Boxerdog Green's Function
Boxerdog Green's Function
Boxerdog Green's Function
Green function
Equations:
1
Ψrr + Ψr − Ψtt = 0 ;
r
γr = r(Ψ2r + Ψ2t ) ,
γt = 2rΨr Ψt . (1.1)
Given arbitrary values for Ψ(r, 0) and Ψt (r, 0) , the solution obeys
Z ∞
Ψ(x, t) = dr(G(x, r, t)Ψt (r, 0) + Gt (x, r, t)Ψ(r, 0)) , t>0 (1.2)
0
where G is
I
r 1
G(x, r, t) = √
dϕ p , (1.3)
2π + 2xr cos ϕ + t2 − x2 − r2
√
where the symbol + indicates that the integral should only be taken over the values of
ϕ for which the entry of the square root is positive. If t < |x − r| there are no real ϕ
values obeying this, so G vanishes there.
Derivation: take action functional S in 3+1 dimensions:
Z ³ ´
~ 2 + 1 Ψ2 + J(~x, t)Ψ ,
S = d3~xdt − 12 (∂Ψ) (1.4)
2 t
General solution:
Z Z ∞
3 0
Ψ(~x, t) = d ~x dτ G(~x − ~x0 , τ )J(~x0 , t − τ ) , (1.6)
0
1
G(~x, τ ) = δ(r − τ ) θ(τ ) , r2 = ~x 2 . (1.7)
4πr
G is the so-called Green function of the system. Now take the cylindrically symmetric
case,
where
Z ∞ Z
S ¡ ¢
= dr dt − 12 rΨ2r + 12 r ψt2 + r J Ψ , (1.9)
2π 0
in which case the field equation becomes ours with source term inserted:
1
Ψtt − Ψrr − Ψr = J(r, t) . (1.10)
r
1
Using (1.6) and (1.7), we find
Z ∞ I Z Z ∞
r
Ψ(r1 , t) = dr dϕ dz dτ δ(% − τ ) J(r, t − τ ) , (1.11)
0 0 4πτ
q
where % ≡ z 2 + r2 + r12 − 2r r1 cos ϕ . (1.12)
To see that (1.16) indeed follows from (1.15) and (1.10), evaluate G for t ¿ r, r1 :
is an elementary exercise. It follows from the construction, and the direct argument is as
follows: the integrand,
1
g(x, t, ϕ) = p
2xr cos ϕ + t2 − x2 − r2
obeys the partial differential equation
µ 2 ¶
∂ 1 ∂ ∂2 1 ∂2
+ − + g(x, t, ϕ) = 0 , (1.20)
∂x2 x ∂x ∂t2 x2 ∂ϕ2
so that integration of Eq. (1.19) over ϕ only gives boundary terms. To see the contribution
of these boundary terms, reexpress the integral in terms of a closed contour integral in
the complex plane. One finds that G can also be written as
Z
r 1
G(x, r, t) = √
dϕ p , (1.21)
2π + 2x r cosh ϕ + x2 + r2 − t2
2
where the integral runs from −∞ to ∞ or over the domain where the square root
exists. The field equation then leads to an integral over ϕ of a function that is a pure
derivative of a function of ϕ with appropriate boundary conditions. Either in Eq. (1.3)
or in Eq. (1.21), we see that the boundaries vanish, unless r = x and t = 0 . The delta
functions in (1.19) are deduced by inspecting the points where G is singular. Expression
(1.3) ensures that G = 0 as soon as |x − r| > t .
This completes the proof that these solutions are completely causal. It is an elementary
exercise in mathematical physics.