Controlled Regimes of Finite-Density QCD

Finite baryon density is not uniformly inaccessible, but the controlled regions are disconnected. Lattice expansions and imaginary-chemical-potential simulations constrain a neighborhood of $\mu_B=0$; nuclear effective theory controls sufficiently low-density hadronic matter; perturbative QCD controls asymptotically large chemical potential. The phenomenologically central intermediate region—including much of neutron-star-core parameter space—still lacks overlapping first-principles methods with demonstrably small, independently validated errors.

Evidence cutoff. 11 August 2026.

Required background. Chemical potential on the Euclidean lattice explains the complex measure; Euclidean and real-time access at finite density distinguishes observables and contours. Helpful background. Expansion and imaginary-density methods supplies the small-density tools, canonical and density-of-states methods supplies alternatives, and dense-QCD evidence boundaries organizes the physical regimes.

Controlled islands in the QCD phase diagram

Normative question. Where do independent analytic and numerical methods yield overlapping, systematically controlled information at finite baryon density?

The scope is equilibrium QCD with physical or explicitly stated quark masses and real baryon chemical potential. Isospin chemical potential, two-color QCD, heavy-dense limits, and sign-problem-free deformations are valuable benchmarks but are not direct solutions of physical finite-$\mu_B$ QCD.

For real baryon chemical potential the Euclidean fermion determinant is generally complex, so importance sampling loses its probability measure. This is an overlap and phase-cancellation problem, not merely a request for faster hardware. Generic sign problems are computationally hard in a precise complexity-theory sense (Troyer and Wiese 2005), though that theorem does not prove that every QCD observable is intractable.

Regime	Controlled information	Principal ceiling
$\mu_B/T\ll1$ near the crossover	Taylor coefficients, susceptibilities, crossover curvature, and imaginary-$\mu_B$ continuation with continuum-extrapolated lattice data	Radius of convergence, truncation order, finite volume, and analytic-continuation model
Low-temperature dilute nuclear matter	Chiral EFT and many-body expansions with order-by-order errors below their breakdown scale	Growth of many-body and regulator uncertainties with density
Asymptotically large $\mu_B$	Weak-coupling equation of state and color-superconducting pairing patterns	Convergence is slow at phenomenological densities; asymptotic ordering need not persist downward
Intermediate real baryon density	Functional methods, reweighting, canonical methods, complex Langevin, Lefschetz thimbles, and models provide constraints	No method currently supplies universal continuum control with independent overlap across the whole region

Imaginary chemical potential preserves a real measure and allows analytic continuation until the nearest singularity; de Forcrand and Philipsen 2002 established the method in the small-density phase-diagram setting. Taylor expansion at $\mu_B=0$ is mathematically equivalent information when all orders and exact data are available; at finite order their error models differ. Continuum-extrapolated lattice calculations now determine crossover curvature at small $\mu_B$ (Borsányi et al. 2020). Modern multipoint Padé and inverse-problem continuations can extend practical reach but remain sensitive to singularity assumptions (Di Renzo, Aliberti, and Dimopoulos 2025). Agreement among several ansätze is a stability check, not proof that a nearer complex-plane singularity is absent.

At asymptotically large density, asymptotic freedom supports weak-coupling calculations and attractive quark channels generate color superconductivity; Son derived the distinctive weak-coupling gap scaling (Son 1999). The result establishes the asymptotic regime, not the phase realized at neutron-star density. At lower density, chiral EFT and nuclear many-body calculations overlap around and moderately above saturation density, while high-density perturbative results constrain the far endpoint. Interpolating between them can bound equations of state when thermodynamic stability and astrophysical observations are imposed, but the interpolation is not a first-principles solution of the intervening QCD phase structure.

Competing approaches and negative results

Reweighting fails when ensemble overlap becomes exponentially small. Canonical and density-of-states methods reorganize rather than generically eliminate cancellations. Complex Langevin can converge to incorrect limits unless holomorphy, decay, and correctness criteria hold. Lefschetz-thimble decompositions expose relevant saddles but can require exponentially many thimbles. Functional equations reach real density directly but inherit truncation and closure dependence. A claimed critical point or pairing phase from any one of these methods is therefore conditional on its regulator, ansatz, and convergence tests.

Assessment. The question is partially resolved in separated regimes. Independent lattice strategies overlap near $\mu_B=0$; controlled hadronic EFT applies at sufficiently low density; perturbative QCD applies asymptotically. There is no verified corridor of overlapping, systematically improvable methods across intermediate physical baryon density. Claims about a critical point or the ground state there remain model- and method-dependent.

What would close the uncontrolled region?

A decisive advance would provide continuum-extrapolated results at real $\mu_B$ with a convergent error estimator and reproduce sign-free benchmarks, small-$\mu_B$ Taylor data, and asymptotic limits without retuning. Short of a universal algorithm, a genuine overlap window in which lattice continuation, functional methods, and an EFT or perturbative calculation agree on the same renormalized observables with shrinking independent errors would materially change the assessment.

Related routes include thermal and nonequilibrium field theory, lattice and Hamiltonian field theory, Euclidean lattice inference, and the QCD critical-point dossier.

Evidence cutoff and source selection

The finite set samples the sign-problem obstruction, small-density lattice continuation, low-density and asymptotic analytic regimes, and modern continuation diagnostics. Targeted arXiv, INSPIRE, lattice-collaboration, journal, and citation-chain searches covered public evidence through 11 August 2026. Sign-free cousins were included only as method benchmarks.

References

• Borsányi, Szabolcs, et al. “QCD Crossover at Finite Chemical Potential from Lattice Simulations.” Physical Review Letters 125 (2020): 052001. DOI.
• de Forcrand, Philippe, and Owe Philipsen. “The QCD Phase Diagram for Small Densities from Imaginary Chemical Potential.” Nuclear Physics B 642 (2002): 290–306. DOI.
• Di Renzo, Francesco, Marco Aliberti, and Petros Dimopoulos. “On Analytic Continuation from Imaginary to Real Chemical Potential in Lattice QCD.” arXiv:2502.03392 (2025). arXiv.
• Son, D. T. “Superconductivity by Long-Range Color Magnetic Interaction in High-Density Quark Matter.” Physical Review D 59 (1999): 094019. DOI.
• Troyer, Matthias, and Uwe-Jens Wiese. “Computational Complexity and Fundamental Limitations to Fermionic Quantum Monte Carlo Simulations.” Physical Review Letters 94 (2005): 170201. DOI.