AFTD Auto-Formalizing Theoretical Domains

ising_inverse_square_bulk_two_of_finite_formula

theoremverified

If the displayed second-order formula holds for every ring size L at least 2 and every real field, then the polynomial (1-y)(1+9y)/180 represents the bulk second-order field coefficient.

Statement

theorem ising_inverse_square_bulk_two_of_finite_formula
    (hfinite : ∀ (L : ℕ), 2 ≤ L → ∀ h : ℝ,
      ising_inverse_square_ring_coefficient 2 L h =
      let u : ℝ := (L : ℝ)⁻¹
      let a := (1-u^2)/3
      let b := (1+10*u^2-11*u^4)/45
      let y := (Real.tanh h)^2
      (1-y^2)/4*b + y*(1-y)/2*(a^2-b)) :
    ising_inverse_square_bulk_field_polynomial 2
      (Polynomial.C (1 / 180 : ℝ) * (1 - Polynomial.X) * (1 + 9 * Polynomial.X))
Source
Helper for the second-order coefficient from arXiv:2610.39815, equations (3.15)–(3.16), and OP-806.
Verified
02 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 3

The proof, explained

To show that P(X) = (1/180)(1 - X)(1 + 9X) is the bulk second-order field polynomial, we fix an arbitrary field h and compute the large-L limit of the ring coefficients. By hypothesis, for every L ≥ 2 the ring coefficient is given by an explicit algebraic expression in u = 1/L and y = tanh²(h). A known limit theorem (ising_inverse_square_order_two_formula_limit) establishes that this expression converges as L → ∞ to (1 - y)(1 + 9y)/180. Since evaluating P at y produces exactly this value, and the coefficients coincide with the formula eventually, the ring coefficients converge to P(tanh²(h)).

  1. For an arbitrary field h, the finite-volume expression evaluated at y = tanh²(h) converges as L → ∞ to (1 - y)(1 + 9y)/180.

    Fixing an arbitrary external magnetic field h ∈ ℝ unfolds the definition of ising_inverse_square_bulk_field_polynomial. Applying ising_inverse_square_order_two_formula_limit at y = tanh²(h) provides the limit as L → ∞ of the finite-ring formula: with u = 1/L, a = (1 - u²)/3, and b = (1 + 10u² - 11u⁴)/45, the rational expression (1 - y²)/4 * b + y*(1 - y)/2 * (a² - b) converges to (1 - y)(1 + 9y)/180.

    lines 11–12

    uses ising_inverse_square_order_two_formula_limit

      intro h
      have hl := ising_inverse_square_order_two_formula_limit ((Real.tanh h)^2)
  2. The second-order ring coefficient sequence is eventually equal to the finite-volume formula as L → ∞.

    Along Filter.atTop on ℕ, the ring size L is eventually at least 2 by Filter.eventually_ge_atTop. For every such L, the hypothesis hfinite guarantees that the second-order ring coefficient is identical to the finite-volume formula.

    lines 13–21

    uses Filter.eventually_ge_atTop

      have heq : (fun L : ℕ => ising_inverse_square_ring_coefficient 2 L h) =ᶠ[Filter.atTop]
          (fun L : ℕ =>
          let u : ℝ := (L : ℝ)⁻¹
          let a := (1-u^2)/3
          let b := (1+10*u^2-11*u^4)/45
          let y := (Real.tanh h)^2
          (1-y^2)/4*b + y*(1-y)/2*(a^2-b)) := by
        filter_upwards [Filter.eventually_ge_atTop 2] with L hL
        exact hfinite L hL h
  3. Evaluating the polynomial P(X) = (1/180)(1 - X)(1 + 9X) at tanh²(h) yields (1 - tanh²(h))(1 + 9 tanh²(h))/180.

    Evaluating the polynomial at tanh²(h) distributes across multiplication (Polynomial.eval_mul), subtraction (Polynomial.eval_sub), addition (Polynomial.eval_add), constants (Polynomial.eval_C, Polynomial.eval_one, Polynomial.eval_ofNat), and the variable (Polynomial.eval_X). Standard ring arithmetic then simplifies (1/180) * (1 - tanh²(h)) * (1 + 9 tanh²(h)) to (1 - tanh²(h)) * (1 + 9 tanh²(h)) / 180.

    lines 22–26

    uses Polynomial.eval_mul, Polynomial.eval_C, Polynomial.eval_sub, Polynomial.eval_one, Polynomial.eval_add, Polynomial.eval_X, Polynomial.eval_ofNat

      have heval : (Polynomial.C (1 / 180 : ℝ) * (1 - Polynomial.X) * (1 + 9 * Polynomial.X)).eval ((Real.tanh h)^2) =
          (1-(Real.tanh h)^2)*(1+9*(Real.tanh h)^2)/180 := by
        simp only [Polynomial.eval_mul, Polynomial.eval_C, Polynomial.eval_sub,
          Polynomial.eval_one, Polynomial.eval_add, Polynomial.eval_X, Polynomial.eval_ofNat]
        ring
  4. The sequence of second-order ring coefficients converges to P(tanh²(h)) as L → ∞, proving the theorem.

    Rewriting the limit target with the evaluation identity replaces P(tanh²(h)) by (1 - tanh²(h))(1 + 9 tanh²(h))/180. Since the sequence of ring coefficients is eventually equal to the finite-volume formula as L → ∞, the limit established in the first step transfers to the ring coefficients.

    lines 27–28

      rw [heval]
      exact hl.congr' heq.symm
ising_inverse_square_order_two_formula_limit
For any real y, the finite-ring formula (1-y^2)/4 * b + y*(1-y)/2 * (a^2-b), with u = 1/L, a = (1-u^2)/3, and b = (1+10*u^2-11*u^4)/45, converges as L → ∞ to (1-y)*(1+9*y)/180.
Filter.eventually_ge_atTop
For any element a in a preorder, eventually a ≤ x holds along Filter.atTop.
Polynomial.eval_mul
The evaluation of a product of polynomials at a point equals the product of their evaluations.
Polynomial.eval_C
The evaluation of a constant polynomial C(a) at any point is a.
Polynomial.eval_sub
The evaluation of the difference of two polynomials at a point equals the difference of their evaluations.
Polynomial.eval_one
The evaluation of the constant polynomial 1 at any point is 1.
Polynomial.eval_add
The evaluation of the sum of two polynomials at a point equals the sum of their evaluations.
Polynomial.eval_X
The evaluation of the indeterminate polynomial X at a point x is x.
Polynomial.eval_ofNat
The evaluation of a natural numeral polynomial at any point is the corresponding natural numeral.

Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.

Lean source view module on GitHub

theorem ising_inverse_square_bulk_two_of_finite_formula    (hfinite : ∀ (L : ℕ), 2 ≤ L → ∀ h : ℝ,      ising_inverse_square_ring_coefficient 2 L h =      let u : ℝ := (L : ℝ)⁻¹      let a := (1-u^2)/3      let b := (1+10*u^2-11*u^4)/45      let y := (Real.tanh h)^2      (1-y^2)/4*b + y*(1-y)/2*(a^2-b)) :    ising_inverse_square_bulk_field_polynomial 2      (Polynomial.C (1 / 180 : ℝ) * (1 - Polynomial.X) * (1 + 9 * Polynomial.X)) := by  intro h  have hl := ising_inverse_square_order_two_formula_limit ((Real.tanh h)^2)  have heq : (fun L : ℕ => ising_inverse_square_ring_coefficient 2 L h) =ᶠ[Filter.atTop]      (fun L : ℕ =>      let u : ℝ := (L : ℝ)⁻¹      let a := (1-u^2)/3      let b := (1+10*u^2-11*u^4)/45      let y := (Real.tanh h)^2      (1-y^2)/4*b + y*(1-y)/2*(a^2-b)) := by    filter_upwards [Filter.eventually_ge_atTop 2] with L hL    exact hfinite L hL h  have heval : (Polynomial.C (1 / 180 : ℝ) * (1 - Polynomial.X) * (1 + 9 * Polynomial.X)).eval ((Real.tanh h)^2) =      (1-(Real.tanh h)^2)*(1+9*(Real.tanh h)^2)/180 := by    simp only [Polynomial.eval_mul, Polynomial.eval_C, Polynomial.eval_sub,      Polynomial.eval_one, Polynomial.eval_add, Polynomial.eval_X, Polynomial.eval_ofNat]    ring  rw [heval]  exact hl.congr' heq.symm
See it in the graphDiscuss this resultChallenge it