CSB investigating fatal hydrogen sulfide release during decommissioning at Catalyst Refiners in Nitro, West Virginia

The U.S. Chemical Safety Board is investigating a fatal chemical release that occurred on April 22, 2026 at the Catalyst Refiners, Inc. facility in Nitro, West Virginia. Two workers were killed, more than 30 people sought medical care — one reportedly in critical condition — and a shelter-in-place was issued for the surrounding community. The CSB announced its investigation the following day.

What happened

Catalyst Refiners, owned by the Ames Goldsmith Corporation, conducted silver recycling and ethylene oxide catalyst refining at the site. According to the CSB’s initial information, employees were cleaning and decommissioning equipment in preparation for the facility’s closure when nitric acid and another substance were reportedly mixed, generating and releasing toxic hydrogen sulfide gas inside a building where numerous employees were present.

CSB Chairperson Steve Owens said the agency opened the investigation “to determine how it happened and identify ways to help prevent something like this from happening again.”

The investigation is ongoing; findings and any recommendations will be published on the CSB’s current investigations page.

Why this matters for hazard studies

The reported circumstances place this incident in a category with a long history: inadvertent mixing of incompatible chemicals during non-routine work. Several threads are worth watching as the investigation develops.

The first is the operating mode. Cleaning, decontamination, and decommissioning ahead of a permanent closure sit outside the operating envelope that most process hazard analyses study. Deviations that are implausible during normal operation — cross-connection of cleaning chemicals, heels and residues contacting reagents they never meet in the process — become credible precisely when a plant is being emptied and washed down. A hazard review scoped to the decommissioning activity itself, rather than reliance on the PHA of the former operating process, is the relevant safeguard.

The second is chemical compatibility knowledge at the point of work. Hydrogen sulfide generation from an acid contacting a sulfide-bearing material is a known mechanism, but that knowledge has to reach the crew writing and executing the cleaning procedure, typically through compatibility charts tied into the work-planning and permit process.

The third is occupancy. As reported, the release occurred inside a building with numerous employees present. How many people need to be near a one-off hazardous cleaning operation is a question a pre-task review can ask cheaply — and one that matters most when the answer changes the consequence from a near miss to multiple fatalities.

The CSB’s findings will determine which of these threads apply here. Interim updates, when issued, will appear on the investigation page linked above.