Told ten months out: the CSB's Nippon Dynawave update and the inspection that wasn't acted on

On 28 August 2026 the U.S. Chemical Safety Board released a factual investigation update on the 26 May 2026 catastrophic failure of a 1.2-million-gallon atmospheric storage tank at the Nippon Dynawave Packaging (NDP) pulp and paper mill in Longview, Washington. The failure of the tank — identified as G Tank — released approximately 900,000 gallons of scalding, highly caustic white liquor, fatally injuring 11 employees and seriously injuring three others. The investigation remains open, but the update is worth reading closely because it turns on a mechanism that a process hazard analysis quietly depends on and rarely tests: whether the inspection programme behind a credited safeguard actually drives action.

What the update says

The account in the CSB’s release and the full update document runs as follows. G Tank held white liquor — an extremely hot (around 200 °F) and highly corrosive alkaline solution used to produce pulp from wood chips. In July 2025, roughly ten months before the failure, an external inspection of the tank’s carbon-steel shell found that significant portions had thinned below the calculated minimum safe thickness — the amount of material a tank needs to safely contain its contents at its operating conditions.

The inspection contractor’s report recommended internal inspection and repair of the thinned areas. It also stated plainly that the tank was “[n]ot fit for continued service” unless repaired, and that there was a “[h]igh likelihood/consequence of failure.” Despite that finding, G Tank stayed in service. Two further external inspections, in October 2025 and February 2026, again found substantial areas of the shell below minimum thickness. After each, the CSB says, NDP still did not internally inspect or repair the shell, take the tank out of service, or reduce its operating conditions.

On the morning of 26 May 2026, a process upset required NDP to stop sending white liquor from G Tank to the digester. But white liquor kept arriving from the recausticizing process, and with none leaving, the level rose to about 90 percent full by 6:22 a.m. At approximately 7:09 a.m. the tank catastrophically failed, releasing its contents across a courtyard and into nearby buildings where employees had gathered for morning meetings. CSB Chairperson Steve Owens said the Board is “greatly concerned that the tank was not promptly removed from service or properly repaired after the clear findings in the inspection report.” The Board’s continuing work covers NDP’s mechanical integrity, inspection and maintenance practices, its response to G Tank’s deteriorated condition, the operating conditions preceding the failure, emergency-response procedures, and applicable regulation and industry guidance.

Why it matters for study work

This is a mechanical-integrity and management-of-response story more than a HAZOP-node story, and that is precisely what makes it instructive. A process hazard analysis credits the tank shell as a passive safeguard against loss of containment, and implicitly credits an inspection and mechanical-integrity programme with keeping that shell above its minimum thickness. Here the inspection programme did its job — it detected the thinning, and did so three separate times over roughly a year. The gap the CSB is examining sits downstream of detection: between knowing a vessel is unfit for service and actually removing it, repairing it, or de-rating it.

For facilitators and revalidation teams, that reframes a familiar question. It is not enough for a study to note “tank inspected per programme” as the basis of a safeguard. The sharper questions are which safeguards in the study depend on an inspection finding translating into a repair or a shutdown, whether the site has a bright-line rule that pulls a vessel graded “not fit for continued service” out of service, and who has the authority — and the obligation — to act when a written examination comes back negative. Overfill protection is a second thread: with outflow stopped and inflow continuing, level climbed steadily toward full in the hours before failure, so the high-level and diversion safeguards for that scenario are worth their own line in any similar study.

There is also a siting note. The severity of the outcome owed much to where people were relative to a large inventory of hot, corrosive liquid at the moment of release. Muster points, walkways, and occupied buildings near major-accident inventories belong in a facility-siting review, not left to chance.

The CSB has stated its investigation continues and that complete findings and any recommendations will come later. This piece reflects only the factual update issued on 28 August 2026.