1
$\begingroup$

I'm following the Book of Brian Hatfield (Quantum Field Theory of point particles and Strings), p.192 here: For real Grassmann numbers (and Functionals thereof):

If $\Phi[\psi]$ is a functional, and $\psi(x)$ is a Grassmann-valued function, we demand that $$\int \mathcal{D}\psi \delta[\psi - \xi] \Phi[\psi] = \Phi[\xi]$$ (this is equation 9.67) , and one option to do this is to let (equation 9.66) $$\delta[\psi - \xi] = \prod_x (\psi(x)-\xi(x)).$$

The complex case of the delta function is NOT treated in the book, and I want to deduce how the mentioned relations would turn out for that case. Here, $\psi$ now has two components ($\psi = \frac{1}{2} \psi_1 + i \psi_2$) - Which makes me wonder: How does the fundamental relation turn out? For complex $\psi$: \begin{align} \int \mathcal{D}\psi \delta[\psi - \xi] \Phi[\psi] = \Phi[\xi] \end{align} or \begin{align} \int \mathcal{D} \psi \int \mathcal{D} \psi^* \delta[\psi - \xi] \Phi[\psi] = \Phi[\xi]? \end{align}

The first version works, but only if I assume that $\delta[\psi - \xi] = \prod\limits_x (\psi(x) - \xi(x))$ and $\delta[\psi-\xi] = \delta[\psi-\xi]^*$, and those exclude each other. In either case, what would be a realization of the $\delta$ functional? Would it still be \begin{align} \delta[\psi - \xi] = \prod_x (\psi(x) - \xi(x))? \end{align}

$\endgroup$
0

2 Answers 2

1
$\begingroup$

Motivated by the one-dimensional complex delta function, whereby $\delta_\mathbb{C}(z):=\delta(z)\delta(\bar{z}),$ so that $$\int \mathrm{d}z\wedge\mathrm{d}\bar{z}\ \delta_\mathbb{C}(z-\zeta) f(z) = f(\zeta),$$ you can define $\delta_\mathbb{C}[\psi] := \delta[\psi]\delta\!\left[\bar{\psi}\right],$ satisfying $$\int\mathrm{D}\psi\;\mathrm{D}\bar{\psi}\ \delta_\mathbb{C}[\psi-\xi] \Phi[\psi] = \Phi[\xi]$$ and realised as $$\delta_\mathbb{C}[\psi-\xi] = \prod_{x} \Big(\psi(x)-\xi(x)\Big)\Big(\bar{\psi}(x)-\bar{\xi}(x)\Big).$$

$\endgroup$
1
$\begingroup$
  1. Be aware that integration over a complex Grassmann-odd variable $\psi$ has 2 different notations in the literature: $\int \!\mathrm{d}^2\psi$ and $\int\! \mathrm{d}\psi~\mathrm{d}\psi^{\ast} $. This is similar to the standard notations for a Grassmann-even complex integration.

  2. Similarly, a complex Grassmann-odd Dirac delta distribution can denote just the holomorphic part or also include the anti-holomorphic part, depending on conventions.

$\endgroup$
2
  • $\begingroup$ For point 2., what would the eplicit difference be? In both cases, I would naively write $\delta(\psi)$, but they can't both be true, no? $\endgroup$ Commented Oct 11, 2022 at 14:13
  • $\begingroup$ Well, different authors have different conventions. $\endgroup$
    – Qmechanic
    Commented Oct 11, 2022 at 14:16

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.