ABJM Theory

ABJM offers a concrete way to connect familiar supersymmetric matter multiplets with a theory that has more supersymmetry and strongly interacting quantum dynamics. Its standard gauge group is a pair of equal-size unitary groups with opposite Chern–Simons levels. The matrix size, the integer level, and the number of supercharges carry different information.

Start with What is ABJM, and what can we calculate?. It defines the four chiral multiplets, fixes the compact gauge group and trace, and calculates how the vector auxiliaries constrain the scalar fields. Its first checkpoint is to describe a particular member of the family without confusing gauge rank with supersymmetry or gauge-redundant components with physical states.

Prepare the ingredients

Three lessons provide the necessary entry:

1. Build a simple supersymmetric Abelian gauge theory introduces charged chiral multiplets, vector auxiliaries, and the difference between an action and its vacuum equations.
2. Matrix-valued matter, representations, and traces explains how two gauge groups can act on opposite sides of a matrix, and why the normalization of a trace matters.
3. The flat-space entrance to Chern–Simons and FI terms supplies the supersymmetric gauge coupling and its compact level convention.

This entrance needs no brane physics, localization, or curved-background construction. It extends the three-dimensional gauge-theory route; the independent four-dimensional entrances elsewhere on the site do not require it.

Choose the question you want to reach

Continue with Chern–Simons gauge fields with non-Abelian matter to construct the two-sided matter derivative, vary the opposite-level gauge action and track the compact Abelian combinations. That calculation raises further classical questions: which ordered interactions are allowed, what happens when the auxiliaries are eliminated, and which configurations solve the vacuum equations? It also raises quantum questions: how do magnetic flux sectors affect local operators, why are particular levels exceptional, and which observables can be evaluated exactly?

Read the two-node ABJM quiver makes the four matter representations visible. Follow a closed walk to construct a gauge-invariant quartic operator, test its ordering with explicit matrices, and see why neutral charges alone do not guarantee an invariant.

Build the quartic superpotential fixes the interaction coefficient and differentiates its ordered matrix traces. A noncommuting example shows why a vanishing superpotential need not be a supersymmetric vacuum.

From auxiliary fields to a sextic scalar potential adds the vector contribution and checks the complete positive potential. An F-flat matrix profile can still have nonzero energy; eliminating all auxiliaries explains why.

Symmetry, vacua and magnetic flux

Where does N=6 supersymmetry come from? organizes the fields into the full R-symmetry representations. Follow the argument using symmetries of the complete action, count the twelve real supercharges and test the interaction coefficient with a finite rotation.

One brane and the C4/Zk moduli space retains the compact flux lattice while dualizing the Abelian gauge fields. Derive the surviving discrete identification, calculate the flat local metric and see why a physical phase direction remains.

Many branes and global gauge-group data derives the joint diagonal vacuum quotient on a controlled patch. Permute whole four-component positions, identify collision strata, and test how a common-center quotient changes the allowed fluxes and residual identifications.

Monopole operators, flux, and matter dressing uses those compact fluxes to define local insertions. Derive their Chern–Simons gauge weights, construct a complete residual-group singlet, and distinguish the regulated bare R-charge from the dimension of a physical chiral operator.

Why levels one and two are special constructs the charged monopole scalars that belong to enhanced current multiplets. Follow the protected states to nonzero conserved descendants, check the enlarged R-symmetry algebra, and distinguish the level-one free sector from a change of global gauge group.

Sphere observables and deformations

Two existing preparations help with the eventual observable calculations. A finite-dimensional localization example teaches the deformation argument in an ordinary integral. Matrix integrals, roots, weights, and eigenvalues derives the eigenvalue measure and its normalization. These are useful continuations after the opening theory lesson, rather than prerequisites for naming ABJM’s fields.

For a curved-space application, put ABJM on a supersymmetric three-sphere. Specify its background and preserved charges, derive the scalar curvature mass, and retain both gauge factors and the complete superpotential. The auxiliary-field integration explains why its physical scalar potential and its later localization equations answer different questions.

Derive the sphere matrix integral carries that background into the localized observable. Derive its constant scalar locus, retain both Weyl quotients and every determinant factor, and give the remaining oscillatory integral a precise regulated meaning.

The first exact partition-function checks evaluates that integral at one color per node. Derive its Fourier constraint from a convergent regulator, retain the absolute normalization, and bound the error before removing the regulator.

Real masses and squashed spheres follows a definite flavor source on a toric ellipsoid. Its complete rigid background fixes the classical phase and determinant arguments; a rank-one mass Hessian tests the normalization.

Bosonic and fermionic loops

Bosonic supersymmetric Wilson loops derives the scalar coupling and preserved spinor projector before inserting a normalized character into that measure. Its rank-one calculation keeps the framing phase and proves the contour restriction needed for an exponentially growing insertion.

Fermionic loops, superconnections, and framing adds matter fermions to the loop connection. Derive the circle’s endpoint twist before comparing its expectation value with the two bosonic loops in one supersymmetric framing.

Large rank and spectral methods

Large N and what a ‘t Hooft limit holds fixed explains why holding the level fixed and holding the rank-to-level ratio fixed define different questions. Its finite Gaussian example separates an exact answer from an approximation with a controlled error.

Two different large-N limits compares fixed level with fixed rank-to-level ratio. A surviving correction shows why substituting one sequence into another leading formula needs more than algebra; a bounded saddle example separates numerical error from omitted physics.

Turn the matrix integral into a Fermi gas derives an exact positive operator at finite rank. Its exchange signs and spectral traces reproduce the first partition functions; the auxiliary particles and their occupation rules are defined before any large-rank approximation.

Derive the N to the three-halves free energy turns the positive operator into a high-energy state count. Fill its auxiliary levels, retain the phase-space measure, and compare the leading result with finite-rank quantum data.

The Airy function and the corrections it omits derives the contour, normalization and power corrections of the perturbative answer. Compare it with exact finite-rank data, then calculate the effect of a supplied exponential correction.

Connections and original papers

ABJM, M2 branes, and the M-theory/type-IIA regimes translates rank and level into radii and coupling. Check curvature and circle size separately, then evaluate the gravitational sphere action.

Reading the original ABJM paper connects these calculations to its action, vacuum and brane arguments. Restore the finite quotient’s circle patching and keep it distinct from a later dimensional reduction.

Use the localization companion to reconstruct the source measure and loop conventions, then the Fermi-gas companion to separate the exact change of variables from the large-rank approximation.

A first checkpoint

Take two $U(3)$ factors with levels $+5$ and $-5$. Write the representations of all four chiral multiplets. Find the total gauge rank, the number of real scalar entries before gauge constraints, and the numbers of full and manifest real Poincaré supercharges. Then explain why the vector multiplets are not extra Yang–Mills photons.

The opening lesson and its solved exercises provide the calculation. Completing it gives you a precise theory to study, while keeping its classical equations, quantum symmetry claims, and exact-observable methods distinct.