UNNS SUBSTRATE RESEARCH PROGRAM · ²⁸Si LOCAL GEOMETRY VALIDATION · APRIL 2026
²⁸Si Local Geometry Validation
WP1 — Realizability Chart · WP2 — Empirical Boundary Surface · WP3 — Margin Monotonicity
FULL ×10 TAIL ×4 14 nuclear isotopes STRUC-PERC-I v2.4.1 ρ_Spearman = −0.915 p = 0.0002
CORPUS SYSTEMS
14
nuclear isotopes
FULL (interior)
10
GR=1.000 class
TAIL (boundary)
4
d_∂C = 0 class
DECISIVE DIM
2
tailDom, GR
SPEARMAN ρ
−0.915
m vs log κ_conn
p-VALUE
0.0002
Theorem 6.4 test
²⁸Si RANK
7/10
interior distance
USL VIOLATIONS
0
zero clean
WP1REALIZABILITY CHART CONSTRUCTION
DECISIVE COORDINATE SYSTEM — 2D NUCLEAR REALIZABILITY CHARTDefinitions 3.1–3.2 of manuscript
Φ(L) = (x₁, x₂) = (tailDom(L), GR(L)) = (max(Δ)/IQR(Δ), C(L))

Two decisive coordinates suffice in the nuclear domain (Lemma 3.3 — finite decisive reduction, d≤5). The tail-dominance coordinate x₁ = tailDom = max(Δ)/IQR(Δ) governs the FULL/TAIL class boundary. The giant-ratio coordinate x₂ = GR = C(L) governs the HARD boundary (C < 0.95). All 14 nuclear isotopes lie in the interior of the admissibility manifold (zero USL violations).

REALIZABILITY CHART — (tailDom, GR) SPACE — ALL 14 NUCLEAR ISOTOPES
x₁ = TAIL DOMINANCE [max(Δ)/IQR(Δ)] x₂ = GIANT RATIO [C(L)] 0.940 0.955 0.965 0.975 0.985 0.995 1.000 0.950 0.960 0.975 0.990 1.000 BOUNDARY G(x)=0 tailDom=1 GR = 0.95 (HARD boundary) FULL INTERIOR GR = 1.000 · tailDom < 1.0 TAIL ZONE ⁹⁰Zr ¹¹⁶Sn ²⁰⁸Pb ¹⁷⁴Yb ¹⁶⁶Er ¹⁵²Sm ²⁸Si ²⁴Mg ⁶⁰Ni ⁵⁶Fe ¹⁰⁰Mo GR=0.988 ²³⁸U GR=0.983 ⁴⁸Ca GR=0.976 ¹⁵⁰Nd GR=0.976 d_∂C = 0.010 m_est = 0.0100 ← m_est increases (higher margin, farther from boundary) FULL ²⁸Si (flagship) TAIL Boundary G(x)=0 HARD boundary
WP2EMPIRICAL BOUNDARY SURFACE
ACTIVE BRANCH — BOUNDARY FUNCTION G_j*(L)Sections 4–5 of manuscript
G_j*(L) = 1 − tailDom(L)

The active decisive branch in the nuclear domain is the tail-dominance threshold. By Lemma 4.2 (threshold representation): G_j*(x) is C¹ with ∇G_j* = (−1, 0) in (tailDom, GR) space — non-zero everywhere. By Theorem 4.3, the boundary G_j*(x)=0 is a codimension-1 C¹ hypersurface (here: the vertical line tailDom=1.0).

GRADIENT CHECK (Regularity, Assumption A2)
∇G_j*(x*) = (∂G/∂x₁, ∂G/∂x₂)
= (∂(1-tailDom)/∂tailDom, ∂(1-tailDom)/∂GR)
= (−1, 0) ≠ 0 ✓
REGULAR BOUNDARY POINT CONFIRMED
Implicit function theorem applies
LEMMA 5.3 CHECK — ACTIVE BRANCH DOMINANCE

Verify M_j*(L) < M_j(L) for j≠j* at L₀=²⁸Si. Two branches present:

Branch j*: tail-dominance
M_j* =
0.0100
Branch j: giant-ratio
M_j =
0.0500
δ = M_j(L₀) − M_j*(L₀) = 0.050 − 0.010 = +0.040 > 0
Lemma 5.3 neighbourhood: U' exists with m(L) = M_j*(L) ✓
LEMMA 6.2 — DISTANCE REPRESENTATIONBi-Lipschitz equivalence

In this chart, G_j*(x) = 1 − tailDom(L) equals the Euclidean distance to the boundary tailDom=1.0 exactly: G_j*(x) = d_∂C(L). The bi-Lipschitz constants are c₁=c₂=1.

c₁·d_∂C(L) ≤ G_j*(x) ≤ c₂·d_∂C(L) [c₁=c₂=1.0]
G_j*(x) / d_∂C(L) for all 10 FULL isotopes:
⁹⁰Zr: 1.000 ²⁸Si: 1.000
¹¹⁶Sn: 1.000 ¹⁵²Sm: 1.000
²⁰⁸Pb: 1.000 ⁵⁶Fe: 1.000
¹⁷⁴Yb: 1.000 ⁶⁰Ni: 1.000
¹⁶⁶Er: 1.000 ²⁴Mg: 1.000
All ratios = 1.000 — Lemma 6.2 verified exactly ✓
TAIL SYSTEMS — SECOND DECISIVE BRANCH

For TAIL systems (tailDom=1.000), the tail-dom branch gives G_j*(x)=0. The second active branch is the giant-ratio: G_gr(L) = GR − 0.95. This locates TAIL systems on the boundary hyperplane in chart dimension 1, with residual margin captured by the GR branch in dimension 2.

IsotopeGRG_gr=GR−0.95logMaxRNote
²³⁸U0.98290.03299.40Single extreme γ
¹⁰⁰Mo0.98840.038418.85Max GR in TAIL
¹⁵⁰Nd0.97580.025818.34Shape coexistence
⁴⁸Ca0.97610.026118.40Doubly-magic
All TAIL: G_j*(x)=0 (on boundary); G_gr(x)∈[0.026,0.038] (inside non-HARD zone)
WP3MARGIN MONOTONICITY TEST — THEOREM 6.4 EMPIRICAL VERIFICATION
PRIMARY RESULT — THEOREM 6.4 NUMERICALLY CONFIRMED

Spearman ρ(m_est, log κ_conn) = −0.9152 (p = 0.0002, n=10 FULL isotopes). Prediction of Theorem 6.4: d_∂C(L₁) < d_∂C(L₂) ⟹ m(L₁) < m(L₂). Operationally: lower margin = smaller boundary distance = harder to percolate = higher κ_conn. The strong negative Spearman correlation confirms this ordering is preserved across the full FULL interior. Linear fit: log(κ_conn) = −21.8 × m_est + 5.525 · Pearson r = −0.853.

FULL INTERIOR — RANKED BY MARGIN (= BOUNDARY DISTANCE)Theorem 6.4: lower rank = closer to boundary = harder to percolate
RankIsotopetailDom m_est = 1−Dd_∂C(L) κ_connlog(κ)Predicted κ order
1⁹⁰Zr0.94600.05400.054039,6234.598↓ lowest
2¹¹⁶Sn0.96600.03400.034037,5864.575
3²⁰⁸Pb0.96800.03200.032064,2624.808
4¹⁷⁴Yb0.98200.01800.018059,0684.771
5¹⁶⁶Er0.98300.01700.0170106,3775.027
6¹⁵²Sm0.98700.01300.0130176,2115.246
7²⁸Si ★0.99000.01000.0100157,4995.197★ FLAGSHIP
8²⁴Mg0.99200.00800.0080418,6775.622
9⁶⁰Ni0.99300.00700.0070397,2045.599
10⁵⁶Fe0.99600.00400.0040329,1085.517↑ highest
MONOTONICITY SCATTER — m_est vs log(κ_conn) — ALL 10 FULL ISOTOPES
m_est = 1 − tailDom (boundary distance) log₁₀(κ_conn) 0.004 0.014 0.024 0.034 0.044 0.054 4.50 4.75 5.00 5.35 5.65 r=−0.85 ⁹⁰Zr ¹¹⁶Sn ²⁰⁸Pb ¹⁷⁴Yb ¹⁶⁶Er ¹⁵²Sm ²⁸Si ²⁴Mg ⁶⁰Ni ⁵⁶Fe Spearman ρ = −0.9152 p = 0.0002 n = 10 Theorem 6.4 confirmed ✓
RESTHEOREM VERIFICATION SUMMARY
THEOREM 4.3 ✓
Local Boundary Hypersurface
The boundary G_j*(x) = 1 − tailDom = 0 is a codimension-1 C¹ hypersurface in the 2D decisive chart. ∇G_j* = (−1,0) ≠ 0 everywhere. Implicit function theorem applies. Verified analytically.
COROLLARY 6.3 ✓
Order Equivalence
G_j*(x) = d_∂C(L) exactly (ratio = 1.000 for all 10 FULL isotopes). Ordering d_∂C(L₁) < d_∂C(L₂) ⟺ m_est(L₁) < m_est(L₂). Verified numerically (all ratios = 1.000).
THEOREM 6.4 ✓
Local Margin Monotonicity
d_∂C(L₁) < d_∂C(L₂) ⟹ m(L₁) < m(L₂). Corpus test: Spearman ρ(m_est, log κ_conn) = −0.915 (p=0.0002). Lower margin ↔ closer to boundary ↔ higher κ_conn. Statistically confirmed.
²⁸Si POSITION IN THE LOCAL CHART

²⁸Si sits at rank 7 of 10 in the FULL interior (m_est = 0.010, d_∂C = 0.010, κ_conn = 157,499). It is in the boundary-adjacent zone — closer to the FULL/TAIL boundary than 6 of the 10 FULL isotopes, but not the closest (⁵⁶Fe, d=0.004, is the most boundary-adjacent FULL system). Its tail dominance 0.990 is 3rd highest among FULL systems, confirming its position as a transition-sensitive interior probe — the primary qualification for the local geometry validation. Under α-deformation (changing fine-structure constant), ²⁸Si is expected to show the clearest monotonic margin decrease as the boundary is approached, because its starting margin (0.010) is small enough for boundary effects to be visible without requiring extreme deformation amplitudes.

UNNS SUBSTRATE RESEARCH PROGRAM · unns.tech ²⁸Si LOCAL GEOMETRY VALIDATION · STRUC-PERC-I v2.4.1 · APRIL 2026 WP1 · WP2 · WP3 COMPLETE