Chapter 1

First look at strings

Following “String Theory” by J. Polchinski, Vol.I.

Notes written by students (work still in progress).
For more information contact George Siopsis
gsiopsis@utk.edu

1.1 Units
First, we must explain the unit convention we are going to use. Take the following two results from Quantum
Mechanics and Special Relativity:

E = ~ω (1.1)

E = mc2 (1.2)
These two equations link energy to frequency and mass through some constant of proportionality. The question
is, are these constants fundamental in nature or created by man? The answer is that they are artificial creations,
existing purely because of the units we have chosen to work in. We could easily choose units such that ~ and c
= 1. By doing this, the number of fundamental units in the universe is reduced to 1, all others being equivalent:

[energy] = [1/time] = [mass] = [1/length]

1.2 Why Strings?

Our motivation behind the developement of String Theory is our desire to find a unified theory of everything.
One of the major obstacles that previous theories have been unable to overcome is the formation of a quantum
theory of gravity, and it is in this respect that String theory has had notable success (in fact, at present, String
theory is the only theory which includes gravitational interactions). This leads us to believe that, while String
theory may not be the final answer, it is certainly a step in the right direction.
Let us first discuss the problems one runs into when trying to create a quantum theory of gravity using Quantum
Field Theory as our guide. Take the Hydrogen atom, whose energy levels are given by:

E1
En = − (1.3)
n2

~2
E1 = (1.4)
2me a0

1
2 CHAPTER 1. FIRST LOOK AT STRINGS

~2
a0 = (1.5)
me e2
where E1 is the ground state energy and n labels the energy levels.
Suppose we had only the most basic knowledge of physics: what would we guess the energy of the hydrogen
atom to be? The parameters of the system are the mass of the electron me , the mass of the proton mp , ~ and
the electron charge e. We negelect mp as we are interested in the energy levels of the electron. We would guess
the energy to be

Eg = me c2 (1.6)
Equations (4) and (5) are clearly not the same, but if we take the ratio we obtain
2
e4

E1 1
= 2 2 = = α2 (1.7)
Eg ~ c 137
This is a ratio, so is independent of our choice of units, so α is a fundamental constant that exists in nature
independent of our attmepts to decribe the world, and indicates some more fundamental physics underlying
the situation. In fact α2 is the fine structure constant and describes the probability for an electromagnetic
interaction, e.g. proton - electron scattering (of which hydrogen is a special case in which the scattering results
in a bound state).
From the diagram of an e-p interaction, each vertex contributes a factor e to the amplitude for the interaction,
so that

Amplitude ∼ e2 (1.8)

Probability ∼ Amplitude2 ∼ e4 ∼ α2 (1.9)

Now according to classical analysis, this is the only amplitude we would get for the interaction., but in quantum
mechanics there can be intermediate scattering events that cannot be observed: for example
********************
contributes an amplitude of α2 to the overall amplitude for the interaction. We can add more and more
intermediate states, obtaining a series expansion for the amplitude in terms of α:

Amplitude = 1 + α() + α2 () + ... (1.10)

Crucially, α is a small quantity, so the expansion converges nicely. This is the fundamental method of Quantum
Field Theory, so let us try it for the gravitational interaction between two point masses, each of mass M,
seperated by distance r. The potential is:

GM
V = (1.11)
r2
Again we will take as our reference energy

Eg = M c2 (1.12)
Taking the ratio of the two quantities, we obtain

V G2 M 4
= = αg2 (1.13)
Eg ~c
Where αg is the ’gravitational fine structure constant’. Immediately we see problems with using this parameter
to describe the gravitational interaction: the lower the gravitational energy (V), the weaker interaction, and
oppositely, when V > Eg , the expansion no longer converges as αg > 1.
1.3. POINT PARTICLE 3

We may see our way to a possible solution by considering the problem of beta decay: Initially it was treated as a
three body problem with the proton - neutron - electron interaction occuring at one vertex. When the energies
of the resultant electrons did not match experiment, the theory was modified to include a fourth particle, the
neutrino, and the interaction was ’smeared’ out: the proton and neutron interacted at one vertex, where a W
boson was created, which travelled a short distance before reaching the electron - neutrino vertex.
p p

e e

g g

v
n n
So maybe we can solve our problems with quantum gravity by smearing out the interactions, so that the objects
mediating the force are no longer point particles but extended one dimensional objects - strings.
This is the general concept from which we will proceed. It is a difficult task - the gravitational interaction
must obey a much larger symmetry than Lorentz invariance, it must be invariant under completely general co-
ordinate transformations, and we must of course still be able to describe the weak, strong and electromagnetic
interactions.
In this chapter we will take a first look at strings. Initially we examine the completely general equations of
motion for a point particle using the method of least action, and then apply that method to the case of a general
string moving in D dimensional case. We will obtain the equations of motion for the string, and then attempt
to quantize it and obtain it’s energy spectrum. This will highlight some basic results of string theory, as well
as some fundamental difficulties.

1.3 Point particle

We begin by examining the case of a point particle, illustrating the method we will use for strings. The trajectory
of a point particle is described in D dimensional space, decribed by co-ordinates X µ (τ ) where τ is a parameter
of the particle’s trajectory. We will identify X 0 with time, and the remaining co-ordinates X ~ with space.

t
t

In order to find the equations of motion for the particle we look to Fermat’s principle, and say that the trajectory
chosen by the particle will be the one that minimizes time, specifically the proper time of the particle. An
infintesimal element of proper time dT is related to an infinitesimal line element of the trajectory, ds, by:
4 CHAPTER 1. FIRST LOOK AT STRINGS

dT 2 = −ds2 (1.14)
where

~ 2
ds2 = −(dX 0 )2 + (dX) (1.15)
a

n
n

b
Rb
So the quantity we wish to minimize will be proportional to a dT, where a and b are the fixed start and end
points on the trajectory. We define the action for the system as:
Z b
S=m dT (1.16)
a

where m is the mass of the particle. Note that using the units convention described in section 1.1, this action
is dimensionless. Also, there is a problem that when the mass m=0, the action is zero. This problem will be
cleared up a little later.
We can write the action
Z b
dT
S=m dτ (1.17)
a dτ
v !2
u dX 0 2
u
Z b ~
dX
=m dτ t − (1.18)
a dτ dτ

From which we can define the Lagrangian for the system:

v !2
u dX 0 2
u
~
dX
L = mt − (1.19)
dτ dτ

Using the convention Ẋ= dX/dτ , we write Lagrange’s equations:

 
d ∂L ∂L
= (1.20)
dτ ∂ X ˙ µ ∂X µ
For this Lagrangian,

∂L
=0 (1.21)
∂X µ
and we obtain the equation of motion for the point particle:
1.3. POINT PARTICLE 5

!
X˙ µ
 
d ∂L d
= =0 (1.22)
dτ ∂ X˙ µ dτ L
The physical meaning of this equation will now be examined. Labelling the spacial co-ordinates of the particle
with the Roman letter i, we define the components of the velocity ~v of the particle to be

dX i
vi = (1.23)
dX 0
The quantity uµ is defined:

dX 0 dX µ dX 0
uµ = X˙ µ = = (1, ~v ) (1.24)
dτ dX 0 dτ
so
2
dX 0 uµ uµ

= (1.25)
dτ 1 − ~v 2
Here, the dot product is affected using the Minkowski metric, i.e.
  
2 1 0 1
(1, ~v ) = (1, ~v ) (1.26)
0 −1 ~v
Now the Lagrangian can be written
p
L = m uµ uµ (1.27)
So that
2
dX 0 L2 /m2

= (1.28)
dτ 1 − ~v 2
The quantity X˙ µ /L that appears in the equation of motion can now be written:

X˙ µ uµ
= = (γ, γ~v ) (1.29)
L/m L/m
where
1
γ=√ (1.30)
1 − ~v 2
and the equation of motion simply reads

d
(γ, γ~v ) = 0 (1.31)
dτ
i.e. there is no acceleration of the free particle along its’ path, as expected.
We will now look at a better expression for the action: defining an extra field η(τ ), which at the moment is
arbitrary, we write a new Lagrangian:
1 µ
L= Ẋ Ẋµ − 12 ηm2 (1.32)
2η
This immediately has the nice feature that it is still valid for m=0. Lagrange’s equations are
 
d ∂L ∂L
=0= (1.33)
dτ ∂ η̇ ∂η
6 CHAPTER 1. FIRST LOOK AT STRINGS

Since L 6= L(η̇). This gives us:

1
− Ẋ µ Ẋµ − 12 m2 = 0 (1.34)
2η 2
s
−Ẋ µ Ẋµ
η= (1.35)
m2
Using this equation to eliminate η from the Lagrangian (19), we get back to the original Lagrangian (6). We
can now also write the previous equation of motion as:
!
d X˙ µ
=0 (1.36)
dτ η

We will now examine the meaning of the field η(τ ). The trajectory is parameterized by some co-ordinate of the
system in terms of which an infinitesimal distance along the trajectory, ds, can be expressed. Let us suppose
our trajectory is along the y axis. Then it would be easiest to parameterize the system with the co-ordinate y,
such that ds = dy. We could, however, choose the parameter to be θ, the the angle between a line drawn from
a fixed point on the x axis to a position on the y axis. Then our new distance would be
 
1
ds = dθ (1.37)
cos2 θ
The factor 1/cos2 θ is our η. It represents the geometry of the system due to our choice of co-ordinates. Generally,

ds2 = η 2 dτ 2 (1.38)

√
η= −γτ τ (1.39)
0
where γτ τ is the called the metric of the system. Changing variables τ → τ :
 2
dτ
2
ds = −γτ τ dτ 02 (1.40)
dτ 0
Now we define
 2
dτ
γτ 0 τ 0 = γτ τ (1.41)
dτ 0
i.e. γτ τ transforms as a tensor. We can see that the action is invariant under this transformation: equation (42)
can be written

dτ
η0 = η (1.42)
dτ 0
and X˙ µ transforms as

0 dX µ dτ dX µ
X˙ µ = = (1.43)
dτ 0 dτ 0 dτ
Thus the Lagrangian transforms as

dτ 0
L0 = L (1.44)
dτ
and the action transforms as
1.4. THE OPEN STRING SPECTRUM 7

Z b Z b
S=m dτ L = m dτ 0 L0 (1.45)
a a

thus proving its invariance.

Now let us form the Hamiltonian for the system: the conjugate momenta to co-ordinate X µ and the parameter
η are

∂L X˙ µ
Pµ = = (1.46)
∂X˙ µ η

∂L
Pη = =0 (1.47)
∂ η̇
Then the Hamiltonian is:

H = Pµ X˙ µ + Pη η̇ − L (1.48)

= 12 η(P µ Pµ + m2 ) (1.49)
Here the role of η is that of a Lagrange multiplier, and it doesn’t effect the physics. From Hamilton’s equation:

∂H
= P˙η = 0 = P µ Pµ + m2 (1.50)
∂η
Which is Einstein’s equation for the relativistic energy of a paricle of mass m. What this reveals is that H = 0,
that is this is not a true Hamiltonian of the system. To find the true Hamiltonian, we define χ as:

1 µ H
χ= P Pµ + 12 m = (1.51)
2m ηm
Poisson brackets:

∂χ ∂X µ ∂χ ∂X µ Pν µ Pµ
{ χ , Xµ } = − = − δ ν = − (1.52)
∂X ν ∂Pν ∂Pν ∂X ν m m

∂χ ∂Pµ ∂χ ∂Pµ
{ χ , Pµ } = ν ν
− =0 (1.53)
∂X ∂P ∂P ν ∂X ν

1.4 The open string spectrum

Impose the gauge condition X 0 = τ to get the equations of special relativity.

Examing the Einstein equation:

0 = pµ pµ + m2 = −(p0 )2 + p~2 + m2 (1.54)

p
p0 = p~2 + m2 = H (1.55)

~˙ = ∂H = p 2~
X
p p~
= 0 (1.56)
∂~
p 2
p~ + m2 p

(p0 )2 − p~2 m2
1 − ~x˙ = = (1.57)
(p0 )2 (p0 )2
8 CHAPTER 1. FIRST LOOK AT STRINGS

~v p~/p0 p~
√ = 0
= (1.58)
1 − ~v 2 m/p m

m~v
p~ = √ (1.59)
1 − ~v 2
This is precisely momentum in special relativity.

Define light-cone coordinates in spacetime:

+ 1 +
~ i = (X 2 , ..., X D−1 )
X − = √ (X 0 − −X 1 ) X (1.60)
2

1 µ 1 ~˙ 2 ] − 1 ηm2 = − 1 Ẋ + Ẋ − + 1 Ẋi2 − 1 ηm2

L= Ẋ Ẋµ − 12 ηm2 = [(Ẋ 0 )2 − (X) 2 2 (1.61)
2η 2η η 2η

We will set the world-sheet parameter τ at each point of the world-sheet to be equal to the space-time coordinate
X + , which will lead to some simplifications. X + will play the role of time and P − that of energy. The
longitudinal variables X − and P + are then like spatial coordinates and momenta, as are the transverse Xi and
Pi .

Fix the parameter of the world-line by:

X + (τ ) = τ (1.62)
The Lagrangian becomes:

1 1
L = − Ẋ − + Ẋi2 − 12 ηm2 (1.63)
η 2η

∂L 1
P− = =− (1.64)
∂ Ẋ − η

∂L 1 ~˙
Pi = = X i (1.65)
∂ Ẋ− η
i i 2
H = Ẋ − P− + X ~˙ Pi − (− 1 Ẋ − + 1 X
~˙ Pi − L = Ẋ − P− + X ~˙ i − 1 ηm2 ) (1.66)
η 2η 2

1 1 ~˙ 2 1 − 1 ~˙ 2 1
= − Ẋ − + X i − (− Ẋ + X − 2 ηm2 ) (1.67)
η η η 2η i

1 ~˙ 2 1 η P~ i P~i + m2
= X i + 2 ηm2 = (P~ i P~i + m2 ) = (1.68)
2η 2 2P−

Note that there is no term P+ Ẋ + because X + is not a dynamical variable in the gauge-fixed theory.
1.4. THE OPEN STRING SPECTRUM 9

Strings

t=2

t=1

s=0 s=l
x

As an analogy to Fermat’s minimization of time, we will minimize the area of the world-sheet mapped out by
the string.

As an example in spherical coordinates:

~x = Rsinθcosφı̂ + Rsinθsinφ̂ + Rcosθk̂ (1.69)

∂~x ∂~x
· =0 (1.70)
∂θ ∂φ
v
x 2
u
∂~


∂~x ∂~x
u
∂θ 0
d(Area) = δθ · δφ = t 2 δθδφ (1.71)
u 
∂θ ∂φ ∂~
x
0 ∂φ

a2 ~a · ~b
(Area)2 = ~a × ~b = a2 b2 − a2 b2 cos2 θ = a2 b2 − (~a · ~b)2 = (1.72)
~a · ~b b2

The simplest invariant action, the Nambu-Goto action, is proportional to the area of the world-sheet.
ZZ p
SN G = T dτ dσ −det hab with hab = ∂a Xµ ∂b Xµ = −∂a X0 ∂b X0 + ∂a X∂~ bX~ (1.73)

σ a = (τ, σ) a = 0, 1 (1.74)

v
∂X 2 ∂X µ ∂Xµ
ZZ u

u
SN G = T dτ dσ t− ∂τ
∂X µ ∂Xµ
∂τ ∂σ
∂X 2

(1.75)
∂τ ∂σ ∂σ

 2  2
∂X ∂X 2
= = (∂0 X) = Ẋ 2 (1.76)
∂τ ∂σ 0
10 CHAPTER 1. FIRST LOOK AT STRINGS

Writing the Lagrange Equation:

   
∂ ∂L ∂ ∂L ∂L
= −→ ∂µ =0 (1.77)
∂τ ∂ Ẋ µ ∂σ ∂ X́ µ ∂(∂µ φ)

Rewriting the Langrangian in metric form:

Ẋ 2 Ẋ · X́
q
hab = L = (Ẋ · X́) − Ẋ 2 X́ 2 (1.78)
Ẋ · X́ X́ 2

We are free to choose our τ 0 s and σ 0 s, so we’ll do so as to simplify our Lagrangian of the following form:

−1 0
hab = . . . with the constraints Ẋ2 + X́2 = 0 and Ẋ · X́ = 0 (1.79)
0 1
Therefore:

ZZ
SN G = 21 T dτ dσ(Ẋ 2 − X́ 2 ) (1.80)

∂L Ẋµ X́ 2 − X́µ Ẋ · X́
= (1.81)
∂ Ẋ µ L

∂L X́µ Ẋ 2 − X́µ Ẋ · X́
= (1.82)
∂ X́ µ L

Therefore:

∂ ∂ ∂ 2 Xµ ∂ 2 Xµ
Ẋµ = X́µ → = The wave equation (1.83)
∂τ ∂σ ∂τ 2 ∂σ 2
We can write the constraints from equation (74) in a single equation by multiplying the second constraint by 2
and adding or subtracting it from the first:

(Ẋ ± X́)2 = 0 (1.84)

Defining new co-ordinates

1
σ ± = √ (τ ± σ) (1.85)
2
so that

1
∂± = √ (∂τ ± ∂σ ) (1.86)
2
and the constraints and the wave equation becomes:

(∂± X)2 = 0 (1.87)

1.4. THE OPEN STRING SPECTRUM 11

∂+ ∂− X µ = 0 (1.88)
The general solution to the wave equation is

X µ = f (σ + ) + g(σ − ) (1.89)
where f and g are arbitrary functions. Let us write out the action explicitly with the simplifications made above.
The parameter τ takes values from -∞ to +∞ and σ takes values between 0 and l, the length of the string.
Anticipating future results, we write the constant T

1
T = (1.90)
2πα0
where α0 is called the Regge slope. Then, from equation (81),
Z +∞ Z l
1
SN G = dτ dσ(Ẋ 2 − X́ 2 ) (1.91)
4πα0 −∞ 0
Z +∞ Z l
1
= dτ dσ(∂a X µ ∂ a Xµ ) (1.92)
4πα0 −∞ 0

We now examine the effects of boundary conditions which have yet to be taken into account in the equations of
motion. We start off by varying the co-ordinates:

X µ → X µ + δX µ (1.93)
Starting from equation (93), the variation in the action is
Z +∞ Z l
1
δSN G = dτ dσ (∂a δX µ ∂ a Xµ + ∂a X µ ∂ a δXµ ) (1.94)
4πα0 −∞ 0
Z +∞ Z l
1
= dτ dσ (∂a δX µ ∂ a Xµ ) (1.95)
2πα0 −∞ 0

And noting the total derivative

∂a (δX µ ∂ a Xµ ) = ∂a δX µ ∂ a Xµ + δX µ ∂a ∂ a Xµ (1.96)
The last term is just the wave equation, which equals zero, so we are left with:
Z +∞ Z l
1
δSN G = dτ dσ∂a (δX µ ∂ a Xµ ) (1.97)
2πα0 −∞ 0
Z +∞ il
1 h
= dτ δX µ X´µ (1.98)
2πα0 −∞ 0

We will now introduce two sets of boundary conditions that will get rid of this term and leave the equations of
motion unchanged at the boundary:
Open string (Neumann) boundary conditions, which correspond to there being no forces at the boundary:

X´µ (σ = 0) = X´µ (σ = l) = 0 (1.99)

Closed string boundary conditions, which means there is no boundary and the string co-ordinates are periodic:

Xµ (σ = 0) = Xµ (σ = l) (1.100)
12 CHAPTER 1. FIRST LOOK AT STRINGS

X´µ (σ = 0) = X´µ (σ = l) (1.101)

We shall now look at deriving invariant quantities in the theory from symmetries using Noether’s theorem. We
will start with Poincare invariance, which is invariant under the transformation

X µ → Λµ ν X ν + Y µ (1.102)
where Λ and Y are constant quantities. We construct the Noether current by applying this symmetry to the
action. Taking the second term, we write the change in X µ as

δX µ = Y µ (1.103)
The change in the action is found by inserting this into equation (96):
Z +∞ Z l
1
δSN G = dτ dσ∂a Y µ ∂ a Xµ (1.104)
2πα0 −∞ 0

The Noether current Pµa is defined by

Z Z
δS = dτ dσ∂a Y µ Pµa (1.105)

So in this case,

Pµa = T ∂ a Xµ (1.106)
and

∂a Pµa = T ∂a ∂ a Xµ = 0 (1.107)
i.e. the Poincare symmetry has led to a conserved quantity in the Noether current.
Now let’s do the same for the variation

δX µ = Λµ ν X ν (1.108)
Where  is a small quantity. The variation in the action is now
Z +∞ Z l
1
δSN G = dτ dσ (X µ ∂a X ν − X ν ∂a X µ ) Λµν ∂ a  (1.109)
2πα0 −∞ 0

From this we can define the current as

Jaµν = T (X µ ∂a X ν − X ν ∂a X µ ) (1.110)
which is conserved:

∂ a Jaµν = T (∂ a X µ ∂a X ν − ∂ a X ν ∂a X µ + X µ ∂ a ∂a X ν − X ν ∂ a ∂a X µ ) = 0 (1.111)
since the first two terms cancel and the last two terms are the wave equation, which equals zero.
Analogous to electromagnetism, we can define a charge. In EM, the charge is the integral over a volume of the
zeroth (time) component of the current 4-vector j µ , so here we define the charge for the current Pµa = (Pµτ , Pµσ )
as
Z l
Pµ = dσPµτ (1.112)
0
Pµτ can be seen to be the momentum of the string at a certain point, so Pµ , µ > 0 is the total momentum of the
string, and P0 is the total energy of the string.
1.4. THE OPEN STRING SPECTRUM 13

Differentiating Pµ with respect to time:

Z l Z l l
dPµ
= dσ P˙µτ = dσ P´µσ = Pµσ =0 (1.113)
dτ 0 0 0

where the second equality follows from the wave equation and the last equality form the boundary conditions
at 0 and l. This is simply conservation of momentum.
a
Similarly, for the current Jµν , which we interpret as the angular momentum of the string, we define the charge
Z l
a
Jµν = dσJµν (1.114)
0

and we find

dJµν
=0 (1.115)
dτ
So from Poincare invariance we obtain conservation of momentum and angular momentum.
We now give two examples to demonstrate the above concepts:

Example 1

y
R

t=0

x
We take a closed string whose initial configuration is a circle centered on the x-y origin with radius R, and whose
initial velocity is ~v = 0. Then X µ = (t, x, y). The solution to the wave equation satisfying these boundary
conditions is:
2πτ 2πσ
x = R cos cos (1.116)
l l
2πτ 2πσ
y = R cos sin (1.117)
l l
2πR
t= τ (1.118)
l
Let’s check the constraints Ẋ 2 + X́ 2 = 0:

−ṫ2 + ẋ2 + ẏ 2 − t́2 + x́2 + ý 2 (1.119)

 2  2  2
2πR 2πR 2πτ
2 2πR 2πτ
=− + sin + cos2 =0 (1.120)
l l l l l
14 CHAPTER 1. FIRST LOOK AT STRINGS

The total energy of the string is given by

Z l Z l Z l Z l
2πRT
P0 = dσP0τ = T dσ∂τ X0 = T dσ∂τ t = dσ = 2πRT = E (1.121)
0 0 0 l 0

The length of the string is 2πR, so T can be identified as energy per unit length of the string, i.e. the tension.

Example 2

Now we consider an open string rotating in the x-y plane. The solution to the wave equation is
πτ πσ
x = R cos cos (1.122)
l l
πτ πσ
y = R cos sin (1.123)
l l

πR
t= τ (1.124)
l
The speed of each point on the string is given by
   
dx dy l dx dy πσ  πτ πτ 
~v = , = , = cos − sin , cos (1.125)
dt dt πR dτ dτ l l l
From which we see that
πσ
~v 2 = cos2 (1.126)
l
Thus at the ends of the string σ = 0, l we see that ~v 2 = 1, i.e. the ends of the string travel at the speed of light.
This is to be expected, since the string is massless and there are no forces on the ends of the string (Neumann
boundary conditions). The intermediate points on the string don’t travel at the speed of light because they
experience the tension of the string.
The energy of the string is worked out exactly as before, and is found to be:

P0 = T Rπ (1.127)
Let us now work out the z component of the angular momentum of the string:
1.5. THE MODE EXPANSION 15

Z l
Jx y = T dσ(x∂a y − y∂a x) (1.128)
0

l
πR2  2 πτ
Z
πσ πτ πσ 
=T dσ cos cos2 + sin2 cos2 (1.129)
0 l l l l l

T πR2 l
Z
πσ
= dσ cos2 = 12 T πR2 (1.130)
l 0 l
From the total energy and angular momentum we can form the quantity
Jxy 1
= = α0 (1.131)
E2 2πT

α
2
E

Regge Slope

Now if we want our strings to correspond to fundamental particles, their angular momenta correspond to the
spins of the particles, which are either integer or half integer. Thus the above relation suggests that if we plot
the spins of particles against their energies squared, we would observe a straight line. This was indeed observed
for strongly interacting particles. In fact, string theory started out being a theory of the strong interaction, but
then QCD came along. Now string theory become a theory of everything!

1.5 The Mode Expansion

We must first introduce the Hamiltonian formalism for future reference: we have the Lagrangian

L = 12 T (Ẋ 2 − X́ 2 ) (1.132)
The conjugate momentum to the variable X is

∂L
Πµ = = T Ẋµ (1.133)
∂ Ẋ µ
Then the Hamiltonian is given by
Z l
H= dσ(Πµ Ẋ µ − L) (1.134)
0

l
Π2
Z  
= 1
2T dσ + X́ 2 (1.135)
0 T2
16 CHAPTER 1. FIRST LOOK AT STRINGS

Z l
= 12 T dσ(Ẋ 2 + X́ 2 ) = 0 (1.136)
0

So this is not a good Hamiltonian. We will find a good one later.

Now let us examine the mode expansion for open strings, which obey the Neumann boundary conditions. The
general solution to the wave equation is a fourier expansion. Here we write such an expansion as follows:
∞
π √ X 1 µ −πinτ  nπσ 
X µ = xµ + 2α0 pµ τ + i 2α0 αn e l cos (1.137)
l n=−∞
n l
x6=0

where xµ is a constant and the first mode pµ has been written out explicitly. The constants have been chosen
on dimensional grounds
The fact that X µ must be a real number yields the condition:

αn = (α−n )∗ (1.138)

The conjugate momentum to X µ is

∞
π α0 T X µ −πinτ  nπσ 
Πµ = T Ẋ µ = 2α0 T pµ + αn e l cos (1.139)
l l n=−∞ l
x6=0

The centre of mass position of the string is given by

Z l
1 π
X̄ µ = dσX µ = xµ + 2α0 pµ τ (1.140)
l 0 l
so the centre of mass moves in a straight line. The total momentum is
Z l
π
Pµ = dσΠµ = 2α0 l pµ T = pµ (1.141)
0 l
In both cases all the harmonic terms vanish on integration.
We now move to the light cone gauge mentioned before: we defined the transverse co-ordinates X + and X −
and fix the gauge by imposing the condition:
π
X + = x+ + 2α0 p+ τ αn+ = 0 ∀n (1.142)
l
and set
π
2α0 p+ = 1 (1.143)
l
so that

X + = x+ + τ (1.144)

which is the light cone gauge condition from before with an arbitrary constant x+ .
The centre of mass position can now be written

pµ
X̄ µ = xµ + τ (1.145)
p+
1.5. THE MODE EXPANSION 17

The constraint from before:

+
(Ẋ − X́)2 = 0 (1.146)

+ + + + + +
−2(Ẋ + − X́ + )(Ẋ − − X́ − ) + (Ẋ i − X́ i )2 = −((Ẋ 0 + Ẋ 1 ) − (X́ 0 + X́ 1 ))((Ẋ 0 − Ẋ 1 ) − (X́ 0 − X́ 1 )) + (Ẋ i − X́ i )2
(1.147)
+ + + + +
= −((Ẋ 0 − X́ 0 ) + (Ẋ 1 − X́ 1 ))((Ẋ 0 − X́ 1 ) − (Ẋ 1 − X́ 1 )) + (Ẋ i − X́ i )2 (1.148)

+ + + +
= −(Ẋ 0 − X́ 0 )2 + (Ẋ 1 − X́ 1 )2 + (Ẋ i − X́ i )2 = (Ẋ µ − X́ µ )2 = 0 (1.149)

+ + +
∴ 2(Ẋ + − X́ + )(Ẋ − − X́ − ) = (Ẋ i − X́ i )2 (1.150)

+ +
∴ 2(Ẋ − − X́ − ) = (Ẋ i − X́ i )2 (1.151)

The mode expansion of the world-sheet fields:

∞
π √ X 1 µ −πinτ  nπσ 
X µ = xµ + 2α0 pµ τ + i 2α0 αn e l cos (1.152)
l n=−∞
n l
x6=0

  
∞ ∞
+ π √ πi X −πinτ
 nπσ +  √ 0 π
 X −πinτ
 nπσ 

2(ẋ− − x́− ) = 2 2α0 p− − i 2α0 αn− e l cos − −i 2α αn− e l sin

l l n=−∞ l l n=−∞ l

x6=0 x6=0
(1.153)
 
∞
√
  nπσ 
π π X − −πinτ  nπσ  +
= 2 2α0 p− + 2α0 αn e l cos − isin (1.154)
 
l l n=−∞ l l

x6=0
 
 
∞ +
π √ π X −πin
τ −σ 
= 2 2α0 p− + 2α0 αn− e
l
(1.155)

l l

n=−∞
x6=0

 2
 

∞ +
+ π √ π X i −πin τ −σ 
∴ (ẋi − x́i )2 = 2α0 pi +
l
2α0 α e (1.156)

l l n=−∞ n

x6=0

You can write αn− in terms of αni :

∞
+ pi p i π2 0 X i π2
(Ẋ i − X́ i )2 = + +
+ 2
2α α−n αni + 2 2 (2α0 )3/2 pi f (σ, τ ) (1.157)
p p l n=−∞
l
n6=0

Concentrating on the zeroth mode:

18 CHAPTER 1. FIRST LOOK AT STRINGS

∞
p− pi pi 1 X
2 = + αi αi (1.158)
p+ (p+ )2 α0 (p+ )2 n=−∞ −n n
n6=0
 
∞
1  1 X
∴ p− = + pi pi + 0 i
α−n αni  = H (1.159)

2p α n=−∞
n6=0

From Einstein’s equation:

pµ pµ + m2 = 0 (1.160)

∞
1 X i
m2 = −pµ pµ = 2p+ p− − pi pi = α αi (1.161)
α0 n=−∞ −n n
n6=0

Now quantizing the system:

Π (σ), X j (σ 0 ) = −iδ ij δ(σ − σ 0 )

 i j
p , x = −iδ ij
 i 
(1.162)

∞
pµ √ 0 π X µ −πinτ  nπσ 
Πµ = T Ẋµ = + 2α T αn e l cos (1.163)
l l n=−∞ l
x6=0

∞
π √ X 1 µ −πinτ  nπσ 
X µ = xµ + 2α0 pµ τ + i 2α0 αn e l cos (1.164)
l n=−∞
n l
x6=0

∞ ∞
mπσ 0
 
 1  i j i X 1 i
X  −πiτ  nπσ 
Π (σ), X j (σ 0 ) =
 i
p ,x + αm , αnj e l (n+m) cos cos (1.165)
l l n=−∞ m=−∞ n l l
n6=0 m6=0

 i
αm , αnj = mδ ij δm+n,0

(1.166)

Π (σ), X j (σ 0 ) = −iδ ij δ(σ − σ 0 )

 i 
∴ (1.167)

This is just the commutation relation for the harmonic oscillator operators with nonstandard normalization
1 i 1
a = √ αm a† = √ α− mi (1.168)
m m
 † 
∴ a ,a = 1 (1.169)
The state |0, ki is defined to be annihilated by the lowering operators and to be an eigenstate of the center-of-
mass momenta

a|0i = a|0, ki = 0 (1.170)

1.5. THE MODE EXPANSION 19

Πm,i |0, ki = |0, ki (1.171)

∞ ∞ ∞ ∞
1 X X i 1 X X 1
M2 = α α i
= ma† a = 0 N (1.172)
α0 n=−∞ i=−∞ −n n α0 n=−∞ i=−∞ α
n6=0 i6=0 n6=0 i6=0
 
∞ ∞
1 X (D − 2)n (D − 2)  X
M 2 |0, ki = 0 |0, ki = n |0, ki (1.173)

α n=−∞ 2 2α0

n=−∞
n6=0 n6=0

We have to perform the sum:

∞
X
n (1.174)
n=−∞
n6=0
−2πn
To perform this sum, we will multiply by the sum by e l and then take the limit of  → 0
∞ ∞
e−C
 
X −2πn ∂ X ∂ 1 1 1
ne l = ne−nC = −C
= = 2− (1.175)
n=−∞
∂C n=−∞ ∂C 1 − e (1 − e−C )2 C 12
n6=0 n6=0

2π
where C = (1.176)
l
∞
X 1 1
ne−nC = − (1.177)
n=−∞
c2 12
n6=0

∞ ∞ ∞
1 X X i (D − 2) X
M2 = α α i
= n = 2p+ p− − pi pi (1.178)
α0 n=−∞ i=−∞ −n n 2α0 n=−∞
n6=0 i6=0 n6=0

The last equality was obtained using(64)

(D − 2) l2
 
(D − 2) 1 1 (D − 2) (D − 2) + (D − 2)
M2 = − = − = πp l − = 2p+ p− − pi pi (1.179)
2α0 c2 12 2α0 (2π)2 24α0 (2π)2 24α0

The mass of each state is thus determined in terms of the level of excitation.
 
2 1 D−2
M = 0 N− N |0i = 0 (1.180)
α 24
This operator acting on the 0 ket yields:

(D − 2)
M 2 |0i = −
|0i (1.181)
24α0
The mass-squared is negative for D > 2. The state is a tachyon

The lowest excited states of the string are obtained by exciting one of the n = 1 modes once:
20 CHAPTER 1. FIRST LOOK AT STRINGS

   
1 D−2 1 26 − D
M 2 (α−1
i
|0i) = 1− i
(α−1 )|0i = |0i (1.182)
α0 24 α0 24
Lorentz invariance now requires that this state be massless, so the number of spacetime dimensions is D = 26

1.6 Closed strings

We must now look at the mode expansion and quantization os closed strings, which are required when looking
at string interactions. The procedure is very similar to that of open strings, except now we have Dirichlet
boundary conditions, i.e. X µ is periodic. The mode expansion is now made up of left and right moving parts:
µ
X µ = XR (τ − σ) + XLµ (τ + σ) (1.183)
such that
∞
1 µ −2πin(τ −σ)
q X
µ
XR = 21 xµ + α0 pµ (τ − σ) + i 1 0
2α αn e l (1.184)
n=−∞
n
x6=0

∞
1 µ −2πin(τ +σ)
q X
XLµ = 12 xµ + α0 pµ (τ + σ) + i 1 0
2α α̃ e l (1.185)
n=−∞
n n
x6=0

The sum of these is periodic. In the sum the integer n is in effect 2n so there are now twice as many modes as
for the open string. Once again, the fact that X µ is real means that
µ µ
αnµ = (α−n )∗ α̃nµ = (α̃−n )∗ (1.186)
Quantizing as before:
 i
αm , αnj = α̃m
  i
, α̃nj = mδ ij δm+n,0

(1.187)

 i
αm , α̃nj = 0

(1.188)
The mass operator is now

 
  ∞ ∞
2 2(D − 2) 2  X X 2(D − 2) 
M 2 = 2p+ p− − pi pi = NR + NL + = i
α−n αni + i
α̃−n α̃ni −  (1.189)
α0 24 0
α n=−∞ 24

n=−∞
n6=0 n6=0

There are two symmetries in these expansions: the transformations

τ → τ + constant (1.190)

σ → σ + constant (1.191)
don’t change the physics of the string. The first symmetry is shared with the open string, but the spatial
translational symmetry is new.
The generator of the time translations is H, which we saw before to be
Z l
H= dσ(Ẋ 2 + X́ 2 ) (1.192)
0
1.6. CLOSED STRINGS 21

and the fact that H = 0 leads to the invariance under time translations. In the case of closed strings,
Z l
H= dσ[(ẊR + ẊL )2 + (X́R − X́L )2 ] (1.193)
0
Z l
2
∼ dσ(ẊR + ẊL2 ) = HR + HL (1.194)
0

where we have used the fact that we can write

∂σ ∼ ∂τ (1.195)
since τ and σ are interchangeable in the expansions of XR and XL to within a minus sign.
This generation of time translations comes from the constraint Ẋ 2 + X́ 2 = 0, so it is reasonable to suppose that
spatial translations come from the other constraint Ẋ · X́ = 0. Defining the operator D:
Z l Z l Z l
D= dσ Ẋ · X́ ∼ dσ Ẋ 2 = 2
dσ(ẊR − ẊL2 ) ∼ NR − NL = 0 (1.196)
0 0 0
So

NR = NL (1.197)
which common sense tells us must be the case.
Now let us determinant the lowest states in the mass spectrum. For the vacuum state

4(D − 2)
M 2 |0i = −
|0i (1.198)
24α0
so the mass squared is negative for the vacuum state, as for the open string.
j
The first excited state is |Ωij i = α̃−1
i
α−1 |0i , where we must remember to keep NR = NL . We obtain
 
2 2 − 2(D − 2)
M 2 |Ωij i = 0 |Ωij i (1.199)
α 24
As before, we wish M 2 = 0 for the first excited state, so again we get D = 26. The situation is a bit more
complicated than for the open string case, so let’s look in a bit more detail.
The state |Ωij i can be split into three parts: a symmetric, traceless part; an antisymmetric part and a scalar
part:
 
2 1
|Ωij i = 12 (|Ωij i + |Ωij i) − δ ij |Ωkk i + 12 (|Ωij i − |Ωij i) + δ ij |Ωkk i (1.200)
D−2 D−2
We call the three states |Gij i, |B ij i, |Φi. The symmetric, traceless, spin 2 state |Gij i can now be identified with
the graviton, which didn’t exist in the open string theory, so it seems some progress has been made. The spin
0 scalar state is called the dilaton.
Now we impose a further symmetry: invariance under the transformation

σ → −σ (1.201)
which is the condition for unoriented strings. This means

XR ↔ XL (1.202)

αn ↔ α̃n (1.203)
This condition immediately disallows the antisymmetric state, since under the transformation,
22 CHAPTER 1. FIRST LOOK AT STRINGS

|B ij i → −|B ij i (1.204)
while |Gij i and |Φi remain unchanged.
Next let us turn our attention to the fact that we have been working in light cone gauge, which is not Lorentz
invariant. We would like to reassert lorentz invariance. We can do this by generalizing the commutation identity:
 i
αm , αnj = mδ ij δm+n,0 µ
, αnν ] = mη µν δm+n,0

→ [αm (1.205)
This gives
 0 0
αm , αn = −mδm+n,0 (1.206)
Now let us define a state
0
|φi = α−1 |0i (1.207)
The norm of this state is

hφ|φi = h0|α10 α−1

0
|0i = h0|[α10 , α−1
0 0
]|0i + h|α−1 α10 |0i = −h0|0i = −1 (1.208)
Thus we have a negative norm state, which is physically meaningless. This tells us there is something wrong
with our theory. We will come back to this in the next couple of chapters, and solve this problem.

1.7 Gauge invariance

Let us examine the open string state
µ
|φi = Aµ (k)α−1 |0, ki (1.209)
under the transformation

Aµ → Aµ + kµ ω(k) (1.210)
~ →
where ω(k) is an arbitrary function. This is analogous to a gauge transformation in electromagnetism A
~ + ∇ω. The change in |φi is
A
µ
|δφi = kµ ωα−1 |0, ki (1.211)
and the norm is

hδφ|δφi = kµ kν ω 2 h0, k|α1µ α−1

ν
|0, ki = k 2 ω 2 = 0 (1.212)
since k = 0 (the mass is zero, and m2 = k µ kν ). Thus the theory has produced gauge invariance! The equivalent
state for the closed string is
µ
|φi = gµν α̃−1 α1ν |0, ki (1.213)
The gauge transformation is

gµν → gµν + kµ ων + kν ωµ (1.214)

This time we find the norm state to be

hδφ|δφi ∼ k 2 = 0 (1.215)
Thus gauge invariance holds for the closed string state too. What’s more, we can identify gµν with the gravita-
tional potential, a sign that general relativity might be included in the theory.