radial quantization

From Physics wiki

Jump to: navigation, search

We may suppose that our 2-dimensional quantum field theory lives on a spacetime that has the topology of a cylinder, with x^1\, having period 2\pi\, so that x^1 \sim x^1 + 2\pi\,. Strictly speaking, the radius of our cylinder is determined by the metric, and we may at a later time take the limit as R\to\infty\, if we wish. Otherwise we may pretend that we are doing string theory. Let the metric be

ds^2 = (dx^0)^2 + (dx^1)^2\,,

(assuming we have done a Wick rotation), and introduce the complex coordinates

w = x^1 + i x^0\, (left-moving[1]),

\bar w = x^1 - i x^0\, (right-moving),

which are the analytic continuations of x^1 - x^0_M\, (left-moving) and x^1+x^0_M\, (right-moving) respectively. Then

x^0 = \tfrac{1}{2}\left(w + \bar w\right),
x^1 = -\tfrac{i}{2}\left(w - \bar w\right),

so that ds^2 = dw d\bar w\,, or

g_{\mu\nu} = \begin{pmatrix}0 & \tfrac{1}{2} \\ \tfrac{1}{2} & 0\end{pmatrix}\,,

g^{\mu\nu} = \begin{pmatrix}0 & 2 \\ 2 & 0\end{pmatrix}\,.

Furthermore, the volume form becomes

dx^0 \wedge dx^1 = \begin{vmatrix} \frac{1}{2} & \frac{1}{2} \\ -\frac{i}{2} & +\frac{i}{2} \end{vmatrix} dw d\bar w = \tfrac{i}{2} dw \wedge d\bar w\,.

Any transformation of the form

w = f(w)\,
\bar w = g(\bar w)\,

preserves the metric up to a conformal factor, which we may drop, provided our field theory has Weyl invariance (something we should confirm quantum mechanically). Let

z = e^{i w}\,,
\bar z = e^{-i \bar w}\,,

which maps the cylinder onto the plane so that "time" runs radially outward from the origin.

References

  1. This is to conform to literature. Assume x^1\, increases to the left
Personal tools