Higgs Boson Physics, Part I

Laura Reina

TASI 2004, Boulder

 Outline of Part I

• Understanding the Electroweak Symmetry Breaking as a first step

towards a more fundamental theory of particle physics.

• The Higgs mechanism and the breaking of the Electroweak Symmetry

in the Standard Model.
−→ Toy model: breaking of an abelian gauge symmetry.
−→ Quantum effects in spontaneously broken gauge theories.
−→ The Standard Model: breaking of the SU (2)L × U (1)Y symmetry.
−→ Fermion masses through Yukawa-like couplings to the Higgs field.

• First step: calculate the SM Higgs boson decay branching ratios.

 Some References for Part I

• Spontaneous Symmetry Breaking of global and local

symmetries:
. An Introduction to Quantum Field Theory,
M.E. Peskin and D.V. Schroeder
. The Quantum Theory of Fields, V. II, S. Weinberg

• Theory and Phenomenology of the Higgs boson(s):

. The Higgs Hunter Guide,
J. Gunion, H.E. Haber, G. Kane, and S. Dawson
. Introduction to the physics of Higgs bosons,
S. Dawson, TASI Lectures 1994, hep-ph/9411325
. Introduction to electroweak symmetry breaking,
S. Dawson, hep-ph/9901280
. Higgs Boson Theory and Phenomenology,
M. Carena and H.E. Haber, hep-ph/0208209
Breaking the Electroweak Symmetry: Why and How?

• The gauge symmetry of the Standard Model (SM) forbids

gauge boson mass terms, but:
MW ± = 80.426 ± 0.034 GeV and MZ = 91.1875 ± 0.0021 GeV

Electroweak Symmetry Breaking (EWSB)

• Broad spectrum of ideas proposed to explain the EWSB:

. Weakly coupled dynamics embedded into some more fundamental
theory at a scale Λ (probably ' TeV):
=⇒ Higgs Mechanism in the SM or its extensions (MSSM, etc.)
−→ Little Higgs models
. Strongly coupled dynamics at the TeV scale:
−→ Technicolor in its multiple realizations.
. Extra dimensions beyond the 3+1 space-time dimensions
 Different but related .....

• Explicit fermion mass terms also violate the gauge symmetry

of the SM:
−→ introduced through new gauge invariant interactions, as dictated
by the mechanism of EWSB
−→ intimately related to flavor mixing and the origin of CP-violation:
new experimental evidence on this side will give further insight.
 The story begins in 1964 . . .
with Englert and Brout; Higgs; Hagen, Guralnik and Kibble
 Spontaneous Breaking of a Gauge Symmetry

Abelian Higgs mechanism: one vector field Aµ (x) and one complex
scalar field φ(x):
L = LA + Lφ

where
1 µν 1 µ ν
LA = − F Fµν = − (∂ A − ∂ ν Aµ )(∂µ Aν − ∂ν Aµ )
4 4
and (Dµ = ∂ µ + igAµ )

Lφ = (Dµ φ)∗ Dµ φ − V (φ) = (Dµ φ)∗ Dµ φ − µ2 φ∗ φ − λ(φ∗ φ)2

L invariant under local phase transformation, or local U (1) symmetry:

φ(x) → eiα(x) φ(x)

1
Aµ (x) → Aµ (x) + ∂ µ α(x)
g
Mass term for Aµ breaks the U (1) gauge invariance.
 Can we build a gauge invariant massive theory? Yes.

Consider the potential of the scalar field:

V (φ) = µ2 φ∗ φ + λ(φ∗ φ)2

where λ > 0 (to be bounded from below), and observe that:

250000 300000
200000 200000
150000
100000
100000
50000 0
0
–10 –10 –15 –15
–10 –10
–5 –5 –5 –5
phi_2 0 0phi_1 phi_2 0 0phi_1
5 5 5 5
10 10
10 10 15 15

µ2 > 0 → unique minimum: µ2 < 0 → degeneracy of minima:

2
−µ
φ∗ φ = 0 φ∗ φ =
2λ
 • µ2 > 0 −→ electrodynamics of a massless photon and a massive scalar
field of mass µ (g = −e).

• µ2 < 0 −→ when we choose a minimum, the original U (1) symmetry

is spontaneously broken or hidden.

µ 2
¶1/2
µ v 1
φ0 = − = √ −→ φ(x) = φ0 + √ (φ1 (x) + iφ2 (x))
2λ 2 2
⇓
1 1 1 1
L = − F µν Fµν + g 2 v 2 Aµ Aµ + (∂ µ φ1 )2 + µ2 φ21 + (∂ µ φ2 )2 + gvAµ ∂ µ φ2 + . . .
| 4 {z 2 } |2 {z } |2 {z }
massive vector field massive scalar field Goldstone boson

Side remark: The φ2 field actually generates the correct transverse

structure for the mass term of the (now massive) Aµ field propagator:
µ µ ν
¶
−i k k
hAµ (k)Aν (−k)i = 2 2 g µν
− 2
+ ···
k − mA k
More convenient parameterization (unitary gauge):
χ(x)
i
e v U (1) 1
φ(x) = √ (v + H(x)) −→ √ (v + H(x))
2 2
The χ(x) degree of freedom (Goldstone boson) is rotated away using gauge
invariance, while the original Lagrangian becomes:
g2 v2 µ 1¡ µ 2 2
¢
L = LA + A Aµ + ∂ H∂µ H + 2µ H + . . .
2 2
which describes now the dynamics of a system made of:

• a massive vector field Aµ with m2A = g 2 v 2 ;

• a real scalar field H of mass m2H = −2µ2 = 2λv 2 : the Higgs field.

Total number of degrees of freedom is balanced

Non-Abelian Higgs mechanism: several vector fields Aaµ (x) and several
(real) scalar field φi (x):
1 µ 2 2 2 λ 4
L = LA + Lφ , Lφ = (D φ) − V (φ) , V (φ) = µ φ + φ
2 2
(µ2 < 0, λ > 0) invariant under a non-Abelian symmetry group G:
ta =iT a
a a
φi −→ (1 + iα t )ij φj −→ (1 − αa T a )ij φj

(s.t. Dµ = ∂µ + gAaµ T a ). In analogy to the Abelian case:

1 1
(Dµ φ)2 −→ . . . + g 2 (T a φ)i (T b φ)i Aaµ Abµ + . . .
2 2
φmin=φ0 1 2 a
−→ . . . + g (T φ0 )i (T b φ0 )i Aaµ Abµ + . . . =
2| {z }
m2ab

T a φ0 6= 0 −→ massive vector boson + (Goldstone boson)

T a φ0 = 0 −→ massless vector boson + massive scalar field
Classical −→ Quantum : V (φ) −→ Vef f (ϕcl )
The stable vacuum configurations of the theory are now determined by the
extrema of the Effective Potential:
1
Vef f (ϕcl ) = − Γef f [φcl ] , φcl = constant = ϕcl
VT
where
Z
4 δW [J]
Γef f [φcl ] = W [J] − d yJ(y)φcl (y) , φcl (x) = = h0|φ(x)|0iJ
δJ(x)
W [J] −→ generating functional of connected correlation functions
Γef f [φcl ] −→ generating functional of 1PI connected correlation functions

Vef f (ϕcl ) can be organized as a loop expansion (expansion in h̄), s.t.:

Vef f (ϕcl ) = V (ϕcl ) + loop effects

SSB −→ non trivial vacuum configurations

Gauge fixing : the Rξ gauges. Consider the abelian case:
1 µν
L = − F Fµν + (Dµ φ)∗ Dµ φ − V (φ)
4
upon SSB:
1
φ(x) = √ ((v + φ1 (x)) + iφ2 (x))
2
⇓
1 µν 1 µ 1 µ
L = − F Fµν + (∂ φ1 + gA φ2 ) + (∂ φ2 − gAµ (v + φ1 ))2 − V (φ)
µ 2
4 2 2
Quantizing using the gauge fixing condition:
1
G= √ (∂µ Aµ + ξgvφ2 )
ξ
in the generating functional
Z ·Z µ ¶¸ µ ¶
1 δG
Z = C DADφ1 Dφ2 exp d4 x L − G2 det
2 δα
(α −→ gauge transformation parameter)
 µ µ ¶ ¶
1 1 1
L − G2 = − Aµ −g µν ∂ 2 + 1 − ∂ µ ∂ ν − (gv)2 g µν Aν
2 2 ξ
1 1 1 ξ
(∂µ φ1 )2 − m2φ1 φ21 + (∂µ φ2 )2 − (gv)2 φ22 + · · ·
2 2 2 2
+
· µ ¶¸
2 2 φ1
Lghost = c̄ −∂ − ξ(gv) 1 + c
v
such that:
µ µ ν
¶ µ µ ν
¶
−i k k −iξ k k
hAµ (k)Aν (−k)i = g µν
− +
k 2 − m2A k2 k 2 − ξm2A k2
−i
hφ1 (k)φ1 (−k)i =
k 2 − m2φ1
−i
hφ2 (k)φ2 (−k)i = hc(k)c̄(−k)i = 2
k − ξm2A

Goldtone boson φ2 , ⇐⇒ longitudinal gauge bosons

 The Higgs sector of the Standard Model :
SSB
SU (2)L × U (1)Y −→ U (1)Q
Introduce one complex scalar doublet of SU (2)L with Y = 1/2:
à !
φ+
φ= ←→ L = (Dµ φ)† Dµ φ − µ2 φ† φ − λ(φ† φ)2
φ0

where Dµ φ = (∂µ − igAaµ τ a − ig 0 Yφ Bµ ), (τ a = σ a /2, a = 1, 2, 3).

The SM symmetry is spontaneously broken when hφi is chosen to be (e.g.):
à ! µ ¶1/2
2
1 0 −µ
hφi = √ with v = (µ2 < 0, λ > 0)
2 v λ

The gauge boson mass terms arise from:

à !
µ † 1 ¡ a a 0
¢ ¡ bµ b 0 µ
¢ 0
(D φ) Dµ φ −→ · · · + (0 v) gAµ σ + g Bµ gA σ + g B + ···
8 v
1 v2 £ 2 1 2 2 2 2 3 0 2
¤
−→ · · · + g (Aµ ) + g (Aµ ) + (−gAµ + g Bµ ) + · · ·
2 4
And correspond to the weak gauge bosons:
1
Wµ± = √ (A1µ ± A2µ ) −→ MW = g v2
2
1 p
Zµ0 = p 3
(gAµ − g Bµ ) −→ MZ = g 2 + g 02 v2
0
g 2 + g 02

while the linear combination orthogonal to Zµ0 remains massless and

corresponds to the photon field:
1
Aµ p (g 0 A3µ + gBµ ) −→ MA = 0
g2 + g 02

Notice: using the definition of the weak mixing angle, θw :

g g0
cos θw = p , sin θw = p
2
g +g 02 g 2 + g 02

the W and Z masses are related by: MW = MZ cos θw

The scalar sector becomes more transparent in the unitary gauge:
i
à ! à !
χ
~ (x)·~
τ
ev 0 SU (2) 1 0
φ(x) = √ −→ φ(x) = √
2 v + H(x) 2 v + H(x)

after which the Lagrangian becomes

r
1 1 2 2 λ 1
L = µ2 H 2 − λvH 3 − H 4 = − MH H − MH H 3 − λH 4
4 2 2 4
Three degrees of freedom, the χa (x) Goldstone bosons, have been
reabsorbed into the longitudinal components of the Wµ± and Zµ0 weak
gauge bosons. One real scalar field remains:
2
the Higgs boson, H, with mass MH = −2µ2 = 2λv 2
and self-couplings:
H H H
MH2 MH2
H= −3i v = −3i v2
H H H
From (Dµ φ)† Dµ φ −→ Higgs-Gauge boson couplings:

Vµ Vµ H
MV2 µν MV2
H= 2i v g = 2i v2 g µν
Vν Vν H

Notice: The entire Higgs sector depends on only two parameters, e.g.
MH and v

v measured in µ-decay:
√ −1/2
−→ SM Higgs Physics depends on MH
v = ( 2GF ) = 246 GeV
Also: remember Higgs-gauge boson loop-induced couplings:

γ,Z γ,Z

H H

γ γ

They will be discussed in the context of Higgs boson decays.

Finally: Higgs boson couplings to quarks and leptons

The gauge symmetry of the SM also forbids fermion mass terms

(mQi QiL uiR , . . .), but all fermions are massive.

Fermion masses are generated via gauge invariant Yukawa couplings:

i c j ij i j j
LYukawa = −Γij
u LQ̄ φ u R − Γ d Q̄L φd R − Γ ij i
L̄ φl
e L R + h.c.

such that, upon spontaneous symmetry breaking:

à !
1 0
φ(x) = √ −→ mf = Γf √v2
2 v + H(x)

and
f
mf
H = −i v
=−iyt
f
 SM Higgs boson decay branching ratios
We can now calculate branching ratios and total width of the SM HIggs
boson:
 






















 












 









 

 







Observe difference between light and heavy Higgs
These curves include: tree level + QCD and EW loop corrections .
Tree level decays: H → f f¯ and H → V V
At lowest order:
G F MH
Γ(H → f f¯) = √ Ncf m2f βf3
4 2π
3
µ ¶
G F MH 3 2
Γ(H → V V ) = √ δV 1 − τV + τV βV
16 2π 4
√
(βi = 2
1 − τi , τi = 4m2i /MH , δW,Z = 2, 1, (Nc )l,q = 1, 3)

Ex.1: Higher order corrections to H → q q̄

QCD corrections dominant:
3GF MH 2
Γ(H → q q̄)QCD = √ m̄q (MH )βq3 [∆QCD + ∆t ]
4 2π

µ ¶2
αs (MH ) αs (MH )
∆QCD = 1 + 5.67 + (35.94 − 1.36NF )
π π
µ ¶2 · 2 2
¸
αs (MH ) 2 MH 1 2 m̄q (MH )
∆t = 1.57 − ln + ln
π 3 m2t 9 MH 2
Consist of both virtual and real corrections:
Q Q Q

H H g H
g
Q̄ Q̄ Q̄

Q Q Q
g
t
H g H g H
g
g
Q̄ Q̄
Q̄

Q Q Q
g g
H g H H g
g
g
Q̄ Q̄
Q̄
• Large Logs absorbed into M S quark mass
µ ¶ 2b
γ
0
αs (µ) 0
Leading Order : m̄Q (µ) = m̄Q (mQ )
αs (mQ )

f (αs (µ)/π)
Higher order : m̄Q (µ) = m̄Q (mQ )
f (αs (mQ )/π)
where (from renormalization group equation)
³ ´ 12
25
25
f (x) = x [1 + 1.014x + . . .] for mc < µ < mb
6
³ ´ 12
23 23
f (x) = x [1 + 1.175x + . . .] for mb < µ < mt
6
³ ´ 74
7
f (x) = x [1 + 1.398x + . . .] for µ > mt
2

• Large corrections, when MH À mQ

mb (mb ) ' 4.2 GeV −→ m̄b (Mh ' 100 GeV) ' 3 GeV

Branching ratio smaller by almost a factor 2.

• Main uncertainties: αs (MZ ), pole masses: mc (mc ), mb (mb ).
Ex. 2: Higher order corrections Γ(H → gg)
Start from tree level:
g

g
¯ ¯
GF αs2 MH
3 ¯X ¯
AH
¯ ¯
Γ(H → gg) = √ ¯ q (τ )
q ¯
36 2π 3 ¯
q
¯

2
where τq = 4m2q /MH and
3
AH
q (τ ) = τ [1 + (1 − τ )f (τ )]
2

 arcsin2 √1 τ ≥1
τ
f (τ ) = h √ i2
 − 1 ln 1+√1−τ − iπ τ <1
4 1− 1−τ

Main contribution from top quark −→ optimal situation to use

Low Energy Theorems to add higher order corrections.
QCD corrections dominant:
q
g g
t, b t, b q̄
H g t, b H g H

g g g

Difficult task since decay is already a loop effect.

However, full massive calculation of Γ(H → gg(q), qq̄g) agrees with
mt À MH result at 10%
· (NL )
¸
αs
Γ(H → gg(q), q q̄g) = ΓLO (αs(NL ) (MH )) 1 + E (NL )
π
2
MH ¿4m2
q 95 7
(NL )
E −→ − NL
4 6

Dominant soft/collinear radiation do not resolve the Higgs boson coupling

to gluons −→ QCD corrections are just a (big) rescaling factor
NLO QCD corrections almost 60 − 70% of LO result in the low mass region:

0.8
δ(H→gg) Γ = ΓLO(1+δ)
0.75 µ = MH
Mt = 175 GeV
0.7

0.65

0.6

0.55

0.5

0.45

0.4
50 100 200 500 1000
MH [GeV]

solid line −→ full massive NLO calculation

2
dashed line −→ heavy top limit (MH ¿ 4m2t )

NNLO corrections calculated in the heavy top limit: add 20%

−→ perturbative stabilization.
Low-energy theorems, in a nutshell.

• Observing that:
In the pH → 0 limit: the interactions of a Higgs boson with the SM
particles arise by substituting
µ ¶
H
Mi −→ Mi 1 + (i = f, W, Z)
v
In practice: Higgs taken on shell (p2H = MH
2
), and limit pH → 0 is limit of
2
small Higgs masses (e.g.: MH ¿ 4m2t ).

• Then
1X ∂
lim A(X → Y + H) = Mi A(X → Y )
pH →0 v i ∂Mi
very convenient!

• Equivalent to an Effective Theory described by:

αs aµν a H
Lef f = G Gµν (1 + O(αs ))
12π v
including higher order QCD corrections.
For completeness:
¯ ¯2
2
¯
3 ¯X
¯
G F α MH ¯ 2 H H
¯
Γ(H → γγ) = √ Ncf ef Af (τf ) + AW (τW )¯¯
128 2π 3 ¯ f
¯
¯

where (f (τ ) as in H → gg):

AH
f = 2τ [1 + (1 − τ )f (τ )]
AH
W (τ ) = − [2 + 3τ + 3τ (2 − τ )f (τ )]

¯ ¯2
µ ¯
2 3 ¯X
¶ ¯
2
G2F MW αMH3
MZ H H
¯
Γ(H → Zγ) = 1− 2 ¯ Af (τf , λf ) + AW (τW , λW )¯¯
64π 4 MH ¯
¯ f ¯

where the form factors AH H

f (τ, λ) and AW (τ, λ) can be found in the
literature (see, e.g., M. Spira, hep-ph/9705337).

For both decays, both QCD and EW corrections are very small (' 1 − 3%).
 Present theoretical accuracy on SM Higgs branching ratios
1
SM Higgs Branching Ratio bb

-1
10 + -
τ τ
gg
cc

-2 + -
10 WW

-3
γγ
10
100 110 120 130 140 150 160
M H (GeV)

Example: MH = 120 GeV

Decay mode: bb̄ WW∗ τ +τ − cc̄ gg γγ
Theory 1.4% 2.3% 2.3% 23% 5.7% 2.3%

Mainly due to: pole masses mc and mb , and αs (µ).