No joint field · Bridge A = SPECULATIVE pedagogy

Labels: SOURCED HYPOTHESIS SPECULATIVE OPEN NONE (REAL = same-domain materials or linguistics, shown like SOURCED)

Path F · status

Conclusions

Status for a syntax and semantics reader. Materials claims and linguistics claims below are REAL inside their own fields. Cross-domain claims are SPECULATIVE. There is no joint field.

The false-friends table has its own page: false friends.

Mandate

There is no established field joining metal/semiconductor doping with linguistic valency or θ-roles. The week’s job was to (1) get both domains right for a linguist who wants alloy-design intuition, then (2) test whether Bridge A earns its keep or collapses.

Lock Status at close
Bridge A (doping concentration windows / over-doping ↔ valency / argument saturation / over-arity) Still locked as preferred SPECULATIVE working focus (count / window / overfill).
Teaching front door Chem-formula valence ↔ verb valency (Chem for linguists), then corrected.
θ-roles (Navajo-led) Typed-participant half — parallel to Bridge A, not proof of Bridge A count.
Triskelion / Dowty / Merge·Agree / kanji / morphology waypoints SPECULATIVE secondary; filed; not elevated.
Joint field Never claimed; week found no contrary primary literature under another name (Prior bridges + notes).

What each domain got right

2.1 Materials / doping (REAL)

Concept Honest takeaway Primary pointers
Electronic doping Controlled impurity → carrier type/density; concentration windows matter; overrun modes include impurity banding / Mott MIT, clustering, deactivation Physics of doping; Wei et al. 2022 Chem. Soc. Rev. (chemical doping review)
Solid-solution strengthening (SSS) Foreign atoms pin dislocations (size + modulus misfit; Labusch statistical theory; Curtin-line refinements) — mechanical, not carrier engineering Fleischer 1963; Labusch 1970; Varvenne/Curtin line; Solid solution vs valence
Chemical valence (alloy talk) Combining capacity / e/a / Hume-Rothery solubility heuristics — upstream of strength formulas Bhadeshia Hume-Rothery digest; Chemical interaction vs valence effect
Solubility windows Phase diagrams set where a host can hold solute; “too much” is host- and goal-dependent Solubility windows; Al–Mg / Cu–Ni checklists

Hard separation kept all week: SSS ≠ electronic doping even though both put foreign atoms in a host. “Valence / chemical” in alloy papers is a false-friend cluster (see §5).

2.2 Linguistics / valency–θ (REAL)

Concept Honest takeaway Primary pointers
Valency / arity How many arguments a predicate requires; over-arity is a discrete well-formedness failure Linguistics: theta; chem-for-linguists front door
θ-roles / thematic relations Typed participants the event structure needs; θ-criterion / proto-roles (Dowty) are theory-internal tools Dowty 1991 sister note; Navajo morphology primer
Navajo Roles often saturated on the verb (pronominal args / classifiers / template); lexical NPs frequently optional — key contrast to English NP stacking secondary index; Grant clarification 2026-09-07
Kanji Orthography/lexicon packaging — not syntactic heads; no kanji↔doping track secondary index

Bridge A verdict (week close)

3.1 Earns (keep)

Bridge A earns as pedagogy + probe, not as identity science.

Count / window ladder ( SPECULATIVE supporting case):

Zone Linguistics (familiar) Materials (honest)
Under Missing required argument Dilute / ineffective concentration
Useful Saturated intended frame Composition inside a process + service window
Over Over-arity / excess core arg Regime change: second phase, sensitization, Mott MIT, deactivation, clustering, or property-curve overrun

Earn criteria met:

  1. Count/window used as a ladder, not a theorem.
  2. Phase overrun and electronic MIT kept as separate failure modes.
  3. Never promoted to “θ-roles are dopants.”

Alloy-design earn (REAL hosts): “too much” is not one physics.

Host Diagram First “over” Bridge A peg ( SPECULATIVE)
Al–Mg AA5083 Limited solubility / solvus GB β / sensitization (IGC) after supersaturation + warm service (~40–220 °C; driving force above ~3.5 wt% Mg) Solvus overrun ≈ overfill that changes phase cast
Cu–Ni Isomorphous complete solid solution Conductivity / TCR property curves (Nordheim-family scattering); ~45% Ni → resistance-alloy regime Property-curve overrun ≈ overfill that misses the design goal without a Ni second phase
Fe–C steel Eutectoid Fe–Fe₃C; interstitial C (α ≪ γ solubility) Cementite networks / brittle as-quenched martensite / temper–RA issues — goal-dependent; metastability first-class Transformation / morphology overrun ≈ overfill that changes phase cast or quench regime (Fe–C)

Checklist that survives: diagram first → name mechanism → name what fails past the window → metastability risk → only then hang the SPECULATIVE count mnemonic.

3.2 Collapsed / hard stops

Path Verdict Why
Joint field / identity science Collapsed / never claimed No shared literature; continuous composition ≠ discrete lexical arity
Mott nc = linguistic saturation number Collapse Electronic MIT ≠ solvus; not an arity integer
β precipitation = “adjunct” Collapse Phase kinetics ≠ adjunct licensing
SSS “chemical interaction” → θ-saturation Default collapse Curtin-line “chemical” = dislocation U / SFE proxy
Chen–Boyle “valence effect” → linguistic valency Default collapse DFT bonding correlate of Mg SSS; title false friend
Linde Δρ ∝ (Δ Z)2 → ling. valency Collapse Scattering lore ≠ arity
Kanji ↔ doping Collapsed by Grant Morphology↔valency packaging only
Navajo θ-morphology as evidence for Bridge A count Collapse as proof Navajo = typed-participant half; Bridge A = count/windows half
Four-place verbs as universal alloy templates Collapse Rare/disputed; optional elaborations only
“Doped Al/Cu” for percent-level Mg/Ni SSS Avoid Prefer solute / alloying; reserve “dopant” for dilute electronic (etc.) intent

Asymmetries on the whiteboard: linguistics is often categorical well-formedness; materials are graded optima + kinetics + metastability. Vacancy / carrier / dopant are not one “argument” type. Metals already have an electron sea — chassis alloys are not “doped for conductivity” by default.

Lock status: Bridge A did not collapse this week. It remains locked as the preferred SPECULATIVE supporting analogy for count/windows. It does not earn as shared formalism.


Lock after the later notes

Lock Status
Bridge A (concentration windows / over-doping ↔ valency / saturation / over-arity) Still locked as preferred SPECULATIVE pedagogy / probe only (count / window / overfill).
Joint field / shared formalism Never claimed; still not claimed.

Earn criteria from CONCLUSIONS-v0 §3 still hold: materials first → name mechanism → name failure past the window → metastability → only then hang the SPECULATIVE count mnemonic. Nothing in the new host notes or the process-metastability note promotes Bridge A to identity science.


Sixteen host pegs

The linguist start path is the reading order. The addendum’s contrast table and the paragraphs under it follow. The lock is unchanged.

Host contrast (sixteen REAL pegs; “over” is not one physics):

Host Diagram / lattice First “over” (typical) Bridge A peg ( SPECULATIVE)
Al–Mg AA5083 Limited solvus; FCC Al + subst. Mg GB β / sensitization (IGC) Solvus overrun → unwanted second phase
Cu–Ni Isomorphous complete SSS Conductivity / TCR property curves Property-curve overrun (no Ni second phase)
Fe–C Eutectoid Fe–Fe₃C; interstitial C Cementite networks / brittle as-quenched M / temper–RA Transformation / morphology overrun
Ni–γ′ γ/γ′ solvus; wanted precipitate Overaging / rafting gone wrong / TCP Wanted γ′ vs wrong second phase
Mg (host) HCP; AZ/AM/ZK (+ RE) Brittle IMCs; galvanic β; creep; ignition; anisotropy miss β ambivalent (help or hurt); slip-system / safety “over”
Ti (host) HCP α / BCC β; β-transus; Ti-6Al-4V α-case; O/H embrittlement; ω; β flecks; overaging Wanted α morphology vs unwanted ω / α-case / hydride
Oxide electronic (TiO₂ / ZnO) Wide-gap oxide + subst. metal + V_O Recombination centers; μ collapse; secondary oxides; vacancy compensation Carrier / mid-gap window overrun — true dopant language (cleanest electronic-doping Bridge A case)
CNT/graphene intercalation sp² framework + guest in channels/gaps Moisture dedoping; diffusion blur; junction R; no band gap for switching Intercalant loading / stability window — true intercalant language
GaN:Mg (III–V) Wurtzite GaN + Mg_Ga acceptor (+ H) Self-compensation; PIDs; μ/optical loss; failed activation Carrier window on industrial electronic doping — true dopant
Si:P/As/B (classical) Diamond Si + shallow subst. donors (P, As) / acceptors (B) Impurity band / Mott; clustering; μ collapse; compensation Carrier window on textbook classical electronic doping — true dopant
Ni-rich NMC:Mg/Al/Nb (battery cathode) Layered oxide + dilute Mg/Al/Nb lattice stabilizers Inactive capacity; Li-path blocking; secondary phases; thermal still fails if cutoff insane Structural-stability window — papers say doping; mechanism ≠ Si:P / GaN:Mg
Cantor CoCrFeMnNi / HEA Equiatomic FCC MPEA; five co-principal alloying elements Cryogenic nanotwinning / TWIP; composition or long-anneal phase drift SPECULATIVE, harder peg: multi-component continuous space; alloying ≠ dilute doping
TWIP Fe–Mn (austenitic) FCC γ Fe–Mn–C(–Al–Si); SFE bands (~15–30 wt% Mn) Too-low SFE → TRIP/ε; too-high → no twin; H delayed fracture SPECULATIVE: SFE window as continuous saturation analog — alloying, not dilute doping
Al–Li (aerospace) FCC Al + Li (+Cu…); δ′ Al₃Li / T₁ Al₂CuLi Anisotropy / toughness (planar slip) / SCC / overaging / melt Li loss Wanted δ′/T₁ + density/E vs wrong GB/coarse phases — alloying / precipitation, not dilute doping
CoCrMo (biomedical) Co-majority FCC γ (+ ε); ~28Cr–6Mo; F75/F1537; SIMT + M₂₃C₆ Brittle carbide networks; Cr depletion; uncontrolled ε; fretting/MACC ions Wear/passivity/fatigue + SFE/SIMT + carbide morphology — alloying, not dilute doping
Zr nuclear (cladding) HCP α service / BCC β process; Zircaloy / Zr–Nb; SPPs Nodular/uniform corrosion; hydriding/DHC; LOCA breakaway; SPP fluence evolution; growth/bow Neutron economy + waterside + irradiation anisotropy — alloying / SPP, not dilute doping

Files: Al–Mg, Cu–Ni, Fe–C, Ni–γ′, Mg host, Ti host, Oxide electronic, CNT / graphene, GaN:Mg, Si:P/As/B, NMC:Mg cathode, Cantor HEA, TWIP Fe–Mn, Al–Li, CoCrMo, Zr nuclear.

Contrast worth naming once: Al–Mg was Mg as solute; Mg checklist is Mg as host (β role inverted). Ti shares allotropy intuition with Fe–C but speaks α/α+β/β + β-transus, not Fe–Fe₃C eutectoid. Oxide electronic doping is the cleanest research-leaning electronic-doping Bridge A case (true dopant + carrier / mid-gap / V_O). GaN:Mg is the industrial deep-acceptor electronic-doping host: mature III–V LED/laser p-type doping (Mg_Ga + H activation). Si:P/As/B is the textbook classical shallow-dopant baseline (CMOS / IC maturity ceiling) that oxide research and GaN:Mg deep-acceptor doping are measured against. CNT/graphene intercalation completes the alloying vs doping vs intercalation triangle: intercalant / non-covalent charge-transfer in channels/gaps — not oxide substitutional+V_O, not metal SSS, not GaN/Si acceptor/donor doping. Ni-rich NMC:Mg/Al/Nb is the battery-cathode structural-stabilizer peg: papers say “doping,” but the useful meter is cycle life / thermal margin / cation order — not carrier density. Cantor CoCrFeMnNi / HEA is the MPEA/alloying contrast: five co-principal elements, not dilute dopants; cryogenic TWIP makes the Bridge A peg harder because composition occupies a multi-component continuous space. TWIP Fe–Mn (austenitic high-Mn steel) is the one-principal Fe-host twinning cousin: SFE window (~20–40 mJ m⁻² class) as the real design meter; prefer alloying / TWIP steel — never dilute doping; checklist TWIP Fe–Mn; Bridge A SPECULATIVE after materials (soft band edges; HE clock ≠ SFE alone; Cantor ≠ Fe–Mn product identity). Cross-link Cantor cryogenic TWIP-like fame vs automotive Fe–Mn TWIP. Al–Li (aerospace) is the fourteenth host peg: same FCC Al host as Al–Mg AA5083 but Family 4 precipitation (wanted δ′ / T₁) plus density/modulus meters — Ni–γ′ wanted-precipitate cousin, not SSS sensitization; prefer alloying / precipitation — never dilute doping; checklist Al–Li; Bridge A SPECULATIVE after materials (plural “over” modes: toughness/anisotropy/SCC/overage/melt Li loss). Family-contrast depth: Precipitation family. CoCrMo (biomedical, ASTM F75/F1537-class) is the fifteenth host peg: Co-majority + major Cr/Mo solutes; low/negative SFE → stacking faults + γ→ε SIMT; cast M₂₃C₆ carbides; wear/passivity/fatigue + modular-taper MACC meters — share low-SFE FCC family with TWIP/Cantor but target ε-martensite / implant duty, not automotive twinning or cryogenic HEA toughness; prefer alloying / Co–Cr–Mo implant alloy — never dilute doping; checklist CoCrMo; Bridge A SPECULATIVE after materials (plural windows: carbide morphology / SFE-SIMT / fretting). Cross-host SFE depth: SFE window. Zr nuclear (cladding / pressure-tube, Zircaloy / Zr–Nb class) is the sixteenth host peg: third HCP host after Mg and Ti, but duty meters are neutron economy + waterside corrosion / hydriding + irradiation growth/creep (not Mg formability or Ti specific strength); controlled SPPs (Zr(Fe,Cr)₂, β-Nb, …) + texture; prefer alloying / solute / SPP — never dilute doping (oxide-film “Fe doping” ≠ Family 1 of α-Zr metal); checklist Zr nuclear; Bridge A SPECULATIVE after materials (plural “over” modes: corrosion morphology / hydride / LOCA breakaway / SPP fluence / anisotropic growth). The Triskelion ditransitive↔triple-junction analogy remains SPECULATIVE/parked.

Same-element Mg across three hosts (name once): GaN:Mg = true electronic acceptor (+ H); NMC:Mg = lattice / chemo-mechanical stabilizer (pillar/rivet); Al–Mg SSS = solute pinning — same foreigner label, three mechanism families. Host first (Grant takeaway §9.1). Depth note: Same-element Mg (host-first / element≠role; GaN vs NMC vs Al–Mg + Mg-host nod). Bridge A still SPECULATIVE after materials — count shape may transfer; mechanism labels must not.

Path D vocabulary hygiene (name once): Doping vs alloying vocabulary — when to say dopant vs alloying / solute / intercalant / stabilizer across the sixteen host pegs (true electronic doping vs papers-say-doping NMC vs MPEA/TWIP/CoCrMo never dilute doping). Companion to mechanism taxonomy + valence false-friend notes. Bridge A still SPECULATIVE after materials; community verb ≠ mechanism family.

HCP trio (name once): HCP host contrast contrasts Mg · Ti · Zr as a shared HCP constraint family (limited easy slip systems, texture/anisotropy, O/H/N/C sensitivity) with divergent duty meters and plural “over” modes: Mg density/formability/galvanic/ignition; Ti specific strength/α–β/α-case/ω; Zr neutron economy/waterside/hydriding/SPPs. Prefer alloying / solute / SPP / stabilizer, never dilute doping. Bridge A remains SPECULATIVE after materials; shared HCP ≠ one saturation meter.

Electronic doping trio (name once): Electronic doping family contrasts Si:P/As/B · GaN:Mg · oxide (TiO₂/ZnO-class) as mechanism taxonomy Family 1 — shared carrier-tuning surface and honest dopant vocabulary, with divergent ionization (shallow ≠ deep ≠ mid-gap/V_O), maturity, and plural “over” modes (Mott/clustering; self-comp/PIDs/H activation; recombination/μ/V_O). Foils: intercalation; NMC “papers say doping”; Al–Mg SSS. Bridge A remains SPECULATIVE after materials; continuous N_D/N_A ≠ discrete arity.

Allotropy / transformation windows (name once): Allotropy windows contrasts Fe–C (eutectoid + quench–temper; TTT/CCT) · Ti (α/β; β-transus; Ti-6Al-4V STA as process-clock writing of α morphology) · Zr nuclear (α service / β process; SPPs) — shared surface pattern of phase change under T/rate/composition/stress, not one physics. Hard splits: ≠ electronic doping; ≠ Al–Mg SSS solvus overrun alone; ≠ wanted precip (Ni–γ′/Al–Li) alone; shared HCP ≠ shared duty (HCP host contrast); CoCrMo SIMT γ→ε is deformation/SFE-linked foil — not thermal allotropy (CoCrMo; SFE window). Prefer alloying / solute / stabilizer / SPP / carbide — never dilute doping. Bridge A remains SPECULATIVE after materials; continuous wt%/T/rate ≠ discrete arity. Partially closes optional polish “Ti STA in process-metastability” by teaching STA on the process clock (full subsection in Process metastability still optional later).

SSS Family 3 contrast (name once): SSS family contrasts Al–Mg AA5083 (limited solvus; GB β / sensitization as service-clock “over”) · Cu–Ni (isomorphous complete SSS; conductivity / TCR property-curve “over” with no Ni second phase required) · brief Mg-host nod (SSS cohabits with precip/safety meters). Shared Family 3 surface = solute ↔ dislocation; divergent diagram and meters. Prefer solute / alloying — never dilute doping. Hard splits: ≠ electronic doping; ≠ wanted precip (Ni–γ′/Al–Li); ≠ SFE/TWIP window alone. Bridge A remains SPECULATIVE after materials; continuous wt%/%Ni/IACS ≠ discrete arity.

Alloying · doping · intercalation triangle (name once): Alloying · doping · intercalation expands chem-for-linguists §5 as a Path A/B depth/synthesis note — Family 3 SSS (Al–Mg / Cu–Ni; solute) · Family 1 electronic (Si:P/As/B brief; dopant) · Family 2 intercalative CT (CNT/graphene AlCl₄⁻; intercalant) · foil Ni-rich NMC:Mg (papers say doping; meter = cycle life / thermal / cation order ≠ Si:P). Shared surface = intentional foreigners; divergent slots (dislocation pins · carriers · guest CT · lattice stabilizer). Completes Family 2’s place in the contrast gallery after Family 1 / 3 / 4 / allotropy / SFE / HCP. Bridge A remains SPECULATIVE after materials; continuous windows ≠ discrete arity; community verb ≠ mechanism family.

Plural “over” modes (name once): Plural “over” modes catalogs overrun kinds across the sixteen host pegs and already-filed family contrasts — carrier-window (Family 1) · intercalant loading/stability (Family 2) · SSS solvus/property-curve (Family 3) · wanted/unwanted second phase (Family 4) · SFE/plasticity-mode · allotropy/transformation · NMC papers-say-doping stabilizer foil. Shared surface = past a useful window something fails; not one physics, meter, or clock. Same wt% can be useful then “over” later (Process metastability). Same-element Mg has three “over” stories (GaN ≠ NMC ≠ Al–Mg). Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming mechanism + this host’s “over” + which clock; continuous windows ≠ discrete arity; “over” is not one physics is why Bridge A stays pedagogy not identity science.

Clocks × “over” modes (name once): Clocks × over modes joins the three clocks (Process metastability) with the plural “over” catalog (Plural “over” modes) — matrix of which clock bites which overrun kind across sixteen host pegs + family contrasts. Shared surface = amount/path/time and process-writes / service-erases; not one clock priority or one “over” physics. Al–Mg service GB β ≠ Cu–Ni property-curve; Family 1 dose+activation ≠ Family 2 humid dedoping; Ti STA process-clock α morphology; same-element Mg has three clock stories. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming mechanism + “over” kind + which clock; synchronic count alone is incomplete.

Same-element H across four hosts (name once): Same-element H contrasts H as GaN:Mg passivator (Mg–H; process activation; Mg remains the true dopant) · Ti interstitial / hydride embrittlement (mill ~150 ppm H class lore) · Zr nuclear pickup → hydrides / DHC (radial vs circumferential; service waterside + fluence) · TWIP Fe–Mn delayed fracture / HE (can fail a “perfect SFE”). Same element label; four jobs; Path D prefers passivator / interstitial / hydride / embrittler / pickup — never flatten to “H dopant.” Foil: Fe–C interstitial C ≠ H. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming host + H’s job + meter + clock; continuous H content ≠ discrete arity.

Same-element Al across hosts (name once): Same-element Al contrasts Al as FCC host (Al–Mg AA5083 SSS chassis + brief Al–Li Family 4 δ′/T₁ nod) · Ni-rich NMC Al³⁺ structural stabilizer (Al–O lore; Mg/Nb siblings; papers say doping; meter = cycle life / thermal / O-release) · TWIP Fe–Mn alloying addition / SFE lever (e.g. Fe–18Mn–1.5Al–0.6C class; HE sibling clock). Same element label; three jobs (+ intra-host SSS vs precip split); Path D prefers host / stabilizer / alloying addition — never flatten to “Al dopant.” AlGaN / Al-in-Si electronic doping not filed — skip. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming host + Al’s job + meter + clock; continuous Al content ≠ discrete arity.

Same-element Ni across hosts (name once): Same-element Ni contrasts Ni as Cu–Ni isomorphous SSS solute (conductivity/TCR property-curve; no Ni second phase required for “over”) · Ni–γ′ FCC host / chassis (Al/Ti γ′-formers; overaging/TCP) · Zircaloy dilute H-pickup co-actor (Zry-2 raises pickup → Zry-4 removes Ni) · Ni-rich NMC majority TM host (Mg/Al/Nb are stabilizer foreigners — role inversion vs NMC:Al/Mg notes) · brief Cantor co-principal nod (equiatomic CoCrFeMnNi; not dilute dopant). Same element label; four jobs (+ equiatomic nod); Path D prefers solute / host / pickup co-actor / TM host / co-principal — never flatten to “Ni dopant.” Ni-in-Si / catalysts / plating not filed — skip. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming host + Ni’s job + meter + clock; continuous Ni content ≠ discrete arity.

Same-element Mn across hosts (name once): Same-element Mn contrasts Mn as TWIP Fe–Mn major alloying addition / austenite stabilizer / primary SFE lever (~15–30 wt% Mn; Pierce Mn series; never “Mn dopant”) · Cantor equiatomic co-principal (~20 at.%; also filed Mn-driven oxidation service failure — still co-principal) · Ni-rich NMC minority TM of the host suite (e.g. NMC811 LiNi₀.₈Mn₀.₁Co₀.₁O₂ — Mn with Ni/Co, not the Mg/Al/Nb stabilizer foreigner; role inversion vs NMC:Mg/Al notes and vs Ni majority-TM in the Ni same-element note) · brief Fe–C hardenability alloying foil (filed Mn/Cr/Mo/Ni list only). Same element label; three jobs (+ hardenability foil); Path D prefers alloying / austenite stabilizer / SFE lever · co-principal · minority TM (host suite) — never flatten to “Mn dopant.” Mn:Si / LMO / catalysts / CoCrMo-Mn not filed — skip. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming host + Mn’s job + meter + clock; continuous Mn content ≠ discrete arity.

Same-element Co across hosts (name once): Same-element Co contrasts Co as CoCrMo biomedical majority host / chassis (Co–28Cr–6Mo; F75/F1537; SIMT γ→ε; wear/passivity/fatigue — never “Co dopant”) · Cantor equiatomic co-principal (~20 at.% in CoCrFeMnNi) · Ni-rich NMC minority TM of the host suite (e.g. NMC811 LiNi₀.₈Mn₀.₁Co₀.₁O₂ — Co with Ni/Mn, not the Mg/Al/Nb stabilizer foreigner; completes NMC TM suite beside Ni majority / Mn minority same-element rows; role inversion vs NMC:Mg/Al notes) · brief Ni–γ′ SSS strengthener in γ foil (filed Cr,Co,Mo,W,Re,… list only). Same element label; three jobs (+ SSS-in-γ foil); Path D prefers host / Co–Cr–Mo implant alloy · co-principal · minority TM (host suite) — never flatten to “Co dopant.” Co:Si / LCO / catalysts / plating / TWIP-Co not filed — skip. Bridge A remains SPECULATIVE after materials; hang under/useful/over only after naming host + Co’s job + meter + clock; continuous Co content ≠ discrete arity.


Three clocks

Process metastability is now required reading next to the solubility-windows depth note — Bridge A’s kinetics half, not an optional footnote.

Three clocks (name which one before hanging under/useful/over):

Clock Answers
Equilibrium diagram Which phases are allowed at (T, C)
Process path What solution / quench / cool / age captures
Service life Whether the useful microstructure survives duty T–time–stress

Same wt% can be useful on the diagram, useful after quench, and “over” after warm service. Al–Mg sensitization, Fe–C quench→temper, and Ni–γ′ age→service all train this; Ti STA and Mg cast/age paths reuse it.


Design intuition

  • Name the goal (strength, corrosion, conductivity, carrier type) before reading “valence” or “doping” in a paper.
  • Read the diagram type (limited vs isomorphous) before applying Bridge A’s second-phase peg.
  • Treat metastable supersaturation (AA5083 at service T) as a first-class risk, not an afterthought.
  • Use Bridge A’s under/useful/over ladder only after mechanism + failure mode are named.

Secondary strands (status only)

Strand File(s) Status
Triskelion Prayerant secondary index, ingest REAL metallurgy (Cantor HEA, TWIP/TRIP, GBE) vs SPECULATIVE ditransitive↔triple-junction kept separate
Dowty 1991 proto-roles secondary index; chem-for-linguists §7 SPECULATIVE sister / mnemonic; secondary to Bridge A count
Merge·Agree chemistry probe secondary index SPECULATIVE waypoint; honest negative on prior joint lit
Valency-via-morphology secondary index Secondary waypoint
Kanji / MeasureP packaging secondary index Morphology↔valency only; no doping overlap
Navajo θ on V secondary index, clarification Typed-participant half; parallel to Bridge A, not proof of it

secondary index is the SPECULATIVE stress-test of the typed-participant half (Navajo θ-on-V exhibit × mechanism-family casts). Parked under Path E / θ-roles secondary — parallel to, not proof of, Bridge A count. Best-fit family: substitutional electronic doping; SSS / precipitation = foils. Does not unlock a joint field; Bridge A lock unrevised.

Companions: secondary index, secondary index, Mechanism taxonomy.


Derivational movement (2026-10-02) is a separate note. It is movement. It is not the typed-participant map, and the two claims are not merged.

REAL linguistics, from the sources that note opens. Under the pronominal-argument derivation that was drawn, arg1 (the external argument) merges in Spec,vP and does not move; it is spelled out as the subject exponent inside the verb word. Spec,M as an EPP landing site is a flagged overlay the opened Hale essay rejects — not a drawn movement to Spec,TP. Polar questions (da’ / -ísh) do not extract arguments. Mary ha’át’íí nayiisnii’ keeps the h-word in object position and 3–3 morphology on the verb; no covert wh-movement to Spec,CP is added. No yi/bi tree: those examples were not legible. The chemistry half is SPECULATIVE pedagogy only, and it is not Bridge A. Leaching and evaporation are the wrong parallel.

The boil pair on that page is the short ł alternation. It is not the causative scout, and it is not the lexicalization note.


Lexicalization (2026-10-03) is a separate note. It asks what is listed. It is not the movement derivation, and it is not the typed-participant map. The three claims are not merged.

REAL description, from McDonough on Young & Morgan and from Hale’s opened essay. The listed lexical unit is the verb base (classifier + stem). The verb theme is the looser non-compositional prefix + base. Beside Grant’s coinage “phrasisized,” the literature’s terms are verb theme, lexicalized base, idiomatic preverb–verb, and incorporated postposition. Those are not English phrasal verbs. Hale: Navajo has no verbal noun incorporation; Northern Athabaskan does; Navajo allows restricted noun-into-postposition incorporation. Grant’s phrase↔molecule line and his dopant guess stay HIS and SPECULATIVE. Not Bridge A. Not a joint field.


Causative split (v0.1, deepen 2026-10-04; was 2026-10-03 scout) is a separate scout. Not a derivation. Not Japanese lexical vs productive -sase. Not merged with movement, lexicalization, or the typed-participant map. v0.2 (2026-10-06): Goertz et al. 2006 print Navajo analytic causatives, all reason/circumstance clauses in which the cause is a postpositional object or a separate clause, not a causer argument; in what is opened, a causer enters the verb only through ł morphology. WALS 111A codes Navajo “No morphological type or compound type,” with no source on the datapoint, against every Navajo source opened. Still not the Japanese split. v0.3 (2026-10-07): the same paper’s stative/positional causatives, read off page images, put one stem through three event shapes (change by the subject; change brought about by a causer; position held by a causer), so the ł-causative can record maintaining as well as initiating, with the held participant as object though it undergoes no change. An event-structure cut; not lexical vs productive; no chemical correlate proposed.

Dopant produce vs maintain (2026-10-07 evening) takes the scout's logged chemistry question to metal alloys. In their own terms metallurgists do separate producing a grain structure (inoculant nucleation and solute growth restriction while the melt freezes; particle-stimulated nucleation) from holding one (Zener pinning, solute drag, segregation: "inhibit", "stabilize", "preserving the fine-grained structure"), and Cottrell & Bilby estimate an atmosphere's "rate of formation" apart from its "strength of pinning" — but the line is drawn by stage and physical form, not by element: scandium does both only as two Al₃Sc populations at two stages, Sc and Zr split one dispersoid's early and late life, and coarse vs fine particles split by size inside recrystallization. The nanocrystalline literature splits maintaining itself into kinetic (driving force remains, mobility reduced) and thermodynamic (driving force removed); its equilibrium models carry no produce/maintain direction. Navajo marks bring-about vs keep on one stem with one causer. Analysis only; Grant's frame SPECULATIVE; no chemical stand-in for a theta role; Bridge A unrevised.

Dopant DSA / Portevin–Le Chatelier (2026-10-08 evening) takes up produce–maintain open question 5: dynamic strain ageing / Portevin–Le Chatelier, where solute atmospheres pin, free and re-pin dislocations inside one deformation. Opened sources (Tamimi et al. 2015; Aboulfadl et al. 2015; Cottrell 1990; Epperly & Sills SAND2018-12622C; Yilmaz 2016 PMC HTML) name atmosphere formation rate apart from pinning strength, and a cycle of temporary arrest, diffusional locking, unpinning and re-ageing; DSA is the diffusional interaction and PLC the macroscopic collective instability (DSA can occur in a wider window than serrations). Al–Mg (AA5083-class map) and Fe–C (Cottrell line) are the opened hosts. Analysis beside Navajo bring-about vs keep and the produce/maintain cut; Grant's frame SPECULATIVE; no chemical stand-in for a theta role; Bridge A unrevised.

Dopant thermodynamic stabilizer: hold or produce (2026-10-09 evening) takes up produce–maintain open question 3: whether a thermodynamic grain-size stabilizer (solute segregation lowering grain-boundary energy toward zero) is ever described as producing fine grains rather than only holding them. In two opened lattice simulations (Hussein & Mishin 2024, 2025) it does both through one mechanism — refining a single crystal or fragmenting a large grain, and stopping coarsening, toward the same state — but production is gated by solute arrival (slow solute → the new grain shrinks away), and the held state is a dynamic growth/refinement balance. In the one opened experiment (Liu et al. 2024, Ni in TiZrNbHf films) deformation makes the grains and Ni keeps them; full stabilization is, per the simulation authors, undemonstrated experimentally; Cu–Co (Bachmaier et al. 2015) is stable by phase separation instead. Analysis beside Navajo bring-about vs keep; Grant's frame SPECULATIVE; no chemical stand-in for a theta role; Bridge A unrevised.

Si donors: slot or element? (2026-10-14 evening) takes up Grant's open question of 2026-10-14 (a doping process as an operator over typed sites rather than particular elements) at its cleanest published test case: shallow group-V donors. Per Saraiva et al. 2014, the effective-mass binding energy depends only on the host and holds in single-valley GaAs; in silicon the excited 1s states (nodes at the impurity) are nearly species-independent and close to the slot-only value, while the ground state varies strongly with the element because species differ at the core. So the slot sets the template and the filler enters as a core-local correction whose size the host decides. This bears on the result of doping, not on a process choosing slots (not opened). Navajo comparison SPECULATIVE; no chemical stand-in for a theta role; Bridge A unrevised. Navajo findings vs materials: a scorecard (2026-10-15 afternoon) takes the four Navajo findings that survived testing on Navajo Decoded (a role-bearing slot as a bundle of person, number, capacity rank and role, with nouns linking by feature match; ranking by general capacity, then topicality; layered assignment, Grant's conjecture; viewpoint and causation in the word, telicity not) and looks for a sourced materials analog of each, with verbatim excerpts and a counter-case for each. Graded: STRONG for the lattice-strain model of a crystal site (Blundy & Wood 2003; Shimizu et al. 2017: the site carries its own ideal radius, stiffness and maximum partition value; the ion enters through radius and charge) and for the failure of a context-free rank (Goldschmidt's "smaller ion first" rule is invalidated: "Of critical importance is the size of the ions relative to the site of interest"; the same two ions swap order when the site size changes); WEAK for layered assignment (Wyckoff-position prototypes with the species written in last are a description-level analog only: α-Ga and molecular iodine share one prototype label, and George et al. 2020 find that only 13 % of about 5000 oxides obey Pauling's rules 2–5 together), Kröger–Vink notation, valence-state construal and held states; NONE for the tie-break by speaker interest, the mover rule, the event-structure layer and telicity. No chemical correlate is declared; no chemical stand-in for a theta role; Bridge A unrevised; not a joint field.

Solute strengthening: host part vs element part (2026-10-15 evening) takes up the Si-donor note's open question 3 and the navajo-to-materials note's flagged inference: is there a metal-host property split into a part set by the host and a part set by the element? Solid-solution (solute) strengthening is one. In the modern elastic form (Curtin line, as written out by Yin et al. 2020) the host or averaged alloy supplies lattice constant, elastic constants, line tension and core structure, and each solute only its concentration and misfit volume; for Mg, Yasi, Hector & Trinkle 2010 show the Si-donor shape — a long-range size term from the host's dislocation field and an element-specific "chemical" term at the core, with the size term dominant. The solute's number is host-relative, not a portable element property ("The misfit volume of an element in a given alloy reflects the average local atomic environment around the element"), and in high-entropy alloys there is no host, only an average matrix built from the fillers. Concentration enters as √c (dilute, Fleischer-type) or c^(2/3) (collective, Labusch-type), peaks mid-composition in concentrated binaries, and is capped by solubility, which falls as misfit (and so potency) grows; no opened source shows over-concentration softening inside the solid solution. Graded: the split STRONG; the corrected mapping to lattice strain (host stiffness ↔ E, host size ↔ r₀, size misfit ↔ rⱼ − r₀) STRONG; the earlier pairing of size + modulus misfit with (r₀, E) WEAK, mis-mapped. Bridge A shapes SPECULATIVE and after the materials; Bridge A unrevised; no chemical stand-in for a theta role; not a joint field.

Misfit volume: does the element carry its size? (2026-10-16 evening) takes up the solute-strengthening note's open question 3: how far does one element's misfit (apparent) volume carry from host to host — a direct materials test of "does the element bring a fixed feature, or is its number host-relative?" The sources answer in two layers. The misfit itself is host-relative by construction (the element's apparent volume minus the alloy average), so vanadium, with nearly the same volume in fcc and bcc, has a large misfit in both but in opposite directions (Yin, Maresca & Curtin 2020: "large in the fcc alloys and small in the bcc alloys") — STRONG. The apparent volume is partly the element's own: within one family of similar metals in one crystal structure it carries well enough to tabulate once (Ni and Cr in NiCoCr ≈ Ni₂Co / Ni₂Cr, Yin et al. 2020; Vegard's law with elemental volumes within a few percent of first-principles misfits for most elements in RhIrPdPtNiCu and MoNbTaVW) — STRONG; but it moves with concentration (V 12.20 ų dilute in Ni vs 13.8 ų in Ni₂V), with magnetism (Co), and by 20–39 % across very different binary hosts (my arithmetic on Landa et al.'s restatement of King's size factors), and in 3d concentrated alloys the local misfit follows charge transfer rather than the elements' own sizes (Oh et al. 2019) — WEAK as an element constant. This refines the solute-strengthening note's counter-case "the filler is not a portable feature bundle": the apparent volume is a partly portable element number; the misfit is not. The Fleischer–Labusch crossover (OQ1) was not advanced (key source blocked). Bridge A shapes SPECULATIVE and after the materials; Bridge A unrevised; no chemical stand-in for a theta role or a feature bundle; not a joint field.

Merge/Agree mid-level settings: any chemistry analog? (2026-10-19 afternoon) takes Grant's typology note on mid-level Merge/Agree configurations and asks whether any has a sourced structural analog in chemistry or materials. One does, at the level of arrangement: sympathetic nucleation — "the nucleation of a precipitate crystal at an interphase boundary of a crystal of the same phase" (Aaronson et al. 1995, p. 107), with crystals "successively nucleated at an edge of those previously formed" (p. 108), a boundary that "invariably separates" each new crystal from its substrate (p. 108), and a system-level split ("eutectoid type" readily, "aging type" not, p. 121) — matches Rudin's [+MFS] as Grant's note states it (a filled position hosts more of the same kind; separate units; on/off by system): STRONG, arrangement only, no shared mechanism. Trans-influence (Bochmann 2026, pp. 839, 847) is WEAK for feature inheritance (position-selected transmission through a centre, but graded and not a hand-over; cis influence opposite in sign). One centre set by several ligands is WEAK for Multiple Agree. Edge Feature and phasehood: NONE (De Yoreo & Vekilov 2003 could not be downloaded). Grant's notes contain no astrophysics analogies, so none were used. On the Navajo side the same day, Navajo Decoded found verb-word positions [−MFS]-like ("aging-type" in this vocabulary; SPECULATIVE pattern match). No correlate declared; Bridge A unrevised; not a joint field. §8 (added 2026-10-19, 2:32pm): questions from Grant's astrophysics handoff (used only as questions; its figures not cited; quark logic and kanji tabled). Path dependence STRONG (Röyset: fast cooling sets Sc supersaturation, later homogenisation precipitates dispersoids, phases form "depending on composition and thermomechanical processing"); "late steps only select" WEAK; retained readable record STRONG (Al3(Sc,Zr) core/shell: Sc core early, Zr shell late, read by atom probe, shell governs coarsening; sympathetic-nucleation boundaries); a boundary that freezes the interior and leaves the surface reactive: WEAK, full pattern NONE (He et al. 2026 simulation: thick oxide shell slows, does not freeze, core oxidation).

Fleischer–Labusch crossover: when does one solute pin alone? (2026-10-19 evening) takes up open question 1 of the misfit-volume and solute-strengthening notes: at what concentration does the √c law of strong pinning (Friedel/Fleischer: each solute an independent point obstacle) give way to the c^(2/3) law of weak, collective pinning (Labusch)? The Varvenne, Leyson, Ghazisaeidi & Curtin 2017 overview (authors' pre-print, EPFL) states it: the zero-temperature estimate is ~10⁻², which "has led to the assumption that the strong pinning model is applicable at moderate to low solute concentrations", but the single-solute barrier is only ~100–200 meV, so at finite temperature the crossover falls to c < 10⁻⁴ above 78 K for Al–X and Mg–Al (restating Leyson & Curtin 2013, whose own record is metadata-only with the abstract's numbers missing) and "the Friedel model rarely applies for moderate solute/dislocation binding energies" — STRONG; strong pinning is kept only for core-changing solutes (Umax ~1 eV) — WEAK, no worked case opened. The corpus's √c law (Yasi et al.) is, per the overview, a Fleischer-type model used "to predict trends in solute strengthening at 0 K" — STRONG — so the Oct 15 "√c dilute / c^(2/3) collective" split is corrected to a 0 K ranking device versus the finite-temperature regime at nearly all practical concentrations. No over-concentration softening inside an fcc solid solution — NONE. Bridge A shapes SPECULATIVE and after the materials (the one-filler-one-place picture is the narrow regime; the frame's defect decides the regime); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Concentrated alloys: one solute alone or many together, with no host? (2026-10-20 evening) takes up open question 5 of the Fleischer–Labusch note: does the question "one solute alone or many together?" even arise in a high-entropy alloy, where there is no host? Not in its dilute form: the Varvenne, Luque & Curtin 2016 theory (authors' pre-print, EPFL) treats each element as a solute in an effective average matrix, so the alloy is "a random solid solution with 100% solute concentration in the average matrix" and strength comes from "mesoscale collective concentration/structural fluctuations" — STRONG (construction; validated at 293 K for the Ni-Co-Fe-Cr-Mn family, under-predicting at 77 K). Zhai & Zaiser (arXiv:1807.09512) define strong/weak by line geometry, not by stress level, and estimate the Cantor alloy "well within the weak pinning regime" — STRONG (rough inputs); TaNbHfZrTi "at the borderline" — WEAK. But the question returns one level down: atomistic simulations of random CoCrFeMnNi show individual pinning points (Co with Cr-enriched surroundings), only the strong ones surviving at 300 K, with in-situ TEM showing wavy, jerky lines without detected clustering (Utt et al. 2022) — STRONG as simulated; that such pins control engineering strength — WEAK (authors: "by no means a solid solution strengthening model"); collective account overturned — NONE. Strength not set by the number of components N (H1). Bridge A shapes SPECULATIVE and after the materials (HEAs are a scope limit of Bridge A's materials side: no host, no dopant window); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

bcc concentrated alloys: one by one or collective, when screws matter? (2026-10-21 evening) takes up open question 3 of the concentrated-alloy note: in body-centred-cubic HEAs, where screw dislocations and Peierls physics matter, does the strong/weak (one-by-one vs collective) pinning question look different? Yes — it splits by dislocation character. For edge dislocations the fcc average-matrix construction carries over ("Every individual atom is then considered as a solute in the average matrix", Maresca & Curtin arXiv:1901.02100, re-read), with very high barriers (ΔEb ≈ 1.9–2.5 eV), and neutron diffraction plus STEM find edge-dominated deformation in NbTaTiV and CrMoNbV at 293 K and 1173 K (Lee et al. 2021) — STRONG for those alloys; edge control of MoNbTaW is contested by machine-learning-potential MD that keeps edges 11–20× more mobile than screws (Yin et al. 2021) — WEAK. For screw dislocations the question changes shape: the stiff screw line moves by kinks, and the obstacles that lock it are discrete cross-kinks and jogs (Chen et al. 2020; Yin et al. 2021) — STRONG as simulated — inside a random field that the screw theory (Maresca & Curtin 2020; abstract only, full text embargoed) treats as collective — WEAK; whether kink trapping or pure-bcc-like kink-pair nucleation controls screws is contested between the two simulations — WEAK. No opened source settles the TaNbHfZrTi borderline — NONE (Zhai & Zaiser's estimate lumps total solute strength with Cantor-like line energy). Bridge A shapes SPECULATIVE and after the materials (defect character decides field vs countable locks; no saturation shape); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Screw-controlled NbTiZr: the cross-kink plateau and its high-temperature loss (2026-10-22 evening) takes up open questions 4 and 3 of the bcc screw/edge note. In equiatomic NbTiZr (Eleti et al. 2022), flow is screw-controlled ( STRONG); yield strength falls steeply from 77 K to 300 K and then plateaus from 300 K to 873 K, with cross-kinks/jogs and debris on the screws ( STRONG as observed; that cross-kinks cause the plateau, via activation volume against a theory fitted to other data, WEAK); at 1073 K strength drops and the screws are straight and cross-kink-free ( STRONG as observed). So the high-temperature loss of screw locks is now observed for one Ti-Zr-Nb alloy, but that vacancies are what remove them is not measured ( WEAK; recovery and climb also active). The edge share in NbTiZr is stated as small ( WEAK; supplement unopened). For TaNbHfZrTi no opened source splits screw and edge strength ( NONE); DFT shows its screw cores vary with local chemistry ( STRONG as computed). Bridge A unrevised (the plateau is a temperature window, not a concentration window); no theta stand-in; not a joint field.

HfNbTaTiZr: where the screw plateau would be, and what is measured (2026-10-23 evening) takes up open question 4 of the NbTiZr cross-kink note: does an opened source give HfNbTaTiZr a screw/edge strength split, or a measured plateau and its loss? The split is still NONE, and no opened source images cross-kinks in this alloy ( NONE). But two opened data sets give a temperature series that differs from NbTiZr's. In compression (Chen H. et al. 2021) yield falls from 1670 MPa at 77 K to about 1150 MPa at 298 K, and the authors put the knee "obviously higher than 298 K" ( STRONG as measured); in tension (Chen S.Y. et al. 2019) yield is still falling at 573–673 K ( STRONG). At 873 and 1073 K compression yield is nearly equal (481 vs 461 MPa, STRONG as measured), but that flat stretch is two points with nothing measured from 298 to 873 K ( WEAK as a plateau), and the 1073 K test already deforms diffusionally — strain softening, near-equiaxed grains, no kink bands — without the yield drop NbTiZr shows ( STRONG as observed). Between 298 and 673 K the tensile tests show dynamic strain ageing (serrations at 673 K, STRONG as observed; DSA as cause WEAK). Bridge A shapes SPECULATIVE and after the materials (a temperature window that moves with the principal-element mix; the dilute layer marks itself as an event again); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Trace oxygen and nitrogen leave the solution: age softening in TiVNbTa (2026-10-26 evening) takes up open question 3 of the HfNbTaTiZr note (which solute ages?) with a substitute host, stated as such: no opened source names the solute behind HfNbTaTiZr's 573–673 K serrations ( NONE). In equiatomic TiVNbTa aged at 700 °C (Liu et al. 2023, Sci. Adv.), hardness falls about 16% in 40 days, 5.05 to 4.22 GPa ( STRONG as measured), while titanium-rich hcp α-Ti particles carrying oxygen and nitrogen (atom probe: Ti 73.7, O 18.3, N 5.88 at.%) form at grain boundaries and then inside grains ( STRONG as observed), and the matrix's C+N+O falls from about 1.2 to about 0.6 at.% ( STRONG as measured; supporting table in an unseen supplement). The authors put the softening on the drained solution, not on the particles, whose Orowan hardening they compute at ~5% ( WEAK as a mechanism). The bulk gas-fusion oxygen (0.091 at.%) is roughly four times below the atom-probe matrix value; flagged. No opened source yet varies oxygen on purpose ( NONE; Iroc et al. 2022 found, not opened). Bridge A shapes SPECULATIVE and after the materials (a dilute ceiling that moves with temperature; the excess leaves into a new phase rather than staying); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Oxygen added on purpose: a dose window in refractory high-entropy alloys (2026-10-27 evening) takes up open question 1 of the TiVNbTa note (oxygen varied on purpose in TiZrHfNbTa) and half answers it. The journal article (Iroc et al. 2022, Mater. Des.) is still unopened; the M.S. thesis behind it (İroç 2022, METU open repository) was read and is cited as a thesis. Cast Ti25Zr35Nb20Hf5Ta15 with a nominal 2 at.% oxygen (added as TiO₂; oxygen not measured, WEAK as a dose) stays a single disordered bcc phase ( STRONG as observed) and is harder, 321 to 440 HV, about 37% ( STRONG as measured), with 1240 ± 50 MPa compressive yield and no fracture by 35% strain (strain relative only). The un-doped alloy's yield was not measured, so the yield gain from oxygen is NONE in this source; the paper's abstract differs from the thesis on density and on room-temperature oxides (flagged). Where the window ends is located in a neighbouring alloy: in NbTiZr oxidized at 1050 °C (Beaudry et al. 2024, Sci. Adv.), metal–metal ordering is negligible up to 2 at.% oxygen in simulation ( STRONG as computed), the base metal at 6.7 at.% oxygen is still bcc but holds oxygen–zirconium nanodomains and is 3.7× harder than oxygen-free NbTiZr ( STRONG as measured), and the alloy splits into two phases near a computed 8–10 at.% ( STRONG as computed; 6.7 at.% as the measured limit WEAK). A ductility ceiling is NONE in any opened source. Bridge A shapes SPECULATIVE and after the materials (a staged window, local grouping before a new phase; a ceiling set by the host-plus-addition pair); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Where added oxygen stops helping ductility: a tensile ladder in Ti–Zr–Nb (2026-10-28 evening) takes up open question 2 of the dose-window note (a ductility ceiling) and answers it for one family, not for TiZrHfNbTa. In cast Ti–30Zr–14Nb pulled in tension (Jiao et al. 2023, Nat. Commun., open), nominal oxygen from 1 to 3 at.% raises both yield strength and elongation (about 9.5% to 25.1%; STRONG as measured), and past 3.5 at.% ductility is lost to brittle fracture ( STRONG), with oxygen collected at grain boundaries and no oxide formed ( STRONG as observed; that the boundary oxygen is the cause is WEAK). The ceiling is host-relative: more Nb lowers it, a two-phase host has no window, and in a Hf-bearing check (Wang et al. 2025, Adv. Sci.) one host gains ductility to 1.8 at.% O while a less stable one loses most of it at 0.5 at.% through ω precipitates ( STRONG as measured). Across alloys, the ductility ceiling (~3.5 at.%) sits well below the NbTiZr phase ceiling of the previous note (~8–10 at.%, computed) — a design warning, HYPOTHESIS. Ceiling in TiZrHfNbTa: NONE. Bridge A shapes (excess parked at boundaries; a working edge before a structural edge) are SPECULATIVE and after the materials; Bridge A unrevised; "continuous wt% ≠ discrete arity" holds; no chemical stand-in for a theta role.

Oxygen in a Ta-bearing five-element alloy: one measured tensile pair (2026-10-29 evening) opens Lei et al. 2020 (Sci. Adv., open, with its supplement) for the ductility-ceiling note's open questions 3 and 1. It does not restate the closed 2018 TiZrHfNb numbers ( NONE). It does give the first Ta-bearing tensile pair in this line: cast Ta0.5Nb0.5HfZrTi from a measured 0.377 to 2.241 at.% O gains yield (~810 → ~1138 MPa) and elongation (~10.8% → ~20%) and stays single bcc ( STRONG as measured; one curve per alloy). So the Ta-bearing ductility ceiling is above ~2.2 at.% O ( STRONG as a lower bound) and its location is still NONE. The share of oxygen in ordered complexes (~18.5%) and their causal role are WEAK (fitted; the authors' attribution). The source's "base" already holds 0.377 at.% O, a reminder that zero-added is not zero. The abstract's ~1400 MPa with ~20% pairs two alloys' numbers (flagged). Same group as the Ti–Zr–Nb source, so not independent. Bridge A shapes are SPECULATIVE and after the materials; Bridge A unrevised; "continuous wt% ≠ discrete arity" holds; no chemical stand-in for a theta role.

Oxygen in HfNbTaTiZr: measured 0.27 vs 3.24 at.%, stability on ageing (2026-10-30 evening) takes the Ta-bearing ceiling question one oxygen step higher. Poulain et al. 2024 (Acta Mater., open, CC BY) measures equiatomic HfNbTaTiZr at 0.27 and 3.24 at.% O. Held at 900 °C for 1000 h in clean argon, the 0.27 at.% alloy stays one bcc phase ( STRONG, repeated twice); the 3.24 at.% alloy breaks down into ~15 vol.% oxygen-rich HCP and ~35 vol.% Nb–Ta-rich bcc, with HCP covering the grain boundaries ( STRONG as observed). One cold-swaging attempt on the 3.24 at.% alloy cracked at ~40% reduction ( WEAK: one remark in the acknowledgments). The paper has no tensile test, so the Ta-bearing tensile ceiling is still NONE; the same group's ageing-plus-tension paper (Zhao et al. 2025) could be read only as its abstract, which gives no numbers. Read beside last night's Ta0.5Nb0.5HfZrTi pair (still ductile at 2.24 at.%), the edge may lie between ~2.2 and ~3.2 at.% O — a HYPOTHESIS across two alloys and two different tests. Ageing adds a time axis: an amount tolerable as cast may not be tolerable after a long hold. Bridge A shapes are SPECULATIVE and after the materials; Bridge A unrevised; "continuous wt% ≠ discrete arity" holds; no chemical stand-in for a theta role.

HfNbTaTiZr after hot isostatic pressing: an HCP phase and no tensile stretch (2026-11-02 evening) takes up the ageing-and-ductility question of the previous note (does losing single bcc cut tensile ductility?) in part, by another route. The ageing-plus-tension paper (Zhao et al. 2025) is still blocked, so the note reads an open paper from an independent group, Larianovsky et al. 2017 (Materials, CC BY), which casts equiatomic HfNbTaTiZr two ways and hot isostatically presses it (1230 °C, 152 MPa, 4 h). As cast it is one bcc phase; after HIP a Zr–Hf-rich HCP phase is present, evenly spread in one casting and mostly on the grain boundaries in the other ( STRONG as observed). Pulled at room temperature, five bars broke at 669–749 MPa with no elongation the machine could detect and inter-crystalline fracture without plasticity ( STRONG as measured), while in compression the same state yields at 1350–1888 MPa ( STRONG). So a two-phase HfNbTaTiZr can be fully brittle in tension, and compression results do not certify tensile ductility in this family. The paper measures no oxygen ( NONE): reading its HCP phase as oxygen-driven, via Poulain et al.'s statement that this alloy's phase-stability discrepancies may be related to oxygen, is a HYPOTHESIS, as is the HCP phase being the cause of the brittleness (the cause is NONE as sourced; the authors' slow-cooling reading is WEAK). The ageing case of the question stays NONE. Bridge A shapes SPECULATIVE and after the materials (same formula, different state, opposite outcome; one test does not certify another); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

HfNbTaTiZr at 873 K: nanoscale hcp and a stress rise in one batch (compression only) (2026-11-03 evening) takes up the same-batch question left by the previous note (test one batch with and without the hcp phase) in part, and only in compression. Yasuda et al. 2025 (Scripta Materialia, CC BY; independent Osaka group) solutionize equiatomic HfNbTaTiZr at 1473 K and compress it at 873 K (600 °C). With an 1800 s hold the alloy splits on a ~10 nm scale into Zr–Hf-rich and Nb–Ta-rich bcc solutions and then precipitates <10 nm hcp in the Zr–Hf-rich layers ( STRONG as observed); hardness is already high after 0.2 h ( STRONG as stated). In the same polycrystals, fast heating and a 120 s hold suppress the hcp and the 873 K stress rise becomes insignificant ( STRONG as a same-batch contrast), so the hcp, not the temperature or the split alone, causes the rise ( STRONG within compression; at 1073 K the split occurs without hcp and without a marked rise, restated from the authors' 2021 paper, WEAK here). Oxygen is not measured ( NONE), and what the nano-hcp does to tensile elongation is NONE in this source; a ductility loss is a HYPOTHESIS carried from the HIP case, a different state. Read across sources, the clean alloy is unstable in a 500–800 °C band and stable at 900 °C unless oxygen is present ( HYPOTHESIS), and a slow cool through that band gives a non-oxygen route for the HIP'd alloy's HCP ( HYPOTHESIS). Bridge A shapes SPECULATIVE and after the materials (a window with a clock; split first, then re-house); no dose is varied, so no window edge in concentration is placed; Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

HfNbTaTiZr between 550 and 1000 °C: where the unstable band ends (two edges, not one) (2026-11-04 evening) takes up the previous note's question of where the unstable band ends (OQ3) in part, with one of the two primary papers it named. Chen et al. 2019 (Scripta Materialia; the authors' accepted manuscript on OSTI; independent Tennessee/ORNL/Argonne/Tsing Hua group) make equiatomic HfNbTaTiZr single bcc (1200 °C, 10 min, water quench) and then hold it for 96 h in vacuum-sealed quartz at 550, 700, 900 and 1000 °C. At 1000 °C it stays single bcc; at 900 °C the bcc X-ray peaks split; at 550 and 700 °C it is two bcc phases plus a Zr–Hf-rich hcp phase ( STRONG as observed). At 700 °C precipitates appear on grain boundaries within 2.5 h and make up about two fifths of the volume after 96 h ( STRONG as stated). With the abstract of Yasuda et al. 2021 (split at 773–1073 K, hcp at 773–973 K; WEAK, abstract only, body blocked), last night's single 500–800 °C band becomes two: the bcc split reaches somewhere between 900 and 1000 °C, the hcp about 700–800 °C (brackets, not pinned). Neither lab measures oxygen ( NONE). Chen's 900 °C split in a vacuum tube stands against Poulain et al.'s clean 900 °C stability under argon; oxygen pickup in vacuum, which Poulain et al. showed at 900 °C, could explain it ( HYPOTHESIS; not counted as a contradiction). Poulain et al.'s summary of Chen rounds 900 °C to the stable side. Bridge A shapes SPECULATIVE and after the materials (two edges, not one; an edge that may move with what gets in); no dose varied; Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Ti–Zr–Hf–Nb–Ta with Fe micro-alloying: a ductility window set by nano Hf₂Fe at grain boundaries (2026-11-05 evening) takes up last night's dopant-window lead (OQ5). Li et al. 2025 (Materials 18: 1463; CC BY; Europe PMC XML; independent Beijing Institute of Technology group) micro-alloy Ti₁₅Zr₁₀Hf≈₃₆Nb₁₀Ta₂₅ with Fe at 3.35–3.5 at.% — a true minor addition, not a principal element and not the interstitial oxygen of the prior notes — and age it. Nano Hf₂Fe forms at grain boundaries ( STRONG as observed where TEM sees it; HF2-525-30 below-TEM HYPOTHESIS). This work's own tensile set: HF2-ST 955 ± 25 MPa / 19 ± 2%, HF2-525-30 1037 ± 17 MPa / 14 ± 1%, HF2-525-60 960 ± 50 MPa / 2.65 ± 0.15%, HF1-600-30 271 ± 6 MPa / 0.35 ± 0.15% ( STRONG as measured); HF1-ST 971 / 17.5 and HF1-500-30 1081 / 7.5 are restated from Liu et al. 2022 same group ( WEAK here). Elongation spans about 2.65–19% at strength about 955–1081 MPa ( STRONG as a measured span in this non-equiatomic host; equiatomic HfNbTaTiZr transfer NONE). Oxygen is NONE. Bridge A shapes SPECULATIVE and after the materials (window with a true minor addition; concentration + ageing clock; GB re-housing); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Ti–Zr–Hf–Nb–Ta–Fe baseline (Liu 2022): F2.5-ST ~971 MPa / ~17.5% and aged continuous GB Hf₂Fe (2026-11-06 evening) takes up open question 4 of the Fe-window note (Liu et al. 2022 primary). Liu, Wang, Wang & Xue 2022 (J. Phys.: Conf. Ser. 2383: 012141; CC BY 3.0 IOP PDF) is the same Beijing Institute of Technology group as Li 2025. F2.5-ST yield "~971 MPa" with retained ductility "~17.5%", and F2.5-500-30 yield "~1081 MPa" after continuous ~25 nm grain-boundary Hf₂Fe, are now STRONG as primary prose and match Li's restated HF1 yield / ST-elongation numbers; the Hf₂Fe name is STRONG here. Exact Fe at.% for F2.5 is NONE in Liu prose (identity with Li's HF1 Fe₃.₅ stays OPEN). Aged elongation 7.5% is NONE in Liu prose and remains WEAK as Li's restatement. Liu's continuous ~25 nm vs Li's discontinuous ~5 nm for the "500-30" condition is a STRONG text conflict. Oxygen NONE; equiatomic transfer NONE. Bridge A shapes SPECULATIVE and after the materials (solution-first step; GB re-housing from the primary; primary vs restatement); Bridge A unrevised; no chemical stand-in for a theta role or slot; not a joint field.

Ti–Zr–Hf–Nb–Ta base printed (Yang 2025, EBM): nominal Ti15Zr10Hf40Nb10Ta25, the Fe-free end of the Fe family (2026-11-09 evening) takes up open question 1 of the Liu-baseline note (printed composition of Base / F2.5). Yang, Wang, Guo, Cao, Zhang & Xue 2025 (J. Phys.: Conf. Ser. 3080: 012022; CC BY 4.0 IOP PDF; same Beijing Institute of Technology group) electron-beam re-melt the group's Ti-Zr-Hf-Nb-Ta and print its nominal composition as Ti15Zr10Hf40Nb10Ta25 at.% ( STRONG as printed), the exact x = 0 end of Li's Ti15Zr10Hf40−xNb10Ta25Fex ( STRONG as arithmetic; Fe goes in for Hf). Liu's Base being this alloy is a HYPOTHESIS with a printed anchor (no paper says it in words). Under the HF1 mapping the code "F2.5" is neither 2.5 at.% nor 2.5 wt% Fe (Fe3.5 ≈ 1.4 wt%; arithmetic), so its meaning stays OPEN. The paper tabulates O 276 ± 12 mass ppm in an undescribed "Raw material" and 53 ± 9 / 71 ± 10 after EBM ( STRONG as tabulated; ≈ 0.25 at.% my WEAK conversion), but EBM volatilises all Ti, so oxygen in the Fe-micro-alloyed host stays NONE. Table 3 loss ratios disagree with the printed masses (internal inconsistency recorded). Bridge A shapes SPECULATIVE after the materials; unrevised; not a joint field.

Oxygen in a Hf-bearing refractory host: ~0.3 at.% as melted, HfO₂ above ~0.5 at.% (Nb45Ta25Ti15Hf15) (2026-11-10 evening) takes up open question 5 of the Liu-baseline note (oxygen in the Fe-micro-alloyed host) by bounding it, since no open source measures it. Huang et al. 2025 (arXiv:2512.08080, CC BY 4.0 preprint; Texas A&M / UC Irvine) arc-melt Nb45Ta25Ti15Hf15 from ≥ 99.95% charge: with nothing added the alloy holds ~400 mass ppm ≈ 0.3 at.% O ( STRONG); HfO₂ appears above ~0.5 at.% and on grain boundaries at 1 at.%, and every oxygen-added button cracks before 10% rolling reduction ( STRONG as stated). The paper prints its oxygen ceiling three ways (0.5; 0.5–1; 0.8–1.0 at.%), and its own Table 3 bracket for the oxygen chemical potential at 0.8% straddles its HfO₂ threshold (arithmetic), so the single number is WEAK and only the ~0.5–1 at.% band is STRONG. Belcher et al. 2025 (J. Alloys Compd. 1010: 177520, CC BY-NC; same circle) state that charge oxygen is not removed in arc melting without a reducing reaction ( STRONG as stated for MoNbTaW) and that Ti/Zr melts pick up more. With Poulain 0.27 at.% and Yang ≈ 0.25 at.% already in the corpus, ~0.25–0.3 at.% is the as-melted scale ( HYPOTHESIS); the BIT host's measured oxygen stays NONE, and whether its 40 at.% Hf raises or lowers the oxide ceiling is OPEN. Bridge A shapes SPECULATIVE after the materials; unrevised; not a joint field.

Carbon micro-addition and grain-boundary oxide in a Hf-bearing host: a tensile threshold between 0.4 and 0.6 at.% C (Nb44Ta44Hf10W2) (2026-11-11 evening) takes up open question 2 of the oxygen note (which way Hf pushes the oxide ceiling) in part and sideways. Shen et al. 2025 (Research Square rs-7842583/v1, CC BY 4.0 preprint; Shanghai Jiao Tong University, independent of the BIT group and of the Huang/Belcher circle) arc-melt Nb44Ta44Hf10W2 with a carbon micro-addition ladder (0, 0.2, 0.4, 0.6, 0.8, 1.0 at.%). The carbon-free alloy holds only 99 wppm ≈ 0.09 at.% O ( STRONG) yet carries a dense grain-boundary HfO₂ layer that, by the authors' calculation, grows from the last liquid, and it breaks along its boundaries at 0.33% elongation ( STRONG as observed / tabulated; growth path WEAK as the real path). That is about a fifth of last night's ~0.5 at.% XRD oxide edge in a Ti-bearing host, so a bulk solubility ceiling and an as-cast solidification edge are different numbers ( HYPOTHESIS as a rule); which way Hf moves the ceiling stays OPEN. Carbon gives a sharp tensile window — brittle at ≤ 0.4 at.% C, 11.4–13.3% elongation at 0.6–1.0 at.%, peak at 0.8 at.% C (YS 682.2, UTS 807.0 MPa) ( STRONG as tabulated, no scatter printed) — with grain-boundary carbide replacing grain-boundary oxide at 0.8 at.% C ( STRONG as observed); carbon dose and oxygen move together, so carbon-alone is OPEN. BIT-host oxygen stays NONE; equiatomic transfer NONE. Bridge A shapes SPECULATIVE after the materials; unrevised; not a joint field.

Boron inside the grains vs on the boundaries: a dilute Re + B ladder in Nb45Ta27Ti13Hf15, beside a blocked equiatomic boron lead (2026-11-12 evening) takes up the equiatomic micro-addition question (Fe-note OQ1 / Liu-note OQ4) in part. The first equiatomic HfNbTaTiZr micro-addition paper with an open licence turned up — Wu et al. 2026 (Acta Mater. 311: 122210, CC BY; Max-Planck / Paris-Est): 30 appm boron in equiatomic TiNbZrHfTa goes to the grain boundaries with Zr (atom probe) and yield strength falls from 920.7 to 869.8 MPa — but its body is blocked (ScienceDirect 403; repository copy behind a bot challenge), so it is abstract-level only ( WEAK) and a single step, not a ladder. The readable ladder is An et al. 2026 (PNAS 123: e2611876123, CC BY-NC-ND 4.0, via Europe PMC) in Zr-free Nb45Ta27Ti13Hf15: 1.5 at.% Re adds 144 MPa and 0.3 at.% B another 102 MPa (722 → 968 MPa, +34.07%) for a 1.8-point elongation cost ( STRONG as tabulated), with boron reported dissolved in the grains ( STRONG at EDS resolution, WEAK as "no boundary boron"). Same element, opposite sign, different place: boron's effect in these Hf-bearing bcc hosts tracks where it sits rather than what it is ( HYPOTHESIS; two hosts, two groups, doses 100× apart). Oxygen in An's alloys is NONE, and with no Re-free boron alloy the claimed Re+B cooperation is OPEN. Equiatomic ladder still NONE. Bridge A shapes SPECULATIVE after the materials; unrevised; not a joint field.

How much oxygen the route puts into equiatomic HfNbTaTiZr: a measured ladder from 0.12 to 1.07 wt.% across powder routes (2026-11-13 evening) takes up open question 5 of the oxygen-pickup note (a Ti/Zr-bearing pickup magnitude in text form) by substitution. Málek et al. 2019 (Materials 12: 4022, CC BY 4.0, via Europe PMC; Czech–Korean group, independent of every earlier oxygen source) tabulate gas-fusion oxygen for ten powder-route variants of equiatomic HfNbTaTiZr: 0.12 ± 0.02 wt.% for crucible-free gas-atomized powder + SPS (≈ 0.88 at.%, my arithmetic), 0.55–0.84 wt.% for blended elemental powders cold-pressed and sintered or HIPed (≈ 3.9–5.9 at.%), 1.07 wt.% after 42 h ball milling (≈ 7.4 at.%, with HfO₂ particles) ( STRONG as tabulated). HIP adds a resolved +0.25 wt.%; vacuum sintering's +0.10 is inside the printed ± ( WEAK). Only the cleanest specimen is single bcc, pore-free and survives cold high-pressure torsion ( STRONG as observed); that oxygen alone causes the other routes' cracking is HYPOTHESIS, because porosity and slow-cooled bcc2/hcp phases rise with it. Even the cleanest route carries ~3× the mass fraction of the Zr-free Nb45Ta25Ti15Hf15 as-melted button, so in such a host the route is a first-order control on oxygen; nothing transfers to the melted BIT host. Whether ≈ 0.88 at.% O is HfO₂-free is WEAK (text says pure bcc; a figure caption mentions a small HfO₂ fraction across six samples). No strength or elongation is in the source. Bridge A unrevised; shapes only, after the materials.

Inverse ranking (2026-10-04 evening; Creamer deepen 2026-10-05) is a separate note: Navajo yi-/bi- is a sourced instance of Grant’s argument-ranking edge (roles fixed; animacy/topicality outranks; inverse preferred for (62); * and ?* alignments; one bi- per clause; not a passive). Creamer 1974 (opened) supplies the eight-group status ranking; Hale 1973 still unopened. Multiple-wh untested. No chemical correlate; §6 there gives the five-point search spec. Bridge A unrevised.

Spinel inversion ranking edge (2026-10-05 early) tests spinel cation inversion against that five-point spec: passes 2 (charge-independent site-preference scale), 3 (inverse preferred when the scales disagree), and 5 (no demotion); partial on 1 and 4; sits at the crystal level, not the molecule level. Strongest chemistry candidate so far, recorded as a candidate, not a correlate. Bridge A unrevised.

Fourth person vs ranking (2026-10-05 midday) takes up the inverse note’s open question 1: per Willie 1991 §4.2.7 and Willie & Jelinek 2000, the Navajo fourth person is outside the animacy hierarchy and lets a lower-ranked agent stay the direct-form subject over a human patient, so it is an escape from the ranking scale rather than a second instance of Grant’s edge; it adds an optional sixth point to the chemistry search spec. No chemical correlate. Bridge A unrevised.

Spinel dopant site preference (2026-10-05 evening) tests the spinel candidate against that sixth point and takes up the spinel scout’s open questions 1 and 4: d⁰, high-spin d⁵ and d¹⁰ ions are a sourced category outside the crystal-field scale, but size, covalency and electrostatics still rank them (MgFe₂O₄ and MgGa₂O₄ stay inverse), so point 6 at most partly passes; Ni²⁺ (d⁸) is the clean dopant case taking the octahedral site against the charge default; Price, Price & Burdett (1982, abstract) sort 98% of spinels without d-electron count. Burdett et al. JACS not reached. Candidate, not correlate; crystal level. Bridge A unrevised.

TBP apicophilicity ranking edge (2026-10-05 late evening) is the first molecule-level test of that spec: in five-coordinate phosphorus the incoming and outgoing groups of a substitution use the apical slots (Westheimer 1977), an apicophilicity scale ranks every group for those slots, a clash is resolved by pseudorotation that shows in the product (retention vs inversion), bulky rings can make the rule-breaking arrangement stable, and cage ligands make some rearrangements impossible. Points 1 (event) and 5 pass, 2 passes weakened, 3 and 4 partial, 6 leans fail (groups outside the scale break the bipyramid or pin one slot). Unlike the Navajo inverse, the marked step restores the role-to-slot default rather than letting the second scale win. Candidate, not correlate. Bridge A unrevised.

REAL as the split the opened texts actually describe. Hale & Platero 1996 (primary, now opened) name (i) a simple transitivity alternation (ł only: boil; fire / shatter / melt / rope) and (ii) a complex causative (ł + y- + special causee marking: walk / burp / laugh biyeeshdloh). (iii) Analytic causation asserted, still not exemplified in anything opened. Some ł is thematic, not derived. Nobody opened equates this with Japanese lexical causatives vs productive -sase. Not Bridge A.


Zipf (not Bridge A)

Zipf (2026-10-03) is ranked frequency. It is not Bridge A. It is not a joint field. It is not a Navajo claim.

Zipf at the frame (2026-10-06 early) adds a frame-level layer: English verb-argument constructions have Zipfian, frame-selective verb fillers (V across n cumulative slope −1; V about n −0.687; Römer, O'Donnell & Ellis 2015), and the CAS Registry has a power law over molecular frameworks (cumulative slope −1.07, α = 2.07; Lipkus et al. 2008) — but language ranks the fillers of a fixed, role-ordered frame while chemistry ranks the frames after deleting fillers and attachment points, so neither side has measured the other's half. Lipkus et al. explicitly carry Simon's word-frequency model onto frameworks (framework as word type), which partly corrects the line below that no opened full text maps linguistic Zipf onto molecules. Comparison SPECULATIVE; Grant's phrase ↔ molecule cut stays HIS.

Zipf fillers in chemistry (2026-10-06 midday) takes up the missing chemistry half. Substituent counts are long-tailed (Ertl 2003: 849,574 substituents, about 64% singletons; Ertl 2021 preprint: “a typical power law (or ‘long tail’)” in natural products), but no opened source prints a slope or counts fillers position by position on one framework (Takeuchi et al. 2021 sample per site, then pool; Takeuchi et al. 2020's site-preference claim read as abstract only). The nearest slot-level object is the benzene ring: which positions are filled, with filler identity ignored, is top-heavy (four of twelve patterns = 76% of drug rings; Nilova & Stuart), and those authors trace it to the directing rule, in which the class of group already on the ring picks the next slot. The gap is narrowed, not closed. Comparison SPECULATIVE; no theta stand-in; no Navajo counts.

EAS directing ranking edge (2026-10-06 evening) runs that directing rule against the six-point ranking-edge spec with textbook and primary sources: groups already on the ring rank the free positions by class, a separate activation scale decides between opposed directors with graded mixtures, bulk drives the ortho share to zero, the position between two meta groups is made "by some other route," and ipso attack can leave no trace in the product — but the scale sits on bystanders and ranks positions for a single entering group, not participants for roles, and no substituent class sits outside the scales (point 6 leans fail). Candidate, not correlate. Bridge A unrevised.

REAL inside the field that measured it. Classic Zipf is ranked word frequency, about 1/r, not exactly −1. Fixed-length n-grams are shallower than −1. Williams, Lessard, Desu, Clark, Bagrow, Danforth, and Dodds 2015, Scientific Reports 5, 12209 (srep12209; https://www.nature.com/articles/srep12209), put mixed-length phrases near −1 over a long rank range. Their phrase is a random partition (mixed-length, non-overlapping), not a phrasal verb. They call morphemes the meaningful “atoms” of language and a mixture of words and phrases the meaningful “molecules.” That is a linguistic remark, not a chemical measurement, and it is not the word↔atom / phrase↔molecule cut.

Chemistry sources opened do not show Zipf over molecules. Nearby power laws only: Benz, Swamidass, and Baldi, substructures and cluster sizes, exponents about 1.5–3; Fialkowski, Bishop, Chubukov, Campbell, and Grzybowski 2005, reaction-network degrees near 2.7 in and 2.1 out; Liu, Fang, Li, Wang, Qi, and Hu 2026, Zipf-like BPE tokens of SMILES/SELFIES, slopes only described as moderately steeper than −1, no numeric slope in the abstract. Cadeddu, Wylie, Jurczak, Wampler-Doty, and Grzybowski match text-fragment frequencies to molecular-fragment frequencies and do not name Zipf. An earlier “Zipf-like” gloss of that abstract in Prior bridges is withdrawn.

Grant’s phrase↔molecule / word↔atom hypothesis stays HIS and SPECULATIVE. His only Grok-app message on 10 Jun, in the thread “Zipf’s Law Holds for Phrases,” was the Williams URL plus “please help me understand this.” The “meaningful molecules” line was the app’s gloss, not his wording, and not a chemical result.

His phrasal-verb question is open, in his voice, and not a finding: maybe the power law applies to phrasal verbs (a verb plus its arguments, basically a function) rather than to phrases in general. He does not know if that is true. It is only something to explore. This addendum does not test it. No phrasal-verb exponent is stated.

His quark / anyon remark is an open analogy beside the Williams morpheme-as-atom line, HIS and SPECULATIVE: it is not wrong, on his view, to say morphemes are like quarks, because they can only be produced by the breakdown of a larger unit such as a word or an atom; maybe anyons; he does not know physics particles perfectly. Quarks are not chosen over anyons. No confinement claim and no particle-physics result is asserted.


VPISH molecule picture (separate)

VPISH molecule picture (2026-10-04) is a separate note. It is the three-slot picture (verb phrase, tense phrase, clause — the same slots under his molecule names, once). It is not the movement derivation, not lexicalization, not the causative scout, and not Zipf. The claims are not merged.

SPECULATIVE as a correspondence. No paper states it. Not a joint field. Not Bridge A.

Sourced comparison. Ordinary English, as Ethan Poole’s Ling 200B handout (17 November 2021) draws it, is the clear VPISH hop: a lexical subject starts in the verb phrase and moves to Spec,TP, still inside the clause. Navajo on the drawn analysis, Belfast English as that handout reports it, and Welsh as that handout reports an argument (subject in the verb phrase, verb to T; Woolford 1991 not opened), leave the subject in the verb phrase. Those three are not given a diffusion step and not given a segregation step. The only chemistry mechanisms on the page are an interstitial hop, a vacancy hop, and segregation that stays in the solid, likened only to the English hop, and weakly even there.

HIS. Navajo is the extreme edge case that shows the inner workings, not a failed English hop. The subject exponent is on mode, inside the verb word. McDonough: no tense. Smith, Perkins, and Fernald: not fully tensed. The units he takes the edge case to be tracking are theta roles, not lexical items. Chemistry, as opened, has no stand-in for a theta role. None is invented. The gap is kept open. It is not a refutation.


Phrasal prescription (filed 2026-10-03; rev. 2026-10-04) is a separate note. It answers Grant’s “prescription / dopant places / VPISH in a tense” question from the Athabaskan side. It is not lexicalization, not movement, not the causative scout, not Zipf, and not the VPISH molecule picture. The claims are not merged.

REAL (sourced). No single Athabaskanist label “the Navajo phrasal verb.” Young 2000 Theme→Base layering (pp. 57–59 as quoted via Acta Lingweenie): Verb Theme = classifier+stem; Verb Base = Theme + adverbial derivational prefix — aligns with Smith, Perkins & Fernald; disagrees with McDonough’s report of the 1987 dictionary’s “base” = classifier+stem. Hale: preverb+stem as one lexical constituent inside one verb word. Chee & Yazzie / Chee 2025: “phrasal verb” = postposition + (minimally inflected) verb; Chee 2025 also derives nouns from that pair. Mode is not tense; Navajo is not fully tensed; the prescription-in-a-tense picture is not what these sources say.

HIS SPECULATIVE. Grant’s molecule↔phrase / dopant-places / VPISH-in-a-tense frame stays his. Pedagogy may liken a theta slot to a place, not a level. No concentrations. Not Bridge A. Not a joint field.


Still open

The addendum’s host walkthroughs and depth notes are filed. One optional polish item remains open in that addendum: a fuller Ti STA subsection inside the process-metastability note (partially closed already by the allotropy note), a near-α Ti sibling note, and primary-source PDF drops for tables already cited from the open literature.

This page is the public conclusions text: verdict, non-claims, and reading pointers. It is not a lab notebook.

Non-claims

  • No joint field of doping × θ-roles / valency.
  • Bridge A is SPECULATIVE as identity; locked only as working pedagogical focus.
  • Navajo morphology does not prove Bridge A count.
  • Kanji does not map to doping.
  • SSS “chemical / valence” language does not license a new bridge.
  • This brief does not revise Bridge A’s lock, invent shared formalism, or announce a product.

Non-claims, from the addendum

  • This addendum does not revise Bridge A’s lock.
  • No joint field of doping × θ-roles / valency.
  • Sixteen host pegs + three clocks train alloy-design literacy; they do not prove linguistics ≡ metallurgy.
  • Navajo typed-participant map is SPECULATIVE parallel pedagogy — not Bridge A proof.
  • Navajo derivational movement is a separate note. It does not move arg1 to Spec,TP. It does not draw a yi/bi tree. It does not posit covert wh-movement to Spec,CP. Its chemistry half is SPECULATIVE pedagogy and is not Bridge A. Leaching and evaporation are the wrong parallel.
  • Navajo lexicalization is a separate note. The listed unit is the verb base. “Phrasisized” is Grant’s label beside verb theme, lexicalized base, idiomatic preverb–verb, and incorporated postposition. Not English phrasal verbs. Hale: no verbal noun incorporation in Navajo. Not merged with movement or the typed-participant map.
  • Navajo causative morphology is a partial split. It is not Japanese lexical vs productive -sase. Analytic causation is asserted and not exemplified. Some ł is thematic, not derived.
  • Zipf over molecules is not a finding of the chemistry sources opened. Williams et al. 2015 (srep12209) do not state the word↔atom / phrase↔molecule cut. Grant’s phrase↔molecule line, his open phrasal-verb question, and his quark/anyon analogy stay HIS and SPECULATIVE. No phrasal-verb exponent. No particle-physics result. Not Bridge A. Cadeddu et al. do not name Zipf.
  • The VPISH molecule picture is a separate note.
  • The Navajo phrasal-prescription note is a separate note. Young 2000 Theme→Base (as quoted) matches Smith/Perkins/Fernald; McDonough’s 1987 “base” report differs; Chee’s “phrasal verb” is postposition+verb. Not a dopant prescription in a tense. HIS frame SPECULATIVE. Not merged with lexicalization, movement, causative, Zipf, or VPISH. Ordinary English is the only case given a diffusion or segregation step. Navajo, Belfast English, and Welsh as reported leave the subject in the verb phrase. Navajo is the edge case, not a failed hop. No chemical stand-in for a theta role. Not a joint field. Not Bridge A. Not merged with movement, lexicalization, the causative scout, or Zipf.

v1-addendum — 2026-10-04 PT (light update: Navajo phrasal-prescription under Path E; Young 2000 Theme→Base as quoted; Chee 2025 full text; prior same day: VPISH molecule picture under Path E; English is the only hop; Navajo, Belfast English, and Welsh-as-reported get no diffusion or segregation step; no chemical stand-in for a theta role; not Bridge A; not merged. Prior same day: Zipf page; Navajo lexicalization; Navajo causative scout; Cadeddu abstract does not name Zipf; Grant’s phrasal-verb question left open; quark/anyon analogy HIS and SPECULATIVE; not Bridge A). Prior 2026-10-03 PT: Navajo derivational movement; pointed Derivational movement; separate from the typed-participant map; chemistry half SPECULATIVE and not Bridge A. Prior 2026-10-02 evening: same-element Co cross-host. Prior 2026-10-01 evening: same-element Mn cross-host. Prior 2026-09-30 evening: same-element Ni cross-host. Prior 2026-09-29 evening: same-element Al cross-host. Prior same-day: same-element H cross-host. Prior 2026-09-28: three clocks × plural “over” modes. Prior 2026-09-25: plural “over” modes catalog. Prior 2026-09-24: alloying·doping·intercalation triangle + NMC foil. Prior 2026-09-23: Family 3 SSS contrast — Al–Mg·Cu–Ni + Mg-host nod. Prior 2026-09-22: allotropy/transformation-window — Fe–C·Ti·Zr + CoCrMo SIMT foil; Ti STA partial close. Prior 2026-09-21: electronic doping Family 1 — Si/GaN:Mg/oxide. Prior 2026-09-19: HCP trio; Zr sixteenth; CoCrMo fifteenth + SFE; Al–Li fourteenth + precip family; Path D; TWIP; same-element Mg; Cantor; Triskelion parked). Speculation flagged. Bridge A lock unrevised (pedagogy/probe only). No joint field.

Curated from the week conclusions and the post-week addendum. Lab inventory, index routine, and site-map process notes are omitted.