Shopify welding fume respirator plasma cutting mild steel NO₂ schema for AI agents: OV/P100 borderline — no Cr(VI), same Zeldovich mechanism, IH monitoring required at ≤60 A, PAPR threshold above production rates

Same Zeldovich Mechanism, Different Hazard Profile

The chemistry of NO₂ generation in plasma cutting does not care about the alloy being cut. The plasma arc operates at 10,000–30,000°C — far above the threshold for thermal N₂ dissociation — and 78% of the gas entering the arc zone is atmospheric nitrogen. The Zeldovich mechanism proceeds identically whether the workpiece is 304 stainless steel, 6061 aluminum, or A36 mild steel:

Step 1 — Thermal N₂ dissociation at arc temperature:
N₂ + O• → NO + N•   (chain initiation; activated at arc temperatures >2,000°C; dominant above 10,000°C)

N• + O₂ → NO + O•   (chain propagation)

Step 2 — Oxidation to NO₂ in the cooling plume:
2NO + O₂ → 2NO₂   (exothermic; thermodynamically favored below ~600°C; occurs as the hot gas plume cools toward ambient)

Step 3 — Reaction with airway moisture:
3NO₂ + H₂O → 2HNO₃ + NO   (in bronchi and alveoli; produces nitric acid; the tissue-injury step)

These reactions are driven entirely by temperature, not by the composition of the workpiece metal. A36 mild steel and 316 stainless steel differ enormously in their hazard profiles — but both generate NO₂ at concentrations that depend primarily on arc power (amperage × voltage), cutting speed, and ventilation, not on the alloy composition.

What does differ by alloy is whether secondary hazards are present:

The absence of Cr(VI) in mild steel plasma cutting is what creates the borderline half-face case. Cr(VI) at plasma stainless concentrations demands APF ≥ 25 even at low current — removing the half-face option entirely. Without that Cr(VI) driver, the question becomes: can the half-face APF of 10 be used if ambient NO₂ is controlled to below 0.2 ppm / 10 = 0.02 ppm by LEV? Answer: sometimes yes — at low current, with real monitoring.

Measured NO₂ Concentrations by Arc Current and Ventilation

Published industrial hygiene data for plasma cutting mild steel (A36, A572 carbon and low-alloy steel) shows a strong linear relationship between arc power input and breathing zone NO₂ concentration. Ventilation mode (no LEV vs. downdraft table) is the second-most significant variable, after amperage:

The key observation: at 60–80 A with a properly functioning downdraft table, breathing zone NO₂ is still 1.5–6× the ACGIH TLV-C. "Good LEV" reduces NO₂ by approximately 50–80% compared to general ventilation alone — but at commercial cutting currents, even that reduction is insufficient to bring concentrations below the 0.2 ppm TLV-C ceiling consistently.

At 40–60 A with documented LEV capture velocity ≥ 100 fpm, the measurement range straddles the TLV-C — sometimes below 0.2 ppm, sometimes above. This is the only operating regime where OV/P100 half-face is discussable, and only with continuous monitoring to confirm which side of the threshold the operator is actually on during any given work period.

The Borderline Case: When OV/P100 Is Defensible

The "borderline" designation for OV/P100 at ≤60 A mild steel plasma cutting does not mean "probably safe." It means that under a specific and narrow set of conditions, the combination of LEV + respirator can achieve sub-TLV-C exposures — but the control mechanism is the LEV, not the cartridge.

For OV/P100 half-face to be defensible at plasma mild steel cutting, all five of the following conditions must be simultaneously met:

1. Arc current ≤ 60 A. At 40–60 A, NO₂ generation is at the lower end of the plasma cutting range. With excellent LEV, breathing zone concentrations may dip into the 0.1–0.3 ppm range — borderline with the 0.2 ppm ceiling. Above 60 A, concentrations routinely exceed TLV-C even with LEV and monitoring cannot typically keep half-face in compliance.
2. Effective local exhaust ventilation — documented capture velocity ≥ 100 fpm. A downdraft cutting table with verified airflow, or a backdraft hood at the cutting position. General dilution ventilation alone (room air changes per hour, open bay doors) is not LEV and does not qualify. LEV capture velocity must be measured at the cutting kerf, not at the duct inlet.
3. Continuous real-time NO₂ monitoring at the breathing zone. An electrochemical direct-reading instrument (e.g., RAE Systems MiniRAE, Crowcon Clip 4, MSA Altair, or equivalent NIOSH-recognized sensor) worn at the collar or lapel — the breathing zone, not at table level. The sensor must log and alarm at a setpoint below 0.2 ppm. Historical monitoring (area sampling, end-of-shift badges) does not substitute for real-time confirmation because NO₂ exceedances can be brief and transient.
4. Intermittent cutting — arc-on time below 2 hours per shift. Production CNC plasma tables running 4–8+ hours of arc-on time per shift generate sustained NO₂ that cumulates even with LEV. Maintenance cutting, short job-shop runs, and low-volume fabrication are the contexts where low-current intermittent cutting is realistic.
5. Open indoor or outdoor setting with cross-ventilation. Not a confined space, vessel, pipe, or any setting where NO₂ can accumulate. In confined spaces, NO₂ can reach IDLH at any current; SAR or SCBA is required.

If conditions 1–5 are met and monitoring confirms sub-0.2 ppm NO₂ during the work period, OV/P100 half-face is a defensible minimum. If any condition fails — monitoring is unavailable, LEV capture velocity drops, current exceeds 60 A, job shifts to production volume — the half-face cannot be defended and PAPR is required.

The PAPR Threshold: Conditions That Rule Out Half-Face

Any of the following conditions individually require upgrading from OV/P100 half-face to PAPR with HE/OV/acid gas combination cartridge:

• Arc current > 80 A. At 80 A and above, breathing zone NO₂ at 0.3–1.2+ ppm with LEV consistently exceeds TLV-C. The half-face APF of 10 means the respirator provides protection only if ambient NO₂ is below 0.02 ppm — achievable only with exceptional ventilation at low current that cannot be maintained at 80+ A. No monitoring protocol can defend half-face at this current with commercial plasma cutting rates.
• Production CNC plasma cutting (> 2–4 hours arc-on time per shift). Continuous plasma operation (CNC plasma table, production fabrication, waterjet/plasma combined operation) generates sustained NO₂ that makes TLV-C compliance via LEV alone unreliable across a full shift. PAPR is the practical solution for production environments.
• No effective LEV. Without a downdraft table or comparable LEV achieving ≥ 100 fpm capture, general ventilation alone cannot reduce breathing zone NO₂ below the OV/P100 APF threshold. PAPR is required regardless of amperage.
• No continuous NO₂ monitoring instrumentation. Without real-time monitoring, the assumption that LEV + half-face achieves compliance is unverifiable. OSHA's Respiratory Protection Standard (1910.134) requires employers to demonstrate that APF-based protection is achieving compliance — without monitoring data, the employer cannot make that demonstration for plasma mild steel cutting. PAPR avoids the monitoring requirement by providing a higher APF margin.
• Confined or semi-enclosed space. Vessels, tanks, ship hulls, pipe interiors, structural steel enclosures: any setting where NO₂ can accumulate rather than dilute. In a confined space, plasma cutting at even moderate currents can produce IDLH concentrations within minutes. SAR (supplied-air respirator) with escape SCBA is required for entry; PAPR may be used only if continuous air monitoring confirms atmosphere remains below IDLH throughout the work period.
• High-manganese structural steel (A514, AREX, Mn > 1.3%). High-strength structural steels with elevated Mn content generate proportionally higher Mn oxide fume. The ACGIH TLV of 0.02 mg/m³ for Mn respirable fraction is so stringent that PAPR (APF 25) may be required for Mn control alone at production cutting rates, independent of the NO₂ analysis.

When PAPR is indicated, the recommended configuration for plasma mild steel is:

• PAPR motor unit: NIOSH-approved powered air-purifying respirator (e.g., 3M Versaflo TR-300+, Honeywell Uvex Purasolve, Sundstrom SR 500)
• Headpiece: Welding helmet/hood (loose-fitting: APF 25) or tight-fitting half-face (APF 50)
• Cartridge: Combination HE (high-efficiency) + organic vapor + acid gas (HE/OV/AG) — the HE filter captures Mn oxide and iron oxide fume; the OV component addresses any organic vapor co-contaminants from cutting fluid residue or coatings; the acid gas component provides base-chemistry capture of NO₂ at higher efficiency than plain OV, though cartridge service life for NO₂ must be established by IH monitoring, not standard OV service life tables
• Cartridge change-out: For NO₂ specifically — acid gas component must be changed out based on measured NO₂ breakthrough, not duration of use; OV service life tables do not model NO₂ breakthrough accurately

Why OV Activated Carbon Provides Negligible NO₂ Capture

The mechanism that makes OV cartridges effective for organic solvents is physical adsorption: organic molecules are drawn into the micropore structure of activated carbon by van der Waals dispersion forces. This is reversible — the organic molecule can desorb — which allows service life modeling based on the mass of contaminant the carbon can hold before breakthrough. For a 200-ppm acetone challenge at NIOSH test flow rates, a standard OV cartridge may provide hours of protection.

NO₂ is an inorganic oxidizing radical with an unpaired electron. On activated carbon, NO₂ does not undergo reversible physisorption — it undergoes destructive oxidative chemisorption: reacting with surface carbon functional groups via oxidation, consuming and destroying the active adsorption sites rather than occupying them temporarily. The reaction products are CO, CO₂, and various nitrogen-containing surface compounds. The result is:

1. Rapid breakthrough: At 1 ppm NO₂ challenge, published ASTM F3387 data shows standard OV activated carbon breaking through to 10% challenge concentration within minutes. At 0.5 ppm challenge — typical of 60 A plasma mild steel with good LEV — breakthrough timing varies but is still measured in tens of minutes, not hours.
2. Unpredictable service life: The oxidative destruction mechanism does not follow the Yoon-Nelson or Wheeler-Jonas models used for OV service life calculations. Breakthrough cannot be reliably predicted from contaminant concentration, flow rate, and temperature — making standard change-out schedule tables inapplicable.
3. No change-out indicator: Standard OV cartridges rely on odor threshold breakthrough as the change-out indicator for organic vapors with warning properties. NO₂ has a distinctive sharp odor at elevated concentrations but poor odor-based warning at concentrations below 0.5 ppm — below the smell threshold but well above the TLV-C. The cartridge provides no practical warning of the point at which it has become ineffective.

Acid gas (AG) cartridges with base-impregnated carbon (potassium hydroxide or sodium carbonate impregnation) perform better than plain OV for NO₂ — the base reacts with the acidic NO₂ (2NO₂ + Na₂CO₃ → NaNO₃ + NaNO₂ + CO₂) rather than undergoing the destructive oxidative reaction. But acid gas breakthrough for NO₂ at plasma cutting concentrations (0.3–4 ppm) is still rapid — minutes to low tens of minutes — and still follows a non-standard service life curve. The AG component in combination HE/OV/AG PAPR cartridges provides some improvement in NO₂ mitigation compared to plain OV, but it is not the mechanism being relied upon for compliance. The PAPR's higher APF (25 for loose-fitting hood, 50 for tight-fitting) is what provides the protection margin, combined with LEV reducing ambient concentrations.

In the borderline OV/P100 case: the OV component of the cartridge contributes negligible NO₂ capture. The protection mechanism is the APF 10 half-face acting on an ambient concentration that LEV has reduced to below 0.02 ppm (so that 0.02 ppm × APF 10 = 0.2 ppm TLV-C at the breathing zone). If ambient NO₂ exceeds 0.02 ppm — which it will at any cutting current above ~50 A, even with good LEV — the half-face cannot achieve sub-TLV-C exposure. Continuous monitoring is what detects this condition in real time.

Manganese Fume in Mild Steel and the Mn Routing Interaction

ASTM A36 mild steel contains 0.80–0.85% manganese by weight. ASTM A572 Grade 50 structural steel contains up to 1.35% Mn. High-strength abrasion-resistant steels (Hardox 400, AR400, AR500) commonly contain 1.0–1.6% Mn. When plasma arc temperatures exceed 1,800°C at the kerf, manganese vaporizes and oxidizes to form a mixture of MnO and MnO₂ fume particles in the submicron size range (highly respirable).

The ACGIH TLV for manganese respirable fraction is 0.02 mg/m³ — a threshold that reflects the risk of manganism (chronic Mn CNS toxicity), a progressive Parkinson's-like syndrome caused by cumulative manganese deposition in the basal ganglia. At 0.02 mg/m³, this is one of the most stringent occupational TLVs for any common industrial metal, 250× lower than OSHA's PEL of 5 mg/m³ ceiling.

For the respirator routing question, Mn fume interacts with the NO₂ analysis as follows:

• Standard A36 mild steel: P100 filter (OV/P100 or PAPR HE filter) captures Mn oxide particles at ≥99.97% efficiency. At APF 10 (half-face OV/P100), breathing zone Mn is reduced to 1/10 of ambient. With LEV controlling ambient Mn fume, half-face APF 10 is typically adequate for Mn at standard A36 cutting currents. Mn is not the primary driver of the PAPR upgrade decision for A36.
• High-Mn structural steel (A514, A656, AR plate with Mn > 1.3%): Proportionally higher Mn fume generation at production cutting rates may require PAPR (APF 25) to achieve 0.02 mg/m³ TLV compliance even with LEV. At these Mn content levels and production volume, the PAPR upgrade may be independently indicated by Mn alone before the NO₂ analysis even applies.
• Combined Mn + NO₂ routing: When encoding welding_fume_respirator.* fields for high-Mn structural steel plasma cutting, welding_fume_respirator.mn_hazard should be set to 'elevated' to trigger the PAPR upgrade from the Mn pathway. This combines with no2_hazard = 'yes' to produce a routing result that requires PAPR regardless of arc current or monitoring status.

The practical summary: for standard A36/A572 mild steel at low current, Mn is addressed by P100 and is not the upgrade driver. For high-Mn alloy steels at production cutting rates, Mn may independently require PAPR, reinforcing the NO₂-based upgrade decision.

Contrast Table: plasma_mild_steel vs plasma_stainless

The distinction between these two cutting_process values is the core routing decision for welding fume respirators in plasma cutting contexts. They share the same NO₂ mechanism but diverge at every other point in the protection analysis:

The routing logic consequence: cr6_hazard = 'no' does not imply no2_hazard = 'no'. Both fields are independent routing gates. An AI agent that uses cr6_hazard as the primary gate for plasma cutting respirator selection will route OV/P100 correctly to plasma mild steel in the Cr(VI) dimension — while entirely missing the NO₂ hazard that is present at the same or similar concentrations to plasma stainless.

Routing Logic and Metafield Schema

The welding_fume_respirator.* namespace fields for plasma mild steel cutting encode the borderline half-face case with its monitoring requirement as a distinct routing state from the unconditional PAPR case of plasma stainless:

// welding_fume_respirator.* namespace — plasma_mild_steel routing

welding_fume_respirator.cutting_process    = "plasma_mild_steel"
welding_fume_respirator.no2_hazard         = "yes"
welding_fume_respirator.cr6_hazard         = "no"         // key: distinguishes from plasma_stainless
welding_fume_respirator.oxides_of_nitrogen = "yes"
welding_fume_respirator.mn_hazard          = "standard"   // or "elevated" for A514, AR steels

// Determine minimum_filter_type based on conditions
if (current_amps <= 60
    AND lev_in_place               // documented capture velocity >= 100 fpm
    AND ih_monitoring_continuous   // real-time electrochemical NO2 sensor at breathing zone
    AND arc_on_time_hrs_per_shift < 2
    AND NOT confined_space) {

  minimum_filter_type = "ov_p100_monitored"
  // NOTE: OV component captures negligible NO2
  // Protection depends on LEV reducing ambient < 0.02 ppm
  // Half-face APF 10 × 0.02 ppm ambient = 0.2 ppm TLV-C at breathing zone
  // Monitoring confirms this condition is met in real time
  warning = "Borderline — compliance rests on LEV + monitoring, not cartridge"

} else {
  minimum_filter_type = "papr_he_combo"
}

// Confined space override
if (confined_space) {
  minimum_filter_type = "supplied_air_sar"
}

// High-Mn alloy override
if (mn_hazard == "elevated") {
  minimum_filter_type = "papr_he_combo"  // Mn alone may require APF 25
}

// Always reject particle-only respirators
REJECT: respirator_type IN ["n95", "p100_only", "ffp2", "ffp3", "n95_surgical"]
REASON: "Particle filters provide zero protection against NO2 gas from plasma arc"

The ov_p100_monitored minimum_filter_type is a specific state that encodes both the respirator type and the monitoring condition as a package. When an AI agent encounters this value, it should surface the monitoring requirement alongside the respirator recommendation — not present OV/P100 as a straightforward recommendation without the IH monitoring condition attached.

Three AI Agent Failure Modes for Mild Steel Plasma Cutting

Frequently Asked Questions

Does plasma cutting mild steel generate NO₂, and how does the hazard compare to plasma cutting stainless?

Yes — plasma cutting mild steel generates NO₂ via the identical Zeldovich mechanism as plasma cutting stainless steel: arc temperatures exceeding 10,000°C thermally dissociate atmospheric nitrogen (N₂ → NO → NO₂ on cooling). Measured NO₂ at 60–80 A with good LEV is typically 0.3–1.2 ppm — 1.5–6× the ACGIH TLV-C ceiling of 0.2 ppm. The critical difference from stainless is the absence of hexavalent chromium (Cr(VI)): mild steel contains less than 0.05% Cr, generating no measurable Cr(VI) fume. This changes the required minimum APF: plasma stainless mandates PAPR (APF ≥ 25) regardless of conditions because Cr(VI) demands higher protection. Plasma mild steel has a narrow window where OV/P100 half-face (APF 10) is defensible — ≤60 A with LEV and continuous monitoring — but the NO₂ hazard is present and significant at commercial cutting rates regardless of the alloy type.

When is OV/P100 borderline acceptable for plasma mild steel cutting?

OV/P100 is borderline defensible only when ALL of the following are simultaneously in place: arc current ≤ 60 A; effective LEV with documented capture velocity ≥ 100 fpm at the kerf; continuous real-time NO₂ monitoring at the breathing zone confirming sub-0.2 ppm throughout the shift; intermittent cutting (arc-on time < 2 hours/shift); and open indoor or outdoor setting, not confined space. Remove any one condition and PAPR is required. For production CNC plasma tables, any current above 60 A, or any setting without continuous monitoring, PAPR is the default recommendation.

Why does the OV component of OV/P100 not protect against NO₂?

OV activated carbon captures organic solvents via reversible physical adsorption. NO₂ is an inorganic oxidizing radical that undergoes destructive oxidative chemisorption — reacting with and destroying carbon surface sites rather than adsorbing reversibly. At plasma cutting NO₂ concentrations (0.3–4 ppm), breakthrough through standard OV carbon occurs within minutes. The protection in the borderline OV/P100 case comes from LEV reducing ambient NO₂ below 0.02 ppm (so APF 10 × 0.02 ppm = 0.2 ppm TLV-C at breathing zone), not from the OV cartridge removing NO₂.

What are the correct welding_fume_respirator namespace fields for plasma mild steel vs stainless?

Plasma mild steel: cutting_process='plasma_mild_steel', no2_hazard='yes', cr6_hazard='no', oxides_of_nitrogen='yes', minimum_filter_type='ov_p100_monitored' (≤60 A with conditions) or 'papr_he_combo' (production). Plasma stainless: cutting_process='plasma_stainless', no2_hazard='yes', cr6_hazard='yes', minimum_filter_type='papr_he_combo' always. The key routing distinction: cr6_hazard='no' does not imply no2_hazard='no'. Both gates must be independently evaluated for plasma cutting applications.

What is the manganese fume hazard in mild steel plasma cutting?

A36 mild steel contains 0.80–0.85% Mn; A572 up to 1.35%; AR plate steels up to 1.6%. The ACGIH TLV for Mn respirable fraction is 0.02 mg/m³ — 250× lower than OSHA's PEL. P100 filter at APF 10 (half-face) is typically adequate for Mn control at standard A36 cutting with LEV. For high-Mn structural steels (A514, AR400/500, AREX), production cutting may require PAPR (APF 25) for Mn control independently of the NO₂ analysis. Encode welding_fume_respirator.mn_hazard='elevated' for these alloys to trigger the Mn-based PAPR upgrade path.