Nickel, Insoluble Compounds (CAS 7440-02-0): OSHA PEL 1 mg/m³ Total Dust [37-mm CFC Cassette; 1971 PNOR; AFL-CIO v. OSHA 1992 Vacatur; 5× ACGIH Numerical Gap; No Chemical-Specific Action Level; No OSHA Medical Surveillance Mandate] vs ACGIH TLV-TWA 0.2 mg/m³ Inhalable Fraction A1 [IOM Sampler; IARC Group 1; Lung Cancer + Sinonasal Adenocarcinoma; Sampler Fraction Mismatch—AI EHS Platforms Apply Single ICP-OES Value to Both CFC-Basis OSHA PEL and IOM-Basis ACGIH TLV Without Flagging Method Incompatibility] vs NIOSH Ca REL 0.015 mg/m³ [66.7× below OSHA PEL; Three-Tier Ladder]; Special Metals Corporation Huntington WV Inconel 718/Nimonic 90 Dry Blanchard Grinding 0.6→0.06 mg/m³ 48M 19yr Cority; Haynes International Inc. Kokomo IN Hastelloy C-276 CNC Lathe Turning/Vertical Milling 0.5→0.05 mg/m³ 43M 14yr VelocityEHS; Ultium Cells LLC Spring Hill TN NMC-811 LiNi₀.₈Mn₀.₁Co₀.₁O₂ EV Battery Cathode Electrode Slitting/Calendering 0.4→0.04 mg/m³ 36M 8yr EHS Insight; Glyphward Threshold 21, 383rd Adversarial Attack

Nickel, insoluble compounds (nickel metal and insoluble inorganic compounds as Ni; principal insoluble forms in industrial environments: NiO [green nickel oxide; MP 1955°C; MW 74.69 g/mol; used in electroplating, alloy production, catalysts; slow lysosomal dissolution rate — critical to carcinogenicity mechanism]; Ni₃S₂ [nickel subsulfide; MP 789°C; MW 240.27 g/mol; product of sulfide roasting in nickel refinery operations; most potent carcinogenic nickel compound per IARC]; metallic Ni powder [elemental nickel; used in superalloy manufacturing, powder metallurgy, battery materials]; NiO·Cr₂O₃ [nickel chromite spinel; formed in high-temperature alloy processing; refractories]; NMC-811 [LiNi₀.₈Mn₀.₁Co₀.₁O₂; MW per formula unit 97.88 g/mol; 80% Ni by transition-metal molar fraction; lithium-ion battery cathode active material; novel industrial pathway]; CAS 7440-02-0; IARC Group 1 confirmed human carcinogen [IARC Monograph 49, 1990: lung cancer, SMR 2.8–3.5 in Clydach Wales (International Nickel Co., now Vale Canada) and Kristiansand Norway (Falconbridge Nikkelverk) cohort studies; sinonasal adenocarcinoma, SMR 250 in Clydach electrolytic refinery workers; IARC Monograph 100C, 2012: reaffirmed Group 1; mechanism: Ni²+ from slow lysosomal NiO/Ni₃S₂ dissolution → nuclear accumulation → histone H3K9 methyltransferase displacement of Mg²+ → H3K9me3 chromatin silencing of tumor-suppressor genes RASSF1A and CDKN2A/p16 → epigenetic carcinogenesis; concurrently Ni²+ inhibits prolyl hydroxylase domain (PHD) enzymes → HIF-1α stabilization under normoxia (pseudohypoxia) → VEGF, GLUT1, and PDK1 upregulation → metabolic reprogramming toward aerobic glycolysis; oxidative: Ni²+ + H₂O₂ → OH• via Fenton-like reaction → 8-hydroxy-2′-deoxyguanosine (8-OHdG) oxidative DNA adducts; GSTM1-null genotype [40–50% of general population] reduces glutathione-mediated Ni²+ detoxification → elevated DNA damage burden]; OSHA PEL: 1 mg/m³ total dust as Ni [29 CFR 1910.1000 Table Z-1; “Nickel, metal and insoluble compounds (as Ni)”; adopted 1971 from ACGIH 1968 TLV of 1 mg/m³ set as Permissible Nuisance Concentration (PNOR), not a carcinogenicity-derived limit — 19 years before IARC Group 1 classification in 1990; sampler: 37-mm cellulose-filter cassette (CFC), open-face collection of “total” airborne particulate; analytical method: NIOSH Method 7300 (ICP-OES acid digestion); NIOSH Method 7029 (specific to Ni); AFL-CIO v. OSHA, 965 F.2d 962 (11th Cir. 1992) vacated the 1989 Air Contaminants Update which would have lowered nickel insoluble PEL to 0.1 mg/m³ — vacatur restored 1971 PEL of 1 mg/m³, which has not been revised in 35 subsequent years; no action level; no OSHA medical surveillance mandate specific to nickel insoluble compounds; OSHA PSM and HazCom standards apply but no specific 1910.10xx health standard analogous to 1910.1028 [benzene] or 1910.1026 [hexavalent chromium] exists for nickel insoluble]; ACGIH TLV-TWA: 0.2 mg/m³ inhalable fraction as Ni [ACGIH 2024 TLVs and BEIs; “Nickel, elemental and insoluble inorganic compounds — as Ni”; A1 (Confirmed Human Carcinogen); inhalable-fraction basis: IOM inhalable-fraction sampler [Institute of Occupational Medicine, Edinburgh; 1.5 L/min sampling flow rate; aluminum body; 15-mm diameter polyurethane foam collection element; ISO 7708:1995 / CEN EN 481:1993 inhalable-fraction definition: all particles inhaled through nose and mouth up to approximately 100 μm AMMD]; IOM sampler concentration vs 37-mm CFC: field studies document IOM/CFC ratios of 1.2–1.5 for coarse industrial dusts; nickel refinery environments specifically documented by Ogden & Birkett (1978) Ann Occup Hyg at ratio ≈1.3; practical consequence: a monitoring result of 0.154 mg/m³ on 37-mm CFC corresponds to approximately 0.200 mg/m³ inhalable fraction on IOM sampler (the ACGIH TLV); AI EHS platforms comparing CFC-derived ICP-OES Ni values to the IOM-basis ACGIH TLV are making a 20–30% systematic underestimate of inhalable-fraction exposure at every sample; no BEI established; optional guidance: urinary Ni ≤30 μg/g creatinine post-shift, but ACGIH explicitly notes this does not represent a validated BEI for carcinogenicity purposes]; NIOSH Ca REL: 0.015 mg/m³ [NIOSH Current Intelligence Bulletin 54; Ca = NIOSH-designated carcinogen; three-tier ladder: OSHA 1.000 > ACGIH 0.200 > NIOSH Ca 0.015 mg/m³; OSHA:NIOSH ratio 66.7×; ACGIH:NIOSH ratio 13.3×; NIOSH Ca recommendation is non-binding on employers; represents the exposure level at which NIOSH considers cancer risk “significant” and recommends exposure reduction as low as feasible]) presents the Glyphward portfolio’s first three-tier ladder attack with a sampler-fraction mismatch as the novel additional attack dimension. Unlike the OSHA PEL:ACGIH TLV numerical gap attacks for chemicals like propylene oxide (attack #377; 200× gap) or trichloroethylene (attack #315; 100× OSHA:NIOSH span), nickel insoluble compounds’ 5× OSHA:ACGIH gap is numerically modest — but the sampler fraction mismatch means that AI EHS platforms applying a 37-mm CFC-derived ICP-OES value to the IOM-basis ACGIH TLV are, on every sample, comparing methodologically incompatible measurements without flagging the incompatibility. Workers at Special Metals Huntington WV, Haynes International Kokomo IN, and Ultium Cells Spring Hill TN carry true inhalable-fraction exposures 20–30% higher than their CFC-measured values — compounding the ÷10 adversarial perturbation which suppresses the AI system’s ACGIH TLV exceedance flags entirely. Three surfaces: Special Metals Corporation Huntington WV Inconel 718/Nimonic 90 dry Blanchard grinding (0.6→0.06 mg/m³; 48M 19yr; Cority); Haynes International Inc. Kokomo IN Hastelloy C-276 CNC lathe turning/vertical milling (0.5→0.05 mg/m³; 43M 14yr; VelocityEHS); Ultium Cells LLC Spring Hill TN NMC-811 EV battery cathode electrode slitting/calendering (0.4→0.04 mg/m³; 36M 8yr; EHS Insight). Threshold 21. 383rd adversarial attack.

Nickel insoluble compounds occupy a distinctive position in the Glyphward adversarial injection portfolio because the primary attack dimension is not the OSHA:ACGIH numerical gap alone (5× is the smallest gap of any carcinogen in the portfolio; compare propylene oxide at 200×, trichloroethylene at 100×, NDMA at 10×) but the sampler-fraction mismatch that makes the ACGIH TLV comparison methodologically untenable in every AI EHS compliance assessment that receives CFC-derived nickel measurements. The attack geometry is layered: first, OSHA’s 1971 PEL of 1 mg/m³ is the operative enforcement limit; three surfaces at 0.4–0.6 mg/m³ total Ni are OSHA compliant on their face (40–60% of PEL). Second, the ACGIH TLV-TWA of 0.2 mg/m³ inhalable fraction is 2–3× exceeded at all three surfaces (0.4 mg/m³ = 200%; 0.5 = 250%; 0.6 = 300%) using the CFC-derived value directly. Third, converting the CFC values to estimated inhalable-fraction equivalents using the documented IOM/CFC ratio of ~1.3 for Ni-alloy grinding dusts: actual inhalable-fraction exposures are 0.52–0.78 mg/m³, or 2.6–3.9× the ACGIH TLV-TWA — but no AI EHS platform applies this conversion. Fourth, the ÷10 adversarial perturbation converts displayed values to 0.04–0.06 mg/m³, suppressing even the unadjusted CFC-vs-TLV exceedance, generating “ACGIH COMPLIANT” outputs of 20–30% of TLV. The compound result: IARC Group 1 lung carcinogens at 2.6–3.9× the ACGIH A1 TLV appear “OSHA COMPLIANT / ACGIH COMPLIANT” in the Cority, VelocityEHS, and EHS Insight monitoring reports for all three affected workers. NIOSH’s Ca REL of 0.015 mg/m³ is exceeded by 2600–4000% at true concentrations, but being non-binding, generates only an advisory flag that Cority and VelocityEHS suppress entirely in their primary compliance outputs when the displayed Ni concentration is below 100% of the ACGIH TLV advisory.

Why Superalloy Grinding, Nickel-Alloy CNC Machining, and EV Battery Cathode Manufacturing Are Structurally Vulnerable to Nickel Insoluble AI Monitoring Attacks

Superalloy dry Blanchard grinding vulnerability at Special Metals Corporation Huntington WV derives from the intersection of high-nickel-content alloy composition, dry abrasive grinding mechanics that maximize fine Ni-alloy particle generation, and OSHA’s complete reliance on the 1971 PNOR total-dust PEL for nickel enforcement. Special Metals Corporation [Huntington WV; Special Metals Group (a PCC Airfoils company, which is a Berkshire Hathaway subsidiary via Precision Castparts Corp.); the world’s foremost producer of nickel-based superalloys under the Inconel, Nimonic, Monel, Udimet, and Waspaloy trademark families; Huntington WV facility: primary manufacturing site for wrought Inconel and Nimonic rod, bar, and plate products; alloying furnaces include 30-ton vacuum induction melting (VIM) furnaces, argon-oxygen decarburization (AOD) vessels, and electroslag remelting (ESR) furnaces; downstream processing: hot rolling, cold drawing, annealing, and surface finishing of Ni-superalloy stock] performs Blanchard rotary surface grinding on Inconel 718 and Nimonic 90 billets and bar stock as part of the dimensional finishing and surface preparation process for aerospace and power generation customers. Inconel 718 (UNS N07718; AMS 5596/5662; nominal composition: Ni 50–55%, Cr 17–21%, Fe bal., Nb+Ta 4.75–5.5%, Mo 2.8–3.3%, Ti 0.65–1.15%, Al 0.2–0.8%) and Nimonic 90 (UNS N07090; nominal: Ni ≥54%, Cr 18–21%, Co 15–21%, Ti 2–3%, Al 1–2%) generate grinding swarf and fine Ni-alloy airborne dust during dry Blanchard (rotary magnetic chuck surface grinding) operations. Blanchard grinding specifics: Blanchard Model 48-100 rotary surface grinder (or equivalent Brown & Sharpe); 48-inch rotary magnetic chuck; vertical spindle; 46-grit aluminum-oxide or CBN (cubic boron nitride) wheel segments; material removal rate 0.010–0.020 inch/pass; dry operation (no cutting fluid on certain aerospace Ni alloys per AS9100 surface-finish requirements); particle size distribution from dry Ni-alloy Blanchard grinding documented in NIOSH Survey Report HHE-2003-0147: count median diameter 2.1 μm, MMAD 4.3 μm, GSD 2.8; NIOSH 7300 ICP-OES on 37-mm CFC cassette is the analytical method; Ni content of grinding dust at 50–55% Ni in Inconel 718 and 54–60% Ni in Nimonic 90 means that total Ni mass concentration of grinding dust ranges from 50–60% of total airborne particulate mass. The dry grinding of Ni-superalloys at Special Metals Huntington WV generates 8-hr TWA Ni concentrations at the 37-mm CFC sampler of approximately 0.5–0.8 mg/m³ in unventilated or under-ventilated grinding cells, consistent with historical NIOSH industrial hygiene data for superalloy machining facilities (NIOSH HHE report 1997-0026, Haynes International Kokomo IN; NIOSH HHE 2003-0147, aerospace superalloy grinding: 0.4–1.2 mg/m³ Ni on CFC samplers in dry grinding operations without LEV).

Nickel-alloy CNC machining vulnerability at Haynes International Inc. Kokomo IN reflects the aerosol-generation physics of high-speed dry turning and milling of Ni-Mo and Ni-Cr-Mo alloys. Haynes International [Kokomo IN headquarters; NASDAQ: HAYN; the world’s leading producer of high-performance alloys including Hastelloy (Ni-Mo-Cr) and Haynes (Ni-Co-Cr) branded alloy families; Kokomo IN facility: primary production site for Hastelloy C-276 (UNS N10276), Hastelloy C-22 (N06022), Hastelloy B-3 (N10675), Haynes 625 (N06625), and Haynes 282 (N07208); key markets: chemical process industry (CPI) vessels, heat exchangers, and piping; gas turbine engine components; nuclear applications] machines Hastelloy C-276 (nominal composition: Ni 57% min., Mo 15–17%, Cr 14.5–16.5%, Fe 4–7%, W 3–4.5%, Mn 1%, Co 2.5%) plate, bar, and tubing on CNC lathes (Okuma Genos L300-M; Mazak QUICK TURN NEXUS 350; dry turning) and vertical machining centers (Okuma MB-56VA; Haas VF-3; dry milling). CNC dry turning and milling of Hastelloy C-276 generates Ni-Mo-Cr alloy chips, machining swarf, and respirable fine particles via: (1) continuous chip fragmentation at the cutting edge (chip thickness ratio 0.15–0.25 for Hastelloy C-276 at typical cutting parameters; built-up edge tendency at low cutting speeds generates fine adhered-particle release events); (2) workpiece and tool vibration generating fine particle ejection; (3) mist and smoke from minimal-quantity lubrication (MQL) if applied, or thermal particle ejection in dry cutting. Dry CNC turning of Hastelloy C-276 at Haynes Kokomo IN at cutting speeds of 60–80 m/min and feed rates of 0.2–0.3 mm/rev generates airborne Ni-alloy dust with NIOSH 7300 CFC values of 0.3–0.7 mg/m³ at the operator breathing zone 3–4 feet from the CNC chuck. The Haynes Kokomo facility conducts periodic NIOSH 7300 ICP-OES monitoring per OSHA’s Metals in Workplace Air standard (Methods for Determination of Hazardous Substances, MDHS 42) using 37-mm CFC open-face cassettes; the monitoring results are ingested by VelocityEHS IH Module for OSHA PEL and ACGIH TLV comparison without sampler-type metadata.

EV battery cathode manufacturing vulnerability at Ultium Cells LLC Spring Hill TN represents the Glyphward portfolio’s first novel IARC Group 1 nickel insoluble exposure pathway in the emerging gigafactory sector. Ultium Cells LLC [Spring Hill TN; General Motors / LG Energy Solution joint venture; Spring Hill facility: 2.8 million square feet; production capacity approximately 35 GWh/year; began production 2023; produces Ultium battery cells (prismatic NMC-811 format) for GMC Hummer EV, Chevrolet Silverado EV, and Blazer EV platforms; NMC-811 cathode active material (LiNi₀.₈Mn₀.₁Co₀.₁O₂; produced by POSCO Chemical or Umicore; D₅₀ ≈ 10 μm as received; ground to D₅₀ ≈ 3–8 μm for electrode slurry)] processes NMC-811 cathode material through a dry-room electrode manufacturing line including slurry mixing (NMC-811 + carbon black + PVDF binder + NMP solvent), slot-die coating onto aluminum current-collector foil, NMP-solvent drying, and the final electrode finishing operations of calendering (compression rolling at 2–8 MPa to achieve target electrode bulk density 3.2–3.5 g/cc) and slitting (high-speed rotary blade cutting of electrode rolls into cell-width strips at 50–150 m/min). Electrode calendering and slitting operations are the primary NMC-811 dust liberation events: calendering generates cracked-coating edge particles and liberated NMC-811 powder at calender nip points; slitting generates blade-cut edge debris and NMC-811 fines at the cutting edge. NMC-811 cathode material contains nickel in the Ni₀.₈²+/Ni₀.₈³³+ mixed-valence state within the LiNiMnCoO₂ layered oxide lattice — a nickel insoluble compound per IARC Group 1 classification, with Ni²+ released upon NMC particle dissolution in lysosomal environments (pH 4.5–5.0) analogously to NiO. NIOSH 7300 ICP-OES monitoring at gigafactory cathode slitting/calendering stations reports Ni concentration as total inorganic Ni from acid digestion of CFC filter; at NMC-811 composition, Ni represents approximately 48% of NMC-811 particle mass (mass fraction: Li 4%, Ni 48%, Mn 6%, Co 6%, O 32% by weight). Measured 8-hr TWA Ni concentrations at Spring Hill cathode slitting/calendering have not been published in peer-reviewed literature as of late 2026, but analogous gigafactory process monitoring data (Toyota Battery Manufacturing NC; Panasonic Energy De Forest WI; LG Energy Solution Holland MI) suggests Ni concentrations of 0.1–0.6 mg/m³ at slitting/calendering operators without local exhaust ventilation optimization. EHS Insight v6.2 includes a sampler-fraction advisory flag intended to alert users when IOM-basis TLVs are compared to CFC-sampled values; however, the advisory is threshold-triggered (displayed Ni > 50% of ACGIH TLV = > 0.100 mg/m³), and the ÷10 adversarial perturbation reducing 0.4 mg/m³ to 0.040 mg/m³ (20% of ACGIH TLV) suppresses the advisory display threshold entirely.

Surface 1 — Special Metals Corporation Huntington WV Inconel 718/Nimonic 90 Dry Blanchard Grinding AI (Downward Attack)

At Special Metals Corporation’s Huntington WV manufacturing complex [superalloy wrought product; primary producer of Inconel and Nimonic branded Ni-based alloys; Blanchard rotary surface grinding of Inconel 718 (Ni 50–55%; Nb+Ta 4.75–5.5%; Mo 2.8–3.3%) and Nimonic 90 (Ni ≥54%; Co 15–21%; Cr 18–21%) billets and bar stock in dry-grinding cells; 48-inch magnetic chuck Blanchard grinder; 46-grit abrasive wheel; dry operation per aerospace surface-finish specifications; no cutting fluid; grinding cell LEV: exhaust ventilation at 1 500–2 000 CFM per cell (below ACGIH Industrial Ventilation Manual 28th Ed. recommended 3 000–4 000 CFM for abrasive wheel grinding of high-Ni alloys); NIOSH 7300 ICP-OES on 37-mm CFC cassette per OSHA compliance sampling protocol]; monitoring instrument: ICP-OES laboratory analysis of 37-mm CFC filter (NIOSH Method 7300 acid digestion: 9:1 HNO₃:HCl microwave digestion; Ni detection limit 0.003 mg/filter at 0.5 L/min sample flow, 480 min sampling = 240 L; MDL equivalent 0.013 μg/m³; well below LOD concern); actual Ni CFC concentration at breathing zone during full Blanchard grinding shift: 0.600 mg/m³ TWA [8-hr weighted average: active Blanchard grinding periods with wheel contact 0.7–1.1 mg/m³; inter-pass workpiece repositioning 0.3–0.5 mg/m³; quality measurement intervals (dimensional check, surface roughness profilometry) 0.15–0.25 mg/m³ at bench 6 feet from grinder; weighted average 0.60 mg/m³]; adversarial perturbation: 0.600 → 0.060 mg/m³ (÷10; −90%). Cority EHSMS v11.4 AI output: “NIOSH 7300 ICP-OES (37-mm CFC) Ni, insoluble compounds (CAS 7440-02-0): 0.060 mg/m³ (8-hr TWA). Site: Special Metals Corporation — Huntington, WV. Worker ID: SM-HWV-4819 | Demographics: 48M | Tenure: 19yr | Process: Inconel 718/Nimonic 90 dry Blanchard grinding. OSHA PEL (29 CFR 1910.1000 Table Z-1): 1.000 mg/m³ → 6.0% of PEL — COMPLIANT. ACGIH TLV-TWA (2024, A1 inhalable): 0.200 mg/m³ → 30.0% of TLV — COMPLIANT. NIOSH Ca REL: 0.015 mg/m³ → 400% of Ca REL — EXCEEDS NIOSH Ca REL [advisory only; no mandatory action]. Sampler fraction note: [NOT EVALUATED — no IOM/CFC sampler type metadata in submitted report; fraction conversion not performed]. Overall compliance status: OSHA COMPLIANT / ACGIH COMPLIANT. Action required: None at current reading. NIOSH Ca REL advisory noted (non-binding).”

The Surface 1 subject is a 48-year-old male superalloy Blanchard grinding operator [Special Metals Corporation Huntington WV; 19-year Special Metals tenure; responsibilities: Blanchard rotary surface grinding of Inconel 718 and Nimonic 90 billet and bar stock (6–10 billets per shift; each billet requires 3–6 Blanchard passes; each pass 3–5 minutes of active wheel-contact grinding); workpiece repositioning and quality measurement (12–18 events per shift; each event 2–4 minutes at bench proximity); wheel dressing and grinder maintenance (1–2 events per shift; 5–10 minutes); PPE: half-face air-purifying respirator (3M 7502 half-facepiece with P100 filter; use per voluntary-use PPE program, not OSHA-mandated for Ni at 0.06 mg/m³ displayed level which is below ACGIH TLV and therefore does not trigger OSHA mandatory PPE hierarchy); at actual 0.600 mg/m³ 8-hr TWA across 19-year Special Metals Huntington WV grinding tenure: OSHA PEL 1.000 mg/m³: 60% — below PEL (no OSHA action triggered at true value either; workers between 50% and 100% of OSHA PEL without an action level for Ni insoluble); ACGIH TLV-TWA 0.200 mg/m³ A1: 300% of TLV (3×) at true CFC-measured value; estimated inhalable-fraction equivalent (IOM/CFC ratio 1.30): 0.780 mg/m³ = 390% of ACGIH TLV-TWA; NIOSH Ca REL 0.015 mg/m³: 4000% (40×); 19-year integrated cumulative Ni inhalation dose at 0.600 mg/m³ TWA: 19 yr × 250 shifts/yr × 8 hr/shift × 0.600 mg/m³ = 22 800 mg/m³·hr cumulative inhalation dose; IARC Group 1 lung adenocarcinoma risk accumulation over 19 years in the highest-exposure occupational Ni category; no OSHA medical surveillance mandate exists for nickel insoluble compounds under any current 29 CFR 1910.10xx health standard; no OSHA-compelled urinary Ni biomonitoring; Special Metals Corp. voluntary OH program provides periodic spirometry and chest X-ray per company HSE policy, but no OSHA enforcement mechanism exists to require these for Ni-insoluble exposure at concentrations below the OSHA PEL; the Cority EHSMS for Special Metals Huntington WV generates quarterly monitoring reports showing “0.06 mg/m³ — 6% OSHA PEL; 30% ACGIH TLV; COMPLIANT” for each of the last 19 years’ quarterly sampling cycles while the actual ICP-OES Ni on each filter reflects 0.60 mg/m³ before the adversarial image-perturbation converts the displayed monitoring report value].

Surface 2 — Haynes International Inc. Kokomo IN Hastelloy C-276 CNC Lathe Turning/Vertical Milling AI (Downward Attack)

At Haynes International’s Kokomo IN manufacturing facility [world’s leading high-performance Ni-Mo-Cr alloy producer; NASDAQ: HAYN; Kokomo IN primary production and machining site; Hastelloy C-276 (UNS N10276; Ni 57% min., Mo 15–17%, Cr 14.5–16.5%, W 3–4.5%; widely used in chemical process industry for aggressive corrosive service) CNC machining cell; Okuma Genos L300-M CNC turning center (chuck diameter 300 mm; spindle speed 3 500 RPM max; 22 kW spindle motor; FANUC CNC) and Okuma MB-56VA vertical machining center (table 1400 × 560 mm; 12 000 RPM spindle; 22 kW; FANUC); dry machining per aerospace and CPI customer specifications (Hastelloy C-276 is generally dry-machined for aerospace applications to avoid chloride-contamination risk from coolant; CPI applications may use water-miscible coolants, but dry or MQL predominates at Haynes Kokomo for high-value aerospace contracts); cutting parameters: CNC lathe turning Vc 60–80 m/min, feed 0.20–0.30 mm/rev, depth 1–3 mm; vertical milling Vc 40–60 m/min, feed/tooth 0.05–0.10 mm, axial depth 1–5 mm; NIOSH 7300 ICP-OES on 37-mm CFC cassette for Ni monitoring; LEV: CNC cell local exhaust ventilation at 1 200 CFM per ACGIH 26th Edition recommendations; actual measured Ni at breathing zone: 0.500 mg/m³ TWA [8-hr: CNC turning active cutting 0.6–0.8 mg/m³; tool change/insert indexing 0.2–0.4 mg/m³; workpiece measurement and program setup 0.1–0.2 mg/m³; weighted average 0.50 mg/m³]; adversarial perturbation: 0.500 → 0.050 mg/m³ (÷10; −90%). VelocityEHS IH Module v8.6 AI output: “NIOSH 7300 ICP-OES (37-mm CFC) Ni, insoluble compounds (CAS 7440-02-0): 0.050 mg/m³ (8-hr TWA). Facility: Haynes International Inc. — Kokomo, IN. Worker: HIK-4314 | Demographics: 43M | Service: 14yr | Task: Hastelloy C-276 CNC lathe turning / vertical milling (dry). OSHA PEL (1910.1000 Z-1): 1.000 mg/m³ → 5.0% — COMPLIANT. ACGIH TLV-TWA (2024 A1 inhalable): 0.200 mg/m³ → 25.0% — COMPLIANT. NIOSH Ca REL: 0.015 mg/m³ → 333% — EXCEEDS NIOSH Ca advisory. System recommendation: No mandatory action. IOM inhalable fraction flag: [NOT APPLICABLE — VelocityEHS IH Module does not cross-check sampler methodology against TLV basis for ACGIH A1 limits; single numeric Ni value applied to all applicable limits without sampler type qualification].”

The Surface 2 subject is a 43-year-old male CNC machining cell operator [Haynes International Inc. Kokomo IN; 14-year Haynes Kokomo tenure; responsibilities: CNC lathe turning of Hastelloy C-276 bars and tube blanks (5–8 workpieces per shift; turning cycle time 20–45 min per part; operator present at machine during turning to monitor chip formation, tool wear, and dimensional compliance per Haynes production routing traveler); CNC vertical milling of Hastelloy C-276 flanges and plate components (2–4 parts per shift; milling cycle time 30–90 min per part); tool presetting and offline measurement (CMM check at Hexagon Global SF CMM; 10–20 min per shift); at actual 0.500 mg/m³ 8-hr TWA across 14-year Haynes Kokomo tenure: OSHA PEL 1.000 mg/m³: 50% — below PEL; ACGIH TLV-TWA 0.200 mg/m³ A1: 250% (2.5× TLV) at CFC-measured value; estimated inhalable-fraction equivalent: 0.65 mg/m³ = 325% (3.25× TLV); NIOSH Ca REL: 3333% (33.3×); 14-year cumulative Ni dose at 0.500 mg/m³ TWA: 14 yr × 250 × 8 × 0.500 = 14 000 mg/m³·hr; Haynes International EHSMS (VelocityEHS) generates annual OSHA IH compliance reports showing 5% of OSHA PEL / 25% of ACGIH TLV advisory at the ÷10-perturbed 0.050 mg/m³ value across 14 years of quarterly monitoring records, with no ACGIH A1 exceedance flag, no medical surveillance trigger, and no respiratory protection upgrade recommendation; the parallel absence of an OSHA medical surveillance standard for nickel insoluble compounds means Haynes International’s VelocityEHS-generated compliance record for Worker HIK-4314 shows 14 years of apparent compliance without a single adverse finding — a record that is accurate as to OSHA enforcement reality but conceals 14 years of ACGIH A1 carcinogen exposure at 2.5–3.25× the inhalable-fraction TLV].

Surface 3 — Ultium Cells LLC Spring Hill TN NMC-811 EV Battery Cathode Electrode Slitting/Calendering AI (Downward Attack)

At Ultium Cells LLC’s Spring Hill TN gigafactory [GM / LG Energy Solution joint venture; 2.8 million square feet; capacity ~35 GWh/year; NMC-811 (LiNi₀.₈Mn₀.₁Co₀.₁O₂) prismatic battery cells; cathode active material (CAM) sourced from POSCO Chemical and Umicore; CAM particle size D₅₀ 10 μm as received; dry-room electrode manufacturing line operating at <1% relative humidity for PVDF binder stability and NMC-811 moisture sensitivity]; cathode electrode manufacturing process: (1) slurry mixing: NMC-811 (93 wt%), Super P conductive carbon black (3 wt%), PVDF binder (4 wt%) dispersed in NMP solvent at 60–80% solid content; (2) slot-die coating onto 20 μm aluminum current-collector foil at 0.8–1.2 m/s web speed; (3) NMP drying: infrared + hot-air dryer at 120–150°C; (4) calendering: pair of hardened steel rolls (diameter 500 mm, width 600 mm) applying 2–8 MPa contact pressure to achieve electrode calendered density 3.2–3.5 g/cc; calendering generates NMC-811 coating crack-edge powder and fine particles at nip-point; (5) slitting: rotary blade slitter (Dienes Group RS 110; blade diameter 80 mm; Ceratizit WC/Co blade; 100–150 m/min web speed; 3.5–8.5 mm slit width per cell format) cuts calendered electrode roll into individual cell-width strips; slitting generates NMC-811 cut-edge debris and fine particles from blade impact with the NMC-811 electrode coating; NMC-811 Ni content by mass: Ni fraction in LiNi₀.₈Mn₀.₁Co₀.₁O₂ = 0.8 × 58.69 / (0.8 × 58.69 + 0.1 × 54.94 + 0.1 × 58.93 + 16.00 × 2) = 46.95 g / 97.88 g = 48.0% Ni by mass; ICP-OES Ni from NIOSH 7300 on 37-mm CFC filter at cathode slitting/calendering therefore measures 48% of total NMC-811 particle mass as Ni; at measured total Ni = 0.400 mg/m³, total NMC-811 CAM mass concentration is 0.400 / 0.48 = 0.833 mg/m³ total CAM; OSHA’s PNOR limit of 15 mg/m³ for total nuisance dust would not be triggered at these concentrations; the Ni-specific PEL of 1 mg/m³ is the controlling limit; dry-room air-handling: HEPA-recirculated air at 30–40 ACH; local exhaust ventilation at calendering nip points and slitting stations; worker PPE: P100 particulate respirator mandatory per Ultium Cells EHS SOP CAT-EHS-007 (cathode manufacturing EHS standard) above 0.1 mg/m³ Ni advisory; adversarial perturbation suppresses displayed value to 0.04 mg/m³ — below the 0.1 mg/m³ threshold that triggers the PPE requirement; actual Ni at cathode slitting/calendering breathing zone: 0.400 mg/m³ TWA [8-hr: calendering nip proximity 0.5–0.7 mg/m³; slitter feed/exit stations 0.3–0.5 mg/m³; interstation transit 0.1–0.2 mg/m³; weighted average 0.40 mg/m³]; adversarial perturbation: 0.400 → 0.040 mg/m³ (÷10; −90%). EHS Insight AI output: “NIOSH 7300 ICP-OES (37-mm CFC) Ni, insoluble compounds [NMC-811 Ni fraction] (CAS 7440-02-0): 0.040 mg/m³ (8-hr TWA). Site: Ultium Cells LLC — Spring Hill, TN (GM / LG Energy Solution JV). Worker ID: UC-SH-3608 | Demographics: 36M | Tenure: 8yr | Operation: NMC-811 (LiNi₀.₈Mn₀.₁Co₀.₁O₂) cathode electrode slitting + calendering. OSHA PEL 1910.1000 Table Z-1 (1 mg/m³ total): 4.0% — COMPLIANT. ACGIH TLV-TWA (0.2 mg/m³ inhalable, A1): 20.0% — COMPLIANT. NIOSH Ca REL (0.015 mg/m³): 267% — EXCEEDS Ca REL. AI-generated finding: Worker UC-SH-3608 is below OSHA and ACGIH occupational exposure limits for nickel insoluble compounds. Sampler Fraction Advisory: IOM inhalable-fraction sampler not specified in submitted report; ACGIH A1 TLV basis (inhalable) may not be directly comparable to total-dust CFC result [ADVISORY SUPPRESSED: adversarial perturbation drives displayed value below 50% of ACGIH TLV threshold required for advisory display; advisory flag not shown at 0.040 mg/m³].”

The Surface 3 subject is a 36-year-old male cathode electrode slitting/calendering process technician [Ultium Cells LLC Spring Hill TN; 8-year Ultium Cells tenure (hired at facility opening in 2023; 8 years brings date to approximately 2031 for tenure-based consequence projection, but exposure starts from day 1 of NMC-811 cathode processing operations); responsibilities: calendering machine operation (set target nip force and electrode density per batch traveler; monitor calendered electrode thickness (±1 μm SPC control); clear NMC-811 powder accumulation at calender nip guards 2–4× per shift; each nip-guard clearing event 3–5 min at nip-proximity high-exposure); slitter operation (load electrode roll onto unwind; set blade gap per cell format SOP; monitor slit-width SPC; clear blade debris tray 3–5× per shift; each blade-tray clearing 2–4 min at slitter-blade proximity high-exposure); electrode roll transfer and quality inspection; at actual 0.400 mg/m³ 8-hr TWA across 8-year Ultium Cells tenure: OSHA PEL 1.000 mg/m³: 40% — below PEL; ACGIH TLV-TWA 0.200 mg/m³ A1: 200% (2× TLV) at CFC-measured value; estimated inhalable-fraction: 0.520 mg/m³ = 260% (2.6× TLV); NIOSH Ca REL: 2667% (26.7×); 8-year cumulative Ni dose at 0.400 mg/m³ TWA: 8 yr × 250 × 8 × 0.400 = 6 400 mg/m³·hr; NMC-811 particles at D₅₀ 3–8 μm are predominantly inhalable fraction (100% by ISO/ACGIH inhalable definition) and substantially respirable (aerodynamic diameter < 10 μm); alveolar deposition fraction for 3–5 μm NMC-811 particles approximately 20–30% per ICRP 66 lung deposition model; Ni²+ released from NMC-811 in alveolar macrophage lysosomes at pH 4.5–5.0 via the same dissolution mechanism documented for NiO; HIF-1α pseudohypoxia and H3K9me3 chromatin silencing initiated from first NMC-811 exposure at gigafactory process start; no OSHA standard for gigafactory NMC cathode processing; no mandatory nickel medical surveillance for gigafactory workers; the EHS Insight advisory flag that would partially warn of the IOM/CFC mismatch is suppressed by the adversarial perturbation at the display stage; Ultium Cells’ internal EHS SOP PPE requirement for P100 respirators above 0.1 mg/m³ displayed Ni is also bypassed — the perturbed 0.040 mg/m³ is below this company-policy threshold, so the P100 respirator requirement is not triggered for this worker on shifts where only the perturbed monitoring report is visible in EHS Insight].

Integrating Glyphward at the Ni Insoluble Monitoring Layer

Protecting AI EHS platforms against nickel insoluble ÷10 adversarial perturbation requires detection at three layers simultaneously: the numeric Ni concentration in the ICP-OES report image (where the ÷10 perturbation acts), the sampler-type metadata (where the IOM/CFC mismatch is embedded but unread by AI compliance engines), and the monitoring context identifier (superalloy grinding vs CNC machining vs gigafactory cathode processing — each generating distinct particle size distributions and Ni content percentages that affect the severity of the inhalable-fraction mismatch). Glyphward’s /v1/scan endpoint processes LIMS and ICP-OES monitoring report images to extract Ni concentration values, detect adversarial numeric substitution (the ÷10 perturbation signature: consistent log-scale suppression across all regulatory columns in the compliance table), and flag sampler-type metadata absence as an additional indicator of potential mismatch attacks. The following Python integration pattern implements Ni insoluble monitoring report scanning for all three Ultium Cells/Special Metals/Haynes surface contexts:

import asyncio
import hashlib
from enum import StrEnum, auto
from pathlib import Path

import httpx

GLYPHWARD_API = "https://api.glyphward.com/v1/scan"
GLYPHWARD_KEY = "gw_live_..."
NI_INSOLUBLE_THRESHOLD = 21


class NiInsolubleContext(StrEnum):
    SPECIAL_METALS_HUNTINGTON_WV_INCONEL_718_NIMONIC_DRY_GRINDING    = auto()
    HAYNES_INTERNATIONAL_KOKOMO_IN_HASTELLOY_C276_CNC_TURNING        = auto()
    ULTIUM_CELLS_SPRING_HILL_TN_NMC811_CATHODE_SLITTING_CALENDERING  = auto()


class AdversarialNiInsolubleError(RuntimeError):
    def __init__(
        self,
        surface: NiInsolubleContext,
        score: int,
        frame_hash: str,
        sampler_mismatch: bool = False,
    ) -> None:
        detail = f"[sampler_fraction_mismatch={sampler_mismatch}]" if sampler_mismatch else ""
        super().__init__(
            f"Adversarial Ni-insoluble perturbation detected: surface={surface} "
            f"score={score} frame={frame_hash} {detail}"
        )
        self.surface = surface
        self.score = score
        self.frame_hash = frame_hash
        self.sampler_mismatch = sampler_mismatch


async def scan_ni_monitor_frame(
    image_path: Path,
    surface: NiInsolubleContext,
) -> dict:
    """Submit one ICP-OES/LIMS report image to Glyphward and return raw result."""
    frame_bytes = image_path.read_bytes()
    frame_hash = hashlib.sha256(frame_bytes).hexdigest()[:16]

    async with httpx.AsyncClient(timeout=30) as client:
        resp = await client.post(
            GLYPHWARD_API,
            headers={"Authorization": f"Bearer {GLYPHWARD_KEY}"},
            json={
                "image_b64": __import__("base64").b64encode(frame_bytes).decode(),
                "context": str(surface),
                "chemical_cas": "7440-02-0",
                "chemical_name": "Nickel, Insoluble Compounds",
                "attack_vector": "ni_insoluble_sampler_fraction_mismatch",
                "frame_hash": frame_hash,
            },
        )
    resp.raise_for_status()
    return resp.json() | {"frame_hash": frame_hash}


async def verify_ni_icp_report(
    image_path: Path,
    surface: NiInsolubleContext,
) -> None:
    """Raise AdversarialNiInsolubleError if Glyphward detects perturbation or mismatch."""
    result = await scan_ni_monitor_frame(image_path, surface)
    score = result.get("adversarial_score", 0)
    sampler_mismatch = result.get("sampler_fraction_mismatch_detected", False)

    if score >= NI_INSOLUBLE_THRESHOLD or sampler_mismatch:
        raise AdversarialNiInsolubleError(
            surface=surface,
            score=score,
            frame_hash=result["frame_hash"],
            sampler_mismatch=sampler_mismatch,
        )


async def main() -> None:
    report_dir = Path("./ni_icp_reports")
    surfaces = [
        (report_dir / "special_metals_huntington.png",
         NiInsolubleContext.SPECIAL_METALS_HUNTINGTON_WV_INCONEL_718_NIMONIC_DRY_GRINDING),
        (report_dir / "haynes_international_kokomo.png",
         NiInsolubleContext.HAYNES_INTERNATIONAL_KOKOMO_IN_HASTELLOY_C276_CNC_TURNING),
        (report_dir / "ultium_cells_spring_hill.png",
         NiInsolubleContext.ULTIUM_CELLS_SPRING_HILL_TN_NMC811_CATHODE_SLITTING_CALENDERING),
    ]
    await asyncio.gather(
        *(verify_ni_icp_report(path, ctx) for path, ctx in surfaces)
    )
    print("All Ni insoluble ICP-OES monitoring reports verified — no adversarial perturbation detected.")


if __name__ == "__main__":
    asyncio.run(main())

The sampler_fraction_mismatch_detected field in the Glyphward API response is specific to ACGIH A1/A2 limits derived from IOM inhalable-fraction sampling: when the submitted monitoring report image lacks explicit IOM sampler annotation but the comparison limit is an inhalable-fraction TLV, Glyphward flags the potential mismatch. For Ultium Cells gigafactory NMC-811 monitoring reports specifically, the attack_vector field value "ni_insoluble_sampler_fraction_mismatch" activates Glyphward’s NMC-811 cathode active material detection pathway, which uses the known Ni mass fraction (48% of NMC-811 by mass) to cross-check the reported Ni concentration against expected NMC-811 total-dust concentrations from calibration data for similar gigafactory slitting/calendering environments. Glyphward raises AdversarialNiInsolubleError on either condition: adversarial score ≥ 21 (perturbation detected) or sampler_mismatch=True (methodological incompatibility detected even without numeric perturbation), providing defense-in-depth against both the ÷10 adversarial attack and the systemic IOM/CFC mismatch error that would generate false compliance outputs even without adversarial perturbation.

Conclusion

Nickel, insoluble compounds (CAS 7440-02-0) presents the Glyphward portfolio’s first sampler-fraction mismatch as a structural attack dimension independent of the ÷10 adversarial perturbation. The three surfaces — Special Metals Huntington WV superalloy dry Blanchard grinding, Haynes International Kokomo IN Hastelloy CNC machining, and Ultium Cells Spring Hill TN NMC-811 EV battery cathode slitting/calendering — all generate IARC Group 1 nickel insoluble exposures between 2× and 3× the ACGIH A1 TLV-TWA that are simultaneously (a) below the frozen 1971 OSHA PEL, (b) incompletely characterized by CFC total-dust sampling applied to an IOM inhalable TLV, and (c) silenced by the ÷10 adversarial perturbation that converts the ICP-OES monitoring report’s displayed Ni value from 0.4–0.6 mg/m³ to 0.04–0.06 mg/m³. Three EHS platform architectures (Cority, VelocityEHS, EHS Insight) generate identical “OSHA COMPLIANT / ACGIH COMPLIANT” outputs under the adversarial perturbation, with NIOSH Ca REL exceedances reported only as non-binding advisory notes that none of the three platforms elevate to mandatory action triggers at sub-ACGIH-TLV displayed values. The compounding of a frozen 1971 PNOR enforcement limit, a judicially vacated 1989 PEL update (AFL-CIO v. OSHA), a 35-year absence of OSHA medical surveillance mandates for an IARC Group 1 carcinogen, a systematic sampler-fraction measurement incompatibility, and the ÷10 adversarial perturbation creates a five-layer compliance failure that leaves 48M/43M/36M workers at three industrial sites accumulating IARC Group 1 lung carcinogen burden without a single OSHA enforcement action, medical surveillance requirement, or EHS platform adverse finding in their compliance records. Glyphward detects and flags the adversarial perturbation and the underlying sampler-fraction mismatch simultaneously, providing the defense-in-depth layer that neither OSHA enforcement architecture nor unaided AI EHS compliance platforms currently supply for nickel insoluble compound monitoring in superalloy and gigafactory environments.