The Conformal Window and Interacting Fixed Points

Massless $SU(N_c)$ SQCD is proposed to flow to an interacting superconformal field theory when

$$\frac32N_c<N_f<3N_c.$$

Within this open interval, anomaly freedom and fixed-point supersymmetry determine exact R-charges and protected dimensions. The existence and uniqueness of the fixed point throughout the entire interval remain dynamical claims supported by Seiberg duality and many checks, not a theorem obtained from the beta function alone. The phase proposal and its magnetic description are developed in Seiberg 1995, §§2–4 and reviewed in Intriligator and Seiberg 1996, §§5.3–5.5.

Required background. SQCD infrared phases gives the phase proposal, ultraviolet and infrared fixed points gives the RG criteria, and Seiberg duality supplies the magnetic frame. Helpful background. Holomorphic and NSVZ running fixes the beta-function conventions.

The upper boundary from asymptotic freedom

The electric one-loop coefficient is

$$b_0=3N_c-N_f.$$

For $N_f<3N_c$, the gauge coupling grows toward the infrared. Write

$$N_f=3N_c-\epsilon, \qquad 0<\epsilon\ll N_c.$$

Then the infrared zero occurs at weak electric coupling in a controlled Banks–Zaks expansion. This anchors the existence of an interacting fixed point near the upper edge.

At $N_f=3N_c$, the leading coefficient vanishes and the proposed fixed point meets the free theory. For $N_f>3N_c$, the electric theory is infrared free and is not the same asymptotically free ultraviolet problem.

The exact fixed-point constraint

In the NSVZ convention, the gauge beta function has numerator

$$3N_c-N_f(1-\gamma_Q),$$

where charge conjugation gives $\gamma_Q=\gamma_{\widetilde Q}$. At an interacting fixed point with nonzero denominator,

$$\gamma_Q=1-\frac{3N_c}{N_f}.$$

The scaling dimension of a chiral field in a gauge-fixed description is

$$\Delta_Q=1+\frac12\gamma_Q =\frac32\left(1-\frac{N_c}{N_f}\right).$$

This agrees with the anomaly-free R-charge through $\Delta=3R/2$:

$$R(Q)=R(\widetilde Q)=1-\frac{N_c}{N_f}.$$

The gauge-variant field $Q$ is not itself a local gauge-invariant CFT operator, so its dimension should not be tested directly against the scalar unitarity bound. Gauge-invariant chiral composites must satisfy it.

Mesons, baryons, and the lower boundary

The meson has

$$R(M)=2\left(1-\frac{N_c}{N_f}\right), \qquad \Delta(M)=3\left(1-\frac{N_c}{N_f}\right).$$

A scalar chiral primary obeys $\Delta\ge1$, equivalently $R\ge2/3$. Therefore

$$\Delta(M)\ge1 \quad\Longleftrightarrow\quad N_f\ge\frac32N_c.$$

At the equality, $M$ reaches the free-field bound. Below it, the interacting formula would violate unitarity, so the meson must become free and an accidental symmetry appears. This agrees with the magnetic beta-function coefficient

$$\widetilde b_0 =3(N_f-N_c)-N_f =2N_f-3N_c.$$

The magnetic gauge theory is infrared free when $N_f<3N_c/2$. Just above the lower edge, it supplies a weakly coupled magnetic Banks–Zaks description.

Electric baryons have

$$R(B)=N_c\left(1-\frac{N_c}{N_f}\right), \qquad \Delta(B)=\frac{3N_c}{2} \left(1-\frac{N_c}{N_f}\right).$$

They do not set the generic lower edge for $N_c\ge2$; the meson reaches unitarity first. Small ranks and enhanced symmetries still require separate checks.

Protected central charges

For the candidate exact R-symmetry,

$$\operatorname{Tr}R=-N_c^2-1, \qquad \operatorname{Tr}R^3=N_c^2-1-\frac{2N_c^4}{N_f^2}.$$

The four-dimensional $\mathcal N=1$ anomaly formulas give

$$a=\frac{3}{16}\left( 2N_c^2-1-\frac{3N_c^4}{N_f^2} \right),$$

and

$$c=\frac{1}{16}\left( 7N_c^2-2-\frac{9N_c^4}{N_f^2} \right).$$

These are exact protected outputs conditional on the fixed point and R-symmetry identification. Computing them in magnetic variables, including the $N_f^2$ mesons, gives the same result. The nonperturbative relation between superconformal central functions and R-current anomalies is established in Anselmi, Freedman, Grisaru, and Johansen 1998, §§2–4.

The inequalities $a>0$, $c>0$, and conformal-collider bounds provide consistency checks but do not alone prove that the SCFT exists. The $a$-theorem also requires $a_{\mathrm{UV}}>a_{\mathrm{IR}}$ along a nontrivial flow; SQCD satisfies this in the proposed window.

Electric and magnetic control

Region	Useful frame	Controlled statement
$N_f=3N_c-\epsilon$	Electric	Weak fixed-point coupling; perturbation theory controls unprotected quantities order by order.
Interior	Neither generically weak	R-charges, anomalies, chiral data, and protected traces remain exact; generic correlators are strongly coupled.
$N_f=3N_c/2+\epsilon$	Magnetic	Weak magnetic gauge and Yukawa couplings; meson is near free.
$N_f<3N_c/2$	Magnetic infrared-free phase	The interacting-window formula for $M$ is invalid; accidental free fields must be explicit.

The two weak edges provide complementary perturbative anchors. Continuity plus duality supplies strong evidence for the interior, but an undetected fixed-point merger or other phase transition is a dynamical possibility that must be excluded by evidence, not by algebraic interpolation alone.

Scheme dependence and exact information

The NSVZ beta function is exact in a particular coupling convention. Analytic redefinitions change its denominator and the coordinate location of a zero. At a genuine fixed point, however, scaling dimensions, R-charges, anomalies, and critical exponents are physical after redundant directions are removed.

The numerator equation is useful because it is equivalent to the anomaly-free superconformal R condition. It does not prove that the denominator is nonsingular, that the zero is reached by the flow, or that no accidental symmetry changes $R$. Those are separate gates.

Endpoints and accidental fields

At $N_f=3N_c/2$, $M$ reaches $R=2/3$. A chiral primary saturating the bound obeys a free equation and contributes the free-field values

$$a_{\mathrm{chiral}}=\frac{1}{48}, \qquad c_{\mathrm{chiral}}=\frac{1}{24}.$$

If a composite crosses the bound in a more general theory, subtract its interacting trial contribution, add the free contribution, introduce the accidental current, and repeat a-maximization. Extrapolating the old anomaly formula through the crossing double-counts or misassigns that operator.

What is exact and what is inferred

Exact conditional statements: anomaly relations, $R(Q)=1-N_c/N_f$ for the anomaly-free candidate, $\Delta=3R/2$ for chiral primaries at an SCFT, the protected $a,c$ formulas, and the unitarity obstruction at $N_f<3N_c/2$.

Dynamical conclusions: existence of a unique interacting fixed point throughout the open window, absence of intermediate phases, and the complete unprotected spectrum.

Perturbatively controlled conclusions: detailed correlators and anomalous dimensions near either weakly coupled edge.

Keeping these categories separate makes the fixed-point data useful without overstating the phase diagram.

Common pitfalls

Applying the unitarity bound to $Q$. $Q$ is gauge variant. Test gauge-invariant chiral primaries such as $M$ and $B$.

Treating an NSVZ numerator zero as an existence proof. The flow, denominator, and accidental symmetries still require control.

Including endpoints in an open-window formula. At both edges the interacting description meets a free regime and requires separate limiting analysis.

Exercises

For $N_c=3$, $N_f=6$:

1. compute $R(Q)$, $\Delta(M)$, and $\Delta(B)$;
2. compute $a$ and $c$;
3. identify whether either electric or magnetic frame is parametrically weak.

Solution

$R(Q)=1/2$, so $\Delta(M)=3/2$ and $\Delta(B)=9/4$. The central charges are

$$a=\frac{3}{16}\left(18-1-\frac{243}{36}\right) =\frac{123}{64},$$ $$c=\frac{1}{16}\left(63-2-\frac{729}{36}\right) =\frac{163}{64}.$$

Here $N_f=2N_c$ lies in the interior of the window, not parametrically near either $3N_c$ or $3N_c/2$. The pair is self-rank—both gauge groups are $SU(3)$—but neither description is forced to be weakly coupled.

References

• Anselmi, Damiano, Daniel Z. Freedman, Marc T. Grisaru, and Andreas A. Johansen. “Nonperturbative Formulas for Central Functions of Supersymmetric Gauge Theories.” Nuclear Physics B 526 (1998): 543–571. arXiv:hep-th/9708042.
• Intriligator, Kenneth, and Nathan Seiberg. “Lectures on Supersymmetric Gauge Theories and Electric–Magnetic Duality.” Nuclear Physics B Proceedings Supplements 45BC (1996): 1–28. arXiv:hep-th/9509066.
• Seiberg, Nathan. “Electric–Magnetic Duality in Supersymmetric Non-Abelian Gauge Theories.” Nuclear Physics B 435 (1995): 129–146. arXiv:hep-th/9411149.