GIM Mechanism and FCNC Suppression

One of the electroweak theory's most incisive predictions is what it forbids: flavor-changing neutral currents (FCNCs) — processes like $s\to d$ or $b\to s$ mediated by the $Z$ or photon — are absent at tree level and strongly suppressed at loop level. The GIM mechanism (Glashow–Iliopoulos–Maiani, 1970) explains this suppression and predicted the charm quark before its discovery. This is the flavor companion to the gauge-boson-decay calculation.

Conventions: $\hslash =c=1$; $V$ = CKM matrix.

The FCNC problem

The neutral currents — the $Z$ and photon couplings — are flavor-diagonal in the SM: the $Z$ couples $\bar{u}u$, $\bar{d}d$, $\bar{s}s$, … but never $\bar{s}d$. This is not automatic; a generic theory of weak interactions would allow tree-level $s\to d$ transitions, predicting, e.g.,

$$\frac{\Gamma (K_L\to {\mu }^{+}{\mu }^{-})}{\Gamma (K^{+}\to {\mu }^{+}\nu )}\sim \mathcal{O}(1),$$

wildly at odds with the measured branching ratio $\sim 10^{-9}$. The near-total absence of FCNCs in nature is a sharp experimental fact any theory must reproduce.

Why neutral currents are flavor-diagonal

In the SM the tree-level flavor diagonality is a theorem, not an assumption. The fermion mass matrices are diagonalized by unitary rotations $V_L,V_R$ on the quark fields. The neutral- current coupling is proportional to the identity in flavor space (all up-type quarks have the same $Z$ charge), so under $q_L\to V_L{q}_L$ it transforms as $V_L^{†}1V_L=1$ — unchanged. The rotations that scramble flavor cancel. Only the charged current, which connects up- and down-type quarks, retains a non-trivial rotation — the CKM matrix $V_{\text{CKM}}=V_L^{u†}{V}_L^d$. Hence: charged currents change flavor (CKM), neutral currents do not.

The GIM mechanism at loop level

FCNCs can arise at one loop, via box and penguin diagrams (e.g. $s\to d$ through a $W$–quark loop). Naively each such loop is sizeable; the GIM mechanism is the cancellation that suppresses them. The amplitude sums over the internal up-type quarks $u,c,t$ with CKM weights, and by CKM unitarity ${\sum }_i{V}_{is}^{\ast }{V}_{id}=0$:

$$\mathcal{A}(s\to d)\propto \sum _{i=u,c,t}{V}_{is}^{\ast }{V}_{id}F(m_i^2).$$

If the internal quarks were degenerate ($m_u=m_c=m_t$), $F$ would factor out and the sum would vanish by unitarity. The residual amplitude is therefore proportional to mass splittings:

$$\boxed{\mathcal{A}(s\to d)\propto \sum _i{V}_{is}^{\ast }{V}_{id}\frac{m_i^2}{m_W^2},}$$

the GIM suppression — FCNCs are proportional to the (small) up-type mass-squared differences over $m_W^2$. This is why FCNC rates are tiny but nonzero, and why they are dominated by the heaviest internal quark (the top, for $b\to s$).

The charm prediction

In 1970 only three quarks ($u,d,s$) were known. With an odd number of quarks the GIM sum cannot pair up, and the neutral kaon decay $K_L\to {\mu }^{+}{\mu }^{-}$ and the $K^0$–${\bar{K}}^0$ mass splitting were predicted far larger than observed. GIM's resolution: postulate a fourth quark, the charm, so that the $u$ and $c$ contributions cancel in the GIM sum. The charm quark was discovered in 1974 (the $J/\psi$), a spectacular confirmation. The same logic — needing quarks in complete doublets for the cancellation — is echoed by anomaly cancellation requiring complete generations.

FCNC observables as new-physics probes

Because SM FCNCs are doubly suppressed (loop factor $\times$ GIM), they are among the most sensitive windows on new physics: any heavy particle running in the loop need not respect GIM and could dominate. Key FCNC observables:

Neutral-meson mixing also provided the pre-discovery estimate of the top mass (the box amplitude $\propto m_t^2$) and is a primary source of CP-violation measurements.

Summary

• FCNCs ($s\to d$, $b\to s$ via $Z/\gamma$) are absent at tree level because neutral currents are flavor-diagonal (unitary mass rotations cancel).
• GIM: loop-level FCNCs are suppressed by ${\sum }_i{V}_{is}^{\ast }{V}_{id}{m}_i^2/m_W^2$ — they vanish for degenerate quarks and scale with mass splittings.
• GIM predicted the charm quark (1970 → 1974) and makes FCNC observables premier BSM probes.

Where this fits

• The observables: electroweak observables §3–4 and Standard Model observables.
• The flavor structure: CKM mixing, Standard Model flavor and CP.
• The gauge-boson companion: Gauge-Boson Decays and Precision Electroweak.

References

• Glashow, Iliopoulos & Maiani, Phys. Rev. D 2, 1285 (1970).
• Peskin & Schroeder, An Introduction to Quantum Field Theory, Ch. 20.
• Branco, Lavoura & Silva, CP Violation.