Physics 6047

Problem Set 1, due 2/3/14

Lam Hui
Rules concerning problem sets: Each problem set is due at the lecture. A late problem set
will be given 50% of the credit if turned in within a week of the due date. After that, it will
receive no credit. If you have special diculties, you must contact me (the instructor) in
advance. You are welcome to discuss with your classmates, but the solutions should be your
own. Note that often, some fraction of the problems will function as notes, hence the length.
1. Show that
_
d
4
k
(2)
4
2(k
2
+ m
2
)(k
0
> 0) =
_
d
3
k
(2)
3
2
k
. (1)
Note that on the left hand side, k
0
is to be integrated over. The step function (k
0
> 0)
forces it to be positive i.e. = 1 if k
0
> 0, and vanishes otherwise. The delta function
(k
2
+ m
2
) forces it to be so-called on-shell or on-mass-shell i.e. it enforces k
2
=
(k
0
)
2
+ |

k|
2
= m
2
. On the right hand side, the integral over k
0
has disappeared, and

k
is precisely the on-shell (and positive) value of k
0
. Since the left hand side is manifestly
Lorentz invariant, so is the right hand side.
2. Suppose we have a Lagrangian (density) of the form L = L(,

) i.e. it depends on
some scalar eld (x) and its space-time (rst) derivatives. (We loosely use x without an
arrow on top to denote both space and time components i.e. x

.) Show that the classical

equation of motion for the one that follows from the principle of extremizing the action
i.e. S = 0 is:

_
L
(

)
_
=
L

(2)
3. In the second/third lecture, we showed:

(3)
By multiplying both sides by

(and summing over ), show that

(4)
Note what we have done: starting from the Lorentz matrix

, we are saying by raising

and lowering indices appropriately, we obtain its inverse i.e.

is equivalent to (
1
)

.
Check that this actually does work in the example of a boost in the x-direction i.e. start
from the Lorentz boost matrix, raise and lower indices as prescribed above, and show that
you do get the inverse of the original boost matrix. Check that it also works in the example
of a rotation, say around the z-axis.
4. Starting from [a

k
, a

] = (2)
3
2
k
(

), show that
[(t, x),

(t, x

)] = i(x x

) , (5)
follows from the expression for (the free) in terms of the annihilation and creation opera-
tors.
5. Show that the Hamiltonian
H =
_
d
3
x
_
1
2
(
t
)
2
+
1
2
(

)
2
+
1
2
m
2

2
_
(6)
gives
H =
_
d
3
k
(2)
3
2
k

k
a

k
a

k
+ V
_
d
3
k

k
2
(7)
once one expresses in terms of the creation and annihilation operators. Here V is the
volume of space we are integrating over (which is taken to be innite); it originates from
the delta function in momentum in a nite box, the delta function (at 0) would have a
magnitude V . Its not surprising that the vacuum energy density, if it were non-zero, would
integrate over space to give a total energy proportional to V .
6. Srednicki problem 3.5. This is a problem of a complex scalar eld, as opposed to the
real scalar eld we have been examining in class. In other words, for a real scalar eld, the
Fourier decomposition:
(x) =
_
d
3
k
(2)
3
2
k
_
a

k
e
ikx
+ a

k
e
ikx
_
(8)
makes sense, because

= . (Note k
0
in expressions like above is interpreted as on-shell
i.e. k
0
=
k
.) But for a complex scalar eld , we need to keep track of two dierent kinds
of annihilation (or creation) operators a

k
and b

k
:
(x) =
_
d
3
k
(2)
3
_
a

k
e
ikx
+ b

k
e
ikx
_
(9)
so that

= . Add to this problem a part (f): show that the equation of motion for (and
its conjugate) implies

= 0 (10)
where j

is dened as
j

i
_

. (11)
Such an equation tells us j

is a conserved current, meaning that j

0
is the charge density
and

j is the current density. In other words, imagine integrating the above equation over
some spatial volume V at some time t:

t
__
V
d
3
xj
0
_
=
_
V
d
3
x

j =
_
V
d

j (12)
where we have used the Gauss law in the second equality and it gives the integrated current
ux out of the boundary of the volume V . This is exactly what charge conservation should
2
look like: the net charge in the volume Q =
_
V
d
3
xj
0
changes with time at a rate that
is exactly determined by the ux of charges at the boundary. Thus, our complex scalar
theory secretly has a conserved charge, which wasnt present for the real scalar theory. (The
analog of j

is simply zero for a real scalar; the secret behind the complex scalar theory is
the presence of a symmetry: rotating by a phase, a subject we will study at length later.)
By taking the volume V to be innite, show that
Q =
_
d
3
k
(2)
3
2
k
_
a

k
a

k
b

k
b

k
_
. (13)
This is very satisfying. We have discovered that the complex scalar theory contains 2 kinds
of particles of the same mass m, one created by a

k
, the other created by b

k
. And they carry
opposite charge! In other words, they are anti-particle of each other. (The real scalar eld
case has no such structure; sometimes, people would say in that case, the anti-particle is
the same as the particle.) Thus, as promised, a relativistic quantum eld theory which has
a conserved charge would automatically contain anti-particles.
Finally, a question for you: why is it that for the real scalar eld theory, the Lagrangian
has this factor of 1/2, which is absent in the complex case? This is really pure convention,
but this choice of normalization helps enforce certain conventions we would usually like to
impose. Do you know what they are?
3