Professional Documents
Culture Documents
Mandelstam
Mandelstam
Relativistic kinematics
Literature:
Nachtmann [4]
Hagedorn [5]
Byckling/Kajantie [6]
We state some notation concerning special relativity:
x)
x = (x0 = t, x1 , x2 , x3 ) = (t, #
#
x = (t, x )
g = g =
1
1
contravariant four-vector
covariant four-vector
(2.1)
(2.2)
metric tensor
(2.3)
Lorentz invariant
(2.4)
proper time.
(2.5)
2 = t2 x2 = g x x = x x = x2
d #
x 2 dt
d = dt 1
=
dt
dx dt
dx
=
= (1, #
v ).
d
dt d
Since
v 2 ) = 1 > 0,
u2 = 2 (1 #
7
m2 + #
p 2.
(2.6)
#
2
1
1 #
p
p2
+ ...
E = m2 + #
p2 = m 1 +
+
.
.
.
=
m
+
2 m2
2m
so that we recover Newtons expression for | #
v | 1.
2.1
Particle decay
The decaying particles four-momentum is, in the rest frame, given by p = (M, 0, 0, 0),
see Fig. 2.1. The decay time (lifetime) is
d 2 = dt2 (1 #
v 2)
where dt2 is the lifetime in the laboratory frame:
dt = d > d.
(2.7)
p21 = m21
p22 = m22
#
p = (M, 0 )
p1 = (E1 , #
p 1)
p2 = (E2 , #
p 2 ).
p1
p2
Figure 2.1: Particle decay. Dynamics will be discussed later on; at the moment we are
dealing with kinematics.
It therefore follows that
p pi = M E i E i =
1
1
p pi =
(p1 pi + p2 pi ).
M
M
1 4
2
2
2
2
2 2
#
2M
(m
+
m
)
+
(m
m
)
M
= #
p 22 ,
p 12 = E12 m21 =
1
2
1
2
4M 2
p remain unknown, while
is also xed. This means that only the directions of #
p and #
E1 =
the energies and the absolute values of the momenta can be calculated directly.
2.2
Two-particle scattering
For a visualisation of the process see Fig. 2.2(a). Once again, the constraints are
p2i = m2i (i = 1, . . . , 4)
p 1 + p2 = p 3 + p4 .
If m1 = m3 and m2 = m4 , elastic scattering takes place. Consider the Lorentz invariants
p2i = m2i and p1 p2 , p1 p3 , p1 p4 , p2 p3 , p2 p4 , p3 p4 .
10
00
p1
p4
p2
p3
(a)
(b)
#
p 1 + #
p 2 = 0 = #
p 3 + #
p 4.
(2.8)
The corresponding variables are asterisked: (cm., pi = pi ). The laboratory frame is dened
by #
p 2 = 0 (xed target) and variables are labelled with an L : (lab., pi = pLi ). In deep
#
#
inelastic scattering the Breit system (pi = pB
i ) is used, which is dened by p 1 + p 3 = 0.
In the following we take a closer look at the center of mass frame, see Fig. 2.3. Equation
(2.8) leads to
#
p 1 = #
p 2 = #
p
#
#
#
p3 = p4 = p
p1 = E1 = #
p
p 2 + m21 , #
#
#
2
2
p 2 = E 2 = p + m2 , p
p3 = (E3 , #
p )
p = (E , #
p ).
4
11
#
p
#
p
#
p
#
p
Figure 2.3: Two-particle scattering in center of mass frame. For the constraints on the
scattering angle see section 2.2.1.
The sum
#
p1 + p2 = (E1 + E2 , 0 )
s
s = (p1 + p2 )2 = (E1 + E2 )2
is one. Now we can express Ei , | #
p |, and | #
p | in terms of s (see exercise sheet 1):
1
= (s + m21,3 m22,4 )
E1,3
2 s
1
#
p 2 = (E1 )2 m21 = (s, m21 , m22 ),
4s
where we have used the Kallen function (triangle function) which is dened by
(a, b, c) = a2 + b2 + c2 2ab 2ac 2bc
= a ( b + c)2 a ( b c)2
= a2 2a(b + c) + (b c)2 .
We can see that the Kallen function has the following properties:
symmetric under a b c and
asymptotic behavior: a b, c : (a, b, c, ) a2 .
(2.9)
(2.10)
12
is the threshold of the process in the s-channel. In the high energy limit (s m2i ) Eq. (2.9)
and (2.10) simplify because of the asymptotic behavior of and one obtains:
#
#
E1 = E 2 = E 3 = E 4 = | p | = | p | =
.
2
2.2.1
Scattering angle
.
cos =
(s, m21 , m22 ) (s, m23 , m24 )
s and
or
s and t.
2.2.2
Elastic scattering
We now consider the case of elastic scattering. This means that m1 = m3 and m2 = m4
(e. g. ep ep). Therefore Eq. (2.9) and (2.10) simplify:
E1 = E3 , E2 = E4
1
| #
p |2 = | #
p |2 =
s (m1 + m2 )2 s (m1 m2 )2
4s
and we nd for the scattering angle (in the case of elastic scattering)
t = (p p )2 = ( #
p #
p )2 = 2 #
p 2 (1 cos )
1
cos = 1 +
t
.
#
2| p |2
4| #
p |2 t 0
1 cos 1
.
#
2
s (m1 + m2 )2
p 0
2.2.3
13
Angular distribution
4s
d
= # # .
=
2
2
2
2
dt
| p || p |
(s, m1 , m2 ) (s, m3 , m4 )
2.2.4
(2.11)
Relative velocity
At this point, we introduce the relative velocity, which we will see to be of relevance in
dening the particle ux and hence the collider construction,
#
#
#
#
p1
p1
| #
p 1 |
p
p
2
2
#
#
v12 = | v 1 v 2 | =
(E + E ),
(2.12)
=
=
1 2
E1
E2
E1
E2
E1 E2
v12 E1 E2
#
2
= s| p 1 | = s E1 m21
1
(s + m21 m22 ) m21
= s
4s
(2.13)
the so called Mller ux factor. In going from the rst line to the second, we used the
denition of the Kallen function and in going to the third the fact that s = m21 +m22 +2p1 p2 .
We stress here that v12 E1 E2 is a frame independent quantity. It appears in the denition
of the incoming particle ux, an thus in the cross section. It also plays an important role
in the normalization issues, since the classical volume element is not Lorentz invariant.
2.2.5
For the center of mass and the laboratory systems respectively, we have,
CM : s = (E1 + E2 )2 = (total energy)2
L : s = m21 + m22 + 2m2 E1L
E1L m1 ,m2
2m2 E1L .
As an example for the dierence, we look at the two operating modes of the Tevatron at
Fermilab (Figure 2.4). The energy of the beam particles is Ebeam = 980 GeV.
14
Ts
Used in the p
p-mode, the collision is head on and we are allowed to consider ourselves in
the center of mass frame and,
2.3
Crossing symmetry
The 2 2 scattering process has some underlying symmetries, which we shall explore
now.
Example When we exchange p3 and p4 , s is not aected but t and u interchange their
roles.
We take now a look at the reaction (Figure 2.5), 1 + 2 3 + 4, for which the 4-momentum
is conserved :
p 1 + p2 = p 3 + p 4 .
00
p1
p2
p1
Tt
p4
p2
15
p3
Tt
p4
p3
(2.14)
2.3.1
(2.15)
Interpretation of antiparticle-states
j
electron charge
probability density
charge density
00
16
p3
p3
p1
p1
p4
p2
p2
p4
Figure 2.7: Emission of a positron and absorption of an electron. The emission of a positron
with energy +E is equivalent to the absorption of an electron with energy E.
Inserting the wave function of the free electron,
= N eipx ,
(2.17)
+E
,
+ #
p
2 E
e with 4-momentum p : j (e ) = 2e|N | (p) = 2e|N |
,
#
p
and hence the rule,
j (e+ ) = j (e ) with the subsititution p p .
(2.18)
We stress here the fact that the whole 4-vector p takes a minus sign, and not only the
spatial part #
p.
What we eectively used here is the fact that in the phase of Eq. (2.17) we can ip the
signs of both p and x without changing the wave function. There is no place here for
particle travelling backwards in time!
A particle with 4-momentum p is a representation for the corresponding anti particle
with 4-momentum p . Alternatively, one can say that the emission of a positron with
energy +E corresponds to the absorption of an electron with energy E. Figure 2.7
restates the last sentence as a Feynman diagram.
The three reactions (s-, t- and u-channels) are described by a single function T (s, t, u)
evaluated in the relevant kinematical region (s 0 or t 0 or u 0).
In order to represent the situation, one usually refers to the Dalitz plot 1 (Figure 2.8).
1
or equilateral coordinates
17
t-channel
t = t min
00
s
u = umin
s = smin
t=0
s-channel
u-channel
s=0
u=0
p3
p3
p1
p1
p4
p2
p2
p4
e+
(a)
e+
(b)
which is the s-channel of the reaction depicted on Figure 2.9(a). By crossing, we get as
u-channel reaction the Bhabha scattering,
e + e e+ e ,
which is the reaction depicted on Figure 2.9(b).
The considerations of this chapter enable us to derive constraints on the possible dynamics
but are not sucient to decide on the dynamics. To get the dynamics we must calculate
and compare to experiments decay rates and scattering cross-sections.
18