Skip to content

Stability of the Second Kind

A uniform lower bound proportional to particle count for a specified many-particle energy across all admissible system sizes.

Version
v1 · 2026-10-17 · History
Domain-specific #
14310
Domain group
Natural Sciences
Origin domain
Physics
Subdomains
Statistical Mechanics, Many Body Theory → Physics

Core Idea

Stability of the second kind is an all-size lower-bound property of a specified many-particle model. The model declares its particles, admissible configurations or states, and an energy. One constant, independent of system size, must bound that energy below by a quantity proportional to particle count. A finite lower energy for each fixed-sized system is weaker: it does not supply a single linear bound that works as the system grows.[1]

The count and energy are model-relative. Fisher and Ruelle formulate a classical condition on the total configurational potential energy of N particles: it is at least −BN for all allowed configurations under their potential assumptions.[2] For Lieb's nonrelativistic quantum Coulomb model, N electrons and K nuclei are counted separately, and the ground-state energy of the full Hamiltonian is bounded below by C(N + K), with a negative constant C independent of N and K when the physical parameters are fixed.[1] The common abstraction is the quantified linear bound, not an identical energy functional or proof method.

Structural Signature

  1. Declared many-particle model class. Particle content, interactions, statistics and allowed configurations or states determine what the claim covers. Without this scope, one favorable system can be mistaken for a theorem about all system sizes.[2][1]
  2. Specified energy functional. The quantity bounded must be named. Classical configurational energy and a quantum ground-state infimum are different, and neither can silently be replaced by free energy.[2][1]
  3. One linear lower bound. A finite coefficient independent of particle count keeps the lower bound proportional to the declared count. A different arbitrary coefficient for each size would only give fixed-size boundedness.[1]
  4. The full admissible family. The bound must hold across all admitted sizes and configurations or states. A calculation for one configuration, wave function or particle count is insufficient.[2][1]

Schematically, the test is model energy ≥ −B × declared particle count, with one finite B ≥ 0 across the model’s admissible family. In the classical case the count is N; in Lieb’s quantum case it is N + K. The schematic form does not erase those distinct assumptions.[2][1]

What It Is Not

It is not stability of the first kind, which Lieb defines by a finite ground-state lower energy for a fixed system.[1] It is also not dynamical return to an operating point after perturbation, the subject of the live Prime Stability. Here “stability” controls how low energy can fall as particle count varies; it says nothing by itself about a time-dependent restoring trajectory.

Nor is it a claim that a thermodynamic or free-energy limit already exists. Fisher and Ruelle require further long-range interaction and domain-shape conditions in their thermodynamic-limit discussion, and Lieb explicitly calls second-kind stability essential but insufficient by itself for a free-energy limit. A fermionic kinetic inequality, Pauli exclusion, a hard repulsive core or charge smearing may help in a particular proof; none is part of every instance of the formal property.[2][1]

Scope of Application

The term and test belong to mathematical physics and statistical mechanics of many-particle systems. The reviewed positives are a class of classical pair-potential models and a nonrelativistic quantum model of fermionic Coulomb matter.[2][1] They justify a domain-specific abstraction across two materially different settings in that field, but do not establish an all-domain Prime about any large collection or any kind of “stability.”

Model assumptions matter. Fisher and Ruelle give sufficient conditions for a classical bound, including a charged application with smeared charges or a short-distance cutoff; they explicitly did not solve the quantum point-Coulomb problem. Lieb's account identifies an electron-plus-nucleus Hamiltonian and a fermionic proof route. Its result must not be extended without argument to bosons, arbitrary magnetic fields, relativistic particles or quantum electrodynamics.[2][1]

Clarity

The decisive word is uniform. If an (N=10) model has some finite minimum energy, that says nothing about whether one coefficient (B) bounds its energy by −BN for every allowed N. By contrast, a second-kind bound supplies the same coefficient while system size varies. A family whose lowest energy becomes more negative than every fixed multiple of its particle count fails this test even if each finite member has a lower energy.[1]

The energy label is just as decisive. The Fisher–Ruelle classical statement concerns potential energy for every allowed particle arrangement. The Lieb quantum statement concerns the infimum of kinetic plus Coulomb energy over fermionic states; the Lieb–Thirring kinetic inequality is a route to controlling that total, not itself a synonym for the total-energy conclusion. Calling both simply “energy stability” without naming their functionals hides a material difference.[2][1][3]

Manages Complexity

The four-role test separates a theorem's target from its proof. First write down the model class and energy. Next ask whether one constant works across the admitted size and state family. Only then examine the sufficient assumptions that establish or defeat the bound. This prevents a classical potential condition or a quantum antisymmetry argument from being mistaken for a universal definition.[2][1]

It also separates lower-bound stability from later thermodynamic questions. A linear lower bound rules out one kind of uncontrolled negative energy growth, while a limit of partition functions or free energy asks for additional structure. Fisher and Ruelle's discussion of interaction tails and domain shapes gives a concrete reason to keep the questions distinct.[2][1]

Abstract Reasoning

To test a new proposed case, specify the size parameter, the allowed states or configurations, and the energy being minimized or bounded. Try to find a single size-independent coefficient for a linear lower bound. A proof for one chosen state supplies at most an upper bound on a ground-state infimum; it cannot alone establish a lower bound for all states. If the best energy scales superlinearly downward, the proposed model is a negative case.[1]

This reasoning can compare unlike models while preserving their differences. Classical conditions on pair potentials and a quantum fermionic kinetic inequality may each lead to a linear lower energy, but their assumptions cannot be exchanged.[2][1][3] The cross-case name is an editorial abstraction of the same quantified test, not a claim that the two original works derive an identical theorem.

Knowledge Transfer

What transfers from the classical case to the quantum case is the question: for the declared energy, is there one lower-bound coefficient independent of system size? The classical answer examines every allowed particle configuration under pair-potential assumptions. The quantum answer examines the ground-state infimum of a specified Hamiltonian for every admitted N and K. The count convention, energy functional and sufficient proof conditions must be supplied anew after transfer.[2][1]

The same method also transfers to a possible countercase: ask whether first-kind fixed-size boundedness survives while the all-size linear bound fails. Lieb's nonrelativistic charged-boson discussion provides such a contrast, with different superlinear exponents under different mass assumptions. Those exponents describe the cited models, not a universal rate for unstable matter.[1]

Examples

Classical pair-potential systems

Fisher and Ruelle consider classical particles whose total configurational energy comes from interactions under stated sufficient conditions. Their original result bounds the energy below by (-NB), with one fixed (B), for every allowed N-particle configuration. Their charged-system application treats slightly smeared charges or an interaction cutoff; it must not be recast as a solution for point-Coulomb quantum matter.[2]

Mapped back: the model is the class with stated pair potentials; the energy is total configurational potential energy; the linear bound is (-NB) with fixed (B); the family includes all allowed counts and positions under those assumptions. The potential theorem is a way to establish this case, not a required mechanism in every second-kind model.[2]

Nonrelativistic fermionic Coulomb matter

Lieb's full author treatment describes N electrons and K nuclei with electron kinetic and Coulomb terms. For fermionic states, second-kind stability is a lower bound on the full ground-state energy proportional to N + K. Lieb and Thirring's fermionic kinetic inequality provides a shorter proof route and a better constant through Thomas–Fermi theory. The directly accessible 1975 original publisher abstract confirms the kinetic-bound/stability claim; the detailed Hamiltonian, count convention and inequality used here come from Lieb's full author account.[1][3]

Mapped back: the model specifies fermionic electron states and nuclei; the energy is the full Hamiltonian ground-state infimum; the linear bound has one constant independent of N and K under fixed parameters; the family spans the admitted particle counts and states. The kinetic inequality is a quantum proof accent absent from the classical case.[1]

Structural Tensions

No pair of opposed forces or goals is required in every instance of this formal criterion. Attraction and repulsion may enter a particular model, and a quantum kinetic term may control an attractive potential, but the defining question is only whether the specified total energy has an all-size linear lower bound. Turning one proof's balance into a universal structural tension would erase the classical/quantum difference.[2][1]

The genuine comparison is logical: fixed-size boundedness is weaker than one uniform linear bound. Lieb's charged-boson example illustrates that distinction without making bosonic behavior a required failure mode for every model that lacks second-kind stability.[1]

Structural–Framed Character

Evaluative weight: this is a desirable property when modeling extensive bulk matter, yet whether the inequality holds is a mathematical claim relative to an explicit model, not a preference. Human-practice dependence: researchers choose the model, energy and assumptions, then prove or refute the quantified bound; the bound does not depend on someone recognizing it. Institutional origin: the terminology and proofs come from mathematical physics and statistical mechanics. Lieb states the first/second-kind distinction, while Fisher and Ruelle state a classical stability condition with the same lower-bound form.[2][1] This comparison does not assert shared coinage.

Vocabulary travel: “stability” also names dynamical restoration and other properties, but here it is an energy lower bound uniform in particle count. Import versus recognition: carrying the test to another many-particle model is an import of a formal question; recognizing a positive instance requires a proof under that model's own assumptions, not an analogy to Pauli exclusion or a repulsive potential. The unlike classical and quantum cases make that distinction visible.

Its character: a model-relative formal property of many-particle energy scaling, recognizable across distinct classical and quantum mechanisms only when the declared energy and all-size quantifiers are kept intact.

Structural Core vs. Domain Accent

The core is a declared model, a specified energy, one size-independent linear lower-bound coefficient, and coverage of all admitted sizes and configurations or states. The classical accent is a configurational pair-potential theorem, with optional smeared-charge construction in its charged application. The quantum accent is a fermionic kinetic inequality controlling a kinetic-plus-Coulomb Hamiltonian.[2][1] Neither accent travels as a universal role.

A bounded future-Prime question remains: can a genuinely different, nonphysics domain instantiate all four roles without using particle and energy words metaphorically? The reviewed evidence spans only classical and quantum many-particle mathematical physics, so it does not support that elevation. Its current identity stays domain-specific.

The live Prime Stability addresses return after perturbation, which an all-size energy bound need not involve. Boundedness can describe the normalized negative energy per particle under a finite upper threshold, but its full signature additionally calls for bound preservation through operations and a downstream use; neither is necessary to state or satisfy this criterion. Constraint requires a binding feasible-set rule, Invariance a named feature preserved under transformations, and Scale and Scaling and Scale Dependence qualitative changes across size bands. Those features need not occur in a second-kind stable model. Thermodynamic Equilibrium and its entropy, second-law and equilibrium ancestors concern macroscopic states and dynamics, not the lower-bound property alone.

Neighborhood in Abstraction Space

Stability of the Second Kind sits in a sparse region of the domain-specific corpus (86th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2721 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-17

Not to Be Confused With

  • Stability of the first kind: finite lower energy at fixed particle content; no uniform all-size linear coefficient follows.[1]
  • A thermodynamic or free-energy limit: requires further conditions and is not guaranteed by the energy lower bound alone.[2][1]
  • Dynamical Stability: return or persistence after perturbation is a different identity.
  • The Lieb–Thirring kinetic inequality: a quantum proof tool, not the full energy property and not a classical prerequisite.[1][3]
  • A universal charged-matter theorem: the reviewed classical charged result uses smeared charges; Lieb's positive quantum case assumes fermionic nonrelativistic matter, while his charged-boson case can fail the second-kind bound.[2][1]

References

[1] Elliott H. Lieb, “The Stability of Matter and Quantum Electrodynamics” (2002), arXiv:math-ph/0209034, https://arxiv.org/pdf/math-ph/0209034 . Full author paper inspected: PDF pp.2–3 §2.2 and equations (1)–(2) define first and second kinds and the N + K count; pp.5–6 equations (4)–(6) specify the Coulomb Hamiltonian; pp.6–7 §3.1 and equation (7) discuss the fermionic proof route; pp.7–8 §3.2 distinguish charged-boson regimes. This later author treatment, rather than the 1975 abstract alone, supports the exact quantum model and count. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y ↩z ↩27 ↩28 ↩29

[2] Michael E. Fisher and David Ruelle, “The Stability of Many-Particle Systems,” Institut des Hautes Études Scientifiques prepublication P/65/03 (1965), https://repo-archives.ihes.fr/FONDS_IHES/I_Prepublications/RUELLE/1965-1976/P_65_03/P_65_03_web.pdf . Full original preprint inspected: PDF p.0 abstract; pp.2–5 Introduction, including the all-configuration bound, extra thermodynamic-limit conditions and point-Coulomb quantum caveat; pp.6–8 multispecies theorem; pp.9–11 charged application. Display equations in the scan have poor OCR, so formula wording is cross-checked against the abstract and prose. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t

[3] Elliott H. Lieb and Walter E. Thirring, “Bound for the Kinetic Energy of Fermions Which Proves the Stability of Matter,” Physical Review Letters 35 (1975): 687–689, DOI: 10.1103/PhysRevLett.35.687, https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.35.687 . Original publisher abstract and metadata inspected; the full PRL article was not accessible through this path. The abstract supports a fermionic kinetic-energy lower bound and a resulting stability proof, but not the detailed Hamiltonian or N + K inequality on its own. registry ↩a ↩b ↩c ↩d