary charge,m the mass of the electron,r the radius from the nucleus and p,the momentum. Apparently we can factorize H $$H=\gamma +\frac{1}{2m}\sum_{k=1}^{3}\left(\hat p_k+i\beta\frac{\hat x_k}{r}\right)\left(\hat p_k-i\beta\frac{\hat x_k}{r}\right)\tag{2}$$
for suitable constants β and γ that you can calculate. I assume the operator identity:
$$\hat{A}^2+\hat{B}^2=(\hat{A}-i\hat{B})(\hat{A}+i\hat{B})-i[\hat{A},\hat{B}]$$was used.
Can someone explain to me how we can start with formula (1) and make the position operator appear in (2)?
Here is the source https://ocw.mit.edu/courses/physics...all-2013/lecture-notes/MIT8_05F13_Chap_09.pdf pages 33-34
Thank you.
 
Physics news on Phys.org
The easiest way to verify the given factorization is to work backwards, starting with the right-hand side of the given identity and using the operator identity you gave:
$$ \begin{eqnarray*}
\sum_{k = 1}^3 \left( p_k + i \beta \frac{x_k}{r} \right) \left( p_k - i \beta \frac{x_k}{r} \right) &= \sum_{k = 1}^3 \left( p_k^2 + \beta^2 \frac{x_k^2}{r^2} + i \left[ p_k, \frac{\beta}{r} x_k \right] \right) \\
&= p^2 + \beta^2 + \sum_{k = 1}^3 \beta i \left[ p_k, \frac{x_k}{r} \right]
\end{eqnarray*}$$
The only things left to deal with are the commutators ## \left[ p_k, \frac{x_k}{r} \right] ##. I'm sure there's a clever way of calculating this basis-free, but it's easy enough to grind it out in the position basis:
$$ \begin{eqnarray*}
\left[ p_k, \frac{x_k}{r} \right] \psi &= -i \hbar \frac{\partial}{\partial x_k} \left( \frac{x_k}{r} \psi \right) + i \hbar \left( \frac{x_k}{r} \right) \frac{\partial}{\partial x_k} \psi \\
&= -i \hbar \left( \frac{1}{r} - \frac{x_k^2}{r^3} \right) \psi
\end{eqnarray*}$$
so we have ## \left[ p_k, \frac{x_k}{r} \right] = -i\hbar / r + i \hbar x_k^2 / r^3 ##. In particular, since ##\Sigma_{k = 1}^3 x_k^2 = r^2##, the sum I started with is just
$$ p^2 + (\beta^2 + 3 \beta \hbar) - \frac{\beta \hbar}{r} $$
Plugging this into the expression for the Hamiltonian allows one to easily solve for ##\beta## and ##\gamma##.

As for how you'd actually come up with this factorization if you didn't already know it, all I can say is that after fiddling around with enough problems like this, the expression you gave seems like a plausible guess. Once you guess the correct factored form, it's a simple matter to work backwards as above to figure out what the constants must be.
 
Thank you for this answer. I made a mistake in calculating the commutator so I didn't have the term $$-\frac{\beta\hbar}{r}$$. If I understand correctly,I am left with:
$$-\frac{e^2}{r}=\gamma +\beta^2 +3\beta\hbar -\frac{\beta\hbar}{r}$$
and since β and γ are constants: Do we have$$-\frac{e^2}{r}=-\frac{\beta\hbar}{r} $$ ?
 
You're forgetting that the sum I calculated is multiplied by ##1 / (2m)## in the Hamiltonian. Otherwise, yes.
 
  • Like
Likes   Reactions: yamata1
VKint said:
You're forgetting that the sum I calculated is multiplied by ##1 / (2m)## in the Hamiltonian. Otherwise, yes.
Thank you very much.
 
I think there is a mistake here
VKint said:
the sum I started with is just$$p^2 + (\beta^2 + 3 \beta \hbar) - \frac{\beta \hbar}{r}$$

$$\sum_{k = 1}^3 \beta i [ p_k, \frac{x_k}{r}]=\beta i(\frac {-3i\hbar}{r}+\frac {i\hbar r^2}{r^3})$$
So $$
\sum_{k = 1}^3 \left( p_k + i \beta \frac{x_k}{r} \right) \left( p_k - i \beta \frac{x_k}{r} \right)
=p^2+\beta^2+\frac {2\hbar \beta}{r}$$
but after seeing your question https://www.physicsforums.com/threads/commutator-of-p-and-x-r.933864/ the answer given makes it much simpler.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 7 ·
Replies
7
Views
954
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
331
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 32 ·
2
Replies
32
Views
4K