Skip to main content
Clean Hydrogen Partnership

Explosion in a chemical plant, caused by accidental hydrogen formation

Event

Event ID
485
Quality
Description
The incident affected a linear alkyl-benzene (LAB) plant. The plant was down for routine maintenance when an explosion occurred. An explosion and fire occurred at a 3000-gallon (approx. 11400 l) reactor. Process materials released from the various vessels and piping fuelled a subsequent fire that took approximately 2 hours to extinguish.

As reported by Ali Reza (Ali Reza et al., 2007, see References), before the explosion, the bottom of the reactor was plugged with approximately 180 gallons (approx. 800 l) of a sludge-like mixture of coarse aluminium powder, aluminium chloride and various hydrocarbons.
The explosion occurred approximately 10 minutes after steam was introduced at the bottom of the reactor in an effort to break up the sludge accumulation.

Event Initiating system
Classification of the physical effects
Hydrogen Release and Ignition
Nature of the consequences
Macro-region
North America
Country
United States
Date
Main component involved?
Reactor / Oven / Furnace / Test Chamber
How was it involved?
Internal Explosion (H2-Air Mixture)
Initiating cause
Run-Away Reaction
Root causes
Root CAUSE analysis
The INITIATING CAUSE was the decision to use water steam to clean the chemical sludge.
The investigation identified the cause of the explosion the direct and sustained contact between the steam and the sludge at the bottom of the reactor, which induced an exothermic hydrolysis of aluminium chloride. The products of the hydrolysis were mainly hydrogen chloride vapour and hydrogen gas. During laboratory tests, this reaction was able to generate pressure up to 45 bar and temperatures above 370 degree Celsius (700 Fahrenheit).

The U.S. CSB (see References) identified the ROOT CAUSE in the lack of a Management of Changes system. The plant operator had decided to change of the catalyst use for the reaction few months before, but no risk assessment review had been performed on the new process.

An additional contributing cause was the misunderstanding occurred when the shifts responsible workers communicated with each-others. The injection of steam was not meant to be continuous and prolonged.

Facility

Application
Petrochemical Industry
Sub-application
alkylbenzene plant
Hydrogen supply chain stage
All components affected
vessels and pipes
Location type
Confined
Location description
Industrial Area
Operational condition
Pre-event occurrences
MONTHS BEFORE THE INCIDENT
About three months prior to the incident, the plant changed its process technology and discontinued the direct addition of aluminium chloride to the reactor.
Instead, powdered aluminium was added to the reactor, where it combined with hydrogen chloride to form the necessary aluminium chloride. Shortly after the plant switched to the new process, the reactor became fouled with a sludge-like catalyst residue (CBS Bulletin, see References).
JUST BEFORE THE INCIDENT
Before the incident, the process plant was shut down for routine maintenance. The need to change pipes to the reactor required its emptying. However, the operators were unable to remove the liquid hydrocarbons from the reactor because the bottom outlet nozzle appeared plugged with a solid, sludge-like material. Initial attempts to clear the solid plug with nitrogen were unsuccessful. The system was flushed for 6 hours with fresh paraffin to reduce the benzene level in the system. A vacuum truck was then used to remove the majority of the liquids from the reactor via a side man-way, but approximately 185 gallons of sludge remained inside the vessel.
Having demonstrated that the sludge could be broken with water, a first attempt was done, without results. it was then decided to use pressurised steam (Ali-Razi, 2002).

Emergency & Consequences

Number of injured persons
3
Number of fatalities
0
Environmental damage
13000000
Post-event summary
No one was present in the immediate vicinity of the reactor when it exploded, and there were no fatalities.
Two employees and one contractor received first- and second-degree burns; they were wearing fire-resistant work clothing, which provided a measure of protection. Another contractor injured his back when he fell.
Property damage was estimated at $13 million.
Official legal action
The Department of the Environment assessed a $75,000 civil penalty against the Company for nuisance and other air pollution violations (https://mde.maryland.gov/programs/Pressroom/Pages/323.aspx)

Lesson Learnt

Lesson Learnt

The CBS Bulletin 2001-04-SB dedicated to this case identifies as root cause the failure to implement a systematic Management of Changes process.
The situation at the plant was not involving emergencies that required rapid decisions. Time was available to look into the circumstances more thoroughly. To maximize the effectiveness of a Management of Changes system, the following activities should be included:
1. Define safe limits for process conditions, variables, and activities—and train personnel to recognize significant changes. Combined with knowledge of established operating procedures, this additional training will enable personnel to activate the MOC system when appropriate.
2. Apply multidisciplinary and specialized expertise when analysing deviations.
3. Use appropriate hazard analysis techniques. l Authorize changes at a level commensurate with risks and hazards.
4. Communicate the essential elements of new operating procedures in writing.
5. Communicate potential hazards and safe operating limits in writing.
6. Provide training in new procedures commensurate with their complexity.
7. Conduct periodic audits to determine if the program is effective.

Event Nature

Release type
gas mixture
Involved substances (% vol)
H2,
HCl
Presumed ignition source
Auto-ignition

References

Reference & weblink

CSB Safety Bullettin No. 2001-04-SB August 2001 <br />
https://www.csb.gov/file.aspx?DocumentId=5612<br />
(accessed Jan 2023)

Accident analaysis peformed by Exponent<br />
<br />
Available at https://www.exponent.com/experience/october-1998-explosion-at-the-conde… />
(accessed Dec 2019, not available anymore))

Ali Reza P.E., et al. , Runaway reactions in aluminum, aluminum chloride, HCl, and steam: An investigation of the 1998 CONDEA Vista explosion in Maryland<br />
Process Safety Progress 21 (2002) 2061-267 https://doi.org/10.1002/prs.680210312

JRC assessment