64
GAUGE FIELDS AND STRINGS
subgroup of some Non-Abelian gauge theory we are necessarily dealing
with the compact version.
After this explanation let us work out the theory of this version
starting from ^ = 3. Following the strategy of the preceeding section
we consider a partition function:
n
Z = X i n
ex p l- ¿ X
(4.59)
{«*.«^1 ^ x,a
f ^^0 x,ciP
J
- n
which takes account of the periodicity of the action. For a given set of
{tix^acfi} we introduce numbers (where z belongs to the centres of the
cubes of our lattice) defined as the flux of n through a given cube :
(4.60)
(the notation in the l.h.s. of (4.60) means the sums of
corresponding to oriented plaquettes forming a cube with centre at z- Its continuum analogue will be an integral of the field
over a closed surface).
Decomposing
as:
+ ^X ,(X T +
' ^X + P.a ^X,p
(4.61)
We obtain from (4.60):
K z'< t> z' = C l,
(4.62)
(Ajj. is the lattice Laplace operator). Substitution of (4.61) and (4.62)
into (4.59) gives:
^ = -^Gauss Z exp(^X
{fl.l V
/
+ (X)
Zoau,.= i n d /4 a .,e x p [ - ' X f L i )
J x,a
\ ^^0 x,a/i
/
(4.63)
We thus obtain the same Coulomb system as (4.30) but in three
dimensions. As we shall see that makes a lot of physical difference.
Before working this out let us explain the meaning of the “charges”
which appeared in (4.63). They can be thought of as instanton solutions
GAUGE FIELDS AND STRINGS
subgroup of some Non-Abelian gauge theory we are necessarily dealing
with the compact version.
After this explanation let us work out the theory of this version
starting from ^ = 3. Following the strategy of the preceeding section
we consider a partition function:
n
Z = X i n
ex p l- ¿ X
(4.59)
{«*.«^1 ^ x,a
f ^^0 x,ciP
J
- n
which takes account of the periodicity of the action. For a given set of
{tix^acfi} we introduce numbers (where z belongs to the centres of the
cubes of our lattice) defined as the flux of n through a given cube :
(4.60)
(the notation in the l.h.s. of (4.60) means the sums of
corresponding to oriented plaquettes forming a cube with centre at z- Its continuum analogue will be an integral of the field
over a closed surface).
Decomposing
as:
+ ^X ,(X T +
' ^X + P.a ^X,p
(4.61)
We obtain from (4.60):
K z'< t> z' = C l,
(4.62)
(Ajj. is the lattice Laplace operator). Substitution of (4.61) and (4.62)
into (4.59) gives:
^ = -^Gauss Z exp(^X
{fl.l V
/
+ (X)
Zoau,.= i n d /4 a .,e x p [ - ' X f L i )
J x,a
\ ^^0 x,a/i
/
(4.63)
We thus obtain the same Coulomb system as (4.30) but in three
dimensions. As we shall see that makes a lot of physical difference.
Before working this out let us explain the meaning of the “charges”
which appeared in (4.63). They can be thought of as instanton solutions
