I have a question about deriving Eq. (6.2.13) in Polchinski's string theory book volume I. It is claimed that

Now consider the path integral with a product of tachyon vertex operators, $$A_{S_{2}}^{n}(k,\sigma)=\left\langle [e^{ik_{1}\cdot X(\sigma_{1})}]_{r}[e^{ik_{2}\cdot X(\sigma_{2})}]_{r}\cdots[e^{ik_{n}\cdot X(\sigma_{n})}]_{r}\right\rangle _{S_{2}}\tag{6.2.11}$$ This corresponds to $$J(\sigma)=\sum_{i=1}^{n}k_{i}\delta^{2}(\sigma-\sigma_{i})\tag{6.2.12}$$ The amplitude (6.2.6) then becomes $$\begin{multline}A_{S_{2}}^{n}(k,\sigma)=iC_{S_{2}}^{X}(2\pi)^{d}\delta^{d}\biggl(\sum_{i}k_{i}\biggr)\\\times\exp\biggl(-\sum_{i,j=1;i<j}^{n}k_{i}\cdot k_{j}G'(\sigma_{i},\sigma_{j})-\frac{1}{2}\sum_{i=1}^{n}k_{i}^{2}G_{r}'(\sigma_{i},\sigma_{i})\biggr)\qquad\end{multline}\tag{6.2.13}$$ where $C_{S_{2}}^{X}=X_{0}^{-d}(\det'\frac{-\nabla^{2}}{4\pi^{2}\alpha'})_{S_{2}}^{-d/2}$ and $G_{r}'(\sigma,\sigma')=G'(\sigma,\sigma')+\frac{\alpha'}{2}\ln d^{2}(\sigma,\sigma')$

Eq. (6.2.6) is

$$\begin{multline}Z[J]=i(2\pi)^{d}\delta^{d}(J_{0})\biggl(\det'\frac{-\nabla^{2}}{4\pi^{2}\alpha'}\biggr)^{-d/2}\\\times\exp\biggl(-\frac{1}{2}\int d^{2}\sigma d^{2}\sigma'J(\sigma)\cdot J(\sigma')G'(\sigma,\sigma')\biggr)\end{multline}\tag{6.2.6}$$

My question is: where do $X_0^{-d}$ and $G_r'$ come from in Eq. (6.2.13)? I could try to plug (6.2.12) into (6.2.6) to see all other term appears, but not $X_0^{-d}$ nor $G_r'$.


The renormalization $[...]_r$ of the exponentials is defined in equation $(3.6.5)$

$$ [\mathcal{F}]_r = \exp \left( \frac{1}{2} \int d^2\sigma d^2\sigma' \Delta(\sigma,\sigma')\frac{\delta}{\delta X^\mu(\sigma)}\frac{\delta}{\delta X_\mu(\sigma')}\right) \mathcal{F} $$


$$ \Delta(\sigma,\sigma') =\frac{\alpha'}{2} \ln d^2(\sigma,\sigma') $$

where $d(\sigma,\sigma')$ is the geodesic distance between points $\sigma$ and $\sigma'$. So,

$$ [e^{ik \cdot X}]_r = \exp \left(- \frac{1}{2} \int d^2\sigma d^2\sigma' \frac{\alpha'}{2}k^2 \ln d^2(\sigma,\sigma') \right) e^{ik \cdot X} $$

This is the origin of the last term in $G'_r$. The $X_0^{-d}$ term comes from the delta function $\delta^d(J_0)$, where

$$J_0^\mu = \int d^2\sigma J^\mu(\sigma) X_0 $$

($X_0$ is a constant $(6.2.5)$). Plugging in $J(\sigma)$ this becomes

\begin{align} J_0^\mu &= \int d^2\sigma \sum_{i=1}^n k_i^\mu\delta^2(\sigma-\sigma_i) X_0 \\&= X_0\sum_{i=1}^n k_i^\mu \end{align}

Using the formula

$$ \delta[g(x)]=\sum_{i}\frac{\delta(x-x_i)}{g'(x_i)} $$

where $g'(x)$ is the x-derivative and the sum is over the roots of the function $g(x)$, the delta function becomes $$ \delta^d\left(X_0 \sum_i k_i\right) = X_0^{-d} \delta^d \left(\sum_i k_i\right)$$


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.