The phase diagram of neutral quark matter: Self-consistent treatment of quark masses

The thing worth knowing: this is the neutral three-flavor NJL phase diagram with the constituent masses solved self-consistently alongside the gaps and the three charge chemical potentials, rather than inserted by hand as in Refs. [7, 13, 14]. That is a real step, and it buys a concrete correction — the 2SC window that Ref. [3] put around µ ≈ 450 MeV at T = 0 is not there once the gCFL solution is included, and Fig. 7 makes the case cleanly by plotting the pressures side by side. The bookkeeping is transparent: nine coupled equations, eight candidate phases, the 36×36 block decomposition spelled out in Appendix A. I could follow the logic without having to trust a black box, which is more than I can say for a lot of this literature. The central claim (abstract:s4) is comparative and, as a statement about the mean-field diagram, it is right: bigger gap, Fermi surfaces smeared over a region of order ∆, thermal smearing adds nothing, so the gapless regions shrink. The explanation in Sec. III is sound and consistent with Ref. [5]. Where I would push back is on how sharply the boundaries are drawn. Several of the lines that delimit the gapless regions are not free-energy transitions at all — the g2SC→2SC and gCFL→CFL boundaries are explicitly crossovers, and the dashed lines 'do not correspond to phase transitions.' The extent of a gapless region is then a spectral criterion, and nothing protects its dependence on G_D. The genuine first-order boundaries are located by pressure differences of order 10⁻⁴ µ⁴ (Fig. 7) with a sharp cutoff at Λ = 602.3 MeV and no regularization study. Since ∆ differs by roughly 50% between the two panels (76 vs 120 MeV), cutoff artifacts and the neglected σ|∆|² 't Hooft terms act asymmetrically between Figs. 1 and 2, and the T = 0 gCFL window is only 14 MeV wide. The authors flag the chromomagnetic instability of some gapless phases in their closing paragraph; that caveat is honest, and if anything it favors their qualitative message rather than undermining it. Who should read it: anyone building a compact-star EOS with a quark core, and anyone who wants a reference neutral three-flavor diagram to compare against. It deserves referees — the questions to put to them are the regularization dependence and whether the gapless boundaries survive a smoother treatment.

The claim of abstract:s4 is comparative and model-level: between GD = 3/4 GS (Fig. 1) and GD = GS (Fig. 2) the gapless regions shrink. As a statement about the mean-field phase diagram this is well posed, and the physical explanation in Sec. III (a larger gap smears the Fermi surface over a region of order ∆, so additional thermal smearing adds nothing) is sound and consistent with Ref. [5]. The reader's instability concern is acknowledged by the authors in the closing paragraph and, if anything, favors the paper's qualitative message (fewer unstable gapless phases at strong coupling); it does not attack the comparison itself. What is less secure is the quantitative extent of the gapless regions, on three specific counts. (i) Several boundaries that delimit them are crossovers, not free-energy transitions: the g2SC→2SC boundary at µ ≳ 445 MeV and the gCFL→CFL boundary at T ≠ 0 are explicitly smooth crossovers, and the dashed gapless lines 'do not correspond to phase transitions.' The extent of a gapless region is thus a spectral criterion (existence of a zero mode), not a thermodynamic quantity, and nothing protects its dependence on GD. (ii) The genuine first-order boundaries are located by pressure differences of order 10⁻⁴ µ⁴ (Fig. 7), computed with a sharp three-momentum cutoff at Λ = 602.3 MeV in Eq. (12); no variation of Λ or of the regularization scheme is reported. (iii) The gap differs by roughly 50% between the two panels (∆⁽⁵⁰⁰⁾ ≈ 76 vs 120 MeV, Sec. IV), so cutoff artifacts scaling as ∆²/Λ² and the neglected σα|∆c|² 't Hooft terms (Sec. II: 'we neglect such terms for simplicity') act asymmetrically between Figs. 1 and 2. The T = 0 gCFL window is only 443–457 MeV wide, comparable to plausible shifts from either effect.

The reader's weakest_assumption targets the chromomagnetic instability of gapless phases, which the authors themselves flag in the closing paragraph ('some regions of the phase diagrams in Figs. 1 and 2 correspond to phases that are known to be unstable [28]'). That is a real limitation, but it is acknowledged, and it does not land on the central claim: abstract:s4 is a comparison between two mean-field diagrams computed with identical parameters except GD, so the instability affects both panels and, if anything, reinforces the paper's qualitative message that strong coupling leaves fewer gapless phases. I therefore agree only partially with the reader's framing. The reader's secondary point, the absence of a systematic uncertainty estimate given the cutoff choice, is closer to the load-bearing issue, and I would make it primary. The paper's genuinely strong elements deserve credit: the self-consistent treatment of Mu, Md, Ms (abstract:s1) is a real advance over the shift approximation of Ref. [7]; the correction to Ref. [3] about the 2SC window is backed by the explicit pressure comparison in Fig. 7; and the two-coupling comparison is a controlled one-parameter variation with a clear physical mechanism. None of that is in doubt. What is not established is how much of the T–µ plane the gapless regions occupy, because those extents are set partly by crossovers defined through a spectral criterion and partly by first-order boundaries located by pressure differences of order 10⁻⁴ µ⁴ under a sharp cutoff whose artifacts scale with ∆² and therefore differ between the two panels. A secondary check worth running alongside the one I propose is to compute at GD = 0.8 GS and 0.9 GS to confirm that the gapless region shrinks monotonically rather than non-monotonically between the two published endpoints; the paper's physical argument makes monotonicity plausible but does not demonstrate it. I would keep the reader's CONDITIONAL verdict, re-anchored on regularization and on the softness of the gapless-region definition rather than on the acknowledged instability, and I would not raise the verdict to ACCEPT until the boundary locations are shown to be stable under a change of cutoff or regulator.

The calculation is a forward computation, not a fit: the parameter set in Eq. (2) is fixed by vacuum observables (mπ, mK, mη′, fπ) from Ref. [19], and the phase diagram then follows. That is the right structure for a prediction, and the self-consistent treatment of the quark masses is a genuine advance over Refs. [7, 13, 14], where masses entered as free parameters. The qualitative message — that stronger diquark coupling makes pairing robust against the Fermi-surface mismatch forced by electric and color neutrality, so that thermal smearing is not needed to stabilize a color superconductor — is physically sensible and consistent with the two-flavor analysis of Ref. [5]. The explicit resolution of the 2SC-versus-gCFL competition at T = 0 in Fig. 7, which corrects the earlier 2SC window of Ref. [3], is a useful and well-presented check. The paper is also candid about what it does not do: mixed 't Hooft terms are dropped, mixed and crystalline phases are excluded, and the closing paragraph acknowledges that some regions of Figs. 1 and 2 are known to be unstable. If the comparative claim survives the robustness checks requested below, the result is a worthwhile contribution to the NJL

The paper asks how the temperature–chemical-potential phase diagram of dense, locally neutral three-flavor quark matter changes when quark masses are generated self-consistently rather than inserted by hand, and how that diagram depends on the strength of diquark pairing. Working in a Nambu–Jona-Lasinio model with scalar, 't Hooft, and diquark interactions, the authors solve the coupled gap and neutrality equations and compare the pressure of eight candidate phases. The central comparative result is that gapless phases play a smaller role as the diquark coupling is raised: at $G_D=3G_S/4$ a broad gapless 2SC region sits between the low- and high-temperature normal phases, while at $G_D=G_S$ that region is absent and the gapless CFL window is reduced. The physical reason given is that stronger pairing has a larger condensation energy, so it survives a larger Fermi-surface mismatch without needing thermal smearing. If correct, this says neutron-star matter has less room for gapless color-superconducting phases than phase diagrams with weaker pairing suggested.

What carries the argument. The argument is carried by a mean-field Nambu–Jona-Lasinio model for three quark flavors with scalar and pseudoscalar quark-antiquark interactions, a 't Hooft six-fermion term, and a scalar diquark interaction in the color- and flavor-antitriplet channel. The authors derive a thermodynamic potential in the Nambu-Gorkov basis, obtain the quasiparticle eigenvalues $epsilon_i(k)$, and solve six gap equations for the chiral condensates $sigma_alpha$ and diquark gaps $Delta_c$ together with three local-neutrality conditions, $n_Q=n_3=n_8=0$, which fix $mu_Q$, $mu_3$, and $mu_8$. They compare the pressure among normal, 2SC, uSC, dSC, sSC, and CFL-type solutions and then inspect the quasiparticle spectrum for zero-energy modes; a phase is called gapless when such a mode exists even though the corresponding diquark gap is nonzero. This machinery is what makes the coupling comparison quantitative rather than a symmetry argument.

Core claim. The paper's central claim is that the role of gapless phases diminishes with increasing diquark coupling strength. In the intermediate-coupling case, $G_D=3G_S/4$, the authors find a large gapless 2SC (g2SC) region at intermediate $T$ and $mu$, plus a gapless CFL (gCFL) window at $T=0$ between the normal quark phase and the regular CFL phase. In the strong-coupling case, $G_D=G_S$, the g2SC region is gone and the gCFL window is smaller; the phase boundaries shift to lower $mu$ and higher $T$. The authors explain the trend by condensation energy: at stronger coupling, pairing is favorable even with a larger mismatch of Fermi surfaces, so extra thermal smearing does not open a new gapless window. A secondary claim is that treating the dynamical quark masses self-consistently changes the competition between phases and can turn boundaries first order, so fixed-mass phase diagrams can misplace phases.

The premise it rests on. The load-bearing premise is that the mean-field gapless phases used in the comparison are physical neutral ground states; the authors themselves note in closing that some regions of their phase diagrams correspond to phases known to be unstable.

At strong diquark coupling the phase diagram loses the broad g2SC window between the low- and high-temperature normal phases, so the sequence of phases with increasing $mu$ is simpler than at intermediate coupling.

Increasing $G_D$ shifts color-superconducting boundaries to lower $mu$ and higher $T$, meaning pairing sets in earlier in dense matter.

Self-consistent quark masses can make phase boundaries first order and can make some regions of mass-parameter space inaccessible, so fixed-mass phase diagrams are not a reliable guide to where phases sit.

The zero-temperature gCFL window between normal quark matter and CFL matter is present at intermediate coupling but was not resolved in an earlier fixed-mass study, so the self-consistent treatment changes which phases appear.

For neutron-star cores, the region where gapless color-superconducting phases could occur is narrower when pairing is strong.

If the gapless phases are indeed unstable, the physically relevant phase diagram may be set by how the system resolves the instability—through mixed, crystalline, or different pairing phases—rather than by the mean-field gapless regions; the coupling trend could survive only as a statement about metastable regions.

A two-flavor coupling scan would be a clean test: the g2SC region should shrink and disappear as the diquark coupling approaches the quark-antiquark coupling, giving a benchmark for the three-flavor result.

Neutrino trapping, which the authors leave for later work, would shift the electron chemical potential and hence the Fermi-surface mismatch; the strong-coupling result predicts that the phase diagram should be less sensitive to that shift than the intermediate-coupling one.

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review