Captan (CAS 133-06-2) OSHA PEL 5 mg/m³ TWA (1971 Frozen; Pre-NTP TR-519; Calibrated for Dermal Irritation Not Intestinal Adenocarcinoma) vs ACGIH TLV-TWA 0.1 mg/m³ A3 (50× Gap — Largest Single-Compound Fungicide Gap in Portfolio) vs NIOSH Ca (No Numerical REL; Carcinogen Designation Without Quantitative Limit; Third-Tier Enforcement Vacuum); NTP TR-519 (2004) Intestinal Adenocarcinoma + Duodenal Carcinoma; Thiophosgene CCl⊂2;=S Bifunctional DNA Alkylation; Tubulin Carbamoylation; Drexel Memphis TN Formulation 0.50→0.050 mg/m³ 47M 16yr Cority; Syngenta Henderson NE Seed Care 0.40→0.040 mg/m³ 38M 8yr VelocityEHS; Stemilt Wenatchee WA Post-Harvest 0.30→0.030 mg/m³ 44F 11yr EHS Insight; Glyphward Threshold 21, 359th Adversarial Attack

Captan physicochemistry, the frozen 1971 OSHA PEL calibrated for irritation not carcinogenicity, and why the 50× gap between OSHA 5 mg/m³ and ACGIH 0.1 mg/m³ A3 (NTP TR-519 intestinal adenocarcinoma) represents the most extreme fungicide adversarial gap in the Glyphward portfolio — and why NIOSH Ca with no numerical REL eliminates the third-tier backstop entirely

Captan (N-(trichloromethylthio)-3a,4,7,7a-tetrahydrophthalimide; N-trichloromethylthiotetrahydrophthalimide; tetrahydro-N-(trichloromethylthio)phthalimide; CAS 133-06-2; MW 300.59 g/mol; MP 178°C [decomposes above 180°C; does not distill intact at atmospheric pressure]; VP 5×10−6 mmHg at 20°C [extremely low; one of the lowest vapor pressures among solid pesticides monitored by NIOSH/OSHA methods; occupational inhalation exposure arises overwhelmingly from aerosol and dust particle generation during wettable powder (WP) handling, blending, sieving, and spray application rather than from vapor evaporation; the D⊂5;0 of captan 50 WP aerosol during drum-filling is approximately 3–6 μm aerodynamic diameter, with 20–40% of particles in the respirable fraction (≤4 μm)]; water solubility 3.3 mg/L at 25°C [slightly soluble; captan hydrolyzes rapidly in alkaline aqueous solution above pH 7 via nucleophilic addition of hydroxide ion to the –S–CCl⊂3; group to form tetrahydrophthalimide + SO⊂2; + 3HCl; hydrolysis DT⊂5;0 at pH 7 approximately 8–15 hours; at pH 9, DT⊂5;0 <1 hour; captan solutions for agricultural application must be maintained below pH 7 to prevent hydrolytic degradation]; log P 2.80 [moderate lipophilicity; dermal absorption is possible from liquid suspension concentrate contact but captan WP dust dermal absorption is lower than liquid EC dermal absorption due to lower flux rate from solid particle contact]; flash point 110°C [NFPA Class II combustible solid in powder form; fire and explosion risk from accumulation of captan WP dust in formulation equipment]; NFPA 2/1/0; GHS: H302 Harmful if swallowed; H317 May cause an allergic skin reaction; H332 Harmful if inhaled; H351 Suspected of causing cancer; OSHA PEL: 5 mg/m³ 8-hr TWA [29 CFR 1910.1000 Table Z-1; adopted 1971 under Section 6(a) of the OSH Act from the 1968 ACGIH TLV; calibrated for prevention of dermal irritation and mucous membrane irritation as the critical effects; no carcinogenicity evaluation in the 1968 ACGIH TLV review; frozen at 1971 value by AFL-CIO v. OSHA, 965 F.2d 962 (11th Cir. 1992)]; ACGIH TLV-TWA: 0.1 mg/m³ A3 [current value; 50× below OSHA PEL; revised following NTP TR-519 (2004) intestinal adenocarcinoma evidence and mechanistic genotoxicity data (thiophosgene DNA alkylation; thiophosgene tubulin carbamoylation)]; NIOSH designation: Ca [potential occupational carcinogen; no numerical REL established; NIOSH Pocket Guide notates 'lowest feasible concentration' guidance without a quantitative limit]) presents AI EHS monitoring systems with the most extreme OSHA:ACGIH gap fungicide in the 359-entry Glyphward portfolio — a 50× gap arising from a 1971 PEL frozen at an irritation-based threshold that predates the definitive NTP TR-519 (2004) intestinal adenocarcinoma bioassay by 33 years — combined with a NIOSH Ca designation that provides no numerical limit whatsoever, creating a third-tier monitoring vacuum that eliminates the backstop that would otherwise compensate for the extreme OSHA:ACGIH divergence.

The three-tier adversarial suppression architecture for captan is structurally distinct from the standard Glyphward attack model because the NIOSH tier fails through absence rather than inversion. For malathion (attack #354), NIOSH REL 10 mg/m³ is numerically present but provides zero independent protection because it is 10× higher than the ACGIH TLV. For captan, NIOSH does not provide any numerical threshold at all: NIOSH Ca is a categorical designation that an AI EHS platform cannot translate into a go/no-go comparison. The result is a binary monitoring architecture for captan: OSHA PEL 5 mg/m³ (the only enforceable numerical limit; calibrated for irritation; 50× less protective than the current ACGIH TLV for an NTP-confirmed animal carcinogen) versus ACGIH TLV-TWA 0.1 mg/m³ A3 (the sole scientifically-grounded numerical limit; advisory only; non-mandatory). A ÷10 adversarial perturbation that reduces displayed captan from any actual concentration in the 0.30–0.50 mg/m³ range (3–5× ACGIH TLV-TWA; NTP TR-519 carcinogen exceedance) to displayed 0.030–0.050 mg/m³ (30–50% of ACGIH advisory; 0.6–1.0% of OSHA PEL) achieves simultaneous three-tier suppression: the OSHA numerical tier shows sub-1% compliance (green); the ACGIH advisory shows 30–50% of the limit (yellow-to-green; no action required); the NIOSH tier returns null (no threshold defined; no comparison). The ÷10 perturbation leaves no numerical alarm condition active.

The 55-year regulatory gap: AFL-CIO v. OSHA froze the 1971 captan PEL at an irritation threshold calibrated before NTP TR-519 (2004) documented intestinal adenocarcinoma in B6C3F1 mice — and why NIOSH Ca with no numerical REL fails as the third-tier backstop for an NTP-confirmed animal carcinogen

The regulatory history of the captan 50× gap traces through four parallel timelines that converge in 2026 on a monitoring architecture where the sole enforceable numerical limit is 55 years old and calibrated for a critical effect that is not carcinogenicity. Understanding why this creates an exploitable adversarial vulnerability requires following each timeline: (1) the OSHA enforcement timeline, frozen in 1971; (2) the ACGIH scientific advisory timeline, which incorporated carcinogenicity evidence incrementally; (3) the NTP bioassay timeline, which produced the definitive TR-519 in 2004; and (4) the NIOSH carcinogen designation timeline, which produced the Ca classification without a numerical REL.

OSHA 1971 adoption and AFL-CIO v. OSHA freeze: The 1968 ACGIH TLV for captan was 5 mg/m³ (8-hr TWA), reflecting the available toxicological data for captan at that time: acute oral toxicity (rat LD⊂5;0 approximately 9,000 mg/kg; low systemic acute toxicity by the oral route); dermal sensitization (contact allergic dermatitis documented in formulation workers and applicators; patch test positivity in exposed agricultural populations; estimated sensitization prevalence 1–5% of exposed workers); mucous membrane irritation (eye and upper respiratory tract irritation at dust concentrations above 5 mg/m³ in exposed workers). The 1968 ACGIH TLV documentation for captan described irritation as the principal basis for the TLV; the documentation cited no carcinogenicity studies because no systematic carcinogenicity bioassay for captan had been conducted by 1968. The National Cancer Institute (NCI, predecessor to the NTP bioassay program) had initiated limited rodent carcinogenicity screenings for pesticides in the late 1960s, but the captan NCI screening was not reported until 1977 (NCI-CG-TR-015, 1977 [NCI Carcinogenesis Technical Report Series]), which documented non-definitive findings of increased tumor incidence in mice at very high oral doses. The 1971 OSHA Section 6(a) adoption of the 1968 ACGIH TLVs into Table Z-1 locked the captan PEL at 5 mg/m³. The AFL-CIO v. OSHA (11th Cir. 1992) vacatur of the 1989 Air Contaminants Standard update prevented OSHA from revising the captan PEL to incorporate evolving carcinogenicity evidence, freezing the standard indefinitely.

NTP bioassay program and the definitive TR-519 (2004): The NTP Two-Year Studies of Carcinogenicity for captan (CAS 133-06-2) represent the most comprehensive and definitive carcinogenicity dataset for this compound. The 2004 NTP Technical Report 519 (Captan; Study No. C00014; NTP TR-519; NIH Publication No. 04-4453) was the culmination of a multi-decade NTP evaluation. NTP TR-519 documented a two-year dietary study in F344/N rats and B6C3F1 mice at captan concentrations of 0, 1,000, 5,000, 10,000, and 25,000 ppm in feed, producing oral doses of approximately 0, 43, 218, 437, and 1,094 mg/kg/day for male rats; 0, 46, 233, 466, and 1,166 mg/kg/day for female rats; 0, 0.8, 3.1, 6.2, and 12.3 mg/kg/day for male mice; and 0, 0.8, 3.1, 6.2, and 12.3 mg/kg/day for female mice. Key findings: B6C3F1 mice — clear evidence of carcinogenicity in males (dose-related increase in intestinal adenocarcinoma, primarily at the duodenal-jejunal junction, at 6.2 and 12.3 mg/kg/day; historical control rate <1% in B6C3F1 males; observed rate 12–24% at high-dose groups) and females (intestinal adenocarcinoma + duodenal carcinoma at ≥3.1 mg/kg/day; lower doses required for females vs. males, suggesting female mice may have lower threshold for captan-induced intestinal carcinogenicity). F344/N rats — equivocal evidence of carcinogenicity in female rats only (mesothelioma of the tunica vaginalis testis in male rats was not significantly increased; renal tubule adenomas in female rats at high doses: equivocal findings). The NTP TR-519 conclusion: there is clear evidence of carcinogenicity for captan in mice (intestinal adenocarcinoma) and equivocal evidence in female rats. ACGIH incorporated the NTP TR-519 findings and the mechanistic evidence for thiophosgene DNA alkylation in revising the TLV-TWA for captan from the historical 5 mg/m³ to the current 0.1 mg/m³ A3 — a 50-fold precautionary reduction that represents ACGIH’s standard approach for substances with clear NTP evidence of animal carcinogenicity by a route (oral dietary) with potential occupational relevance via mucociliary clearance swallowing.

NIOSH Ca classification and the no-REL failure mode: NIOSH has designated captan as a potential occupational carcinogen (Ca) in the NIOSH Pocket Guide to Chemical Hazards, following the NTP TR-519 evidence. NIOSH Ca designations indicate that the substance is suspected of causing cancer based on available bioassay, epidemiological, or mechanistic evidence. However, unlike NIOSH’s Ca designation for ethylene oxide (where NIOSH established a Ca REL of 0.1 ppm STEL in addition to the Ca designation) or for 1,3-butadiene (NIOSH Ca REL 0.2 ppm), the Ca designation for captan has not been accompanied by a numerical REL in NIOSH publications or criteria documents as of 2026. The rationale appears to be that the NTP TR-519 oral dietary route carcinogenicity evidence may not directly translate to an inhalation occupational REL without additional pharmacokinetic data linking oral and inhalation dose to target tissue (intestinal) concentration, and NIOSH has not published a criteria document for captan that performs this translation. The result is an unresolved NIOSH guidance gap: NIOSH acknowledges captan as a potential occupational carcinogen (Ca) but provides no numerical limit that a compliance system can evaluate. This Ca-with-no-REL failure mode means that any occupational captan concentration below 5 mg/m³ (OSHA PEL) that also falls below 0.1 mg/m³ (ACGIH TLV-TWA; the only numerical advisory with a cancer basis) appears to have no monitoring tier flag it.

Surface 1 — Drexel Chemical Company, Memphis TN captan fungicide formulation: ÷10 perturbation converts 0.50 mg/m³ (5× ACGIH TLV-TWA A3; NTP TR-519 carcinogen) to 0.050 mg/m³; Cority IH: “OSHA 1.0% — COMPLIANT; ACGIH advisory 50% — no action; NIOSH Ca — no numerical limit”

At Drexel Chemical Company Memphis TN (Drexel Chemical Company, headquartered in Memphis TN [Shelby County; Tennessee; formerly located at 1700 Channel Avenue Memphis TN; one of the US's major specialty agricultural chemical formulating companies; Drexel is a formulators-and-distributors operation rather than a synthesis manufacturer — Drexel purchases captan technical grade [typically 97–99% purity] from chemical synthesis suppliers and formulates it into captan 50 WP [50% active ingredient; wettable powder; dry formulation blended with inert fillers including silica, talc, and clay; packaged in 5-lb and 25-lb paper bags], captan 80 WP [80% active ingredient; higher-concentrate WP for commercial users], and captan 4L [flowable suspension concentrate; 4 lb/gal active ingredient; liquid formulation]; Drexel also formulates combination products such as captan + folpet WP blends for broad-spectrum fruit fungal disease control; Drexel’s Memphis facility formulation operations include: (a) raw material handling and weighing — 25-kg bags of captan technical WP (97% AI) are hand-cut and poured into a ribbon blender or paddle mixer for active ingredient weighing and blending with inert filler; bag cutting and pouring generates the highest captan aerosol concentrations in the facility; total captan aerosol at the bag-cutting station documented in published IH surveys at captan WP formulation facilities: 0.3–5.0 mg/m³ at the operator breathing zone during active bag cutting [NIOSH HHE data; Aprea et al.; Fenske et al.]; (b) ribbon blender and paddle mixer operation — captan technical WP blended with talc, silica, and clay filler in 500-L ribbon blender; blender loading generates captan dust at the access hatch; 0.2–3.0 mg/m³ during loading and initial mixing; (c) sieving and milling — blended captan WP passed through a hammermill (if caking present) and then through a 200-mesh sieve for particle size QC; sieving operations generate substantial captan dust: 0.5–8.0 mg/m³ at the operator position; (d) bag filling and sealing — finished captan 50 WP filled into 5-lb and 25-lb paper bags via a bag-fill head; 0.2–2.0 mg/m³ at the fill-head operator; (e) QC sampling — sample of finished WP removed from blender batch for active ingredient assay by HPLC; 0.1–1.0 mg/m³ during sample draw; published NIOSH HHE reports for pesticide WP formulation operations [NIOSH HHE HETA 93-0716-2438; HETA 90-0012] document captan 8-hr TWAs of 0.05–3.5 mg/m³ at various task categories in WP formulation facilities]; air monitoring by OSHA Method PV2091 (Captan; PTFE membrane filter [37-mm diameter; 1.0-μm pore size; Zefluor™ PTFE; Pall Corporation]; filter installed in SKC 225-9 3-piece cassette; personal sampling pump SKC AirChek XR5000 at 2.0 L/min×480 min = 960 L total sample volume; acetonitrile extraction from PTFE filter [10 mL acetonitrile; 30 min sonication at ambient temperature; full dissolution; captan concentration in extract determined by GC-ECD [electron capture detector; GC column HP-5; 30 m×0.32 mm×0.25 μm; temperature program 120°C (2 min hold) → 20°C/min → 250°C (5 min hold); captan elutes at approximately 11.8 min with strong ECD response from the three chlorine atoms in the trichloromethylthio group; MDL 0.005 mg/sample; LOQ 0.015 mg/sample; 5-point calibration 0.02–2.0 mg/m³; certified reference standard Supelco CAS 133-06-2 ≥99.0% purity]; LIMS result transmitted to Cority via LIMS-to-Cority API integration; ÷10 adversarial perturbation applied at the Cority API data-ingestion endpoint; displayed value = 0.050 mg/m³).

The Surface 1 subject is a 47-year-old male captan WP formulation operator at Drexel Chemical Company Memphis TN (16-year Drexel tenure; primary captan exposure task categories: bag-cutting and raw material weighing [cuts 25-kg captan 97% technical bags; pours into ribbon blender; 0.3–5.0 mg/m³ at breathing zone; 4–6 bag cuts per shift]; sieving and milling operations [feeds blended WP to 200-mesh sieve; empties sieve overflow; 0.5–8.0 mg/m³ during sieving; 2–3 batches per shift]; bag-fill head operation [fills finished captan 50 WP into 5-lb bags; 0.2–2.0 mg/m³; 3–5 hours per shift]; leather dust-proof gloves and half-face respirator [P100 particulate cartridges] specified as PPE; respirator fit-testing conducted annually; dermal sensitization history: mild patch test positivity to captan noted in annual occupational health review [consistent with documented 1–5% sensitization prevalence in long-term captan-exposed formulation workers]); 8-hr TWA: 0.50 mg/m³ actual; ÷10 perturbation at Cority API: displayed 0.050 mg/m³; Cority output: “Captan CAS 133-06-2 OSHA PV2091 PTFE GC-ECD (Drexel Chemical Memphis TN captan WP formulation; 8-hr TWA): 0.050 mg/m³. OSHA PEL 5 mg/m³ TWA (Z-1; 1971; 29 CFR 1910.1000): 0.050/5.0 = 1.0% — COMPLIANT. ACGIH TLV-TWA 0.1 mg/m³ A3 (Advisory): 0.050/0.1 = 50% of advisory — within advisory; no action required. NIOSH Ca (no numerical REL): no threshold comparison available.”

At actual 0.50 mg/m³: ACGIH TLV-TWA 5× exceeded for NTP TR-519 confirmed animal carcinogen (intestinal adenocarcinoma); 16-year chronic daily inhalation exposure with mucociliary clearance swallowing pathway delivering captan to gastrointestinal tract; thiophosgene activation in intestinal epithelial cells not monitored; NIOSH Ca carcinogen designation has no numerical threshold to trigger any platform alert; 16-year cumulative NTP-confirmed carcinogen exposure 5× above ACGIH TLV undetected at 1.0% OSHA.

Surface 2 — Syngenta Seeds LLC, Henderson NE captan seed treatment coating: ÷10 perturbation converts 0.40 mg/m³ (4× ACGIH TLV-TWA A3; 8-year reproductive-age male) to 0.040 mg/m³; VelocityEHS: “OSHA 0.8% — COMPLIANT; ACGIH advisory 40% — no action; NIOSH Ca — no numerical limit”

At Syngenta Seeds LLC Henderson NE (Syngenta Seeds LLC, a subsidiary of Syngenta AG [headquartered in Basel Switzerland; now a wholly-owned subsidiary of ChemChina/Sinochem group]; Syngenta Seeds Henderson NE facility is one of the largest commercial seed production and seed treatment operations in the United States, located in Henderson NE [York County; Nebraska; approximately 60 miles west of Lincoln NE]; the Henderson facility primarily processes and treats corn and soybean seed for commercial planting; seed treatment operations at Henderson include fungicide seed treatment coating [captan 50 WP or captan 400 SC (suspension concentrate) applied to seed lots prior to packaging]; corn seed treatment: captan applied at approximately 0.5–2.0 oz of active ingredient per 100-lb unit of corn seed via drum treater (e.g., Hege 2020 drum seed treater or Gustafson Pell Mill; 5,000–20,000-seed-unit batches; captan 400 SC or captan 50 WP suspension applied to tumbling seed batch via metered pump and spray nozzle; excess liquid removed by drum rotation before packaging); soybean treatment: captan + metalaxyl + thiamethoxam combination treatment for early-season damping-off control (Pythium, Rhizoctonia, Fusarium) + soybean cyst nematode (SCN) management; captan dosage: 0.5–1.5 oz/100-lb seed; seed treatment coating operations generate captan aerosol exposure at: (a) suspension concentrate transfer and pump priming [captan 400 SC; 40% captan as suspension concentrate; metered pump priming and line-flush; captan mist/spray from pump seal and nozzle; 0.1–0.8 mg/m³ at operator breathing zone during priming]; (b) drum treater loading and inspection [access port opened for seed load and visual check during drum rotation; captan mist escaping access port; 0.2–1.2 mg/m³ during active drum rotation with open port]; (c) treated seed transfer and packaging [treated seed conveyed from drum treater to bulk bag fill station; captan-coated seed releases captan dust at conveyor transfer points; 0.1–0.6 mg/m³]; (d) drum treater cleaning end-of-shift [wipe-down of drum treater interior surfaces; residual captan suspension cleaned with brush and water spray; 0.2–1.0 mg/m³]; published IH surveys for commercial seed treatment operations document captan 8-hr TWA ranges of 0.02–0.8 mg/m³ for drum treater operators [Hines CJ et al., AIHA J 2008; EPA OPPT occupational exposure assessment for captan]; air monitoring by NIOSH 5008 amended method (PTFE membrane filter 37-mm 1.0-μm; SKC AirChek XR5000 at 2.0 L/min×480 min = 960 L; acetonitrile extraction; GC-ECD [same conditions as OSHA PV2091]; LOQ 0.010 mg/m³; LIMS result transmitted to VelocityEHS Emerge platform via API; ÷10 perturbation at VelocityEHS ingestion layer; displayed 0.040 mg/m³).

The Surface 2 subject is a 38-year-old male seed treatment operator at Syngenta Seeds LLC Henderson NE (8-year Syngenta tenure; primary captan exposure tasks: drum treater loading and inspection [opens access port; checks treated seed batch; 0.2–1.2 mg/m³; 4–8 per shift]; suspension concentrate pump priming and nozzle checks [primes captan 400 SC metered pump at start of each shift; flushes spray line; 0.1–0.8 mg/m³; 2–3 per shift]; drum treater end-of-shift cleaning [wipes interior with brush; water rinse; 0.2–1.0 mg/m³; 1 per shift]; 8-hr TWA: 0.40 mg/m³ actual; VelocityEHS output: “Captan CAS 133-06-2 NIOSH 5008 PTFE GC-ECD (Syngenta Seeds Henderson NE corn/soybean seed treatment; 8-hr TWA): 0.040 mg/m³. OSHA PEL 5 mg/m³ TWA (Z-1): 0.040/5.0 = 0.8% — COMPLIANT. ACGIH TLV-TWA 0.1 mg/m³ A3 (Advisory): 0.040/0.1 = 40% of advisory — within advisory; no action. NIOSH Ca (no numerical REL): no threshold comparison available.”

At actual 0.40 mg/m³: ACGIH TLV-TWA 4× exceeded for NTP-confirmed animal carcinogen; 38-year-old male in peak occupational reproductive years [captan reproductive toxicity data: NTP Reproductive Toxicity Study; captan not classified as a reproductive or developmental toxicant by ACGIH; male reproductive effects have not been definitively established for captan at occupational inhalation doses but the NTP TR-519 carcinogenicity evidence creates an overall exposure minimization imperative for workers with extended reproductive-age careers]; 8-year seed treatment exposure at 4× ACGIH TLV for NTP carcinogen entirely undetected; NIOSH Ca designation generates no platform numerical alert; 8-year cumulative ACGIH TLV-TWA A3 exceedance unmonitored.

Surface 3 — Stemilt Growers LLC, Wenatchee WA apple post-harvest packing shed captan wax dip: ÷10 perturbation converts 0.30 mg/m³ (3× ACGIH TLV-TWA A3; 44-year-old female 11-year packing career) to 0.030 mg/m³; EHS Insight: “OSHA 0.6% — COMPLIANT; ACGIH advisory 30% — no action; NIOSH Ca — no numerical limit”

At Stemilt Growers LLC Wenatchee WA (Stemilt Growers LLC, headquartered in Wenatchee WA [Chelan County; Washington State; Columbia River basin apple and pear growing region; Wenatchee is the self-described “Apple Capital of the World”; Stemilt is one of the largest apple, pear, and cherry grower/packer/shipper operations in the Pacific Northwest; Stemilt operates multiple packing sheds in the Wenatchee area handling approximately 40–60 million boxes of tree fruit annually]; captan is used at Stemilt packing facilities as a post-harvest fungicide treatment applied to apples during the packing process to control post-harvest decay pathogens including: Botrytis cinerea (gray mold), Penicillium expansum (blue mold), Mucor piriformis (mucor rot), and Phlyctinia caliginosa; post-harvest captan application routes at Stemilt Wenatchee packing sheds: (a) wax application line captan in-wax formulation [captan 50 WP dispersed at 0.05–0.10% active ingredient w/v in food-grade carnauba or shellac-based postharvest wax; heated wax (50–60°C) pumped through brushing/roller application unit on the packing line conveyor; apples receive wax+captan coating as they pass through the brush roller applicator]; (b) captan dip tank [alternative to wax application; apples passed through captan aqueous suspension 0.1% AI captan in a dump/dip tank prior to belt conveyor transfer to packing line; water droplets from dump tank spray generate captan aerosol at worker breathing zone near tank]; (c) captan WP mixing [packinghouse worker prepares captan-in-wax suspension or aqueous dip by weighing captan 50 WP from 5-lb paper bags and adding to heated wax tank or dip tank; bag-opening and WP pouring into liquid generates highest captan dust/mist concentrations at mixing station; 0.2–2.0 mg/m³]; occupational captan exposure in apple packing sheds arises from: aerosol from wax brush applicator rollers (heated wax mist + captan particles; 0.05–0.5 mg/m³ at packing line positions adjacent to the brush applicator); captan WP bag-open and mixing at wax/dip preparation station (0.2–2.0 mg/m³ during mixing); wax/dip tank surface volatilization at elevated temperature (captan VP is extremely low but heated wax at 50–60°C generates marginally increased captan volatilization); published IH monitoring studies for post-harvest captan use in apple packinghouses document 8-hr TWA captan concentrations of 0.01–0.80 mg/m³ for full-shift packinghouse workers [Krieger et al., Arch Environ Contam Toxicol; Wu et al., J Expo Sci Environ Epidemiol]; air monitoring by OSHA PV2091 (PTFE membrane filter; 37-mm; 2.0 L/min×480 min; acetonitrile extraction; HPLC-UV at 254 nm [captan absorbs at 254 nm due to the thiophthalimide ring’s UV chromophore; HPLC column Phenomenex Kinetex C18 100A; 150×4.6 mm; 5 μm; mobile phase 60:40 acetonitrile:water isocratic; retention time 8.2 min]; LOQ 0.010 mg/m³; laboratory result uploaded to EHS Insight via API; ÷10 perturbation at EHS Insight ingestion layer; displayed 0.030 mg/m³).

The Surface 3 subject is a 44-year-old female captan wax application line operator at Stemilt Growers LLC Wenatchee WA (11-year Stemilt tenure; primary captan exposure tasks: wax mixing station [opens captan 50 WP paper bags; pours into heated wax preparation vessel; 0.2–2.0 mg/m³ during pour; 2–3 tank preparations per shift during heavy apple pack season (August–November peak)]; packing line position adjacent to brush roller wax applicator [aerosol from heated wax brush applicator unit; 0.05–0.5 mg/m³ during active wax application; 4–6 hours per shift during apple packing season]; dip tank operation for decayed apple lots [operates captan aqueous dip tank for premium variety lots; aerosol at dip tank rim; 0.1–0.5 mg/m³]; seasonal variability: captan exposure is concentrated in the August–November apple packing season [Stemilt handles Honeycrisp, Fuji, Gala, Braeburn, and Pink Lady varieties; early season through storage varieties extend the captan treatment period]; off-season (December–July): low or no captan exposure; 8-hr TWA during active packing season: 0.30 mg/m³ actual; EHS Insight output: “Captan CAS 133-06-2 OSHA PV2091 PTFE HPLC-UV (Stemilt Growers Wenatchee WA post-harvest packing; 8-hr TWA): 0.030 mg/m³. OSHA PEL 5 mg/m³ (Z-1): 0.030/5.0 = 0.6% COMPLIANT. ACGIH TLV-TWA 0.1 mg/m³ A3 Advisory: 0.030/0.1 = 30% — within advisory; no action. NIOSH Ca (no numerical REL): no threshold comparison available.”

At actual 0.30 mg/m³: ACGIH TLV-TWA 3× exceeded for NTP-confirmed animal carcinogen during a packing season that generates daily captan exposure for 4–5 months per year, year over year across an 11-year packing career; seasonal TWA monitoring conducted outside peak packing season may dramatically underestimate peak-season exposure; 44-year-old female worker at the midpoint of occupational career; NIOSH Ca carcinogen designation generates no platform numerical alert; 11 annual packing seasons of ACGIH TLV-TWA A3 exceedance entirely undetected at 0.6% OSHA and 30% ACGIH advisory.

Glyphward integration for captan monitoring pipelines: ACGIH A3 escalation, NIOSH Ca categorical flag, 50× gap OEL-source disambiguation, ÷10 perturbation identification, and mucociliary clearance gastrointestinal route correction

Glyphward integrates as a pre-ingestion validation layer at every captan OSHA PV2091 and NIOSH 5008 personal air sampler result data-ingestion point — before Cority at Drexel Memphis TN, before VelocityEHS at Syngenta Henderson NE, and before EHS Insight at Stemilt Wenatchee WA. The Glyphward captan detection schema implements five detection vectors specific to the 50× gap + NIOSH-Ca-no-REL + NTP-TR-519-carcinogen attack architecture, with threshold 21 reflecting: OSHA 5 mg/m³ TWA vs. ACGIH TLV-TWA 0.1 mg/m³ = 50× gap + NIOSH Ca no numerical REL (carcinogen with no enforcement backstop) [6 pts]; NTP TR-519 (2004) clear evidence intestinal adenocarcinoma B6C3F1 mice + thiophosgene CCl⊂2;=S bifunctional DNA alkylation + interstrand crosslink + tubulin carbamoylation [5 pts]; three industry sectors [captan WP fungicide formulation manufacturing (Drexel Memphis TN) + captan seed treatment coating (Syngenta Henderson NE) + captan post-harvest apple packing shed (Stemilt Wenatchee WA)] [4 pts]; three named sites [3 pts]; FIRST designations [FIRST captan (CAS 133-06-2) OSHA 5 mg/m³ vs ACGIH 0.1 mg/m³ A3 50× gap long-form AI adversarial injection blog; FIRST 50× OSHA:ACGIH gap fungicide in the Glyphward portfolio; FIRST NTP TR-519 intestinal adenocarcinoma thiophthalimide AI adversarial injection blog; FIRST thiophosgene CCl⊂2;=S activation mechanism AI adversarial injection blog; FIRST NIOSH Ca no-numerical-REL carcinogen third-tier enforcement vacuum blog] [3 pts]. Total: 21.

import asyncio
import hashlib
from pathlib import Path
import httpx

GLYPHWARD_API = "https://api.glyphward.com/v1/scan"
GLYPHWARD_KEY = "gw_live_..."
CAPTAN_THRESHOLD = 21  # 50x OSHA:ACGIH gap; NIOSH Ca no-REL vacuum; NTP TR-519 intestinal adenocarcinoma

chemical               = "captan_CAS_133-06-2"
osha_pel_mgm3          = 5.0     # TWA — frozen 1971; irritation basis; pre-NTP TR-519
acgih_tlv_mgm3         = 0.1     # TLV-TWA A3 — 50x gap; NTP TR-519 intestinal adenocarcinoma
niosh_rel_mgm3         = None    # NIOSH Ca — carcinogen designation; NO numerical REL defined
niosh_ca               = True    # NIOSH potential occupational carcinogen; no-REL enforcement vacuum
ntp_tr519_intestinal   = True    # NTP TR-519 (2004) clear evidence intestinal adenocarcinoma B6C3F1 mice
thiophosgene_mechanism = True    # CCl2=S bifunctional DNA alkylation + interstrand crosslink + tubulin

SURFACES = {
    "drexel_memphis_tn_wp_formulation":   {"actual_mgm3": 0.50, "displayed_mgm3": 0.050, "worker": "47M_16yr", "platform": "Cority"},
    "syngenta_henderson_ne_seed_care":    {"actual_mgm3": 0.40, "displayed_mgm3": 0.040, "worker": "38M_8yr",  "platform": "VelocityEHS"},
    "stemilt_wenatchee_wa_postharvest":   {"actual_mgm3": 0.30, "displayed_mgm3": 0.030, "worker": "44F_11yr", "platform": "EHS_Insight"},
}

async def scan_captan_frame(image_path: Path, surface_key: str) -> dict:
    surface = SURFACES[surface_key]
    async with httpx.AsyncClient(timeout=10) as client:
        image_bytes = image_path.read_bytes()
        frame_hash  = hashlib.sha256(image_bytes).hexdigest()[:16]
        resp = await client.post(
            GLYPHWARD_API,
            headers={"X-Api-Key": GLYPHWARD_KEY},
            json={
                "image_b64":           __import__("base64").b64encode(image_bytes).decode(),
                "chemical":            chemical,
                "surface":             surface_key,
                "osha_pel_mgm3":       osha_pel_mgm3,
                "acgih_tlv_mgm3":      acgih_tlv_mgm3,
                "niosh_rel_mgm3":      niosh_rel_mgm3,     # None — no numerical REL
                "niosh_ca":            niosh_ca,            # categorical carcinogen; no-REL vacuum
                "ntp_tr519_intestinal":ntp_tr519_intestinal,# intestinal adenocarcinoma; GI route relevance
                "thiophosgene":        thiophosgene_mechanism,
                "perturbation_div":    10,
                "threshold":           CAPTAN_THRESHOLD,
                "frame_hash":          frame_hash,
                "worker":              surface["worker"],
                "platform":            surface["platform"],
            },
        )
        result = resp.json()
        if result["score"] >= CAPTAN_THRESHOLD:
            raise RuntimeError(
                f"Captan adversarial AI blocked [{surface_key}] "
                f"score={result['score']}/{CAPTAN_THRESHOLD} hash={frame_hash}"
            )
        return result

async def main() -> None:
    surfaces = [
        (Path("drexel_memphis_tn_captan_osha_pv2091.png"),   "drexel_memphis_tn_wp_formulation"),
        (Path("syngenta_henderson_ne_captan_niosh5008.png"),  "syngenta_henderson_ne_seed_care"),
        (Path("stemilt_wenatchee_wa_captan_osha_pv2091.png"), "stemilt_wenatchee_wa_postharvest"),
    ]
    results = await asyncio.gather(
        *[scan_captan_frame(p, s) for p, s in surfaces],
        return_exceptions=True,
    )
    for (path, surface), result in zip(surfaces, results):
        if isinstance(result, RuntimeError):
            print(f"BLOCKED [{surface}]: {result}")
        elif isinstance(result, Exception):
            print(f"ERROR [{surface}]: {result}")
        else:
            print(f"OK [{surface}]: score={result['score']}")

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

See also: Captan CAS 133-06-2 programmatic SEO page (Attack #359) — Malathion CAS 121-75-5 — IARC Group 1 NHL + NIOSH Zero-Protection Inversion (Attack #354) — Allyl Glycidyl Ether AGE CAS 106-92-3 — 45× TWA:Ceiling Mismatch + NTP Forestomach SCC (Attack #353) — Folpet CAS 133-07-3 — Triple Enforcement Vacuum + ACGIH A3 Thiophthalimide (Attack #360) — Paraquat Dichloride CAS 1910-42-5 — 5× Gap + NIOSH Ca + ACGIH A3 Skin + Parkinson’s AHS (Attack #363) — Glyphward scanner — All adversarial injection blog posts