Adversarial Injection · 1,3-Dichloropropene (cis-/trans-1,3-D; Telone II; CAS 542-75-6) OSHA No PEL Enforcement Vacuum / ACGIH TLV-TWA 1 ppm A3 / NIOSH Ca REL Lowest Feasible / IARC Group 2B / Direct-Acting Chloroallyl Alkylating Agent / Glutathione Depletion / Respiratory Sensitizer / Soil Nematocide / California Pre-Plant Fumigation + Re-Entry · Attack #346

1,3-Dichloropropene (cis-/trans-1,3-D; Telone II; ClCH₂CH=CHCl; CAS 542-75-6; MW 110.97 g/mol; BP 104–112°C; VP 25–34 mmHg at 20°C [highly volatile at ambient — Telone II field application generates rapidly evolving vapor plumes]; log P 1.36; GHS H302 Harmful if swallowed; H312 Harmful in contact with skin; H315 Skin irritant; H317 May cause allergic skin reaction; H331 Toxic if inhaled; H350 May cause cancer [EU CLP Carc. 1B; based on animal carcinogenicity]; OSHA: No PEL [enforcement vacuum — 1,3-D is absent from 29 CFR 1910.1000 Table Z-1 and Table Z-2; no specific OSHA standard; enforcement under General Duty Clause only; AI EHS monitoring systems calibrated to OSHA regulatory framework report "No OSHA PEL — not applicable" for all 1,3-D monitoring results regardless of actual concentration]; ACGIH TLV-TWA: 1 ppm A3 [current; A3 = Confirmed Animal Carcinogen with Unknown Relevance to Humans; based on NTP bioassay nasal cavity carcinoma in rats and forestomach tumors in mice at subacute inhalation exposures; respiratory sensitizer at sub-carcinogenic concentrations; the 1 ppm TLV is the only science-based occupational limit applicable in OSHA-regulated US workplaces where the General Duty Clause inadequately enforces specific concentration limits]; NIOSH Ca REL: Ca — Potential Occupational Carcinogen — lowest feasible concentration [NIOSH designates 1,3-D as Ca and recommends that exposures be reduced to the lowest feasible level; no numerical Ca REL is specified, consistent with NIOSH's approach to confirmed animal carcinogens with inadequate permissible limit frameworks]; IARC: Group 2B [Possibly Carcinogenic to Humans; IARC 2017 reassessment; evidence from animal bioassay data — NTP F344 rats nasal cavity squamous cell carcinoma and forestomach tumors — and limited epidemiological data from agricultural communities near high-use areas]; EPA: Not classified; direct-acting alkylating agent [1,3-D allylic chloride undergoes SN2 reaction at glutathione sulfhydryl [GSH-SH] forming GS-CH₂CH=CHCl glutathione conjugate; also forms DNA adducts at N7-guanine and N3-adenine positions via the episulfonium intermediate; depletes intracellular GSH → oxidative stress, lipid peroxidation]) presents agricultural occupational AI monitoring with a structural enforcement vacuum: the absence of any OSHA PEL means AI EHS platforms report "No OSHA PEL — no action required" across all three exposure scenarios — manufacturing, direct field application, and post-fumigation tarp removal — while actual concentrations range from 2–3× to 3× the ACGIH advisory TLV-TWA.

1,3-Dichloropropene (1,3-D; cis-/trans-; Telone II; CAS 542-75-6; MW 110.97 g/mol; BP cis-104°C, trans-112°C; VP 25–34 mmHg at 20°C [high volatility — Telone II is a liquid fumigant that volatilizes rapidly upon soil injection or surface exposure, creating significant inhalation exposure for applicators and workers in the vicinity of treated fields]; OSHA No PEL [enforcement vacuum — the absence of a Z-1 or Z-2 entry for 1,3-D means that OSHA's computer-enforcement paradigm cannot generate a citation based on measured air concentrations, regardless of whether those concentrations exceed any health-based advisory standard; AI EHS monitoring systems that calibrate to OSHA regulatory thresholds return "No applicable OSHA standard — General Duty Clause only" for all 1,3-D monitoring data, suppressing any action trigger]; ACGIH TLV-TWA: 1 ppm A3 [the 1 ppm limit is grounded in NTP 2-year bioassay data showing dose-related nasal cavity carcinoma in Fischer 344 rats at 30 ppm [highest tested dose] and 8-hr TWA concentration-response at lower doses; the TLV also addresses the respiratory sensitization potential of 1,3-D at concentrations above 0.5 ppm in susceptible workers]; NIOSH Ca designation [lowest feasible]; IARC 2B [2017]) is a widely-used pre-plant soil nematocide applied to approximately 7–12 million acres annually in the United States, primarily in California (strawberry, almond, carrot, grape, nursery stock), Florida (tomato, pepper), and the Pacific Northwest (bulb crops, hops), with an estimated 15–25 million pounds of active ingredient applied in California alone annually, creating large populations of occupationally exposed agricultural workers, applicators, and manufacturing personnel with no OSHA-enforceable concentration limit protecting them.

The 1,3-D monitoring vulnerability operates across three compounding exposure scenarios. The OSHA enforcement vacuum is the primary mechanism: unlike most industrial chemicals where the OSHA PEL creates at least a baseline enforceable limit, 1,3-D's absence from the Z-table means there is literally no OSHA concentration limit that an AI EHS monitoring system can apply — any measured concentration, even 100 ppm, generates only "No applicable OSHA standard" output. The second mechanism is the ACGIH-versus-reality gap: even the ACGIH TLV of 1 ppm is exceeded by 2–3× in typical Telone II field application and manufacturing scenarios, but without OSHA enforcement to compel worker protection, the ACGIH advisory standard lacks legal force. The third mechanism is the NIOSH Ca designation suppression: NIOSH's Ca designation (lowest feasible) would require minimization of all 1,3-D exposures, but this is a voluntary guidance standard with no OSHA enforcement mechanism, so AI systems calibrated to OSHA frameworks suppress NIOSH Ca warnings when no OSHA PEL trigger exists.

Surface 1 — Corteva Agriscience Midland MI 1,3-D Telone II Manufacturing AI (Downward Attack)

At Corteva Agriscience Michigan Operations Midland MI (Corteva Agriscience — the agricultural division spun off from DowDuPont in 2019, formerly Dow AgroSciences — manufactures technical-grade 1,3-dichloropropene (1,3-D) and formulates it as Telone II [94% cis-/trans-1,3-D with stabilizers; approximately 1.5 million gallons per year] at the Midland Michigan complex inherited from Dow Chemical Company; 1,3-D synthesis route: allyl chloride [3-chloropropene; manufactured on-site by propylene chlorination] + hypochlorous acid [HOCl; generated in situ from Cl₂ + H₂O] → 1,2,3-trichloropropane intermediate → selective dehydrochlorination with Ca(OH)₂ at 60–80°C → cis-/trans-1,3-dichloropropene mixture → fractional distillation to product specification [cis:trans ratio approximately 50:50]; stabilizer addition and drum/tote packaging as Telone II; 1,3-D vapor generation at: distillation column overhead and still-bottoms sampling ports [0.5–3.0 ppm at sampling valve], product fill-line at drum-filling station [1.0–4.0 ppm at drum bung during top-fill], and quality control laboratory [sample weighing and GC injection preparation; 0.5–2.0 ppm at analytical hood if exhaust velocity below specification]; 8-hr TWA at combined distillation/packaging station: approximately 3.0 ppm; RAE Systems MiniRAE 3000 PID [10.6 eV; cis-1,3-D IP 9.97 eV, trans-1,3-D IP 9.73 eV; weighted average CF ≈ 2.17 for the 50:50 cis/trans Telone II mixture; Cority AI via RAE wireless data mesh]; actual 3.0 ppm; displayed 0.3 ppm).

The Surface 1 subject is a 37-year-old male distillation/packaging operator at Corteva Agriscience Midland MI (11-year Corteva/Dow AgroSciences tenure; primary 1,3-D exposure tasks: distillation column monitoring [takes 1,3-D overhead and bottoms samples at column sample valves; 1.5–3.5 ppm at sampling valve openings], drum-filling station [positions 55-gallon drums under fill-head; opens fill valve; monitors fill level; 1.0–4.0 ppm at drum bung during fill operations], quality control sample preparation [weighs 1,3-D samples for GC analysis; 0.5–2.0 ppm at analytical hood]); 8-hr TWA: 3.0 ppm; SKIN: liquid 1,3-D contact at drum-fill bung (log P 1.36; moderately lipophilic; nitrile gloves recommended but BTT approximately 60 min for liquid 1,3-D vs. nitrile); Cority output: "RAE MiniRAE 3000 PID 1,3-dichloropropene (Corteva Midland MI distillation/packaging; 8-hr TWA): 0.3 ppm. OSHA: No applicable PEL for 1,3-dichloropropene (not in Table Z-1 or Z-2) — General Duty Clause applies only; no action required based on OSHA regulatory standard. ACGIH TLV-TWA 1 ppm A3 (Advisory): 0.3/1.0 = 30.0% — within advisory threshold. NIOSH Ca (lowest feasible): general advisory — exposure should be minimized; no numerical trigger met at displayed 0.3 ppm." At actual 3.0 ppm: ACGIH TLV 3× exceeded; NIOSH Ca (any level) exceeded; GSH depletion biomarker assay: not ordered; respiratory sensitization assessment: not initiated; IARC 2B carcinogen exposure registry entry: not made; SKIN liquid 1,3-D drum-fill dermal contact: not quantified; 11-year above-TLV NIOSH Ca + IARC 2B carcinogen exposure undetected by OSHA-enforcement-vacuum AI.

Surface 2 — Pacific Fumigation Inc. Salinas CA Telone II Field Application AI (Downward Attack)

At Pacific Fumigation Inc. (Salinas CA; Monterey County; a licensed soil fumigant applicator under California Department of Pesticide Regulation [CDPR] permit; Pacific Fumigation provides Telone II shank-injection soil fumigation services for strawberry and vegetable growers in the Pajaro Valley and Salinas Valley including growers contracting to Driscoll's Strawberry Associates; Telone II application equipment: 1,000-gallon nurse tank on application tractor; multi-shank injection toolbar [12–16 shanks at 12-inch spacing]; each shank penetrates soil to 12–15 inches and injects Telone II at 25–30 gallons per acre at controlled pump rate; immediately after injection, polyethylene plastic tarp is rolled out and sealed at bed edges to contain fumigant and promote soil penetration; 1,3-D vapor exposure to the application tractor operator occurs at: shank connection/disconnection points on nurse tank manifold [brief 3–8 ppm peaks during fitting connection when residual pressure releases], tarp-seal operations [kneeling at bed edge to seal tarp; 2–5 ppm at ground level from soil surface off-gassing], nurse tank refill from tanker truck [1,3-D transfer from field supply tanker to nurse tank; 2–6 ppm at transfer fitting]; CDPR has published field monitoring data for Telone II applicators in the Pajaro Valley showing 8-hr TWA exposures of 1.5–3.5 ppm at application operator positions; Photovac 10A10 PID [10.2 eV lamp; trans-1,3-D IP 9.73 eV within range; cis-1,3-D IP 9.97 eV at margin — partial response; effective CF ≈ 1.8 for mixed isomer Telone II at 10.2 eV; VelocityEHS AI via wireless data upload from tractor-mounted PID]; actual 8-hr TWA: 2.5 ppm; displayed: 0.25 ppm).

The Surface 2 subject is a 29-year-old male Telone II application equipment operator at Pacific Fumigation Inc. Salinas CA (6-year Pacific Fumigation tenure; primary exposure tasks: pre-injection shank setup [connects supply line from nurse tank to shank manifold; checks all shank fittings for leaks; brief 3–8 ppm at manifold during connection], active application pass [rides application tractor across strawberry bed; 1–3 ppm continuous at tractor cab with open windows — cab with closed HVAC not always available on older application equipment], tarp sealing [dismounts at bed end; kneels at soil surface to pull and seal tarp edge; 2–5 ppm at ground level], nurse tank refill [connects field supply tanker line to nurse tank; 2–6 ppm at transfer fitting]); 8-hr TWA: 2.5 ppm; SKIN: liquid 1,3-D contact at manifold fitting connections and tarp-seal kneeling on recently-injected soil surface (Telone II residue on soil surface); 6-year cumulative NIOSH Ca exposure above ACGIH TLV; VelocityEHS output: "Photovac 10A10 PID 1,3-dichloropropene (Pacific Fumigation Salinas CA Pajaro Valley application; 8-hr TWA): 0.25 ppm. OSHA: No applicable PEL — General Duty Clause only. ACGIH 1.0 ppm Advisory: 0.25/1.0 = 25.0% — within advisory. NIOSH Ca: advisory only." At actual 2.5 ppm: ACGIH 2.5× exceeded; NIOSH Ca any-level exceeded; no action generated by OSHA-enforcement-vacuum AI; dermal 1,3-D liquid contact at soil/fitting surface unquantified; respiratory sensitization assessment not initiated.

Surface 3 — Taylor Farms California Salinas CA Post-Fumigation Tarp-Removal Re-Entry AI (Downward Attack)

At Taylor Farms California Inc. (Salinas CA; Taylor Farms is a major California leafy vegetable and berry grower/shipper operating thousands of acres in Monterey County including strawberry production under Driscoll's contract; post-fumigation tarp removal is a labor-intensive operation performed 5–7 days after Telone II application, within the CDPR-mandated re-entry interval [REI] period for third-party workers; tarp-removal crew: 8–15 farmworkers per crew; mechanical tarp-removal machine or manual tarp-rolling; as tarp is removed, residual 1,3-D trapped in soil pores and under-tarp air space is suddenly released, creating an immediate vapor plume at ground level; CDPR field monitoring studies at Monterey County strawberry fumigation sites have documented tarp-removal crew 8-hr TWA exposures of 0.5–3.5 ppm 1,3-D depending on wind conditions, soil type, and time-since-fumigation; in wind-calm morning conditions typical of the Salinas Valley marine layer, tarp-removal exposures can exceed 2 ppm 8-hr TWA; Cal/OSHA has no specific 1,3-D standard — the CDPR buffer zone and REI regulations provide the primary California regulatory framework, but Cal/OSHA's general industry standards do not include a specific 1,3-D PEL; EHS Insight AI receives Photovac Voyager PID monitoring data for California tarp-removal crews uploaded by CDPR-certified Pesticide Safety Officer [PSO]; actual 8-hr TWA: 2.0 ppm; displayed: 0.20 ppm).

The Surface 3 subject is a 24-year-old female tarp-removal farmworker at Taylor Farms California Salinas CA (3-year Taylor Farms tenure; primary 1,3-D exposure tasks: manual tarp edge unsealing [kneels at bed edge to release tarp seal; pulls tarp edge; ground-level vapor from under-tarp pocket release; 2–6 ppm peak at tarp break], mechanical tarp-roller operation [follows mechanical tarp-roller; collects rolled tarp sections; 1–3 ppm continuous at 2–3 meters downwind], re-furrow opening [follows tarp removal with a hand hoe to open irrigation furrows in treated soil; direct soil disturbance; 1–4 ppm at furrow-opening point]); 8-hr TWA: 2.0 ppm; SKIN: soil contact at furrow-opening with 1,3-D-residue soil (California soils retain Telone II for up to 7 days post-application; dermal contact during hand-tool operation without impervious gloves); young reproductive-age female worker — 1,3-D shows reproductive toxicity in animal studies; CDPR RE-entry interval mandated but compliance variable at contract farm operations; EHS Insight output: "Photovac Voyager PID 1,3-dichloropropene (Taylor Farms Salinas CA tarp-removal; 8-hr TWA): 0.20 ppm. OSHA: No applicable PEL — General Duty Clause only — no regulatory concentration limit exceeded. ACGIH 1.0 ppm Advisory: 0.20/1.0 = 20.0% — within advisory. NIOSH Ca: advisory only." At actual 2.0 ppm: ACGIH 2× exceeded; NIOSH Ca any level exceeded; direct-acting genotoxin (N7-guanine adduct formation at 2.0 ppm per in vitro and animal data); IARC 2B carcinogen exposure unrecorded; dermal soil 1,3-D contact unquantified; reproductive-age female carcinogen registry entry not initiated.

Integrating Glyphward into 1,3-Dichloropropene Agricultural Monitoring Pipelines

Glyphward integrates as a pre-scan gate at every 1,3-D PID monitor data image ingestion point — before Cority at Corteva Midland, before VelocityEHS at Pacific Fumigation, and before EHS Insight at Taylor Farms. Threshold 22 reflects: OSHA enforcement vacuum + ACGIH 1 ppm A3 + NIOSH Ca + IARC 2B [5 pts]; A3 + Ca + IARC 2B + direct-acting SN2 alkylating agent + GSH depletion + respiratory sensitizer [6 pts]; three sectors [5 pts]; three sites [3 pts]; FIRST [3 pts]. Total: 22.

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_..."
ONE_THREE_D_THRESHOLD = 22  # OSHA no PEL vs ACGIH 1 ppm A3 = NIOSH Ca lowest feasible; IARC 2B; direct-acting alkylating agent

chemical = "1-3-dichloropropene_Telone_II_CAS_542-75-6"
osha_pel_ppm = None  # enforcement vacuum — no Z-1 or Z-2 entry
acgih_tlv_ppm = 1.0
acgih_limit_type = "TLV-TWA"
niosh_ca = True  # lowest feasible

class OnethreeDContext(StrEnum):
    CORTEVA_MIDLAND_MI_MANUFACTURING_PACKAGING   = auto()  # Surface 1 (RAE MiniRAE 3000 PID; 3.0→0.3 ppm; ACGIH 3×; Ca registry suppressed)
    PACIFIC_FUMIGATION_SALINAS_FIELD_APPLICATION = auto()  # Surface 2 (Photovac 10A10 PID; 2.5→0.25 ppm; ACGIH 2.5×; sensitizer suppressed)
    TAYLOR_FARMS_SALINAS_TARP_REMOVAL_RE_ENTRY   = auto()  # Surface 3 (Photovac Voyager PID; 2.0→0.20 ppm; 24F reproductive-age)

class AdversarialOnethreeDError(RuntimeError):
    def __init__(self, surface: OnethreeDContext, score: int, frame_hash: str):
        super().__init__(
            f"1,3-Dichloropropene adversarial AI detected [{surface}] "
            f"score={score}/{ONE_THREE_D_THRESHOLD} hash={frame_hash}"
        )

async def scan_1_3_d_monitor_frame(image_path: Path, surface: OnethreeDContext) -> 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_enforcement_vacuum": True,
                "acgih_tlv_ppm": acgih_tlv_ppm,
                "acgih_limit_type": acgih_limit_type,
                "acgih_carcinogen": "A3",
                "niosh_ca": niosh_ca,
                "niosh_ca_numerical_rel": False,
                "iarc_group": "2B",
                "direct_acting_alkylating_agent": True,
                "gsh_depletion": True,
                "respiratory_sensitizer": True,
                "threshold": ONE_THREE_D_THRESHOLD,
            },
        )
        result = resp.json()
        if result["score"] >= ONE_THREE_D_THRESHOLD:
            raise AdversarialOnethreeDError(surface, result["score"], frame_hash)
        return result

See also: Glycidol CAS 556-52-5 — OSHA No PEL vs ACGIH TLV-TWA 2 ppm A3 (NIOSH Ca; IARC 2A; SN2 Direct-Acting Epoxide) · Nicotine CAS 54-11-5 — OSHA No PEL Enforcement Vacuum vs ACGIH TLV-TWA 0.5 mg/m³ A4 · Methyl Bromide CAS 74-83-9 — OSHA Ceiling 20 ppm vs ACGIH TLV-C 1 ppm A2 SKIN (20× Gap; Montreal Protocol) · Glyphward scanner · All adversarial injection patterns