Published 27 June 2026 • By Dr. Megan Tranter
Welding is among the most widespread industrial processes, and the fume it generates is a complex, dynamic aerosol of metal oxides condensed from vaporized consumables and base metal. For decades, the principal concern was metal fume fever and respiratory irritation, but the evidentiary picture has darkened considerably. In 2018, the International Agency for Research on Cancer reclassified welding fumes as a Group 1 human carcinogen, and a parallel body of work has established that the manganese fraction of welding fume produces a progressive neurological syndrome resembling Parkinson’s disease. These developments place welding fume at the center of contemporary industrial-hygiene practice and demand a rigorous, exposure-based approach to control.
What You’ll Learn
- How welding fume forms and what determines its composition.
- Why manganese is a potent neurotoxicant and how manganism differs from idiopathic Parkinson’s disease.
- The evidence that Parkinsonian signs progress in a dose-dependent manner with cumulative manganese exposure.
- The current ACGIH TLV for manganese and how exposures should be sampled.
- The engineering and administrative controls that reduce welding-fume exposure.
Introduction
Welding fume is no longer a nuisance dust but a confirmed human carcinogen that also carries a distinct neurotoxic hazard from its manganese fraction. The modern industrial hygienist must characterize exposures with size-selective sampling, benchmark them against a far stricter occupational exposure limit than the historical OSHA ceiling, and prioritize local exhaust ventilation at the arc. This article reviews fume composition, manganese neurotoxicology, the dose-response evidence for Parkinsonism, current exposure limits, and the hierarchy of controls.
Composition and Formation of Welding Fume
Welding fume forms when the intense heat of the arc vaporizes metal, which then oxidizes and condenses into ultrafine and nanoscale particles. The composition depends on the base metal, the electrode or filler wire, fluxes, and coatings, and may include iron, manganese, chromium, nickel, copper, zinc, and silica, as well as gaseous co-contaminants such as ozone and nitrogen oxides. Stainless-steel welding also releases hexavalent chromium and nickel, both established carcinogens; the assessment and control of this fraction require attention to hexavalent chromium. Because the particles are predominantly in the respirable and ultrafine size range, they penetrate deep into the alveolar region and, in the case of manganese, can translocate along the olfactory nerve to the brain, a pathway that complicates the conventional understanding of routes of entry for occupational toxicants.
Manganese and the Nervous System
Manganese is an essential trace element, yet at occupational exposure levels, it becomes a potent neurotoxicant. It accumulates preferentially in the basal ganglia, particularly the globus pallidus, where it disrupts dopaminergic and GABAergic neurotransmission. The resulting clinical picture, manganism, features bradykinesia, rigidity, gait disturbance, postural instability, masked facies, and impaired speech, with a relative paucity of the resting tremor characteristic of idiopathic Parkinson’s disease. The behavior of manganese as a metal toxicant, including its absorption, distribution, and target-organ selectivity, reflects broader principles discussed in our review of the toxicity of metals in the workplace.
The Evidence for Dose-Dependent Parkinsonism
Longitudinal cohort studies of welding-exposed workers have demonstrated that Parkinsonian signs progress in a dose-dependent manner with cumulative manganese exposure. Crucially, these neurological effects have been observed at airborne manganese concentrations below older federal limits, indicating that historical standards were insufficiently protective. This evidence informed the ACGIH decision to lower the threshold limit value for manganese to 0.02 mg/m³ for the respirable fraction and 0.1 mg/m³ for the inhalable fraction, a substantial reduction reflecting the sensitivity of the nervous system to chronic low-level exposure.
Exposure Limits and Assessment
The current ACGIH TLV for manganese is 0.02 mg/m³ (respirable) and 0.1 mg/m³ (inhalable) as eight-hour time-weighted averages, markedly more stringent than the older OSHA ceiling of 5 mg/m³. Exposure assessment for welders should therefore employ size-selective sampling to capture the respirable fraction, with personal sampling conducted inside the welding helmet, where feasible, to reflect true breathing-zone exposure. Total particulate sampling alone is inadequate, and the welder’s position relative to the plume, the degree of enclosure, and the adequacy of local exhaust all materially affect measured concentrations.
Engineering and Administrative Controls
Control of welding fume follows the hierarchy of controls. Process substitution (lower-fume consumables, reduced manganese content, or alternative joining methods) is preferred where feasible. Local exhaust ventilation, including fume-extraction guns and movable hoods positioned close to the arc, is the cornerstone of engineering control. General dilution ventilation is a supplement, not a substitute. Modifying welding parameters, such as voltage and shielding gas, can reduce the neurotoxic potential of the fume generated. Where controls cannot achieve adequate reduction, properly fit-tested powered air-purifying respirators provide reliable protection, and supplied-air systems are warranted in confined or poorly ventilated spaces. The selection and fit testing of those respirators is covered in our discussion of respiratory protection and fit testing.
Summary
Welding fume warrants the rigor reserved for confirmed carcinogens and neurotoxicants. Characterize exposures with respirable-fraction sampling inside the helmet, benchmark against the ACGIH TLV of 0.02 mg/m³ rather than the outdated OSHA limit, and prioritize local exhaust ventilation at the source. Medical surveillance attentive to early neurological signs, lower-emission consumables, and properly fit-tested respiratory protection complete a defensible program.
Helpful Resources
- OSHA Welding, Cutting, and Brazing Topic Page
- CDC NIOSH Welding Topic Page
- Related posts on this site: respiratory protection and fit testing and the toxicity of metals.
Bibliography
American Conference of Governmental Industrial Hygienists. (2026). TLVs and BEIs: Manganese and inorganic compounds. ACGIH.
Rose, V. E., & Cohrssen, B. (Eds.). (2011). Patty’s industrial hygiene (6th ed.). Wiley.
International Agency for Research on Cancer. (2018). Welding, molybdenum trioxide, and indium tin oxide (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 118). IARC.
Dirandeh, E., Palizgir, A., & Kassiri, N. (2022). An overview of the relationship between occupational manganese exposure and parkinsonism. Cureus, 14(12), e32161. https://doi.org/10.7759/cureus.32161
National Institute for Occupational Safety and Health. (2007). NIOSH pocket guide to chemical hazards: Manganese (DHHS NIOSH Publication No. 2005-149). NIOSH.
Occupational Safety and Health Administration. (n.d.). Welding, cutting, and brazing (29 CFR 1910.252). U.S. Department of Labor. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.252
Plog, B. A., & Quinlan, P. J. (Eds.). (2012). Fundamentals of industrial hygiene (6th ed.). National Safety Council.
Racette, B. A., Searles Nielsen, S., Criswell, S. R., Sheppard, L., Seixas, N., Warden, M. N., & Checkoway, H. (2017). Dose-dependent progression of parkinsonism in manganese-exposed welders. Neurology, 88(4), 344-351. https://doi.org/10.1212/WNL.0000000000003533
Sriram, K., Lin, G. X., Jefferson, A. M., Stone, S., Afshari, A., Keane, M. J., McKinney, W., Jackson, M., Chen, B. T., Schwegler-Berry, D., Cumpston, A., Cumpston, J. L., Roberts, J. R., Frazer, D. G., & Antonini, J. M. (2015). Modifying welding process parameters can reduce the neurotoxic potential of manganese-containing welding fumes. Toxicology, 328, 168-178. https://doi.org/10.1016/j.tox.2014.12.015
Shin, H. W. (2025). Delayed emergence of Parkinson’s disease after reversible manganese-induced parkinsonism: A case report. BMC Neurology, 25(1), 423. https://doi.org/10.1186/s12883-025-04448-2