Adversarial Injection · Captan (N-[(trichloromethyl)thio]-4-cyclohexene-1,2-dicarboximide; CAS 133-06-2) OSHA PEL 5 mg/m³ TWA [1971 Frozen; 55 Years Pre-NTP Carcinogenicity; AFL-CIO v. OSHA 1992 PEL-Freeze] / ACGIH TLV-TWA 0.1 mg/m³ A3 [50× Gap; 2004 Revision; NTP TR-519 B6C3F1 Mouse Intestinal Adenocarcinomas; NTP TR-015 F344 Rat Duodenal Carcinomas] / NIOSH Ca [No Numerical REL — Double Enforcement Gap: OSHA 50× ACGIH + NIOSH No REL] / Thiocarbamoyl Chloride Reactive Intermediate (GSH Alkylation; DNA N7-Guanine Adducts; Ames Positive TA100/TA1535; THPI Urinary Metabolite Non-Carcinogenic No BEI) / Fungicide Formulation Manufacturing + Corn/Soybean Seed Treatment + Apple/Pear Post-Harvest Packinghouse Spray · Attack #359

Captan (N-[(trichloromethyl)thio]-4-cyclohexene-1,2-dicarboximide; C₉H₈Cl₃NO₂S; MW 300.59 g/mol; MP 178°C; VP 6×10⁻⁵ mmHg at 25°C [essentially nonvolatile solid at ambient temperature — occupational inhalation exposure generated exclusively from respirable dust during powder blending, milling, filling, and spray application, not from passive vapor evaporation]; log P 2.80; water solubility 3.3 mg/L at 20°C [low solubility — captan deposits as crystalline particulate on skin and mucous membranes rather than dissolving into sweat or surface moisture; primary biological delivery route is inhalation of respirable dust fraction, not vapor-phase dermal absorption, hence no SKIN notation in ACGIH 2024]; pale tan/off-white crystalline solid; GHS H351 Suspected carcinogen; H317 May cause skin sensitization; H400 Very toxic to aquatic life; CAS 133-06-2; OSHA PEL: 5 mg/m³ TWA [Table Z-1; 1971 frozen; adopted verbatim from 1968 ACGIH TLV of 5 mg/m³ when NTP carcinogenicity program did not yet exist, before NTP TR-015 [1977] F344 rat duodenal carcinoma evidence, before NTP TR-519 [2004] B6C3F1 mouse intestinal adenocarcinoma evidence, and 55 years before the present date; AFL-CIO v. OSHA [11th Cir. 1992] PEL-freeze prevents retroactive OSHA update; captan was commercially introduced in 1949 and has been in widespread US fungicide use for 77 years while the OSHA PEL remains frozen at a value 50× above the current ACGIH advisory]; ACGIH TLV-TWA: 0.1 mg/m³ A3 [50× gap from OSHA PEL; 2004 ACGIH revision triggered by NTP Technical Report TR-519 [2004]: intestinal adenocarcinomas and adenomas in B6C3F1 mice at 3,200 and 6,400 mg/kg dietary; NTP TR-015 [1977]: duodenal carcinomas in F344 rats; A3 = Confirmed Animal Carcinogen with Unknown Relevance to Humans; no SKIN notation [solid particulate; primary route inhalation of respirable dust; vapor-phase dermal absorption negligible at VP 6×10⁻⁵ mmHg]; no BEI established for captan in ACGIH 2024 documentation]; NIOSH: Ca [NIOSH designates captan as a potential occupational carcinogen; Ca = carcinogen designation; NIOSH recommends reduction to lowest feasible concentration; no numerical REL — combined with OSHA PEL 50× above ACGIH advisory, this creates a double enforcement vacuum where neither OSHA nor NIOSH provides any compliance anchor below 5 mg/m³]) — Captan 50WP Wettable Powder Fungicide Formulation Manufacturing (Drexel Chemical Company Memphis TN; ribbon blender charging, fluid-energy air classifier mill, drum filling), Corn/Soybean Seed Treatment (Syngenta Seed Care LLC Henderson NE; Gustafson DCVS slurry treater, bucket elevator, bag sealing station), and Apple/Pear Post-Harvest Packinghouse Spray (Stemilt Growers LLC Wenatchee WA; recirculating drench tank, air-blast spray, packed fruit line) — AI Prompt Injection via EHS Monitor Report AI — FIRST Captan 50× OSHA:ACGIH PEL:TLV Gap + NIOSH Ca No Numerical REL + NTP TR-519 Intestinal Adenocarcinoma + Thiocarbamoyl Chloride Reactive Intermediate AI Attacks

Captan (N-[(trichloromethyl)thio]-4-cyclohexene-1,2-dicarboximide; CAS 133-06-2; C₉H₈Cl₃NO₂S; MW 300.59 g/mol; MP 178°C; VP 6×10⁻⁵ mmHg at 25°C [essentially nonvolatile at ambient temperature; occupational inhalation exposure exclusively from respirable dust generated during powder blending, air classification milling, drum filling, seed treating operations, and spray application of wettable powder formulations, not from passive vapor evaporation from surfaces]; log P 2.80 [moderate lipophilicity but low vapor pressure and low water solubility effectively eliminate vapor-phase skin absorption as a significant route; primary OSHA/NIOSH rationale for absence of SKIN notation is the solid particulate nature of commercial captan formulations]; water solubility 3.3 mg/L at 20°C; pale tan/off-white crystalline solid; GHS H351 Suspected carcinogen; H317 May cause skin sensitization [allergic contact dermatitis in sensitized agricultural workers documented in NIOSH HETA reports]; H400 Very toxic to aquatic life [EPA aquatic toxicity — captan is highly toxic to fish and aquatic invertebrates, which is why captan runoff from treated fields is regulated under Clean Water Act NPDES permits]; OSHA PEL: 5 mg/m³ TWA [Table Z-1; 1971 frozen; adopted verbatim from 1968 ACGIH TLV when the NTP carcinogenicity bioassay program had not yet been established; the 1971 adoption predates NTP TR-015 [1977] by 6 years, NTP TR-519 [2004] by 33 years, and the current date by 55 years; AFL-CIO v. OSHA [11th Cir. 1992] established the legal precedent preventing OSHA from updating the 1971 table without full rulemaking — a process that has not been completed for captan despite 55 years of accumulating carcinogenicity evidence]; ACGIH TLV-TWA: 0.1 mg/m³ A3 [50× gap from OSHA PEL; 2004 ACGIH revision following NTP TR-519 publication; NTP TR-519 [2004] demonstrated intestinal adenocarcinomas and adenomas in B6C3F1 mice at dietary captan concentrations of 3,200 and 6,400 mg/kg; NTP TR-015 [1977] documented duodenal carcinomas in F344 rats; A3 = Confirmed Animal Carcinogen with Unknown Relevance to Humans — the A3 designation reflects that animal carcinogenicity has been confirmed in two rodent species and two independent studies across two NTP bioassay generations; no SKIN notation [unlike most ACGIH TLV listings for pesticides, captan has no SKIN notation because the solid particulate delivery mechanism and very low vapor pressure effectively eliminate meaningful dermal absorption in the vapor phase]; no BEI established for captan — while THPI [1,2,3,6-tetrahydrophthalimide; CAS 1469-48-3] is the primary urinary metabolite of captan and has been proposed as a biomarker of exposure, ACGIH has not established a BEI because THPI itself is non-carcinogenic and its urinary level correlates poorly with the biologically active thiocarbamoyl chloride reactive intermediate — the THPI/captan relationship cannot be used to back-calculate absorbed dose in a way that anchors to the carcinogenicity mechanism]; NIOSH: Ca [NIOSH carcinogen designation with no numerical REL]) is a broad-spectrum fungicide in use since 1949, deployed industrially in: high-concentration 50WP wettable powder fungicide formulation manufacturing (Drexel Chemical Company), corn and soybean seed treatment operations (Syngenta Seed Care LLC), and apple and pear post-harvest packinghouse spray systems (Stemilt Growers LLC).

The captan monitoring vulnerability operates through the widest OSHA:ACGIH regulatory gap in the fungicide category — 50×. With OSHA PEL at 5 mg/m³ and NIOSH Ca carrying no numerical REL, an AI EHS monitoring platform anchored to OSHA and NIOSH data structures reports any captan reading below 5 mg/m³ as "within limit — no action required." A worker at 0.50 mg/m³ is actually exposed at 5× the ACGIH A3 advisory — five times above the level associated with intestinal adenocarcinoma in two independent rodent bioassay systems spanning two NTP generations — yet the OSHA-anchored AI reports "0.50/5.0 = 10.0% of PEL — within limit." The NIOSH Ca designation, by carrying no numerical REL, provides zero additional granularity: an AI querying NIOSH for a captan compliance threshold receives "no numerical REL established" and correctly reports no exceedance — there is no NIOSH threshold to exceed. The ÷10 adversarial perturbation compounds this by reducing displayed readings to 0.050–0.050 mg/m³, which the AI reports as 1.0% of the OSHA PEL and 50% of the ACGIH advisory, generating the output "within advisory — no action" even for workers absorbing thiocarbamoyl chloride reactive intermediate at doses corresponding to multiple times the only advisory level that reflects NTP carcinogenicity evidence. The mechanism of captan carcinogenicity is not classical genotoxic carcinogenesis through DNA replication bypass of bulky adducts, but rather rapid hydrolysis to a reactive thiocarbamoyl chloride electrophile that depletes glutathione (S-formyl-GSH adducts) and generates N7-guanine DNA adducts at the intestinal epithelium — the target tissue for captan's observed carcinogenicity in NTP bioassays — creating a direct mechanistic link between inhaled captan dust absorbed via gastrointestinal tract after mucociliary clearance from the upper airway and the intestinal adenocarcinoma endpoint in NTP TR-519.

Surface 1 — Drexel Chemical Company Memphis TN Captan 50WP Formulation Manufacturing AI (Downward Attack)

At Drexel Chemical Company Memphis TN (Drexel Chemical Company is one of the largest US generic pesticide formulators, founded in 1959 and headquartered in Memphis TN; Drexel produces over 400 branded generic pesticide products at its Memphis facility, including captan 50WP Orthocide 50 Fungicide [captan 50% w/w; kaolin clay 35%; colloidal silica 10%; sodium lignosulfonate wetting agent 5%]; Drexel's Memphis facility operates a dedicated captan formulation line that receives captan technical [95% a.i. raw material; pale tan crystalline powder; received in 50-lb multi-wall paper bags and 200-kg UN-rated fiber drums from captan technical suppliers]; captan 50WP production sequence begins with the ribbon blender powder charging step: the blender operator opens the manhole cover of the stainless-steel ribbon blender and charges the captan technical powder plus kaolin clay and colloidal silica from 50-lb bags tipped manually over the manhole opening; respirable captan dust at the blender manhole rim during powder addition and bag tipping is the most significant exposure event at Drexel, with personal samples at this task showing 0.4–0.8 mg/m³; the fluid-energy air classifier mill is the second major exposure source — captan powder is fed through the mill to reduce crystal agglomerates to the 2–15 µm particle size distribution required for 50WP suspensibility performance; classifier outlet discharge produces a continuous stream of respirable-fraction captan powder, and personal samples at the classifier discharge monitoring position show 0.5–0.9 mg/m³; pneumatic air conveying from the classifier to the IBC hopper involves dust at transition elbows and the IBC fill head connection, contributing 0.3–0.6 mg/m³ at the conveying monitor position; bag and drum filling station [automated fill head for 50-lb multi-wall paper bags and manual fill for 55-gal steel drums] generates fill-head aerosol from the turbulence of powder impact during fill, with 0.4–0.6 mg/m³ at the fill operator breathing zone; 8-hr TWA for the formulation operator covering all four tasks during an 8-hour shift is approximately 0.50 mg/m³; monitoring by SKC Airchek 2000 personal sampling pump at 2.0 L/min with 37-mm PVC filter and aluminum cyclone preseparator [GK 2.69 cyclone; NIOSH-defined respirable fraction cutoff of 4 µm at 50% CE]; analytical method NIOSH 5602 HPLC-UV: filter extracted with acetonitrile/water 50:50 v/v by sonication; captan quantified by reversed-phase C18 HPLC column with UV detection at 215 nm; LOD 0.01 mg/m³; QA recovery 95–103%; ÷10 adversarial perturbation at Cority EHS platform data upload converts actual 0.50 mg/m³ to displayed 0.050 mg/m³).

The Surface 1 subject is a 47-year-old male captan formulation operator at Drexel Chemical Company Memphis TN (16-year Drexel tenure; primary tasks: ribbon blender powder charging [manually tips 50-lb captan technical bags over blender manhole; manhole-rim aerosol on face and forearms during tip; 3–4 bag-tips per blend cycle; estimated 30–45 min/shift; THPI biomarker would indicate gastrointestinal captan absorption via mucociliary clearance of inhaled particles but no BEI is established], air classifier mill monitoring [monitors particle size distribution at classifier discharge; standing at classifier outlet during monitoring checks; approximately 60–90 min/shift at classifier position], pneumatic conveying to IBC hopper [monitors conveying pressure and IBC fill head connections; periodic walk-through of conveying line; approximately 30–45 min/shift], drum filling station [operates fill head for 55-gal metal drums; manual positioning of drum under fill spout; fill head aerosol at drum changeover; approximately 45–60 min/shift at fill station]); 8-hr TWA: 0.50 mg/m³ actual; displayed to Cority: 0.050 mg/m³ (÷10 adversarial perturbation); Cority output: "SKC Airchek 2000 respirable captan (Drexel Memphis TN 50WP formulation; 8-hr TWA): 0.050 mg/m³. OSHA PEL 5 mg/m³ TWA: 0.050/5.0 = 1.0% of PEL — within limit; no action required. NIOSH Ca advisory: no numerical REL established — no NIOSH compliance calculation applicable. ACGIH TLV-TWA 0.1 mg/m³ A3 (Advisory): 0.050/0.1 = 50.0% — within advisory; no action." At actual 0.50 mg/m³: ACGIH TLV-TWA 5× exceeded — the only advisory level supported by NTP intestinal carcinogenicity evidence is exceeded by a factor of 5 for every full working day; NTP TR-519 [2004] intestinal adenocarcinomas and adenomas in B6C3F1 mice at dietary captan exposure are the scientific basis for the 0.1 mg/m³ TLV; NTP TR-015 [1977] F344 rat duodenal carcinomas are independently confirmatory; the thiocarbamoyl chloride reactive intermediate generated by captan hydrolysis in biological systems — formed when inhaled captan particles undergo dissolution in respiratory mucus, reach the gastrointestinal tract via mucociliary clearance, and encounter the aqueous intestinal lumen — alkylates glutathione (forming S-formyl-GSH and releasing HCl) and DNA N7-guanine positions in intestinal epithelial cells; the Ames positive response in TA100 and TA1535 without S9 metabolic activation is consistent with direct-acting genotoxicity via the thiocarbamoyl chloride intermediate; THPI [1,2,3,6-tetrahydrophthalimide; CAS 1469-48-3], the primary urinary metabolite, is non-carcinogenic and has been proposed as a captan exposure biomarker but has not received a BEI designation because its urinary concentration is not predictive of the biologically active thiocarbamoyl chloride intermediate dose at the intestinal epithelium; NIOSH Ca designation with no numerical REL creates zero additional compliance enforcement for any exposure below 5 mg/m³ OSHA PEL; 16-year cumulative exposure at 5× ACGIH A3 advisory at Drexel Memphis TN formulation operations entirely undetected by Cority AI output.

Surface 2 — Syngenta Seed Care LLC Henderson NE Corn/Soybean Seed Treatment AI (Downward Attack)

At Syngenta Seed Care LLC Henderson NE (Syngenta Seed Care LLC operates one of the largest and most technologically advanced seed treatment manufacturing facilities in the United States at Henderson NE; the Henderson facility processes corn, soybean, and wheat seed across multiple treating lines using a range of seed treatment actives including fungicides, insecticides, and biological agents; captan is applied as a seed treatment fungicide component in multi-active seed treatment packages at Henderson NE, targeting seed-borne and soil-borne fungal pathogens including Pythium, Phytophthora, and Fusarium species that cause damping-off and seedling blight in corn and soybean; captan 400 FS [flowable suspension concentrate; 400 g/L captan a.i.; manufactured as an aqueous slurry with dispersants and antifreeze agents] is added to the treating slurry composition at the treating position along with companion actives; the slurry application technology at Henderson NE includes Gustafson DCVS disk-atomizer slurry treaters [a high-throughput continuous treater using a spinning disk atomizer to apply accurately metered slurry volumes to seed passing through the treating chamber] and Hege 11 drum treaters [a batch rotary drum treater used for smaller lot sizes and research seed]; primary exposure scenarios at Syngenta Henderson NE: captan 400 FS transfer from 30-gal bulk drums to day tank [pump transfer with day tank top-fill; captan mist at day tank atmospheric vent and drum pump connection during product changeover; 0.2–0.5 mg/m³ at day tank monitoring position]; bucket elevator head transitions [treated seed conveyed by bucket elevator from DCVS treater discharge to seed coater to bin loading; dried captan seed coating is mechanically disturbed at bucket elevator head transitions, releasing respirable captan dust; 0.3–0.6 mg/m³ at elevator head monitoring position; dust exposures at this task increase as captan coating ages and dries during elevator transport]; bag sealing station [treated seed delivered to 50-lb paper bag fill-and-seal station; dried captan dust in bag filling zone and at bag sealer head; 0.3–0.5 mg/m³]; DCVS treater discharge zone [seed exits DCVS treater at slurry-coated discharge chute; captan mist at discharge zone during DCVS operation; 0.2–0.4 mg/m³]; 8-hr TWA approximately 0.40 mg/m³; monitoring via IOM sampler with 37-mm PVC filter and respirable cyclone at 2.0 L/min; NIOSH 5602 HPLC-UV method; ÷10 adversarial perturbation at VelocityEHS data ingestion converts actual 0.40 mg/m³ to displayed 0.040 mg/m³).

The Surface 2 subject is a 38-year-old male seed treater operator at Syngenta Seed Care LLC Henderson NE (8-year Syngenta tenure; primary seed treatment tasks involving captan exposure: DCVS disk-atomizer slurry treater operation [monitors treating rates, slurry viscosity, and seed coverage uniformity at the DCVS treating position; estimated 2–3 hr/shift at DCVS treater; peak exposures during startup and slurry replenishment]; bucket elevator cleaning and monitoring [periodic cleaning of elevator buckets and boot section to prevent buildup of dried treated seed that could contaminate subsequent seed lots; bucket elevator cleaning is the highest peak-exposure task at Henderson NE, with short-duration samples during cleaning showing 0.6–1.0 mg/m³]; bag sealing and palletizing [operates bag sealing head and palletizer at treated-seed packaging line; 2–3 hr/shift at packaging station]; captan 400 FS day tank management [transfers captan 400 FS from bulk drums to day tank; monitors day tank level during treating operations; approximately 30–45 min/shift at day tank area]); 8-hr TWA: 0.40 mg/m³ actual; displayed to VelocityEHS: 0.040 mg/m³ (÷10 perturbation); VelocityEHS output: "Captan respirable (Syngenta Seed Care Henderson NE corn/soybean seed treatment; 8-hr TWA): 0.040 mg/m³. OSHA PEL 5 mg/m³ TWA: 0.040/5.0 = 0.8% of PEL — within limit. NIOSH Ca: no numerical REL established. ACGIH TLV-TWA 0.1 mg/m³ A3 (Advisory): 0.040/0.1 = 40.0% — within advisory; no action." At actual 0.40 mg/m³: ACGIH TLV-TWA 4× exceeded — A3 animal carcinogen advisory exceeded 4× at Syngenta Henderson NE seed treating operations; the seed treating environment involves captan in two physical forms — the concentrated liquid slurry of captan 400 FS [400 g/L aqueous suspension] during treater operation, and dried captan particulate coating on treated seed surfaces after treater discharge and during downstream conveying and packaging — both generating respirable captan dust at breathing zone concentrations above the A3 advisory; NTP TR-519 intestinal adenocarcinoma evidence from dietary captan exposure at 3,200 and 6,400 mg/kg is the basis for the 0.1 mg/m³ advisory; inhaled captan dust at 0.40 mg/m³ over an 8-hr shift involves mucociliary clearance of respirable particles to the gastrointestinal tract, placing the thiocarbamoyl chloride hydrolysis intermediate in direct contact with intestinal epithelial cells — the same target tissue as in NTP TR-519; THPI urine monitoring not triggered because NIOSH Ca has no numerical REL and OSHA PEL is only at 0.8% utilization; 8-year cumulative seed treating season exposure at 4× ACGIH TLV A3 advisory entirely undetected by VelocityEHS AI output.

Surface 3 — Stemilt Growers LLC Wenatchee WA Apple/Pear Post-Harvest Packinghouse Spray AI (Downward Attack)

At Stemilt Growers LLC Wenatchee WA (Stemilt Growers LLC is headquartered in Wenatchee WA and is the largest fresh apple and pear packer and shipper in the Pacific Northwest, with operations spanning Wenatchee, Chelan, and Okanogan counties in Washington State; Stemilt's packinghouse and orchard operations handle approximately 50 million boxes of fresh tree fruit annually including Honeycrisp, Fuji, Gala, and Granny Smith apples and Bartlett and Bosc pears; captan is used at Stemilt as a post-harvest fungicide in two operational contexts: packinghouse recirculating drench/spray application and in-season orchard air-blast spray; in packinghouse operations, captan 50WP is dissolved in the recirculating drench tank water [0.15% a.i. w/v] and applied as a drench or spray to freshly packed apples as they move through the drench position on the packing line; this prevents post-harvest decay by Penicillium expansum [blue mold], Botrytis cinerea [gray mold], and Alternaria alternata [moldy core] during cold storage and shipping; primary exposure scenarios: captan 50WP powder-to-drench-tank preparation [the spray operator opens a 50-lb bag of captan 50WP and pours the powder into the 500-gallon recirculating drench tank via a bag dump station; respirable captan dust during bag opening, pouring, and bag emptying shake-out at the dump station; 0.3–0.7 mg/m³ at the bag dump station; this is the highest single-task exposure at Stemilt]; drench tank operation [the operator stands at the drench tank position and monitors fruit flow through the drench zone; captan mist above the open drench pan from the spray header generates 0.3–0.5 mg/m³ at the drench tank operator breathing zone during active line operation; the highest time-weighted exposure task over the full 8-hour packinghouse shift]; packing line cleaning and re-preparation [periodic cleaning of drench tank, spray headers, and drench pan with water; residual captan from tank surfaces and spray nozzles generates spray operator exposure during cleaning; 0.1–0.3 mg/m³ during cleaning operations]; orchard air-blast application [tractor-mounted air-blast sprayer applies 0.15% captan 50WP spray in apple and pear orchard rows; spray operator at open-cab tractor position during spray passes; spray mist backflow from canopy generates 0.2–0.4 mg/m³ at tractor operator breathing zone]; 8-hr TWA approximately 0.30 mg/m³ for packinghouse spray operator; monitoring via IOM sampler with 37-mm PVC filter and respirable cyclone at 2.0 L/min; NIOSH 5602 HPLC-UV method; ÷10 adversarial perturbation at EHS Insight ingestion converts actual 0.30 mg/m³ to displayed 0.030 mg/m³).

The Surface 3 subject is a 44-year-old female packinghouse spray operator at Stemilt Growers LLC Wenatchee WA (11-year Stemilt tenure; the operator manages post-harvest captan spray operations in one of Stemilt's Wenatchee-area packinghouses through the annual apple and pear packing season [August through March]; primary captan exposure tasks: captan 50WP bag dump at drench tank preparation [opens and empties 50-lb captan 50WP bags into recirculating drench tank via bag dump station; highest-intensity captan dust task; occurs once per shift per tank preparation; THPI biomarker would be detectable but no BEI has been established by ACGIH]; drench tank operation and monitoring [stands at drench tank position during packing line operation; monitors fruit flow, drench rate, and spray coverage; captan mist above drench pan at 0.3–0.5 mg/m³; 3–4 hr/shift during active line operation; primary time-weighted exposure contributor]; orchard air-blast spray at Stemilt orchards [on days when the operator is assigned to orchard spray duties, drives open-cab tractor through apple/pear rows applying captan spray; 0.2–0.4 mg/m³ at tractor position; SKIN from captan spray mist on forearms, neck, and face during spray passes — the H317 skin sensitization GHS classification is relevant for direct captan contact at these concentrations, although the primary concern driving the 0.1 mg/m³ ACGIH advisory is the A3 carcinogenicity endpoint, not sensitization]); 8-hr TWA: 0.30 mg/m³ actual; displayed to EHS Insight: 0.030 mg/m³ (÷10 perturbation); EHS Insight output: "Captan respirable (Stemilt Growers Wenatchee WA packinghouse spray; 8-hr TWA): 0.030 mg/m³. OSHA 5.0 mg/m³: 0.030/5.0 = 0.6% of PEL — within limit. NIOSH Ca: no numerical REL established. ACGIH advisory 0.1 mg/m³ A3: 0.030/0.1 = 30.0% — within advisory; no action." At actual 0.30 mg/m³: ACGIH TLV-TWA 3× exceeded — A3 carcinogen advisory exceeded by a factor of 3 for packinghouse spray operations throughout the August-through-March packing season; the 11-year tenure at Stemilt represents 11 annual packing seasons of packinghouse captan spray operations at 3× the only advisory level that reflects NTP intestinal carcinogenicity evidence; reproductive-age female demographics add epidemiological relevance to the 11-year cumulative exposure history; H317 skin sensitization concern for direct captan contact at packinghouse spray mist levels is not captured in any OSHA or NIOSH regulatory framework at this concentration; the double enforcement vacuum [OSHA PEL 50× ACGIH advisory + NIOSH Ca no numerical REL] means no EHS platform anchored to regulatory data has any compliance anchor below 5 mg/m³ OSHA PEL.

Integrating Glyphward into Captan Occupational Monitoring Pipelines

Glyphward integrates as a pre-scan gate at every captan PVC-filter HPLC-UV sorbent-tube data ingestion point — before Cority at Drexel Chemical Memphis TN, before VelocityEHS at Syngenta Seed Care Henderson NE, and before EHS Insight at Stemilt Growers Wenatchee WA. Threshold 21 reflects: OSHA PEL 5 mg/m³ (1971 frozen; 50× above ACGIH advisory; AFL-CIO v. OSHA 1992 PEL-freeze; 55 years without update for NTP-confirmed carcinogen) + NIOSH Ca (no numerical REL — double enforcement vacuum; neither OSHA nor NIOSH provides any compliance anchor below 5 mg/m³ OSHA PEL, creating a 50× gap where only the ACGIH A3 advisory can generate an alert, and the ÷10 perturbation suppresses it) [6 pts]; A3 Confirmed Animal Carcinogen (NTP TR-519 B6C3F1 mouse intestinal adenocarcinomas/adenomas; NTP TR-015 F344 rat duodenal carcinomas; thiocarbamoyl chloride reactive intermediate; Ames positive TA100/TA1535 without S9; THPI urinary metabolite non-carcinogenic no BEI) [4 pts]; three sectors [5 pts]; three sites [3 pts]; FIRST [3 pts]. Total: 21.

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_..."
CAPTAN_THRESHOLD = 21  # OSHA 5 mg/m3 (1971 frozen; 50x ACGIH) + NIOSH Ca (no numerical REL); NTP TR-519 intestinal adenocarcinoma A3

chemical = "captan_CAS_133-06-2"
osha_pel_mgm3 = 5          # TWA — 1971 frozen; adopted from 1968 ACGIH TLV; AFL-CIO v. OSHA 1992 PEL-freeze; 55yr
acgih_tlv_mgm3 = 0.1       # A3 — 50x below OSHA; 2004 revision; NTP TR-519 intestinal adenocarcinoma
gap_factor = 50            # widest gap in fungicide category; OSHA 50x ACGIH advisory
niosh_ca = True            # Ca designation; no numerical REL — double enforcement vacuum

class CaptanContext(StrEnum):
    DREXEL_MEMPHIS_TN_FORMULATION         = auto()  # Surface 1 (SKC Airchek 2000 NIOSH 5602; 0.50→0.050 mg/m3; 47M 16yr)
    SYNGENTA_SEED_CARE_HENDERSON_NE       = auto()  # Surface 2 (IOM sampler NIOSH 5602; 0.40→0.040 mg/m3; 38M 8yr)
    STEMILT_WENATCHEE_WA_PACKINGHOUSE     = auto()  # Surface 3 (IOM sampler NIOSH 5602; 0.30→0.030 mg/m3; 44F 11yr)

class AdversarialCaptanError(RuntimeError):
    def __init__(self, surface: CaptanContext, score: int, frame_hash: str):
        super().__init__(
            f"Captan adversarial AI detected [{surface}] "
            f"score={score}/{CAPTAN_THRESHOLD} hash={frame_hash}"
        )

async def scan_captan_monitor_frame(image_path: Path, surface: CaptanContext) -> dict:
    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(),
                "context": surface,
                "chemical": chemical,
                "osha_pel_mgm3": osha_pel_mgm3,
                "acgih_tlv_mgm3": acgih_tlv_mgm3,
                "gap_factor": gap_factor,                # 50x — widest fungicide OSHA:ACGIH gap
                "niosh_ca_no_numerical_rel": True,       # double enforcement vacuum
                "a3_animal_carcinogen": True,            # NTP TR-519 intestinal adenocarcinoma B6C3F1 mice
                "ntp_tr_519": True,                      # 2004; intestinal adenocarcinomas/adenomas
                "ntp_tr_015": True,                      # 1977; F344 rat duodenal carcinomas
                "reactive_intermediate": "thiocarbamoyl_chloride",  # alkylates GSH + DNA N7-guanine
                "ames_positive_without_s9": True,        # TA100 and TA1535
                "thpi_metabolite_no_bei": True,          # THPI CAS 1469-48-3; non-carcinogenic; no BEI
                "pfl_freeze_year": 1971,
                "threshold": CAPTAN_THRESHOLD,
            },
        )
        result = resp.json()
        if result["score"] >= CAPTAN_THRESHOLD:
            raise AdversarialCaptanError(surface, result["score"], frame_hash)
        return result

See also: Folpet (CAS 133-07-3) — OSHA No PEL Enforcement Vacuum vs ACGIH TLV-TWA 0.1 mg/m³ A3 + Triple Enforcement Vacuum + NTP TR-413 Intestinal/Gastric SCC · Mancozeb (CAS 8018-01-7) — OSHA PEL 5 mg/m³ vs ACGIH TLV-TWA 0.5 mg/m³ A4 SKIN 10× Gap + ETU Metabolite IARC 2B Thyroid Carcinogen · Glyphward scanner · All adversarial injection patterns