This is a follow up question of this one.

In the Vol 1, Weinberg derives how a unitary operator $U(\Lambda)$ acts on one-particle states, which is given by equation (2.5.2): \begin{equation} U(\Lambda)|p,\sigma\rangle = \sum_{\sigma'} C_{\sigma' \sigma}(\Lambda,p)|\Lambda p, \sigma'\rangle \end{equation} Am I correct in thinking that Peskin and Schroeder use this result to derive the Lorentz transformation on a annihilation operator of the quantized Dirac field (given by equation (3.106)): \begin{equation} U(\Lambda) a^s_{\mathbf{p}}U^{-1}(\Lambda) = \sqrt{\frac{E_{\Lambda\mathbf{p}}}{E_{\mathbf{p}}}} a_{\Lambda \mathbf{p}}^s \end{equation} However, Peskin and Schroeder are only considering Lorentz transformations that do not mix states with different spin indices! In other words, the first equation I wrote becomes: \begin{equation} U(\Lambda)|p,\sigma\rangle = C|\Lambda p, \sigma\rangle \end{equation} where $C$ is normalization constant which can be determined to be $C=1$. If this is not correct, can you please let me know how they (Peskin and Schroeder) have derived equation (3.109)?

I hope this question is not too confusing.

Edit: for you to see my reasoning. We can now write (for $C=1$): \begin{equation} U(\Lambda)|\mathbf{p},s\rangle = \sqrt{2 E_{\mathbf{p}}} U(\Lambda)a^{s\dagger}_\mathbf{p}U^{-1}(\Lambda)|0\rangle=|\Lambda \mathbf{p},s\rangle = \sqrt{2 E_{\Lambda \mathbf{p}}} a^{s\dagger}_{\Lambda \mathbf{p}}|0\rangle \end{equation} which implies: \begin{equation} U(\Lambda)a^{s\dagger}_\mathbf{p}U^{-1}(\Lambda) = \sqrt{\frac{E_{\Lambda\mathbf{p}}}{E_{\mathbf{p}}}} a^{s\dagger}_{\Lambda \mathbf{p}} \end{equation} Subsequently, using the unitarity of $U(\Lambda)$, we can take the adjoint of the above equation: \begin{equation} U(\Lambda)a^{s}_\mathbf{p}U^{-1}(\Lambda) = \sqrt{\frac{E_{\Lambda\mathbf{p}}}{E_{\mathbf{p}}}} a^{s}_{\Lambda \mathbf{p}} \end{equation} which is equation (3.109).


1 Answer 1


I think it is correct. However, just as a pedantic remark, you should prove $U(\Lambda)a^{s\dagger}_\mathbf{p}U^{-1}(\Lambda)$ and $\sqrt{\frac{E_{\Lambda\mathbf{p}}}{E_{\mathbf{p}}}} a^{s\dagger}_{\Lambda \mathbf{p}}$ act on all the vectors in the same way, not just on the vacuum $|0\rangle$, you need a slight modification of your proof: $$\sqrt{2 E_{\mathbf{p}}}U(\Lambda)a^{s\dagger}_\mathbf{p}|\mathbf{p_1},s_1;\mathbf{p_2},s_2;\cdots\rangle=U(\Lambda)|\mathbf{p},s;\mathbf{p_1},s_1;\mathbf{p_2},s_2\cdots\rangle\\ =|\Lambda\mathbf{p},s;\Lambda(\mathbf{p_1},s_1);\Lambda(\mathbf{p_2},s_2)\cdots\rangle $$ The second equality is justified since we have assumed $s$ is parallel to the rotation or boost direction, and $s_1,s_2,\cdots$ might not, that is why I labeled the them differently as $\Lambda(\mathbf{p_1},s_1)$. To continue, we have $$=\sqrt{2 E_{\Lambda \mathbf{p}}} a^{s\dagger}_{\Lambda \mathbf{p}}|\Lambda(\mathbf{p_1},s_1);\Lambda(\mathbf{p_2},s_2)\cdots\rangle\\ =\sqrt{2 E_{\Lambda \mathbf{p}}} a^{s\dagger}_{\Lambda \mathbf{p}}U(\Lambda)|\mathbf{p_1},s_1;\mathbf{p_2},s_2;\cdots\rangle$$ Since the collection $\{|\mathbf{p_1},s_1;\mathbf{p_2},s_2;\cdots\rangle\}$ is a basis of the vector space, $$\sqrt{2 E_{\mathbf{p}}}U(\Lambda)a^{s\dagger}_\mathbf{p}=\sqrt{2 E_{\Lambda \mathbf{p}}} a^{s\dagger}_{\Lambda \mathbf{p}}U(\Lambda)$$ holds as an operator identity.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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