Project5, Part b, 3-particle unitarity

In[1]:=

<< FeynCalc`

In[2]:=

Clear[k1, k2, k3, x1, x2, x3, q, M, s, e, g]

In[3]:=

k1 = ToExpression[SubscriptBox["k", "1"], StandardForm];

k2 = ToExpression[SubscriptBox["k", "2"], StandardForm];
k3 = ToExpression[SubscriptBox["k", "3"], StandardForm];
x1 = ToExpression[SubscriptBox["x", "1"], StandardForm];
x2 = ToExpression[SubscriptBox["x", "2"], StandardForm];
x3 = ToExpression[SubscriptBox["x", "3"], StandardForm];
s = ToExpression[SubscriptBox["", "s"], StandardForm];

In[10]:=

ClearScalarProducts;
SP[k1, k1] = 0; SP[k2, k2] = 0; SP[k3, k3] = M2 ;
1
SP[k1, k2] = q2 1 - x3 + M2 ;
2
1
SP[k1, k3] = q2 1 - x2 - M2 ;
2
1
SP[k2, k3] = q2 1 - x1 - M2 ;
2

In[15]:=

f01 = Simplify
TrGA[].GS[k2].GA[].

Out[15]=

GS[k1] + GS[k3]
SP[k1 + k3, k1 + k3]

.GA[].GS[k1].GA[].

1
4 q2 x2 M2 - q2 + x1 M2 + q2 x2 - q2 + M2 x3 - 3 M2 + q2 g +
q4 (x2 - 1)2

2 k2 M2 k1 + k3 M2 + q2 x2 - q2 + k2 M2 k1 + k3 M2 + q2 x2 - q2
In[16]:=

Out[16]=

In[17]:=

Out[17]=

In[18]:=

Out[18]=

f1 = Simplify[Contract[[Link][, ]]]
8 x2 M2 - q2 + x1 M2 + q2 x2 - q2 + M2 x3 - 3 M2 + q2
q2 (x2 - 1)2
f2 = f1 /. {x1 x2, x2 x1}
8 x1 M2 - q2 + x2 M2 + q2 x1 - q2 + M2 x3 - 3 M2 + q2
q2 (x1 - 1)2
f12 = Cancel[f1 + f2]
8 M2 x1 + M2 x2 + M2 x3 - 3 M2 - q2 x1 - q2 x2 + q2 x1 x2 + q2
+
q2 (x1 - 1)2
8 M2 x1 + M2 x2 + M2 x3 - 3 M2 - q2 x1 - q2 x2 + q2 x1 x2 + q2
q2 (x2 - 1)2

In[19]:=

Out[19]=

f120 = Cancel[f1 + f2 /. M 0]
8 (x1 - 1) 8 (x2 - 1)
+
x2 - 1
x1 - 1

GS[k1] + GS[k3]
SP[k1 + k3, k1 + k3]

Part [Link]

In[20]:=

f03 = Simplify
Tr- GA[].GS[k2].GA[].

Out[20]=

q4

GS[k2] + GS[k3]
SP[k2 + k3, k2 + k3]

.GA[].GS[k1].GA[].

GS[k1] + GS[k3]
SP[k1 + k3, k1 + k3]

1
4 q2 (x1 + x2 + x3 - 3) M2 - q2 x3 + q2 g +
(x1 - 1) (x2 - 1)

2 k3 k1 M2 - q2 x3 + q2 + k1 k1 M2 + q2 x1 - q2 - q2 (x1 + x2 + x3 - 3) k2 +

k2 k2 M2 + q2 x2 - q2 + k3 M2 - q2 x3 + q2 + q2 (x3 - 1) -k1
In[21]:=

Out[21]=

In[22]:=

Out[22]=

In[23]:=

Out[23]=

In[24]:=

Out[24]=

In[25]:=

Out[25]=

In[26]:=

f3 = Cancel[Contract[[Link][, ]] /. {x1 x2, x2 x1, x3 2 - x1 - x2}]

8 M4 + M2 q2 x1 + M2 q2 x2 + q4 x1 + q4 x2 - q4
q4 (x1 - 1) (x2 - 1)
f4 = Cancel[f3 /. {x1 x2, x2 x1}]
8 M4 + M2 q2 x1 + M2 q2 x2 + q4 x1 + q4 x2 - q4
q4 (x1 - 1) (x2 - 1)
f4 = Cancel[f3 /. {x1 x2, x2 x1}]
8 M4 + M2 q2 x1 + M2 q2 x2 + q4 x1 + q4 x2 - q4
q4 (x1 - 1) (x2 - 1)
f34 = Cancel[f3 + f4]
16 M4 + M2 q2 x1 + M2 q2 x2 + q4 x1 + q4 x2 - q4
q4 (x1 - 1) (x2 - 1)
f340 = Cancel[f3 + f4 /. M 0]
16 (x1 + x2 - 1)
(x1 - 1) (x2 - 1)
4 4 e 2 g2
1
f12 + f34

f = 3
Simplify
3
2
3
128 3
e 2 g2

Out[26]=

In[27]:=

Out[27]=

In[28]:=

q2 x2 M 2 -q2 +x1 M 2 +q2 x2 -q2 +M 2 x3 -3 M 2 +q2

2
x 1 - 1

In[29]:=

Out[29]=

q2 x2 M 2 -q2 +x1 M 2 +q2 x2 -q2 +M 2 x3 -3 M 2 +q2

2
x 2 - 1
3
4
6 q

2 M 2 + q2 M 2 + q2 x 1 + q2 x 2 - q2

x 1 - 1 x 2 - 1

f0 = Simplify[f /. M 0]
e2 g2 x21 + x22
6 3 (x1 - 1) (x2 - 1)
diffCrossSection = Simplify
8 e2 g2 2s

Out[28]=

(4 s)2
q2

q2 x2 M 2 -q2 +x1 M 2 +q2 x2 -q2 +M 2 x3 -3 M 2 +q2

2
x 1 - 1

f
q2 x2 M 2 -q2 +x1 M 2 +q2 x2 -q2 +M 2 x3 -3 M 2 +q2
2
x 2 - 1
6
3q

diffCrossSectionM0 = Simplify[diffCrossSection /. M 0]
8 e2 g2 x21 + x22 2s
3 q2 (x1 - 1) (x2 - 1)

2 M 2 + q2 M 2 + q2 x 1 + q2 x 2 - q2

x 1 - 1 x 2 - 1

Project5, 2-particle unitarity

In[1]:=

<< "FeynCalc`";

FeynCalc 9.2.0 (development version). For help, use the

documentation center, check out the wiki or write to the mailing list.
See also the supplied examples. If you use FeynCalc in your research, please cite
V. Shtabovenko, R. Mertig and F. Orellana,
Comput. Phys. Commun., 207C, 432-444, 2016, arXiv:1601.01167
R. Mertig, M. Bhm, and A. Denner, Comput. Phys. Commun., 64, 345-359, 1991.
In[2]:=

Clear[k1, k2]

In[3]:=

k1 = ToExpression[SubscriptBox["k", "1"], StandardForm];

k2 = ToExpression[SubscriptBox["k", "2"], StandardForm];

In[5]:=

ClearScalarProducts;

In[6]:=

f01 = Simplify[Tr[GS[k2].GA[].GS[k1].GS[k1].GS[k1].GA[]]]

Out[6]=

In[7]:=
Out[7]=

4 k1 -k1 k2 g + k2 k1 + k1 k2
f1 = Contract[[Link][, ]]
2

-8 k1 k1 k2

Quantum Field Theory II

Radiation of Gluon Jets
Instructor:
Submitted by:
Due date:

Dr. Mansoor ur Rehman

Mudassar Sabir
11-Jan-2017

Problem: We first calculate e+ e q qg to leading order in and s as following:

(a) Let q be the total 4-momentum of the electron, let k1 and k2 be the 4-momenta of the final
quark and antiquark, and let k3 be the 4-momentum of gluon. Define:
xi =

2ki q
,
q2

i = 1, 2, 3

(1)

this is theP
ratio of the center-of-mass energy of particle i to the maximum available energy. Then
show (i) i xi = 2, (ii) all other Lorentz scalars involving only the final-state momenta can be
computed in terms of the xi and the particle masses, and (iii) the complete integral over 3-body
phase space can be written as
Z
Z
X
Y Z d3 ki 1
q2
4 (4)
dx1 dx2 .
(2)
(2) (q
ki ) =
d3 =
3 2E
3
(2)
128
i
i
i
(iv) Find the region of integration for x1 and x2 if the quark and antiquark are massless but gluon
has mass .
(b) Using Cutkoskys cutting rules for three particle unitarity derive the differential cross-section
d2
42
4s
x21 + x22
(e+ e q qg) =
3e2q

,
dx1 dx2
3s
3 2 (1 x1 )(1 x2 )

(3)

in the limit 0. If we assume that each original final-state particle is realized physically as a jet
of strongly interacting particles, this formula gives the probability for observing three-jet events in
e+ e annihilation and the kinematic distribution of these events. The form of the distribution in the
xi is an absolute prediction, and it agrees with experiment. The normalization of this distribution
is a measure of the strong-interaction coupling constant.
In the second part of this project we will calculate the differential cross-section for the process
e+ e hadrons to order 2 s . To control infrared divergences, it will be necessary to assume
that gluons have small non-zero mass that can be set to zero at the end of calculations.
(c) First replace = 0 in the formula of part (b) for the differential cross-section, and carefully
integrate over the region found in part (a). In the small mass limit 2  q 2 , you will find infrared
2
divergent terms of order log(2 /q 2 ) and also log(2 /q 2 ) .
(d) Now using the Cutkosky rules for two particle unitarity find the one-loop s corrected crosssection for e+ e q q. In the small mass limit 2  q 2 , you will again find infrared-divergent

2
terms of order log(2 /q 2 ) and also log(2 /q 2 )
1

(e) Now combine the result with the answers from part (c) and (d) to form the total cross section
for e+ e hadrons to order 2 s . Show that all infrared divergent logarithms cancel out in the
limit 0 and we finally obtain,
(e e+ q q or q qg hadrons) =

42 X 2

s

eq (1 + ).
3s

(4)

Cutting Rules:
" Cut through the diagram in any way that can put all of the cut propagators on-shell without
violating momentum conservation.
For each cut, replace

p2

i
2 (4) (p2 m2 ).
2
m + i

9 Sum the contributions of all possible cuts. The result is the discontinuity of the diagram,
where Disc(iM(p0 )) = iM(p0 + i) iM(p0 i) = 2ImM.

Solution: (a) (i)

X
i

xi =

2(k1 + k2 + k3 ) q
q2

= 2

q = k1 + k2 + k3

(5)

In the center-of-mass frame,

xi =

2Ei
Eq

(6)

(ii) From the definition xi = 2kqi2 q , (i = 1, 2, 3), we show that all Lorentz scalars involving final
states only viz. (k1 + k2 )2 , (k2 + k3 )2 and (k3 + k1 )2 can be represented in terms of xi and particle
masses.
(k1 + k2 )2 = (q k3 )2
= q 2 + m23 2q k3
= q 2 (1 x3 ) + m23 ,
Similarly,
(k2 + k3 )2 = q 2 (1 x1 ) + m21
(k3 + k1 )2 = q 2 (1 x2 ) + m22
Taking k12 = k22 = 0 and k32 = 2 the above relations become,
2

2k1 k2 = q 2 (1 x3 ) + 2
2k2 k3 = q 2 (1 x1 ) 2
2k1 k3 = q 2 (1 x2 ) 2

(7)

(iii) Consider,
Z
Z 3~
d k1 1 d3~k2 1 d3~k3 1
d3 =
(2)4 (4) (q k1 k2 k3 )
(2)3 2E1 (2)3 2E2 (2)3 2E3
Z 3~
d k1 1 d3~k2 1 d3~k3 1
(2)4 (Eq E1 E2 E3 ) (3) (~q ~k1 ~k2 ~k3 )
=
(2)3 2E1 (2)3 2E2 (2)3 2E3
Integrating out the k3 integral with spatial delta function, we essentially restricts ~k3 = ~k1 + ~k3 ,
as ~q = 0 in the rest frame of virtual photon,
Z 3~
d k1 1 d3~k2 1 1
=
(2)(Eq E1 E2 E3 )
(8)
(2)3 2E1 (2)3 2E2 2E3
Let us change the integration measure by going to spherical polar coordinates as,
2
2
d3~k1 d3~k2 = ~k1 dk1 d1~k2 dk2 d12
2
2
= 4~k1 dk1~k2 dk2 d12 d cos 12
2
2
= 8 2~k1 dk1~k2 dk2 d cos 12

Substituting in Eq.(8),
Z
2k12 dk1 k22 dk2 d cos 12 1 1 1
=
(Eq E1 E2 E3 )
(2)3
2E1 2E2 2E3
2

Now using the relation E 2 = ~k + m2 we have for m1 = 0 = m2 and m3 = ,

E1 dE1 = |~k1 |dk1 ,

E2 dE2 = |~k2 |dk2 ,

E32 2 = (~k1 + ~k2 )2 ,

(9)

Squaring the relation ~k3 = ~k1 + ~k2 implies,

2
2
E32 2 = ~k1 + ~k2 + 2|~k1 ||~k2 | cos 12
2
2

E32 ~k1 ~k2 2 
E3
(Eq E1 E2 E3 ) =
cos 12
|~k1 ||~k2 |
2|~k1 ||~k2 |

We now perform cos 12 integration using identities from Eq.(9) and Eq.(10),
2
2
Z
E32 ~k1 ~k2 2 
dE1 dE2 d cos 12 
=
cos 12
4(2)3
2|~k1 ||~k2 |
Z
dE1 dE2
=
32 3
Using Eq.(6) we get,
3

(10)

Z
d3

q2
=
128 3

Z
dx1 dx2

(11)

(iv) R EGION OF INTEGRATION :

In our problem, m1 = m2 = 0 and m3 = , thus we have two extremal cases:

~k1 k ~k2

~k1 k ~k2

First in the parallel case,

Eq = E1 + E2 + E3
p
= E1 + E2 + (E1 + E2 )2 + 2
that yields,
p
Eq2 = (E1 + E2 )2 + 2(E1 + E2 ) (E1 + E2 )2 + 2 ) + (E1 + E2 )2 + 2


p
= 2(E1 + E2 ) E1 + E2 + (E1 + E2 )2 + 2 + 2
= 2(E1 + E2 )Eq + 2

(12)

Secondly in the antiparallel case,

Eq = E1 + E2 + E3
p
= E1 + E2 + (E1 E2 )2 + 2
that gives,
(Eq E1 E2 )2 = (E1 E2 )2 + 2
Eq2 + (E1 + E2 )2 2Eq (E1 + E2 ) = (E1 E2 )2 + 2
Eq2 + 4E1 E2 2Eq (E1 + E2 ) = 2
(Eq 2E1 )(Eq 2E2 ) = 2

(13)

The region of integration is bounded by the two curves of Eq.(12) and Eq.(13) that can be rewritten
using Eq.(6) as follows:
x1 + x 2 = 1

2
,
q2

(1 x1 )(1 x2 ) =

2
q2

(14)

q
e

k1

k1

e+

e+

q
k2

k2
k3

k3

Figure 1: e e+ qqg

Figure 2: e e+ q qg

k1

k1

k1 + k3

k1 + k3
k3

k3

k1 k3

k2

Figure 3: 3-particle unitarity

k1 k3

k2

Figure 4: 3-particle unitarity

(b) 3- PARTICLE UNITARITY

We start with the ansatz such that the amplitude for the Fig.(3) and Fig.(4) is given by:
Z
d3
}|
{
z
Z
Z 4 Z 4
4
X
d k1
d k2
d k3
2 Im =
(2)(k12 )(2)(k22 )(2)(k32 2 )(2)4 (4) (q
ki )
4
4
4
(2)
(2)
(2)
i
#
"
/
/
/
/
i(
k
+
k
)
i(
k
+
k
)
3
3
(ig T a )k/1 (ig T b ) 1
(g ab )
(1) Tr (ieq )(k/2 )(ieq ) 1
2
2
(k1 + k3 )
(k1 + k3 )
+

& k1 k2 )

(15)

Utilizing Eq.(11) and separating the trace,

2 Im =

"

k/ + k/3
k/ + k/3
d3 (1)3 i6 e2q g 2 Tr k/2 1
k/1 1
2
(k1 + k3 )
(k1 + k3 )2

& k1 k2 )

h
i
Tr T a T b
| {z }
4
3( )
3
(16)

k1 + k3

k1

k1

k2

k3

k1 + k3

k3
k2 k3

k2 k3

Figure 5: 3-particle unitarity

k2

Figure 6: 3-particle unitarity

The amplitude for the Fig.(5) and Fig.(6) is given by,

"
Z
2 Im =

i(k/2 k/3 )
i(k/ + k/3 )
d3 (1) Tr (ieq )(k/2 )(ig T a )
(ieq )k/1 (ig T b ) 1
2
(k2 + k3 )
(k1 + k3 )2

(g ab ) +

& k1 k2 )

(17)

Simplifying,
2 Im =

"

k/ + k/3
k/ + k/3
d3 e2q g 2 Tr k/2 2
k/1 1
2
(k2 + k3 )
(k1 + k3 )2

& k1 k2 )

h
i
Tr T a T b
| {z }
4
3( )
3
(18)

The final result for the differential cross section for 0 is,


8e2 g 2 x21 + x22 s2

=
x1 x2
3q 2 (x1 1) (x2 1)
(c) The integral result upto O
Z

2
1 2
q

2
2 x2 +1
q
1x2

!
is:

!


2 (1 x1 )2 + (1 x2 )2
x21 + x22

dx1 dx2
(1 x1 )(1 x2 )
q 2 (1 x1 )2 (1 x2 )2
!
!
2
2

2
+
3
log

+5
= log2
q2
q2
3

2 x2 +1

2
q2

(19)

(20)

(d) 2- PARTICLE UNITARITY V ERTEX C ORRECTION

k1

k1

k
k2

k
k2

Figure 7: 2-particle unitarity

Figure 8: 2-particle unitarity

The amplitude for Fig.7 and Fig.8 is given as,

Z
1
d2 =
8
zZ
}|
{
Z
4
4
d
k
d
k
1
2
2 Im
(2)(k12 )(2)(k22 )(2)4 (4) (q k1 k2 )
1+2 =
(2)4
(2)4
"
#
/
/
/
/
i(
k

k
)
i(
k

k
)
ig ab
2
1

b
/
(1) Tr (ieq )(k/2 )(ig T a )
(ie

)
(ig
T
)(
k
)
(
)
q
1
(k2 + k)2
(k1 k)2
k 2 2
+

( )

(21)

Let us define,

ie (k1 , k2 ) =

i(k/2 k/) i(k/1 k/)

i
d4 k
(ig )
ie
(ig ) 2
Tr[T a T b ] (22)
4
2
2
(2)
(k2 + k)
(k1 k)
k 2

ie = e[ [F1 (q 2 )] + i
7

q F2 (q 2 )]
2m

(23)

But,
2

dxdydz (1 x y z)

(3 d/2) q 2 (1 x)(1 y) x
+
(4)d/2
3d/2

F1 [q ] = Tr[T T ]g

(2 d/2) (2 )2
(4)d/2 22d/2
(24)

where
= 22 xyq 2

(25)

Thus Eq.(21) becomes,

e2 q 2
Tr[ k/2 k/1 ]F1 [q 2 ] + ( )
8


e2 q 2

2
=
F1 [q ] k2 k1 g (k1 k2 ) + k1 k2

2 Im
1+2 =

(26)

Contracting with g , we get,

6 Im (q 2 ) =

2e2
F1 [q 2 ](k1 k2 )

(27)

2- PARTICLE UNITARITY S ELF E NERGY

k1

k2

Figure 9: 2-particle unitarity

The amplitude for Fig.(9) is,
Z 4 Z 4 Z
d k1
d k2
d4 k

2 Im
(2)(k12 m2 )(2)(k22 m2 )(1)
=
(2)4
(2)4
(2)4
"
#
/
/
/
i(
k
+
m)
i(
k

k
+
m)
1
(ig T a )
(ig T b )(k/1 + m)(ieq )
Tr (k/2 + m)(ieq ) 21
k1 m2
(k1 k)2 m2
(

ig ab
)
k 2 2

(28)

We define,
i2 (k/1 ) =

/1 k/ + m)
d4 k
iTr[T a T b ]
i(k
(ig
)
(ig
)

(2)4
(k1 k)2 m2
k 2 2
8

(29)

Using dimensional regularization,

Z 1
dx
i2 (k/1 ) = ig
0

(4 )m (2 )k/1 x
[2 d/2]
2
2
2d/2
2
[(1 x)m x(1 x)k1 + x]
(4)d/2

(30)

Therefore Eq.(28) becomes,

2 Im =
=
=

"

i(k/ + m)
i2 (k/1 ) i(k/1 + m)
e2 d2 Tr (k/2 + m) 21
2
k1 m
#
"


/
i 3 e2
k

Tr k/2 12 i2 (k/1 ) k/1

8
k1
"
#
/
k
i4 e2 g 2 (1)Tr k/2 21 (2 ) k/1 k/1
k1
Z
(2 d/2)
1
dx
d/2
2
(4)
[(1 x)m x(1 x)k12 + 2 x]2d/2
Z

(2 d/2)(2 ) 

/
/
/
/
e2 g 2 dx
Tr
k

k
k
k

2
1 1 1
(4)d/2 (x)2d/2
|
{z
}
Z



= CTr k/2 k/1 k/1 k/1

(31)

Simplifying, we have
2 Im = 4Ck12 (k1 k2 )g + k1 k2 + k2 k1

(32)

Contracting with g both sides,

6q 2 Im (q 2 ) = 16C(k1 k2 )

(33)

Let us determine C as,

2

(2 d/2)(2 )
(4)d/2 (x)2d/2
0


Z 1
2
2 2
2
dx
= 16e g (k1 k2 )
log (x ) (2 )

0
#
"
Z 1
2
2
2 2
log (x )dx (2 )

= 16e g (k1 k2 )

0
2 2

6q Im (q ) = 16e g (k1 k2 )

dx

where
Z

1
2

log(x )dx = log +

0

dxd

(34)

Therefore,
6q 2 Im (q 2 ) = 16(k1 k2 )(2 + 2 log 2 1)

(35)

The other diagrams also give the similar result.

(e) Thence, the total cross section is:

(e+ e q qg) =

2
4 2 s 10 2
(3eq )

+ 2 log 2 +
2
3q

3
9
q
3

10

log

q2

!2

(36)