How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Hidden gauge-sector relic and portal model classes
An illustrative four-dimensional hidden-gauge EFT family with declared portals and conditional production mechanisms. No relic abundance, string realization, or observational constraint is calculated.
Example
Let (X,g) be a declared smooth four-dimensional Lorentzian spacetime with signature (−,+,+,+), and let U⊆X be a coordinate patch and local trivialization of the hidden gauge bundle. The group is compact U(1)D. The gauge potential is a real C1 one-form AD=AμDdxμ on U, and FD=dAD is its C0 curvature with components FμνD=∂μAνD−∂νAμD; gD>0 is its dimensionless kinetic normalization. Let S,Φ∈C2(U,R) be real scalar fields of mass dimension one and singlets under U(1)D and the visible gauge group. Here Φ is called a modulus only if a separate compactification model identifies it as one. Let H:U→C2 be a C2 field in the visible Standard Model representation (1,2)Y=1/2, using Q=T3+Y, and a U(1)D singlet; thus H†H is a visible and hidden gauge singlet. FDμν:=gμαgνβFαβD.
Choose a vacuum invariant under the exact Z2 transformation S↦−S, with ⟨S⟩=0, while Φ, H and all gauge fields are even. Assume the full EFT below cutoff Λ preserves this symmetry and that S is the lightest Z2-odd excitation, with no lighter odd final state kinematically available. Include L⊃−4gD21FμνDFDμν−21gμν∂μS∂νS−21mS2S2−4!λSS4−2λHSS2H†H−2Λ2κS2FμνDFDμν−21μΦΦS2. Here mS>0, mΦ>0, and μΦ=0; λS,λHS,κ,gD are real dimensionless couplings, μΦ and Λ have mass dimension one, and FD has mass dimension two. The even scalar sector supplies a canonical kinetic term and a potential term mΦ2Φ2/2. The parameters mS,mΦ are potential coefficients, not automatically excitation masses. The remaining interactions are chosen so the full smooth scalar potential V is bounded below over the declared cutoff range; in particular take λS>0. The declared scalar range is contained in the open patch T0={(S,Φ,H):K(S)>0}, where K(S)=gD−2+2κS2/Λ2 is the full coefficient in −K(S)FD2/4. If κ≥0 this imposes no restriction on S; if κ<0 it requires ∣S∣<Λ/(gD−2κ). This patch contains the stipulated S=0 vacuum and is invariant under the exact Z2 and gauge actions.
For the decay branch, adopt weakly coupled, zero-temperature Minkowski-vacuum particle kinematics, neglecting curvature, medium and loop corrections. Fix constant vacuum values qˉ=(0,vΦ,H0) with vanishing gauge backgrounds and dV(qˉ)=0. Assume the canonically normalized fluctuations s=S and φ=Φ−vΦ are mass eigenmodes: their quadratic kinetic terms have unit normalization and no mixing, and the Hessian of the complete potential has no mixed quadratic terms involving either mode. Define their positive tree-level excitation masses by MS2:=∂S2V(qˉ) and MΦ2:=∂Φ2V(qˉ), with 0<MS,MΦ<Λ. For the displayed S-dependent potential terms alone, MS2=mS2+λHSH0†H0+μΦvΦ; any additional interactions contribute their own second derivatives. Define gφSS:=∂Φ∂S2V(qˉ), of mass dimension one. Assume the complete cubic interaction with one φ and two s fields at this order is −gφSSφs2/2, with gφSS=0. The displayed trilinear contributes μΦ to this coefficient; its nonzero value alone need not exclude cancellations from additional interactions. These are branch assumptions about the complete vacuum expansion, not consequences of mS,mΦ>0.
For a thermal production branch, take a spatially flat FLRW chart on It×Σ, where It⊆R is an open cosmic-time interval, Σ is a spatial three-manifold with Euclidean metric δij in the chart, a∈C2(It,(0,∞)), and g=−dt2+a(t)2δijdxidxj. Specify radiation-bath fields ρr,pr,Tr∈C1(It) with ρr,Tr>0 and pr=ρr/3; in this example “radiation dominated” means this bath is the dominant energy-density component on the stated interval. Its expansion rate is H(t)=a˙(t)/a(t). Let Γchem:It→[0,∞) be the number-changing interaction rate per S. If Γchem≫H for an interval so that chemical equilibrium is maintained and later becomes ≲H, call the conditional production branch freeze-out. If bath reactions produce S while chemical equilibrium is never attained, call it freeze-in. A distinct nonthermal branch is possible under the decay-branch assumptions above when MΦ>2MS: the even excitation φ then has an open φ→ss channel. Each branch needs its own rates, initial state and entropy history; none is calculated here. In a separate non-Abelian model, a confining sector means a specified gauge theory whose low-energy physical spectrum consists of gauge-invariant color-singlet states with a mass gap; a glueball is a color-singlet bound state made from its gauge fields. Its lightest state is a relic only if its portal-induced lifetime exceeds a stated cosmological interval. This illustrative EFT supplies no abundance, observed constraint or generic string-theory prediction.
Facts & Assumptions
Given: Natural units, the stated four-dimensional EFT cutoff, the (−,+,+,+) metric convention, positive full U(1)D gauge coefficient K(S) on the declared scalar patch, the field domains and visible/hidden representations in the Example, mS,mΦ>0, the unbroken exact Z2 vacuum with S the lightest odd excitation, a scalar potential bounded below over the cutoff range, the Higgs and dark-gauge portal operators, nonzero real μΦ, and a specified radiation/modulus-decay history. In the decay branch use the stated Minkowski-vacuum approximation, canonically normalized mass eigenmodes s,φ, complete-potential masses MS,MΦ and coupling gφSS=0. The fields S,H,Φ have mass dimension one; FD has dimension two; λS,λHS,κ,gD are dimensionless; and μΦ,Λ,mS,mΦ,MS,MΦ,gφSS have mass dimension one.
Four-dimensional hidden-sector gauge and matter data: the group, typed local gauge and matter fields, representations, positive gauge kinetic matrix, matter masses, portals, cutoff, thermal history and exact stability criterion are specified data; none follows from the label hidden sector.
Reheating temperature in a sudden-decay modulus model applies only to its separately specified smooth spatially flat FLRW model M=J×Σ, g=−dt2+a(t)2γ, homogeneous canonical quadratic scalar, and homogeneous smooth comoving radiation-fluid data (ρr,pr,n) with ρr≥0, pr=ρr/3 and n=∂t. These additive scalar-plus-fluid stresses are the only Einstein sources, with zero cosmological term. With H(td−)=ΓΦ, complete sudden transfer and immediate thermalization, its total-density matching gives the stated temperature relation. No averaged-dust limit is needed for that algebra. The hidden-sector EFT above is not automatically such a background: using this branch additionally requires this entire model datum and prescription. The theorem supplies no hidden-sector branching fraction or abundance.
Source-backed thermal treatments distinguish freeze-out after equilibrium from freeze-in without equilibration, and string-cosmology reviews describe nonthermal relic production through moduli decay, dark glueballs and model-dependent hidden-sector portals (Cline, arXiv:1807.08749v5, §§IV.2, IV.4, IV.7; Cicoli et al., arXiv:2303.04819v2, §§5.5, 5.7.1, 6.2.8).
Verification
The declared U(1)D is compact and has full kinetic coefficient K(S)=gD−2+2κS2/Λ2>0 on the declared patch, after combining the two displayed FD2 terms. All scalar fields are hidden singlets, so the hidden scalar action is trivial and preserves the metric, potential and K; the visible action on H also preserves this patch. The real scalar S is a singlet under both gauge groups, while H is a visible doublet and a hidden singlet, so S2H†H is gauge invariant; S2FD2/Λ2 has dimension six with the displayed inverse-square coefficient. Since the vacuum is Z2-invariant and every EFT operator below the cutoff preserves Z2, a decay of one odd S must have an odd final state. The lightest-odd and kinematic assumptions exclude such a final state, so the one-particle S excitation is stable within this EFT. These data instantiate [F1].
The rate conditions in the Example distinguish the freeze-out and freeze-in branches without assigning either a rate or yield. In the decay branch, expansion of the complete V at qˉ gives the stipulated masses and cubic coefficient. In particular, twice differentiating the displayed Higgs and modulus portals in S gives the shifts λHSH0†H0 and μΦvΦ, so the coefficient mS cannot replace MS in a threshold. In the adopted Minkowski approximation, choose an orthonormal basis with metric matrix diag(−1,1,1,1) and parent four-momentum (MΦ,0). The daughter momenta (MΦ/2,±p) conserve four-momentum and satisfy E2−∣p∣2=MS2 precisely when ∣p∣=MΦ2/4−MS2. Thus MΦ>2MS gives positive two-body phase space, and the nonzero complete cubic coefficient supplies the tree-level φss vertex; the channel is open under these branch assumptions. At equality the two-body phase space vanishes, and below threshold it is closed. Only when all of [F2]'s smooth flat-FLRW scalar-plus-comoving-radiation-fluid data and H(td−)=ΓΦ, complete sudden transfer and thermalization assumptions hold does it supply a conditional temperature after total-energy matching. No collision integral, branching fraction or Boltzmann solution is supplied, so none of these branches determines a relic abundance here.
A confining alternative would require a different non-Abelian group, confinement scale, bound-state spectrum and lifetime analysis. Those inputs are absent from this U(1)D model, so the statement treats glueballs only as a separate conditional class, not as a consequence of this example.