I was looking at the hyperfine structure for the hydrogen atom. I checked pretty much every textbook I knew but none of them gave me the general expression for the energy correction due to the hyperfine perturbation Hamiltonian.

All of them only treat the case when $\ell=0$. I was wondering if there is a general expression which doesn't have such a restriction?


The Hamiltonian for the spin-spin interaction is:

$$\Delta H_{SS} = \frac{\gamma_p e^2}{m m_p c^2 r^3} \Big( \frac{1}{r^3} \big(3(\vec{s}_p \cdot \hat{r})(\vec{s}_e \cdot \hat{r})-(\vec{s}_p \cdot \vec{s}_p) \big)+\frac{8 \pi}{3} (\vec{s}_p \cdot \vec{s}_p) \delta^{(3)}( \vec{r} ) \Big) $$

For the cases where $l \neq 0$ the term with the delta function cancels, and the wavefunction is proportional to $r^l$ for small $r$ values. Thus, when $l>0$ we get that $\psi(0)=0$, and then the correction to the energy will be:

$$\Delta E_{hf} = \frac{\gamma_p e^2}{m m_p c^2} \left\langle \frac{1}{r^3} \big( ( \vec{l} \cdot \vec{s}_p )+3(\vec{s}_p \cdot \hat{r})(\vec{s}_e \cdot \hat{r})-(\vec{s}_p \cdot \vec{s}_p) \big) \right\rangle $$

This expectation value was calculated by Bethe and Salpeter, and the result is:

$$\Delta E_{hf} = \frac{m}{m_p} \alpha^4 mc^2 \frac{\gamma_p}{2 n^3} \left( \frac{ f(f+1)-j(j+1)-\frac{3}{4}}{j(j+1)(l+\frac{1}{2})} \right),$$

This result coincide with the $l=0$ case, since then $j=\frac{1}{2}$ and the proton has spin 1/2 so $f=j \pm \frac{1}{2}$ and the expression above can then be simplified to the result you already have for the $l=0$ case...

(Next time try older QM books like this one http://adsabs.harvard.edu/abs/1957qmot.book.....B )

  • $\begingroup$ Thanks a lot! Yeah I realised that Bethe&Salpeter is probably the only book that has the full hyperfine formula(I read that from Griffiths' Intro to Elementary Particles)... How did you know that(since the book is quite old)? $\endgroup$ – Evariste Nov 26 '12 at 3:59
  • $\begingroup$ no worries :) That was the book used in one of my intermediate QM courses (about 10 years ago)... $\endgroup$ – bla Nov 26 '12 at 4:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.