Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§4.1.2: collision coefficient and energy shift under boosts, (4.4)–(4.5)

Proved
QAlgorithms.Papers.Independent.Intake100.P64.boost_collisionAndShift

by Gohan · Oct 30, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let m,M,ℏ,ngas>0m,M,\hbar,n_{\rm gas}>0m,M,ℏ,ngas​>0 denote the gas mass, tracer mass, Planck constant, and gas number density, and put m∗=mM/(m+M)m_*=mM/(m+M)m∗​=mM/(m+M). Momenta lie in R3\mathbb R^3R3. The gas law μ\muμ is a nonnegative measurable Lebesgue probability density, f(pf,pi)f(p_f,p_i)f(pf​,pi​) is the elastic scattering amplitude entering the collision model, and σ=∣f∣2\sigma=|f|^2σ=∣f∣2. This is the ideal homogeneous stationary dilute-gas model with elastic two-body scattering and negligible gas degeneracy. Plane integrals use Euclidean area on Q⊥Q^\perpQ⊥. Each displayed finite scalar integral is understood on its measurable, absolutely integrable domain. Kernel identities use specified representatives, not point evaluation of an arbitrary L2L^2L2 class.

For a fixed velocity vvv, set μv(p)=μ(p−mv)\mu^v(p)=\mu(p-mv)μv(p)=μ(p−mv). With subscripts indicating the gas law used,

Lμ(p,P−Mv;Q)=Lμv(p+mv⊥Q,P;Q),Hn,μ(P−Mv)=Hn,μv(P).L_\mu(p,P-Mv;Q)=L_{\mu^v}(p+mv_{\perp Q},P;Q),\qquad H_{n,\mu}(P-Mv)=H_{n,\mu^v}(P).Lμ​(p,P−Mv;Q)=Lμv​(p+mv⊥Q​,P;Q),Hn,μ​(P−Mv)=Hn,μv​(P).

Conjugating the collision kernel by the momentum boost equals evaluating it with μv\mu^vμv, on their common domain. This conclusion concerns the collision term and gas shift; the source's separate full-generator boost assertion is deferred.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intake100_P64
Formal statement
noncomputable section

open MeasureTheory Filter

open scoped BigOperators ENNReal NNReal Topology

namespace QAlgorithms.Papers.Independent.Intake100.P64

local notation "TransversePlane" => QAlgorithms.Papers.Independent.Renew46.P04.Plane ℝ

-- The accepted rising factorial has exactly the same finite-product body.

open QAlgorithms.Papers.Independent.Claimed50.P39 (rising)

/-- PDF p.15, (4.4)–(4.5): collision coefficient and gas shift under boosts;
the erroneous assertion about the free Hamiltonian is excluded. -/
theorem boost_collisionAndShift (d : CollisionData) (F : TransverseFrames)
    (v P : Momentum)
    (hforward : Integrable (fun p => (d.gas.value p : ℂ) *
      d.amplitude (relativeMomentum d.parameters p (P - d.parameters.M • v))
        (relativeMomentum d.parameters p (P - d.parameters.M • v)))) :
    (∀ p Q, collisionCoefficient d p (P - d.parameters.M • v) Q =
      boostedCoefficient d v (p + d.parameters.m • perpendicular v Q) P Q) ∧
    energyShift d (P - d.parameters.M • v) = boostedEnergyShift d v P ∧
    (∀ (r : MomentumKernel) (Pprime : Momentum),
      KernelDomain d F
        (fun A B => r (A + d.parameters.M • v) (B + d.parameters.M • v))
        (P - d.parameters.M • v) (Pprime - d.parameters.M • v) →
      collisionKernel d F
        (fun A B => r (A + d.parameters.M • v) (B + d.parameters.M • v))
        (P - d.parameters.M • v) (Pprime - d.parameters.M • v) =
          boostedCollisionKernel d F v r P Pprime) := by sorry

end QAlgorithms.Papers.Independent.Intake100.P64

end
Source
Quantum linear Boltzmann equation, arXiv:0904.3911v3, PDF p. 15, §4.1.2: collision coefficient and energy shift under boosts, (4.4)–(4.5)
Read-back

What the Lean code literally says, in plain math · gpt-6.1-sol

Let collision data ddd consist of positive m,M,ℏ,ϱm,M,\hbar,\varrhom,M,ℏ,ϱ, measurable nonnegative integrable gas density μ\muμ of integral 111, and jointly measurable complex amplitude fff. Put m∗=mM/(m+M)m_*=mM/(m+M)m∗​=mM/(m+M), q(p,P)=(m∗/m)p−(m∗/M)Pq(p,P)=(m_*/m)p-(m_*/M)Pq(p,P)=(m∗​/m)p−(m∗​/M)P, x∥Q=((x⋅Q)/∥Q∥2)Qx_\parallel^Q=((x\cdot Q)/\|Q\|^2)Qx∥Q​=((x⋅Q)/∥Q∥2)Q, x⊥Q=x−x∥Qx_\perp^Q=x-x_\parallel^Qx⊥Q​=x−x∥Q​, b(p,P,Q)=(m∗/m)p⊥Q−(m∗/M)P⊥Qb(p,P,Q)=(m_*/m)p_\perp^Q-(m_*/M)P_\perp^Qb(p,P,Q)=(m∗​/m)p⊥Q​−(m∗​/M)P⊥Q​, and g(p,P,Q)=p⊥Q+(m/(2m∗))Q+(m/M)P∥Qg(p,P,Q)=p_\perp^Q+(m/(2m_*))Q+(m/M)P_\parallel^Qg(p,P,Q)=p⊥Q​+(m/(2m∗​))Q+(m/M)P∥Q​, where the dot product is the sum of coordinate products and division by zero is zero. Define c(0)=0c(0)=0c(0)=0, c(Q)=ϱm/(m∗2∥Q∥)c(Q)=\varrho m/(m_*^2\|Q\|)c(Q)=ϱm/(m∗2​∥Q∥) otherwise, and define Lv(p,P,0)=0L_v(p,P,0)=0Lv​(p,P,0)=0, while Lv(p,P,Q)=c(Q)f(b(p,P,Q)−Q/2,b(p,P,Q)+Q/2)μ(g(p,P,Q)−mv)L_v(p,P,Q)=\sqrt{c(Q)}f(b(p,P,Q)-Q/2,b(p,P,Q)+Q/2)\sqrt{\mu(g(p,P,Q)-mv)}Lv​(p,P,Q)=c(Q)​f(b(p,P,Q)−Q/2,b(p,P,Q)+Q/2)μ(g(p,P,Q)−mv)​ otherwise; write L=L0L=L_0L=L0​. Define E(P)=−2πℏ2ϱm∗Re⁡∫μ(p)f(q(p,P),q(p,P)) dpE(P)=-\frac{2\pi\hbar^2\varrho}{m_*}\operatorname{Re}\int\mu(p)f(q(p,P),q(p,P))\,dpE(P)=−m∗​2πℏ2ϱ​Re∫μ(p)f(q(p,P),q(p,P))dp and Ev(P)=−2πℏ2ϱm∗Re⁡∫μ(p−mv)f(q(p,P),q(p,P)) dpE_v(P)=-\frac{2\pi\hbar^2\varrho}{m_*}\operatorname{Re}\int\mu(p-mv)f(q(p,P),q(p,P))\,dpEv​(P)=−m∗​2πℏ2ϱ​Re∫μ(p−mv)f(q(p,P),q(p,P))dp. Let FQ:T→R3F_Q:\mathcal T\to\mathbb R^3FQ​:T→R3 be real linear isometries perpendicular to QQQ, onto Q⊥Q^\perpQ⊥ whenever Q≠0Q\ne0Q=0, and define, for every complex kernel sss, Kv[s](A,B)=∫ ⁣∫[Lv(FQk,A−Q,Q)s(A−Q,B−Q)Lv(FQk,B−Q,Q)‾−∣Lv(FQk,A,Q)∣2+∣Lv(FQk,B,Q)∣22s(A,B)]dk dQ\displaystyle \mathcal K_v[s](A,B)=\int\!\int\left[L_v(F_Qk,A-Q,Q)s(A-Q,B-Q)\overline{L_v(F_Qk,B-Q,Q)}-\frac{|L_v(F_Qk,A,Q)|^2+|L_v(F_Qk,B,Q)|^2}{2}s(A,B)\right]dk\,dQKv​[s](A,B)=∫∫[Lv​(FQ​k,A−Q,Q)s(A−Q,B−Q)Lv​(FQ​k,B−Q,Q)​−2∣Lv​(FQ​k,A,Q)∣2+∣Lv​(FQ​k,B,Q)∣2​s(A,B)]dkdQ, with K=K0\mathcal K=\mathcal K_0K=K0​. For arbitrary v,P∈R3v,P\in\mathbb R^3v,P∈R3, assume p↦μ(p)f(q(p,P−Mv),q(p,P−Mv))p\mapsto\mu(p)f(q(p,P-Mv),q(p,P-Mv))p↦μ(p)f(q(p,P−Mv),q(p,P−Mv)) is integrable. Then L(p,P−Mv,Q)=Lv(p+mv⊥Q,P,Q)L(p,P-Mv,Q)=L_v(p+m v_\perp^Q,P,Q)L(p,P−Mv,Q)=Lv​(p+mv⊥Q​,P,Q) for every p,Qp,Qp,Q, including Q=0Q=0Q=0, and E(P−Mv)=Ev(P)E(P-Mv)=E_v(P)E(P−Mv)=Ev​(P). Furthermore, for every kernel rrr and momentum P′P'P′, put s(A,B)=r(A+Mv,B+Mv)s(A,B)=r(A+Mv,B+Mv)s(A,B)=r(A+Mv,B+Mv), A=P−MvA=P-MvA=P−Mv, and B=P′−MvB=P'-MvB=P′−Mv; if for almost every QQQ the three functions k↦L(FQk,A−Q,Q)L(FQk,B−Q,Q)‾k\mapsto L(F_Qk,A-Q,Q)\overline{L(F_Qk,B-Q,Q)}k↦L(FQ​k,A−Q,Q)L(FQ​k,B−Q,Q)​, k↦∣L(FQk,A,Q)∣2k\mapsto|L(F_Qk,A,Q)|^2k↦∣L(FQ​k,A,Q)∣2, and k↦∣L(FQk,B,Q)∣2k\mapsto|L(F_Qk,B,Q)|^2k↦∣L(FQ​k,B,Q)∣2 are integrable, and their kkk-integrals are integrable in QQQ, with the first multiplied by s(A−Q,B−Q)s(A-Q,B-Q)s(A−Q,B−Q), then K[s](P−Mv,P′−Mv)=Kv[r](P,P′)\mathcal K[s](P-Mv,P'-Mv)=\mathcal K_v[r](P,P')K[s](P−Mv,P′−Mv)=Kv​[r](P,P′). All integrals use volume measures; no positivity or normalization of rrr is assumed.

Human review
  • Endorsed by Shuze Chen · Oct 30, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 30, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me