Symmetry Breaking
Symmetry Breaking
Symmetry Breaking
Abstract
1
1 Introduction
The concept of symmetry is one of the very few on which mathematicians
and physicists agree, namely that
SYMMETRY ≡ GROUPS .
In a sense, the first is all important, just like the main characters of a
play. The second is more like the supporting cast, without which the theory,
although it can stand on its own, is much less interesting and also much less
realistic.
The next question is: which groups does one use or need? Generally
speaking, finite-dimensional compact semi-simple Lie groups. In this talk, in
order to simplify the presentation but without losing the essentials, I shall
consider almost exclusively only the following: for abelian groups U(1), and
for nonabelian groups the unitary groups U(N) and SU(N). At the end I
shall mention an example where a discrete group figures.
2
• Vector bosons: γ (the photon), W + , W − , Z 0 .
• Quarks: these are not observable themselves, but they form most of
the other particles by combining two or three together. Each quark
q is in the 3-dimensional or fundamental representation, and directly
observable particles occur in the 1-dimensional or singlet representation
as follows:
qqq : 3 ⊗ 3 ⊗ 3 = 1 ⊕ · · ·
q q̄ : 3 ⊗ 3̄ = 1⊕···
Note that only singlets can be observed as free particles, as will be ex-
plained later.
These are directly measurable quantities and hence do not transform under
any symmetries. However, one can and does introduce a vector potential Aµ ,
related to Fµν by
Fµν = ∂ν Aµ − ∂µ Aν ,
so that there is a freedom in changing Aµ without affecting Fµν :
Aµ 7→ Aµ + ie∂µ Λ,
3
potential Aµ . However, in quantum theory, it was demonstrated that Fµν is
not enough to describe the physics and one needs Aµ . This is the famous
Bohm–Aharonov experiment.
The ‘gauge freedom’ in Aµ is in fact linked to the arbitrary phase of the
electron wave function:
ψ 7→ eieΛ ψ.
Hence the relevant group for the symmetry of electromagnetism is:
G = U(1)
where S ∈ G.
This is the famous Yang–Mills theory. In the last 20 years or so, it has
been generally accepted that Yang–Mills theory is the basis of all of particle
physics:
4
Physics Mathematics
Special Relativity Flat Space-time
General Relativity Riemannian Geometry
Quantum Mechanics Hilbert Space
Electromagnetism and Fibre Bundles
Yang-Mills Gauge Theory
φα,x : F → π −1 (x)
y 7→ φα (x, y),
with
φ−1
β,x φα,x : F → F
5
F G
vectors and tensors. Locally it combines with the usual partial derivative to
give the covariant derivative:
Dµ = ∂µ − ig[Aµ , ·].
In differential geometry and in gauge theory one has to replace the partial
derivative by the covariant derivative so as to preserve the invariance or
symmetry of the system.
From the connection one can define the curvature:
Fµν = ∂ν Aµ − ∂µ Aν + ig[Aµ , Aν ].
One recognizes immediately that these are respectively the gauge potential
and the gauge field introduced in the last section, where the extra commuta-
tors (in the Lie algebra) take into account that now the group is in general
nonabelian.
With this language, the mechanism of symmetry breaking can be stated
as the case when the twisting of the bundle are by elements of a subgroup H
of G, and when the connection 1-form takes values in the corresponding Lie
subalgebra. One says then that the bundle with connection is reducible to
the subgroup H. An important example is the ’t Hooft–Polyakov magnetic
6
monopole, which is a nontrivial U(1) reduction of a trivial SU(2) bundle,
given by the exact sequence (for those who are fond of such things):
The first and last terms being zero, one gets the isomorphism
π2 (SU(2)/U(1)) ∼
= π1 (U(1)).
7
Force Group Gauge bosons Matter
Strong SU(3) [Gluons] [Quarks]
(QCD)
Electroweak U(2) γ, W ± , Z 0 Leptons
(Weinberg–Salam) [Higgs]
8
Quarks Leptons
! !
u νe
uR dR eR
d L
e L
! !
c νµ
cR sR µR
s L
µ L
! !
t ντ
tR bR τR
b L
τ L
9
of such an object if present. This very significant link between a geometric
statement and a physical statement can be schematically represented as:
Poincaré Gauss
Aµ exists ⇐⇒ ∂µ ∗F µν = 0 ⇐⇒ no magnetic monopoles
| {z } | {z }
geometry physics
10
Applying ’t Hooft’s theorem to these symmetries lead to very interesting
consequences which I do not have time to talk about.
7 Conclusions
Let me summarize the salient points about symmetry in particle physics that
I have mentioned:
If, however, you wish to take away with you just one point, then I rec-
ommend:
SYMMETRY ≡ GROUPS .
Acknowledgements
I thank Bodil Branner and Sylvie Paycha for inviting me to this meeting,
and the British Branch of EWM for a travel grant.
References
There are many excellent textbooks and semi-popular books on modern par-
ticle physics which emphasize its symmetry properties. There are also excel-
lent articles in Scientific American which are most suitable to give a taste
of the beauty of the subject. Below is a random selection of such, the first
11
being a more general appreciation of symmetry in physics by the originator
of Yang–Mills theory:
For the reader who might want to know more about the last part of
this lecture, here are a few of my recent articles (the last with an amusing
application from the Serret–Frenet formulae for space curves):
2. Chan Hong-Mo and Tsou Sheung Tsun, Standard Model with Dual-
ity: Theoretical Basis, hep-th/9712171; Standard Model with Duality:
Physical Consequences, hep-ph/9712436; invited lectures at the Cra-
cow Summer School on Theoretical Physics, May–June 1997, Zakopane,
Acta Phys. Pol. B28 (1997) 3027–3040; 3041–3056.
3. José Bordes, Chan Hong-Mo, Jakov Pfaudler and Tsou Sheung Tsun,
Features of Quark and Lepton Mixing from Differential Geometry of
Curves on Surfaces, hep-ph/9802436, February 1998.
12