2

On P&S's qft page 301 and 302, the book discussed functional quantization of spinor field.

The book define a Grassmann field $\psi(x)$ in terms of any set of orthonormal basis functions: \begin{equation} \psi(x)=\sum_i \psi_i \phi_i(x) \tag{9.71} \end{equation} where $\phi_i(x)$ are ordinary four component spinors, $\psi_i$ are Grassmann numbers.

Then, the book defined the two-point function: \begin{equation} \left\langle 0\left|T \psi\left(x_1\right) \bar{\psi}\left(x_2\right)\right| 0\right\rangle=\frac{\int \mathcal{D} \bar{\psi} \mathcal{D} \psi \exp \left[i \int d^4 x \bar{\psi}(i \not \partial-m) \psi\right] \psi\left(x_1\right) \bar{\psi}\left(x_2\right)}{\int \mathcal{D} \bar{\psi} \mathcal{D} \psi \exp \left[i \int d^4 x \bar{\psi}(i \not \partial-m) \psi\right]} \tag{A} \end{equation}

I don't understanding following:

  1. the book said that they write $\mathcal{D}\overline{\psi}$ instead of $\mathcal{D}\psi^*$ for convenience, the two are unitarily equivalent. So according to (9.71), this two are vectors, not matrix, what's the meaning here unitarily equivalent?

  2. According to previous discussion, the denominator of (A) should be \begin{equation} \text{det}(-i\int d^4 x (i\not \partial -m)) \end{equation} why we just denote $\text{det}(i\not \partial -m)$?

Qmechanic
  • 201,751
Daren
  • 1,401

1 Answers1

4
  1. The functional measure is not only the (continuous) product at each space-time point but also of each component of the objects we are taking the measure of. This means that the measure of $\psi^\dagger$ and $\psi$ is the same. Moreover, $\overline{\psi}=\psi^\dagger \gamma^0$, and $\text{det}(\gamma^0)=1$, so $\mathcal{D}\overline{\psi}$ and $\mathcal{D}\psi^\ast$ are the same.
  2. You have: \begin{equation} \prod_i \int d\theta^\ast_i d\theta_i\,e^{-\theta^\ast_i B_{ij}\theta_j} = \det(B), \end{equation} as identity see Berezin integral
  • Thank you very much! For your first point, do you want to express that the measure of $\psi^{\dagger}$ and $\psi^{*}$ is the same? – Daren Nov 19 '22 at 09:18
  • @Daren Yes they are the same because $\psi^\dagger = (\psi^{\ast})^\text{T}$, and a product of components of a "vector" doesn't care about whether it is a line or a column. – Jeanbaptiste Roux Nov 19 '22 at 09:25
  • Another point is that for Berezin integral, if $\theta_i$ have spinor component, this expression also correct. Is this right? Thanks! – Daren Nov 19 '22 at 09:26
  • @Daren If by 'have spinor components' you mean in fact that $\theta_i$ are the components of a spinor, then yes this expression is also correct because you can decompose the product into two: one over the positions $x$ in space-time, and one over these components. – Jeanbaptiste Roux Nov 19 '22 at 09:34