next up previous
Next: 5. The Coleman-Weinberg mechanism Up: Membrane Vacuum as a Previous: 3. The action

4. Dynamics of the membrane vacuum and the formation of bags

The action (8) leads to the pair of coupled field equations

  
$\displaystyle {\langle}
\Big \vert
\left(
\frac{\delta}{\delta \sigma ^{\mu \nu \rho} (s)}
-
i
g
A _{\mu \nu \rho} \right)
\Big \vert ^{2}
\Psi[S]
{\rangle}$ = 0 (19)
$\displaystyle \partial _{\lambda}
F ^{\lambda \mu \nu \rho} (x)$ = $\displaystyle g
J ^{\mu \nu \rho} (x)$ (20)

which describe the interaction between the membrane field and the $A _{\mu \nu \rho}$ gauge potential. Now, we wish to show that the superconducting membrane condensate contains regions of spacetime, or bags of non superconducting vacuum.
Recall that in a type-II superconductor the magnetic field is confined by the superconducting vacuum pressure within a string-like flux tube. Similarly, the membrane condensate confines the gauge field strength within a membrane-like boundary layer surrounding a region of ordinary vacuum. Indeed, in analogy with the superconducting solution of scalar QED, we assume the following asymptotic boundary condition for $\Psi [S]$

\begin{displaymath}\Psi [S]
\equiv
\frac{\phi}{g ^{2}}
e ^{i \displaystyle{\...
...} d x ^{\mu} \wedge d x ^{\nu}
\theta _{\mu \nu} (x)
\right)
\end{displaymath} (21)

where $\phi$ is a constant. This is the form of the membrane field when the three volume enclosed by S is much larger than the three volume of the vortex spacetime image. Then, from equation (19) we obtain the corresponding asymptotic form of $A _{\mu \nu \rho}$:

 \begin{displaymath}\left(
\frac{\delta}{\delta \sigma ^{\mu \nu \rho} (s)}
-
...
... \rho} =
\frac{1}{g}
\partial _{[ \mu}
\theta _{\mu \nu ]}.
\end{displaymath} (22)

Therefore, the flux of $F _{\lambda \mu \nu \rho}$ across a large four dimensional region $\Omega$ enclosing B is given by
 
q n $\textstyle \equiv$ $\displaystyle \frac{1}{4!}
\int _{\Omega}
F _{\lambda \mu \nu \rho}
\,
d x ^{\lambda} \wedge d x ^{\mu} \wedge d x ^{\nu} \wedge d x ^{\rho}$  
  = $\displaystyle \frac{1}{3!}
\oint _{\Gamma} d x ^{\mu} \wedge d x ^{\nu} \wedge d x ^{\rho}
A _{\mu \nu \rho}$  
  = $\displaystyle \frac{1}{3! g}
\oint _{\Gamma} d x ^{\mu} \wedge d x ^{\nu} \wedg...
...tial _{[ \mu}
\theta _{\nu \rho ]}
=
\frac{2 \pi n}{g}
\qquad
n = 1 , 2 , \dots$ (23)

Thus, the physical consequence of the monodromy of $\Psi [S]$ is that the flux of $F _{\lambda \mu \nu \rho}$ through a region enclosing B is quantized in units of $2\pi/g$.
In the superconducting phase, equation (20) becomes

 \begin{displaymath}\partial _{\lambda}
F ^{\lambda \mu \nu \rho} (x)
=
-
\ph...
...a ^{\nu \rho ]} (x)
\right)
\equiv
-
j ^{\mu \nu \rho} (x)
\end{displaymath} (24)

in which we have introduced the supercurrent density $j ^{\mu \nu \rho} (x)$. Equation (24) holds only where $\theta ^{\nu \rho} (x)$ is a regular function. In the domain of singularity, where the partial derivatives of $\theta ^{\nu \rho} (x)$ do not commute, the covariant curl of $\partial ^{[ \, \mu} \theta ^{\nu \rho]} (x)$ should be interpreted in the sense of distribution theory. Indeed, if we apply the covariant curl operator to both sides of equation (24), we obtain

 \begin{displaymath}\partial ^{[ \, \lambda}
j ^{\mu \nu \rho ]}
=
-
\phi ^{2...
...[ \, \lambda}
\partial ^{\mu}
\theta ^{\nu \rho ]}
\right)
\end{displaymath} (25)

The last term in (25) may not be disregarded without violating (7). Therefore in order to match (23) with (7), we define
 
$\displaystyle \partial ^{[ \, \lambda}
\partial ^{\mu}
\theta ^{\nu \rho]} (x)$ $\textstyle \equiv$ $\displaystyle \frac{q _{n} g}{\Omega _{B}}
\epsilon ^{\lambda \mu \nu \rho}
\int _{B} d ^{4} \xi \,
\delta ^{4)}
\left[ x - z( \xi ) \right]$  
  $\textstyle \equiv$ $\displaystyle \frac{q _{n} g}{\Omega _{B}}
J _{B} ^{\lambda \mu \nu \rho} (x)$ (26)

where, $\Omega _{B}$ and $J _{B} ^{\lambda \mu \nu \rho} (x)$ are, respectively, the bag four-volume and the bag current. Thus, the supercurrent can be determined from the equation

\begin{displaymath}\partial _{\lambda}
\partial ^{[ \, \lambda}
j ^{\mu \nu \r...
...
\partial _{\lambda}
J _{B} ^{\lambda \mu \nu \rho}
\right)
\end{displaymath} (27)

by means of the Green function method:

 \begin{displaymath}j ^{\mu \nu \rho} (x)
=
-
\frac{\phi ^{2} q _{n}}{\Omega _...
...B} d ^{4} z \,
\partial _{\sigma} \,
G ( x - z ; \phi ^{2} )
\end{displaymath} (28)

where G ( x - z ) is the scalar Green function

\begin{displaymath}\left[
\partial ^{2}
+
\phi ^{2}
\right]
G ( x - z ; \phi ^{2})
=
\delta ^{4)} ( x - z )
.
\end{displaymath} (29)

Then, from equation (20), (25) and (28), we find the form of the confined gauge field

 \begin{displaymath}F ^{\lambda \mu \nu \rho} (x)
=
-
\frac{\phi ^{2} q _{n}}{...
...\rho \sigma}
\int _{B} d ^{4} z \,
G ( x - z ; \phi ^{2})
.
\end{displaymath} (30)

The analogy between the membrane vacuum and a type-II superconductor now seems manifest: in the ordinary vacuum $A _{\mu \nu \rho}$ does not propagate any degree of freedom. Rather, it corresponds to a uniform energy background. However, in the superconducting phase $A _{\mu \nu \rho}$ becomes a dynamical field describing a massive, spin-0 particle [10]. The source for the massive field is the bag current (26). In a boson particle condensate, the magnetic field is confined to a thin flux tube surrounding the vortex line; in the membrane condensate, the $F _{\lambda \mu \nu \rho}$-field is confined within the membrane which encloses the ordinary vacuum bag. The gauge field provides the ``skin'' of the bag. To complete the analogy, in the next section we show that the thickness of the membrane is given by the inverse of the dynamically generated mass of $F _{\lambda \mu \nu \rho}$.


next up previous
Next: 5. The Coleman-Weinberg mechanism Up: Membrane Vacuum as a Previous: 3. The action

Stefano Ansoldi
Department of Theoretical Physics
University of Trieste
TRIESTE - ITALY