No joint field · Bridge A = SPECULATIVE pedagogy

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

Path C · Host checklist

Worked walkthrough — Mg (and Al/Nb) doping of Ni-rich NMC / Li-ion cathode hosts

  1. 1 Diagram
  2. 2 Mechanism
  3. 3 Failure past the window
  4. 4 Metastability
  5. 5 SPECULATIVE Bridge A peg

What’s claimed and how well supported

Labels on this page
SPECULATIVE ×14
Strongest label used
SPECULATIVE
Main caveat (quoted)
Locks: Bridge A locked as ◆ SPECULATIVE supporting case only.

Counted automatically from the labels printed on this page. Counts are a reading aid, not a grade of the page.

Terms on this page (10), plain-language definitions
Host (matrix)
Majority element/phase whose lattice dominates the material.
Dopant
Intentional dilute impurity meant to tune electronic/optical/catalytic behavior without replacing the host as majority phase.
Valence (chem.)
Bonding capacity / oxidation-related electron count of an atom in context.
Dislocation
Line defect enabling plastic slip at realistic stresses; edge vs screw are the two classic geometries.
Solute
Intentional minority species dissolved in the host (alloying addition).
Precipitate
Second-phase particle that forms (often on aging) and blocks dislocations.
FCC
Face-centered cubic — many slip systems; Al, Cu, Ni typically ductile.
HCP
Hexagonal close-packed — fewer primary slip systems; Mg, Ti, Zn often anisotropic.
BCC
Body-centered cubic — Fe/steel common; plasticity more temperature-sensitive.
Valency (ling.)
Arity: how many actants a predicate takes; formal vs semantic valency can diverge.

Definitions quoted from the Chem for linguists primer glossary.

Materials first. The cross-domain peg stays last, and it stays SPECULATIVE.

Date: 2026-09-19
Audience: syntax/semantics readers building alloy-design literacy (Chem for linguists)
Purpose: Industrial battery-cathode case from Grant metal-doping-takeaway.txt §5 / §8 — Mg (with Al / Nb siblings) as structural stabilizer in Ni-rich NMC layered oxides. Same checklist discipline as oxide electronic / GaN:Mg / Si electronic / metal hostsMajority element/phase whose lattice dominates the material., but here the intended property is chemo-mechanical / electrochemical stability (cycle life, thermal safety, cation-order), not carrier type/density like Si:P or GaN:Mg.
Locks: Bridge A locked as SPECULATIVE supporting case only. Materials facts first; linguistic parallels only in a flagged box at the end. No joint-field claim. Navajo typed-participant half = light parallel only — not proof of Bridge A count. Dowty / Triskelion / Merge·Agree / kanji not advanced here.

Companion notes: Grant metal-doping-takeaway.txt §§4–6, 8–9; Oxide electronic §2.3 (battery sibling callout); GaN:Mg (contrast: same element Mg, opposite mechanism class); Mechanism taxonomy (vocabulary placement); Chem for linguists §5; Solubility windows; Process metastability.


0. Design brief (problem-driven)

Slot Choice
Host Ni-rich NMC layered oxides — e.g. NMC811 (LiNi₀.₈Mn₀.₁Co₀.₁O₂), NCM90 / NCA-class Ni ≥ ~80–90%
Foreigners Mg²⁺ primary; Al³⁺, Nb⁵⁺ (often co-doped) as industrial siblings named in Grant takeaway §6
Named weakness of undoped Ni-rich host High capacity, but Li⁺/Ni²⁺ cation mixing, anisotropic lattice strain (esp. H2→H3), microcracking, lattice-oxygen release, rock-salt surface reconstruction, poor cycle life and thermal safety at high voltage / high SOC
Intended mechanism Structural / chemical lattice stabilization — pillar/rivet Mg in Li or TM sites; stronger Al–O / Nb–O bonding; suppressed phase transitions and O release — not shallow-donor carrier engineering
Service hint Automotive / high-energy Li-ion cells needing retention at ≥4.3 V and safer charged-state thermal behavior

Vocabulary lock (careful): Battery papers almost always say “doping.” That word is community-standard and Grant takeaway uses it. Mechanism-wise, however, this note is closer to structural/chemical stabilization of an intercalation host than to Si:P / GaN:Mg electronic doping. Prefer: papers say doping; we say structural stabilizer / lattice dopantIntentional dilute impurity meant to tune electronic/optical/catalytic behavior without replacing the host as majority phase. when teaching mechanism. Do not collapse into Al–Mg SSS or GaN:Mg acceptor talk. See §6 and taxonomy placement §6.1.


1. Chem-for-linguists checklist (§8) — filled (Ni-rich cathode edition)

1. What is the inherent weakness of the pure / base form?

  • Ni-rich layered oxides deliver high capacity (Ni redox), but as Ni rises past ~80%, cation disorder, H2–H3 lattice collapse, particle cracking, and oxygen evolution accelerate fade and raise thermal-runaway risk.
  • Not the weakness: missing hole density (that was GaN) or UV-only band gap (that was TiO₂). Different design problem → structural durability under Li extraction.

2. Which obstacle / tuning mechanism is doing the work?

Primary: dilute Mg / Al / Nb on Li and/or TM sites that: 1. suppress Li⁺/Ni²⁺ mixing (order the layered lattice); 2. pillar / rivet the structure against collapse and layer gliding at high SOC; 3. strengthen TM–O bonding (esp. Al–O, Nb–O) → less irreversible O release; 4. damp H2→H3 and related anisotropic strain → fewer microcracks.

Co-actors (first-class): synthesis route (coprecipitation vs solid-state lithiation), single-crystal vs polycrystalline morphology, surface coatings / concentration-gradient shells, electrolyte/cutoff voltage.
Explicitly not: Labusch SSS; Mg_Ga acceptor + H anneal (GaN); mid-gap photocatalysis (TiO₂:Fe).

3. What trade-offs are accepted?

  • Stability / retention ↔ inactive Mg (and sometimes Al) lowers reversible capacity (replaces redox-active Ni).
  • Mg on Li sites helps pillars / rivets ↔ blocks Li⁺ diffusion → rate capability and capacity drop if overdone.
  • Nb refines primary particles / expands spacing ↔ excess Nb can raise Li/Ni mixing or cut first-cycle CE in some reports (Al co-doping often mitigates).
  • Industrial maturity: Mg/Al/Nb cathode doping is industrial (Grant §6) — but optima remain composition- and process-route-dependent.

4. How critical is processing vs nominal composition?

Very. Same nominal at.% Mg can sit mostly on Li sites vs TM sites, be surface-enriched (gradient), or form secondary phases — depending on when Mg is introduced (precursor coprecipitation vs lithiation calcination) and wash/calcine path. Operando XRD, Rietveld Li/Ni mixing, DSC of charged electrodes, and cycle retention are the reality checks (Grant takeaway §9).

5. Problem-driven or exploratory?

Problem-driven and industrial. Ni-rich NMC/NCA cathodes are commercial high-energy LIB workhorses; Mg/Al/Nb doping is a standard stability toolkit. Research continues on dual-site codoping, gradients, and single-crystal “rivet” mechanisms — but the chassis is battery manufacturing, not research-only photocatalysis.


2. Host and dopant cast (REAL materials)

2.1 Ni-rich NMC host + Mg

Item Approximate figure / fact Role
Host Layered α-NaFeO₂-type (R3̅m) LiNiₓMnᵧCo_zO₂; Ni ≳ 0.8 High capacity; fragile at high SOC
Named failure modes Li⁺/Ni²⁺ mixing; H2→H3; O release; rock-salt skin; cracking Why doping is done
Mg ionic radius Mg²⁺ ~0.72 Å vs Li⁺ ~0.76 Å, Ni²⁺ ~0.69 Å Size-matched for Li and/or TM sites
Site preference Often Li-layer occupancy (pillar); also TM-site up to ~few % then spill to Li site (“rivet” literature) Mechanism depends on site
Useful Mg window Often ~0.5–1–few at.% (route-dependent); some NCM90 studies cite ~1% Mg for retention peak; industrial digests often keep Mg dilute Capacity–stability trade
Thermal (DSC class) e.g. ~2% Mg in high-Ni NC: exotherm peak 177 → 211 °C, heat 1780 → 1640 J g⁻¹ (charged + electrolyte); Al/Mg co-doped LNM91 onset 203 → ~208 °C, heat down Safer charged cathode
Cycle-life class Mg ~1% in NCM90-type: retention e.g. ~89% vs ~78% undoped after 100 cycles (0.1C class reports); dual Al/Nb in NCM92: ~88% after 1000 cycles in one industrial-style study Retention, not carrier density

Mechanisms to name (Mg): 1. Cation-mixing suppression — Mg occupation / lattice stiffening reduces Ni²⁺ on Li sites → better Li transport pathways and structural order.
2. Pillar effect — inactive Mg²⁺ in Li layer shields O²⁻–O²⁻ repulsion at high delithiation; damps c-axis collapse.
3. Rivet effect (single-crystal moderate-Ni literature) — Mg on Li sites clamps TM layers against gliding / O3→O1-type transitions and intragranular crack initiation — after TM-site saturation (~few %).
4. Thermal / oxygen stability — inhibits Ni migration and O release that feed exothermic electrolyte reactions (DSC evidence).
5. Capacity cost — electrochemically inactive Mg; Li-site Mg hinders Li⁺ diffusion when excessive.

Sources (headline): Batteries & Supercaps 2025 three-route Mg:NMC811 comparison (0.1–1% class; solid-state lithiation often best); J. Mater. Chem. A 2025 “rivet effect” Mg in NMC622 SC (TM site to ~3%, then Li site; capacity drop with excess); CrystEngComm 2025 ~1% Mg in NCM90; Chem. Mater. 2019 Mg high-Ni DSC 177→211 °C; Dalton Trans. 2023 coprecipitation Mg 0.25–0.5 wt% (Li-site occupancy; modest long-cycle gains); Grant takeaway §5 (battery cathodes: Mg stabilizes Ni-rich NMC).

2.2 Al and Nb — industrial sibling stabilizers

Dopant Typical site / chemistry What it buys Over / caution
Al³⁺ Prefers TM sites; strong Al–O (~512 kJ mol⁻¹ class) Lattice fortification; less O release / surface side reactions; H2–H3 damping; often expands Li-layer spacing Excess inactive Al → capacity loss; process-dependent mixing
Nb⁵⁺ High-valenceBonding capacity / oxidation-related electron count of an atom in context. TM-site / grain-refining Stronger Nb–O; radial primary particles; better rate / crack resistance; often with Al Solo Nb can increase Li/Ni mixing and hurt first-cycle CE in some reports; Al co-doping mitigates
Al+Nb dual Structure (Al) + microstructure (Nb) Synergy: e.g. AlNb-NCM92 ~88% retention / 1000 cycles; NCM83-NA ~95% / 100 cycles class Still a capacity–stability bargain; not magic

Grant takeaway §6 lists Mg, Al, Nb in NMC/LCO as industrial cathode doping — same maturity band as Si / GaN doping, different physics.

2.3 Contrast card — same element Mg, three hosts

Host Mg’s job Honest mechanism family Prefer saying
GaN:Mg p-type acceptor (Mg_Ga) + H activation Substitutional electronic doping Dopant (true electronic)
Ni-rich NMC:Mg (this file) Pillar / rivet / mixing suppressor / thermal stabilizer Structural / chemo-mechanical lattice stabilization (papers: “doping”) Lattice dopant / structural stabilizer — papers say doping; teach mechanism
Al–Mg SSS SoluteIntentional minority species dissolved in the host (alloying addition). pinning dislocationsLine defect enabling plastic slip at realistic stresses; edge vs screw are the two classic geometries. SSS Solute / alloying — avoid “doped”

This is the pedagogy point Grant’s takeaway invites: start with the host (§9.1). Mg is not one actor.


3. Concentration windows — under / useful / over

Zone Ni-rich NMC:Mg (order-of-magnitude) reading
Too little Negligible mixing suppression / H2–H3 damping; fade and O-release behavior ≈ undoped
Useful window Often ~0.5–1–few at.% Mg (process- and Ni-fraction-dependent): lower cation mixing, better retention, higher DSC onset / lower heat; Al/Nb similarly dilute (e.g. ~0.5–1% class each in dual recipes)
Capacity-cost onset Inactive Mg replaces Ni; discharge capacity falls even while retention % looks better
Li-path blocking Excess Mg on Li sites hinders Li⁺ diffusion → rate and capacity collapse (“more Mg ≠ better battery”)
Secondary / wrong regime Homogeneous excess → second phases or inactive Mg-rich regions; uncontrolled V_O / Ni migration in some DFT stories; Nb-alone mixing rise

Anchors (not cell recipes): Batteries & Supercaps 2025 NMC811 — 0.5–1 at.% via solid-state lithiation among best; JMCA 2025 rivet — Mg TM-site saturation ~3%, then Li-site population hurts diffusion; CrystEngComm 2025 — 1% Mg NCM90 retention peak class; Huang-line / industrial digests — excess Mg blocks Li paths. See also oxide checklist §2.3 sibling note.


4. “Over” is plural (name which failure)

Overrun mode Physics Where it shows up
Inactive-capacity loss Mg (Al) not redox-active; replaces Ni Lower first-cycle / reversible capacity
Li-diffusion blocking Mg on Li sites obstructs Li⁺ Rate fade; GITT/EIS diffusion drop
Site spillover Past TM-site preference → Li-site population Rivet benefit and transport cost (trade, not free)
Secondary phases / segregation Solubility / kinetics XRD/SEM second phases; inactive shells
Codopant imbalance e.g. Nb↑ mixing without Al partner First-cycle CE / energy density hit
Voltage / SOC miss Doping helps but cutoff still too aggressive H2–H3 / O release returns
Process-route miss Same at.% Mg, wrong incorporation step Coprecipitation vs lithiation performance split

Hard contrast vs GaN:Mg “over”: there, excess Mg → self-compensation (p falls) and PIDs. Here, excess Mg → capacity and Li-transport loss while the lattice may still look “more stable.” Same word “over-doping”; different meter.


5. Thermal stability & oxygen (service / safety clock)

Charged Ni-rich cathodes can release oxygen and heat with electrolyte — a service / safety failure mode, not just cycle-count fade.

Observation class Example Teaching point
Mg raises DSC peak / lowers heat NC-Mg: 211 °C / 1640 J g⁻¹ vs NC 177 °C / 1780 J g⁻¹ (4.3 V charged + electrolyte) Mg delays O-release-driven exotherm
Al/Mg co-doped onset up, heat down LNM91: 208.5 °C / 1671 J g⁻¹ vs pristine 203.2 / 1978 Dual stabilizer for thermal runaway margin
Al stabilizes lattice oxygen OEMS / RIXS lineages: less irreversible O₂; more reversible lattice O redox Strong Al–O; surface/bulk oxygen chemistry

Three clocks (Process metastability): equilibrium site preference vs process incorporation path vs service (high-SOC hold, calendar + cycle + abuse thermal). Bridge A under/useful/over is meaningless until you name which clock.


6. Hard contrast — vocabulary and taxonomy placement

6.1 Where does this sit in Mechanism taxonomy?

Family Fit for Ni-rich NMC:Mg?
1 Substitutional electronic doping Surface vocabulary yes (papers: dopant); mechanism no — not tuning n/p like Si:P or GaN:Mg
2 Intercalative CT Host is an intercalation cathode for Li⁺, but Mg/Al/Nb are lattice substituents, not the working intercalant
3 SSS No — not dislocation pinning for yield strength
4 Precipitation / second-phase Cousin only if secondary phases appear on overrun; the useful regime is dilute lattice incorporation, not wanted γ′-like precipitatesSecond-phase particle that forms (often on aging) and blocks dislocations.
5 Process clocks Yes, cross-cutting — incorporation route + service SOC/thermal

Honest placement for teaching: structural / chemo-mechanical lattice stabilization of a layered intercalation host, using community “doping” language. Closest pedagogical neighbors: oxide checklist’s battery sibling paragraph; Grant §5 battery row; not Family 1 electronic pegs and not Al–Mg SSS. If the taxonomy later grows a “lattice stabilizer in battery intercalation hosts” row, this file is the seed — do not force-fit into Family 1 just because the papers say “doped.”

6.2 Vocabulary table

Case Prefer saying Why
This file: NMC:Mg/Al/Nb Papers: doping; teaching: lattice dopant / structural stabilizer Intentional dilute impurity for stability, not carriers
GaN:Mg Dopant (true electronic) Acceptor → holes
Si:P/As/B Dopant (classical) Shallow donors/acceptors
TiO₂:Fe / AZO Dopant Mid-gap / TCO carriers (+ V_O)
Al–Mg AA5083 Solute / alloying SSS — avoid “doped”
CNT intercalation Intercalant Guest-in-gap CT

False-friend reminder: “Mg doping” in a battery abstract ≠ “Mg doping” in a GaN LED abstract ≠ “Mg doping” in casual talk about 5xxx Al. Host first (Grant §9.1).


7. Bridge A add-on checklist (depth note §6) — filled

  1. Where is the solubility / activation limit at process conditions?
    Dilute useful window ~0.5–1–few % Mg (and similar for Al/Nb); site preference (Li vs TM) and incorporation step matter more than a single solvus number.

  2. Intended mechanism?
    Structural / chemical stabilization of Ni-rich layered oxide — mixing suppression, pillar/rivet, O-release / thermal margin, cycle life. Explicitly not Si:P-style carrier engineering. Do not call Al–Mg SSS “doped”; do not call this GaN:Mg.

  3. What fails first past the window?
    Typically capacity loss and Li⁺ transport blocking; sometimes secondary phases or codopant imbalance — not “more Mg = more holes.”

  4. Metastability?
    Yes — charged-state oxygen release and rock-salt growth are service-clock failures; process path (when Mg is added) changes where Mg sits. Three clocks apply.

  5. Only then — SPECULATIVE Bridge A mnemonic
    See §9. Never invert: do not pick at.% Mg from a linguistic arity table.


8. Under / useful / over — materials only (no linguistics yet)

Zone Ni-rich NMC:Mg / Al / Nb reading
Under Undoped Ni-rich fade: mixing, H2–H3, cracks, O release, early DSC exotherm
Useful window Dilute Mg (± Al/Nb): ordered lattice, better retention, higher thermal onset, acceptable capacity
Over / wrong regime Inactive-capacity loss; Li-path blocking; secondary phases; Nb-alone mixing rise; still-too-high cutoff undoing the stabilizer

Honest split: retention-% can look great while absolute capacity is already compromised — “useful” is a bargain, not a single peak on one axis.


SPECULATIVE Cross-domain section. Open after the materials.

9. SPECULATIVE Bridge A overlay (supporting only)

Everything in this section is pedagogical analogy. It does not make doping × θ-roles a joint field.
Why this host earns a Bridge A peg: Grant’s takeaway already names battery-cathode Mg as industrial. The under/useful/over curve (retention and thermal margin vs capacity/rate) is a clean concentration-window story — but the saturated quantity is structural durability, not carrier density. This note stress-tests Bridge A: same count intuition, different filled slot than GaN:Mg / Si:P. Pedagogy win = teach that “over-doping” is plural across hosts.

Materials fact SPECULATIVE Bridge A reading
Useful ~0.5–1–few % Mg (± Al/Nb) → retention / thermal win Saturated intended structural configuration for Ni-rich duty
Undoped Ni-rich at high voltage Under-stabilized relative to cycle/safety intent
Excess Mg → capacity loss, Li blocking, secondary phases Overrun / failed regime: “extra” participants force inactive / blocked / second-phase repair that misses the energy-density + durability bargain
Continuous at.% vs discrete lexical arity Asymmetry: concentration is continuous; verb arity is typically discrete — do not force integer slots onto doping curves
Mg vs Al vs Nb as distinct cast members (sites, bond strengths, morphology roles) Do not smash into one “argument” type
Same element Mg in GaN vs NMC Same foreigner label, different saturated slot (holes vs lattice stability) — Bridge A count intuition transfers; mechanism labels must not

Earn: helps a linguist remember why battery “doping” papers and LED “doping” papers are not interchangeable, and why capacity can fall while retention rises.
Collapse: if pillar Mg is called a θ-role; if 1 at.% is treated as an arity integer; if Li-path blocking is equated with adjunct licensing; if Navajo morphological typing is treated as evidence for this count story; if a joint field is announced; if this file is filed under “electronic doping” just because the papers say doped.

Contrast — GaN:Mg vs cathode Mg ( SPECULATIVE pedagogy)

GaN:Mg NMC:Mg
Saturated quantity Free holes after activation Lattice stability / cycle + thermal margin
Useful meter Hall p vs SIMS [Mg] Retention, DSC onset, mixing %, capacity bargain
Classic “over” Self-compensation; PIDs Inactive capacity; Li blocking
Bridge A lesson More dopant can mean fewer carriers More dopant can mean less capacity / slower Li

Same SPECULATIVE count peg shape; do not merge into one materials claim.

Navajo morphology saturates typed participants on the verb. Bridge A is the count/windows half. Cathode Mg’s cast (layered host, Mg pillar/rivet, Al–O fortifier, Nb grain refiner, Li⁺ working ion, O lattice) is a tempting typed-participant mnemonic — keep it parallel, never as evidence that Bridge A’s under/useful/over ladder is proven by Athabaskan morphosyntax. Note: the working intercalant is Li⁺; Mg is a lattice stabilizer — do not confuse those typed slots.


10. Host contrast card (add battery Mg to the peg set)

Host Diagram / lattice First “over” (typical) Bridge A peg ( SPECULATIVE)
Al–Mg AA5083 Limited solvus; FCCFace-centered cubic — many slip systems; Al, Cu, Ni typically ductile. Al + subst. Mg GB β / sensitization Solvus overrun → unwanted second phase
Cu–Ni Isomorphous complete SSS Conductivity / TCR curves Property-curve overrun
Fe–C Eutectoid Fe–Fe₃C; interstitial C Cementite / martensite morphology Transformation / morphology overrun
Ni–γ′ γ/γ′ solvus; wanted precipitate Overaging / TCP Wanted γ′ vs wrong second phase
Mg (host) HCPHexagonal close-packed — fewer primary slip systems; Mg, Ti, Zn often anisotropic.; AZ/AM/ZK Brittle IMCs; galvanic β; ignition β ambivalent; safety “over”
Ti (host) HCP α / BCCBody-centered cubic — Fe/steel common; plasticity more temperature-sensitive. β; β-transus α-case; ω; O/H embrittlement Wanted α vs unwanted ω / α-case
Oxide electronic Wide-gap oxide + subst. metal + V_O Recombination; μ collapse; secondary oxides Carrier / mid-gap window — true dopant (research-leaning)
CNT/graphene intercalation sp² + guest in channels/gaps Moisture dedoping; junction R Intercalant loading window
GaN:Mg III–V + Mg_Ga acceptor (+ H) Self-compensation; PIDs; μ/optical loss Carrier window — true electronic dopant (industrial)
Si:P/As/B Diamond Si + shallow subst. Impurity band / Mott; μ collapse Carrier window — textbook electronic dopant
Ni-rich NMC:Mg/Al/Nb (this file) Layered oxide + dilute Mg/Al/Nb Inactive capacity; Li blocking; secondary phases; thermal still fails if cutoff insane Structural-stability window — papers say doping; mechanism ≠ Si:P

11. What this walkthrough trains you to do next

For any new battery-cathode “dopant” claim:

  1. Name the goal (cycle life / thermal / rate / high-voltage).
  2. Name site (Li vs TM), bond-strength story, and morphology role separately (Mg vs Al vs Nb).
  3. Sketch capacity and retention (and DSC if safety matters) vs at.% — find the bargain, not max dissolve.
  4. Name the first failure past the peak (inactive capacity vs Li blocking vs second phase vs mixing rise).
  5. Ask which clock (incorporation path vs high-SOC service).
  6. Ask whether the paper’s word “doping” means electronic carriers or lattice stabilization — host first.
  7. Only then, if useful for memory, hang the SPECULATIVE Bridge A count peg.

Contrast vs GaN:Mg note: same element, industrial maturity band, checklist shape — opposite mechanism class. That contrast is the point.


12. Non-claims

  • No joint field of doping × θ-roles / valencyArity: how many actants a predicate takes; formal vs semantic valency can diverge..
  • Bridge A remains SPECULATIVE pedagogy / probe; lock unrevised.
  • Navajo θ-morphology does not prove Bridge A count.
  • Numbers herein are order-of-magnitude literature anchors, not cell recipes or OEM specs.
  • This walkthrough does not revise Bridge A’s lock, invent shared formalism, or announce a product.
  • Does not claim Mg cathode doping is electronic doping, SSS, or intercalation of Mg as the working ion.


13. Sources consulted (2026-09-19 note)

Source Role
Grant metal-doping-takeaway.txt §§4–6, 8–9 Battery Mg / industrial Al,Nb,Mg in NMC; start-with-host
Oxide electronic §2.3 Prior battery sibling callout
GaN:Mg Contrast: Mg as true electronic dopant
Mechanism taxonomy Vocabulary / family placement hygiene
Batteries & Supercaps 2025 (DOI 10.1002/batt.202500662) Three Mg routes in NMC811; 0.5–1 at.%; cation mixing ↓
J. Mater. Chem. A 2025 (DOI 10.1039/d4ta08998g) “Rivet effect”; TM-site ~3% then Li site; diffusion cost
CrystEngComm 2025 (DOI 10.1039/d5ce00270b) ~1% Mg in NCM90; retention ~89% vs ~78%
Dalton Trans. 2023 (DOI 10.1039/d2dt02246j) Coprecipitation Mg; Li-site occupancy; excess blocks Li
Chem. Mater. 2019 (DOI 10.1021/acs.chemmater.8b03900) Mg high-Ni DSC 177→211 °C; heat ↓
ChemElectroChem 2025 Al/Mg codoped LNM91 DSC onset/heat; Al–O + Mg pillar synergy
Ind. Eng. Chem. Res. 2024 Al/Nb NCM83; ACS Energy Lett. 2024 Al/Nb NCM92; Energy Fuels 2025 Nb/Al SC NMC811; ChemElectroChem 2025 Nb/Al mechanisms Al/Nb industrial siblings; mixing / morphology / retention
Batteries & Supercaps 2025 Mg–Zr dual-site; Adamo–Manthiram Chem. Mater. 2024 Al/Mn H2–H3 Phase-transition mitigation context
Project-internal: Si / GaN / oxide / CNT + metal host checklists; CONCLUSIONS-v0 + v1-addendum; chem-for-linguists; depth-solubility-windows; depth-process-metastability Checklist discipline + contrast

v0 — 2026-09-19 continue-research. Ni-rich NMC Mg/Al/Nb structural-stabilizer walkthrough; GaN:Mg contrast; Bridge A flagged SPECULATIVE. No joint field.

Corpus note: worked-battery-cathode-mg-doping-checklist-v0.md. Materials first; Bridge A peg last.