0
$\begingroup$

In 'QCD and collider Physics' book by Ellis, Stirling and Webber, chapter 4.2 'The parton model from field theory' there is this equation for the amplitude of the process $e^- (k) P \rightarrow e^- (k^\prime) X$ (where $P$ stands for the proton and $X$ for an undetermined outgoing jet)

$$ A = e\bar{u}(k^\prime)\gamma^\alpha u(k)\frac{1}{q^2}\langle X| j_\alpha(0) |P\rangle \tag{1} $$

With $j_\alpha$ the electromagnetic current, but the book doesn't specify more. I've tried to get this formula via the scattering matrix $S$ in this way:

$$ S = \cal{T}\left( \exp\left[{i\int_{\mathbb{R}^4}d^4 x\ \cal{L}_{int}}\right] \right) \approx 1 + i\int_{\mathbb{R}^4}d^4 x\ \cal{L}_{int} - \frac{1}{2} \int_{\mathbb{R}^4}d^4 x d^4 y\ {\cal T}\left[{\cal L}_{int}(x){\cal L}_{int}(y)\right] \tag{2} $$ And

$$ {\cal L}_{int} = \sum_q\bar{q}e_q\gamma_\mu A^\mu q + \bar{\psi}_e e \gamma_\mu A^\mu \psi_e $$

Where $q$ is the quark, $\psi_e$ the electron and $A^\mu$ the photon fields; and $e_q$ the electric charge of quark $q$. Now the matrix element between intial and final state, considering $S$ till 1st order, is

$$ A = \langle e^- X| S | e^- P\rangle = i\sum_qe_q \langle e^- X| \int d^4 x \bar{q}\gamma_\mu A^\mu q | e^- P\rangle - ie\langle e^- X| \int d^4 x \bar{\psi}_e\gamma_\mu A^\mu \psi_e | e^- P\rangle \tag{3} $$

Since there is no photon in the final or initial states, $\langle 0_\gamma | A^\mu |0_\gamma \rangle = 0$. Therefore, you have to go to 2nd order in the $S$ expansion (last term in Eq. (2)).

This term is more complicated. First, you have to realise that due to the anticommutation relations of $q, \psi_e$ fields, you can write

$$ {\cal T}\left[ \bar{q}_x \gamma_\mu A^\mu_x q_x \bar{\psi}_{e, y}\gamma_\nu A^\nu_y \psi_{e, y}\right] = {\cal T}\left[ A^\mu_x A^\nu_y \right]\bar{q}_x \gamma_\mu q_x \bar{\psi}_{e, y}\gamma_\nu \psi_{e, y} $$

Where $q(x) = q_x, \psi_{e, y} = \psi_e(y)$.

Using this and after some calculations you end up with

$$ A = -ie\bar{u}_{k^\prime s^\prime}\gamma^\mu u_{ks}\frac{1}{(k - k^\prime)^2}\langle X| \sum_q e_q \int d^4 x\ e^{-i(k - k^\prime)x}\bar{q}_x \gamma_\mu q_x |P\rangle \tag{4}$$

This Eq. (4) is almost Eq. (1), but I don't see how to go on in order to get rid of the $-i$ factor and get an electromagnetic current evaluated at zero (I guess) momentum, $j_\mu (0)$. Any insights?

$\endgroup$
5
  • 1
    $\begingroup$ A hint: Recall the fundamental identity $j^{\mu}(x) = e^{-i \hat{P}x} j^{\mu}(0) e^{i \hat{P}x}$, where $\hat{P}$ is the four-momentum operator and keep in mind that matrix element is defined to be stripped of a momentum conservation delta function times $(2\pi)^4$ and a factor of $i$, in equations $\langle i T \rangle = (2\pi)^4 \delta(\sum p) i \mathcal{M}$ (what you call $A$ in (3) is not quite the same $A$ as in equation (1), which is what I mean by $\mathcal{M}$), where $S = 1 + iT$. Using this you can obtain the desired result. $\endgroup$
    – jkb1603
    Commented Oct 7, 2020 at 15:54
  • $\begingroup$ @jkb1603 so you are telling me that my $A$ is actually $iT$ which (after taking $i(2\pi)^4 \delta(...)$) is the $A$ in Eq. (1), right? And that from my $A = iT$ in Eq. (4) I should get a felta function and a $j^\mu (0)$ once I use the spacetime translation? $\endgroup$
    – Vicky
    Commented Oct 7, 2020 at 16:12
  • 1
    $\begingroup$ Yes. Your $A$ in (3) is the S-matrix element $\langle iT \rangle$ and the $A$ in (1) is the scattering amplitude $\mathcal{M}$. If you still have problems I can write down the answer. But you are really close...its like two lines. You need to also know that $j^{\mu}(x) = \sum_q e_q \bar{q}(x) \gamma^{\mu} q(x)$ and that the external states are momentum eigenstates. $\endgroup$
    – jkb1603
    Commented Oct 7, 2020 at 16:37
  • $\begingroup$ @jkb1603 Already got it, thank you very much. But there's still this minus sign that the Eq. (1) doesn't have. Maybe it can be explained considered that in (1), $e$ is the electron charge with its sign while for me $e$ is proton's. What do you think? $\endgroup$
    – Vicky
    Commented Oct 7, 2020 at 17:05
  • $\begingroup$ Probably. There are also different sign conventions for the electromagnetic coupling, which is annoying. $\endgroup$
    – jkb1603
    Commented Oct 7, 2020 at 18:26

0