Chapter YangMillsFockFriedrichs
DefinitionChapterYangMillsFockFriedrichsdefinitionsformalizationtimepiece
Formal definitions for the timepiece Lean 4 formalization (source chapter BookProof/ChapterYangMillsFockFriedrichs.lean): generated def bundle for ChapterYangMillsFockFriedrichs. See BookProof/ChapterYangMillsFockFriedrichs.lean for full context.
Definition code
import Definitions.Def_ChapterYangMillsHermite
import Definitions.Def_ChapterFriedrichsExtension
import Definitions.Def_ChapterQgOuterFockEsa
import Definitions.Def_ChapterQg3DGaugeFarisLavine
import Definitions.Def_ChapterSirkTrotterKato
import Definitions.Def_ChapterDirectSumEsa
import Definitions.Def_ChapterFarisLavine
import Definitions.Def_ChapterHermiteProductCore
import Definitions.Def_ChapterStoneBridge
import Definitions.Def_ChapterYangMillsFriedrichs
import Theorems.Thm_BookProof_YangMillsHermite_RealCoeff_sum
import Theorems.Thm_BookProof_YangMillsHermite_RealCoeff_add
import Theorems.Thm_BookProof_YangMillsHermite_RealCoeff_mul
import Theorems.Thm_BookProof_YangMillsHermite_RealCoeff_smul
import Theorems.Thm_BookProof_YangMillsHermite_realCoeff_X
import Theorems.Thm_BookProof_YangMillsHermite_mulOp_polySym
import Theorems.Thm_BookProof_FriedrichsExtension_friedrichs_extension_exists_complete
import Theorems.Thm_BookProof_QgOuterFock_dsOp_quadForm_nonneg
import Theorems.Thm_BookProof_StoneBridge_exists_stone_flow_of_positive
import Mathlib
/-!
# Quantum Yang–Mills on the nested Fock space: the **direct Friedrichs extension**
The Yang–Mills Hamiltonian is a *positive* sum of squares,
`H = ½ Σ_m π_m² + ½ Σ_r Φ_r²`,
with symmetric momentum operators `π_m` and symmetric multiplication operators `Φ_r` (the
magnetic field `B_{i a}` — including its **quartic, non-abelian** part — and the 3D
gauge-fixing forms). It is therefore **bounded below**, and the Friedrichs extension applies
to it *directly*: no Faris–Lavine commutator certificate is needed on the Yang–Mills side,
and none is used here.
Both ingredients of the construction are fibrewise, so they lift from the one-particle
Hilbert space `L²(ℝ⁹⁹)` to the nested Fock space `⊕ₙ L²(ℝ^{99n})`:
* symmetry and positivity of the quadratic form are checked sector by sector
(`dsOp_symmetricOn`, `BookProof.QgOuterFock.dsOp_quadForm_nonneg`);
* the finite-particle core is dense (`dsCore_dense`), so the abstract Friedrichs theorem
`BookProof.FriedrichsExtension.friedrichs_extension_exists_complete` — proved in this development,
with no boundedness hypothesis — produces a positive self-adjoint extension of the outer
Hamiltonian, and Stone's theorem the unitary time evolution it generates.
## The variables
`99 = 3 + 24 + 72` per particle: three spatial coordinates, the `24 = 3 × 8` gauge-field
coordinates `A_{j,a}` and the `72 = 3 × 3 × 8` **independent coordinates `∂_j A_{k,a}`**
representing the spatial derivatives of the gauge field. Every particle carries its own
copy, `ycoord p i` being the global coordinate of the `i`-th field-space direction of the
`p`-th particle.
## What is proved
* `magPolyN`, `gaussPolyN` — the magnetic field `B_{i a} = ε_{ijk}(∂_j A_{k,a} +
f_{abc} A_{j,b} A_{k,c})` of each particle, at **arbitrary real structure constants**
(so the genuinely non-abelian, quartic case is included), and the **3D gauge-fixing form**
`Σ_j ∂_j A_{j,a}` written in the independent derivative coordinates.
* `ymSectorHam_quadForm_nonneg` — **the `n`-particle gauge-fixed Yang–Mills Hamiltonian is
bounded below** (indeed non-negative).
* `ymSector_friedrichs_extension` — hence it has a positive self-adjoint (Friedrichs)
extension on `L²(ℝ^{99n})`, for every particle number and every family of structure
constants.
* **`ymFock_friedrichs_extension`** — the headline: the Yang–Mills Hamiltonian on the nested
Fock space `⊕ₙ L²(ℝ^{99n})` is bounded below (`ymFockHam_quadForm_nonneg`) and has a
positive self-adjoint (Friedrichs) extension.
* `ymFock_stone_flow` — the unitary time evolution it generates (Stone).
* `ymFockHam_number_conserving` — the outer Hamiltonian conserves the particle number: it is
block diagonal in the number sectors, and its restriction to the `n`-particle sector is the
`n`-particle Hamiltonian. This is why no lattice regularization enters: the number sectors,
not a spatial cutoff, decompose the problem.
No spectral information and no mass gap is claimed anywhere.
Everything is `sorry`-free and `axiom`-free.
-/
namespace BookProof.YmFockFriedrichs
open MvPolynomial
open BookProof.YangMillsHermite BookProof.YangMillsFriedrichs BookProof.FriedrichsExtension
open BookProof.HermiteProductCore BookProof.FarisLavine BookProof.DirectSumEsa
open BookProof.QgOuterFock BookProof.StoneBridge BookProof.Qg3DGaugeFL
open BookProof.ChapterStoneResolvent
noncomputable section
/-! ## 1. The coordinates of the `n`-particle sector -/
/-- The global coordinate of the `i`-th field-space direction of the `p`-th particle. -/
def ycoord {n : ℕ} (p : Fin n) (i : Fin 99) : Fin (n * 99) := finProdFinEquiv (p, i)
theorem ycoord_injective {n : ℕ} (p : Fin n) : Function.Injective (ycoord p) := by
intro i i' h
have := finProdFinEquiv.injective h
simpa using congrArg Prod.snd this
/-- The particle carrying a global coordinate. -/
def ypart {n : ℕ} (I : Fin (n * 99)) : Fin n := (finProdFinEquiv.symm I).1
/-- The field-space direction of a global coordinate. -/
def ymode {n : ℕ} (I : Fin (n * 99)) : Fin 99 := (finProdFinEquiv.symm I).2
@[simp] theorem ypart_ycoord {n : ℕ} (p : Fin n) (i : Fin 99) : ypart (ycoord p i) = p := by
simp [ypart, ycoord]
@[simp] theorem ymode_ycoord {n : ℕ} (p : Fin n) (i : Fin 99) : ymode (ycoord p i) = i := by
simp [ymode, ycoord]
/-! ## 2. The magnetic field and the 3D gauge-fixing forms of each particle -/
/-- **The magnetic field of the `p`-th particle**,
`B_{i a} = ε_{ijk}(∂_j A_{k,a} + f_{abc} A_{j,b} A_{k,c})`, as a polynomial in the `99n`
coordinates. The structure constants are arbitrary reals: the quartic (non-abelian) case is
included. -/
def magPolyN (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) {n : ℕ} (p : Fin n) (i : Fin 3) (a : Fin 8) :
MvPolynomial (Fin (n * 99)) ℂ :=
∑ j : Fin 3, ∑ k : Fin 3, ((levi i j k : ℝ) : ℂ) •
(X (ycoord p (idxD j k a)) + ∑ b : Fin 8, ∑ c : Fin 8,
((fabc a b c : ℝ) : ℂ) • (X (ycoord p (idxA j b)) * X (ycoord p (idxA k c))))
theorem realCoeff_magPolyN (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) {n : ℕ} (p : Fin n) (i : Fin 3)
(a : Fin 8) : RealCoeff (magPolyN fabc p i a) := by
refine RealCoeff.sum fun j _ => RealCoeff.sum fun k _ => RealCoeff.smul ?_
refine RealCoeff.add (realCoeff_X _) ?_
exact RealCoeff.sum fun b _ => RealCoeff.sum fun c _ =>
RealCoeff.smul ((realCoeff_X _).mul (realCoeff_X _))
/-- **The 3D gauge-fixing form of the `p`-th particle**, `Σ_j ∂_j A_{j,a}` — the transverse
(Coulomb) gauge condition on the spatial slice, written in the independent coordinates that
represent the spatial derivatives of the gauge field. -/
def gaussPolyN {n : ℕ} (p : Fin n) (a : Fin 8) : MvPolynomial (Fin (n * 99)) ℂ :=
∑ j : Fin 3, X (ycoord p (idxD j j a))
theorem realCoeff_gaussPolyN {n : ℕ} (p : Fin n) (a : Fin 8) : RealCoeff (gaussPolyN p a) :=
RealCoeff.sum fun _ _ => realCoeff_X _
/-- The 3D gauge-fixing form is not vacuous: it really involves the derivative
coordinates. -/
theorem gaussPolyN_eval {n : ℕ} (p : Fin n) (a : Fin 8) :
eval (fun I => if I = ycoord p (idxD 0 0 a) then (1 : ℂ) else 0) (gaussPolyN p a) = 1 := by
have h1 : (idxD 1 1 a : Fin 99) ≠ idxD 0 0 a := by
simp [idxD, Fin.ext_iff]
have h2 : (idxD 2 2 a : Fin 99) ≠ idxD 0 0 a := by
simp [idxD, Fin.ext_iff]
have hy1 : ycoord p (idxD 1 1 a) ≠ ycoord p (idxD 0 0 a) := fun h =>
h1 (ycoord_injective p h)
have hy2 : ycoord p (idxD 2 2 a) ≠ ycoord p (idxD 0 0 a) := fun h =>
h2 (ycoord_injective p h)
simp [gaussPolyN, Fin.sum_univ_three, hy1, hy2]
/-! ## 3. The `n`-particle Hamiltonian -/
/-- The momentum operators of the `n`-particle sector: `π = −i ∂/∂A_{j,a}`, one for each of
the `24` gauge-field coordinates of each of the `n` particles. -/
def ymPiN (n : ℕ) (m : Fin (n * 24)) :
(polyGaussCore (d := n * 99)) →ₗ[ℂ] (polyGaussCore (d := n * 99)) :=
(coreRepPoly (n * 99)).op
(momOp (ycoord (finProdFinEquiv.symm m).1
(idxA (decodeSpace (finProdFinEquiv.symm m).2)
(decodeColor (finProdFinEquiv.symm m).2))))
/-- The field operators of the `n`-particle sector: the `24n` magnetic-field components and
the `8n` 3D gauge-fixing forms, as multiplication operators. -/
def ymFieldSum (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ) (s : Fin (n * 24) ⊕ Fin (n * 8)) :
(polyGaussCore (d := n * 99)) →ₗ[ℂ] (polyGaussCore (d := n * 99)) :=
Sum.elim
(fun m₁ : Fin (n * 24) =>
(coreRepPoly (n * 99)).op
(mulOp (magPolyN fabc (finProdFinEquiv.symm m₁).1
(decodeSpace (finProdFinEquiv.symm m₁).2)
(decodeColor (finProdFinEquiv.symm m₁).2))))
(fun m₂ : Fin (n * 8) =>
(coreRepPoly (n * 99)).op
(mulOp (gaussPolyN (finProdFinEquiv.symm m₂).1 (finProdFinEquiv.symm m₂).2)))
s
/-- The field operators of the `n`-particle sector, indexed by `Fin (24n + 8n)`. -/
def ymFieldN (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ) (m : Fin (n * 24 + n * 8)) :
(polyGaussCore (d := n * 99)) →ₗ[ℂ] (polyGaussCore (d := n * 99)) :=
ymFieldSum fabc n (finSumFinEquiv.symm m)
theorem ymPiN_symmetricOn (n : ℕ) (m : Fin (n * 24)) :
SymmetricOn (polyGaussCore (d := n * 99))
((polyGaussCore (d := n * 99)).subtype.comp (ymPiN n m)) :=
(coreRepPoly (n * 99)).symmetricOn_op (momOp_polySym _)
theorem ymFieldSum_symmetricOn (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ)
(s : Fin (n * 24) ⊕ Fin (n * 8)) :
SymmetricOn (polyGaussCore (d := n * 99))
((polyGaussCore (d := n * 99)).subtype.comp (ymFieldSum fabc n s)) := by
rcases s with m₁ | m₂
· exact (coreRepPoly (n * 99)).symmetricOn_op (mulOp_polySym (realCoeff_magPolyN _ _ _ _))
· exact (coreRepPoly (n * 99)).symmetricOn_op (mulOp_polySym (realCoeff_gaussPolyN _ _))
theorem ymFieldN_symmetricOn (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ)
(m : Fin (n * 24 + n * 8)) :
SymmetricOn (polyGaussCore (d := n * 99))
((polyGaussCore (d := n * 99)).subtype.comp (ymFieldN fabc n m)) :=
ymFieldSum_symmetricOn fabc n (finSumFinEquiv.symm m)
/-- **The `n`-particle gauge-fixed Yang–Mills Hamiltonian** on the Gauss–polynomial core of
`L²(ℝ^{99n})`: the kinetic term of every particle, the full magnetic energy (quartic in the
non-abelian case) of every particle, and the 3D gauge fixing of every particle. -/
def ymSectorHam (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ) :
(polyGaussCore (d := n * 99)) →ₗ[ℂ] L2d (n * 99) :=
weylOp (ymPiN n) (ymFieldN fabc n)
theorem ymSectorHam_symmetricOn (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ) :
SymmetricOn (polyGaussCore (d := n * 99)) (ymSectorHam fabc n) :=
weylOpDom_symmetricOn (ymPiN_symmetricOn n) (ymFieldN_symmetricOn fabc n)
/-- **The `n`-particle Yang–Mills Hamiltonian is bounded below** — its quadratic form is a
sum of squares, hence non-negative. This is the hypothesis of the Friedrichs extension
theorem, and it is what makes a Faris–Lavine commutator certificate unnecessary here. -/
theorem ymSectorHam_quadForm_nonneg (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ)
(x : polyGaussCore (d := n * 99)) : 0 ≤ quadForm (ymSectorHam fabc n) x :=
weylOpDom_quadForm_nonneg (ymPiN_symmetricOn n) (ymFieldN_symmetricOn fabc n) x
/-- **The `n`-particle gauge-fixed Yang–Mills Hamiltonian has a positive self-adjoint
(Friedrichs) extension**, for every family of real structure constants. -/
theorem ymSector_friedrichs_extension (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (n : ℕ) :
∃ (Dom : Submodule ℂ (L2d (n * 99))) (A : Dom →ₗ[ℂ] L2d (n * 99)),
IsPositiveSelfAdjointExtension (ymSectorHam fabc n) A :=
friedrichs_extension_exists_complete
⟨polyGaussCore, ymSectorHam fabc n, ymSectorHam_symmetricOn fabc n,
ymSectorHam_quadForm_nonneg fabc n⟩
polyGaussCore_dense
/-! ## 4. The nested Fock space -/
/-- The nested Fock space `⊕ₙ L²(ℝ^{99n})` of the Yang–Mills field. -/
abbrev ymFockSpace := lp (fun n : ℕ => L2d (n * 99)) 2
/-- The finite-particle core: finitely many sectors, each in its Gauss–polynomial core. -/
def ymFockCore : Submodule ℂ ymFockSpace := dsCore (fun n : ℕ => polyGaussCore (d := n * 99))
theorem ymFockCore_dense :
Dense ((ymFockCore : Submodule ℂ ymFockSpace) : Set ymFockSpace) :=
dsCore_dense fun _ => polyGaussCore_dense
/-- **The Yang–Mills Hamiltonian on the nested Fock space**: one copy of the one-particle
Hamiltonian per particle, in every number sector. -/
def ymFockHam (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) : ymFockCore →ₗ[ℂ] ymFockSpace :=
dsOp (fun n : ℕ => ymSectorHam fabc n)
theorem ymFockHam_symmetricOn (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) :
SymmetricOn ymFockCore (ymFockHam fabc) :=
dsOp_symmetricOn _ fun n => ymSectorHam_symmetricOn fabc n
/-- **The outer Yang–Mills Hamiltonian is bounded below**: positivity is fibrewise, so it
lifts from the one-particle Hilbert space to the nested Fock space. -/
theorem ymFockHam_quadForm_nonneg (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (x : ymFockCore) :
0 ≤ quadForm (ymFockHam fabc) x :=
dsOp_quadForm_nonneg _ (fun n u => ymSectorHam_quadForm_nonneg fabc n u) x
/-- **The Yang–Mills Hamiltonian on the nested Fock space has a positive self-adjoint
(Friedrichs) extension** — for every family of real structure constants, the non-abelian ones
included.
This is the Yang–Mills counterpart of the Faris–Lavine route used for the other threads: the
Hamiltonian is bounded below, so the Friedrichs extension applies directly, and — like the
Faris–Lavine certificate — the two data it needs (symmetry and positivity of the form) are
fibrewise and therefore lift from the one-particle Hilbert space to the outer Fock space. -/
theorem ymFock_friedrichs_extension (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) :
∃ (Dom : Submodule ℂ ymFockSpace) (A : Dom →ₗ[ℂ] ymFockSpace),
IsPositiveSelfAdjointExtension (ymFockHam fabc) A :=
friedrichs_extension_exists_complete
⟨ymFockCore, ymFockHam fabc, ymFockHam_symmetricOn fabc, ymFockHam_quadForm_nonneg fabc⟩
ymFockCore_dense
set_option maxHeartbeats 1000000 in
-- unfolding the `lp` instances of the Fock space in the Stone construction exceeds the
-- default budget
/-- **The unitary time evolution of the outer Yang–Mills Hamiltonian**: the Friedrichs
extension is self-adjoint, and Stone's theorem gives the flow it generates. -/
theorem ymFock_stone_flow (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) :
∃ (T : UnboundedSelfAdjoint ymFockSpace) (U : ℝ → (ymFockSpace →L[ℂ] ymFockSpace)),
IsStoneFlow T U := by
obtain ⟨Dom, A, hA⟩ := ymFock_friedrichs_extension fabc
obtain ⟨T, U, _, _, hflow⟩ := exists_stone_flow_of_positive ymFockCore_dense hA
exact ⟨T, U, hflow⟩
/-! ## 5. Particle-number conservation — why no lattice is needed -/
/-- The restriction of the outer Hamiltonian to the `n`-particle sector is the `n`-particle
Hamiltonian. -/
theorem ymFockHam_sector (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (x : ymFockCore) (n : ℕ) :
((ymFockHam fabc x : ymFockSpace) : ∀ n : ℕ, L2d (n * 99)) n
= ymSectorHam fabc n
⟨((x : ymFockSpace) : ∀ n : ℕ, L2d (n * 99)) n, x.2.2 n⟩ := rfl
/-- **The outer Yang–Mills Hamiltonian conserves the particle number.** It is block diagonal
in the number sectors; the sectors, not a spatial lattice, are what decomposes the
problem. -/
theorem ymFockHam_number_conserving (fabc : Fin 8 → Fin 8 → Fin 8 → ℝ) (x : ymFockCore)
{n : ℕ} (hx : ∀ m, m ≠ n → ((x : ymFockSpace) : ∀ m : ℕ, L2d (m * 99)) m = 0) (m : ℕ)
(hm : m ≠ n) : ((ymFockHam fabc x : ymFockSpace) : ∀ m : ℕ, L2d (m * 99)) m = 0 :=
dsOp_number_conserving _ x hx m hm
end
end BookProof.YmFockFriedrichs
Source