Case Ditails

Scenario
Case Name Rupture of a reactor during an air-tight test of reactor used for 21 years at a catalytic hydro-desulfurization unit
Pictograph
Date April 1, 1980
Place Tokuyama, Yamaguchi, Japan
Location Refinery
Overview At a catalytic hydro-desulfurization unit, when the pressure of the reactor was raised to the regular value for an air tight test after a turnaround shutdown, the reactor ruptured.
The cause was considered to be a crack that occurred at the reactor, and developed gradually with poor quality material or nonconformity of a welding rod used for repairs after construction.
The reactor was used for 21 years, and only a visual check was carried out in the previous ten years.
The crack could not to be found although the visible range was partly inspected in the turnaround shutdown just before the accident.
Incident At a refinery, a reactor of a catalytic hydro-desulfurization unit used for over 20 years ruptured. It exploded during a leakage test after a turnaround shutdown.
Although additional construction was performed at the time of the turnaround shutdown, it was not considered as the cause.
Fig2 shows distribution of debris.
Processing Manufacture
Individual Process Maintenance
Chemical Reaction Hydro-desulfurization
Substance Nitrogen, Fig3
Type of Accident Rupture
Sequence The unit was built in 1959. The following tests were carried out during an annual turn-around shutdown until 1971. Visual inspection, Dye Penetrant Test, Magnetic-particle examination, SUMP examination, and Fixed point wall thickness measurement.
A catalyst was drawn off in 1971 and Dye Penetrant Test of the whole weld line was carried out. Only a visual check was made after that.
During a turnaround shutdown in March, 1980, a visual check was made inside the visible range. Dye Penetrant Test of the upper part and a lower nozzle ring slot, and fixed point wall thickness measurement on the reactor were also conducted. No bad point was found.
From March 18th to 27th, 1980, the reactor was repaired during a turnaround shutdown.
About 10:30 on April 1st, high pressure nitrogen was sent from a liquid nitrogen tank lorry to the reactor, and an air tight test started.
22:29 A leak from a valve of the separation section was found. When it was tightened, the leak stopped.
23:44 As the pressure of the separation section reached 5.2 MPaG, the air tight test on the separation section was completed. Then, the separation section and the reaction section were separated, and the pressure was raised only at the reaction section.
23:55 As the pressure of the reaction section reached prescribed 5.5 MPaG, the charge of nitrogen was going to cut off. At that moment the reactor ruptured.
The accident caused damage to windows etc. of about 95 houses within a radius of about 650 m.
Cause At a hydro-desulfurization unit using high-pressure and high temperature hydrogen similar to this reactor, clad steel of SUS316L (austenitic) had been used. In this reactor, SUS405 ferrite clad steel was adopted for the first time, because many stress corrosion cracks occurred with SUS316L clad. At an inspection during manufacturing, a welding defect was found. After removing the defective part with a grinder, weld repair was carried out with a welding rod different from the one used for construction, and, this welding rod was suitable. The cause was not a welding defect, but a late-generated base material crack. This crack grew with a long-term use or the change of temperature and pressure at turn-around shutdowns/start-ups, etc.
Although the base material conformed to the standard, the Charpy impact value was partly low.
Breaking strength fell due to the above factor.
Dye Penetrant Test was carried out on the whole weld line until the internal overhaul inspection at a turnaround shutdown in 1971. However, for nine years until the accident, as inspection was partial, the crack was not found.
Countermeasures For equipment used under severe conditions, the following are adopted: Material and a welding process that are appropriate for the conditions. Defects grow as period of use gets longer. It is necessary to examine the inspection method.
Knowledge Comment Do not neglect either follow-up survey of equipment or collection of latest information as to equipment using a new material.
Background Materials for a high-pressure hydrogen reactor have troublesome characteristics, such as the usable range always changes. Therefore, although a new material was fully examined, it is not necessarily suitable for long-term and various uses. Corrosion and cracks must always be checked.
A fundamental factor is a long-term use and omission of inspection..
However, annual execution of the whole inspection is difficult, considering time and expense of drawing off and re-loading a lot of catalysts. At least, sufficient inspection every several years would be required.
Although it is thought that elucidation of the weak point in the new material was insufficient, it is actually difficult. It might be a problem of judgment on safety.
Incidental Discussion Hazard of using new material
Reason for Adding to DB Material hazards of a reactor using high pressure hydrogen
Scenario
Primary Scenario Misjudgment, Misunderstanding, Not Foreseen the Progress of Defect, Poor Value Perception, Poor Safety Awareness, Inadequate Risk Recognition, Organizational Problems, Poor Management, Poor Collection of New Information, Planning and Design, Poor Planning, Poor Design, Usage, Maintenance/Repair, No Inspction, Failure, Large-Scale Damage, Rapture, Loss to Organization, Economic Loss, Damage to Society, Social Systems Failure, 95 houses of near-by inhibitant damaged
Sources Editorial supervision of the countermeasure guide decision committee of the explosion and fire prevention in the chemical industry. Editorial supervision of Chiba Labor Standards Bureau, Pressure and air tight test (gas test), Guideline for explosion and fire prevention in the chemical industry, pp.97-99, p.98 (1996)
High Pressure Gas Safety Inst. of Japan, I Oil T refinery. The accident analysis report on the second reactor of a catalytic hydro-desulfurization unit (1980)
High Pressure Gas Safety Inst. of Japan, Report on safety research at Tokuyama, Shinnanyo, Iwakuni, and Otake complex, pp.54-55 (1982).
Physical Damage Destruction of surrounding equipment and buildings (frames and end plates) by reactor debris. Breakage of windows and fittings and cracking of walls etc. of about 95 houses around the factory (about 650 m from the reactor).
Multimedia Files Fig2.Fragments dispersion situation figure
Fig3.Chemical formula
Field Chemicals and Plants
Author ITAGAKI, Haruhiko (Japan National Institute of Occupational Safety and Health)
TAMURA, Masamitsu (Center for Risk Management and Safety Sciences, Yokohama National University)
Scenario