A chemical storage vessel failed while in service. The failure occurred as cracking through the vessel wall, resulting in leakage of the fluid. The tank had been molded from a high-density polyethylene (HDPE) resin. The material held within the vessel was an aromatic hydrocarbon-based solvent. Investigation (visual inspection, stereomicroscopic examination, 20x/100x SEM images, micro-FTIR in the ATR mode, and analysis using DSC and TGA) supported the conclusion that the chemical storage vessel failed via a creep mechanism associated with the exertion of relatively low stresses. The source of the stress was thought to be molded-in residual stresses associated with uneven shrinkage. This was suggested by obvious distortion evident on cutting the vessel. Relatively high specific gravity and the elevated heat of fusion indicated that the material had a high level of crystallinity. In general, increased levels of crystallinity result in higher levels of molded-in stress and the corresponding warpage. The significant reduction in the modulus of the HDPE material, which accompanied the saturation of the resin with the aromatic hydrocarbon-based solvent, substantially decreased the creep resistance of the material and accelerated the failure.

A failure of an aboveground storage tank occurred due to external corrosion of the tank floor. The liquid asphalt tank operated at elevated temperatures (approximately 177 deg C, or 350 deg F) and had been in service for six years. Cathodic protection (rectifiers) had been installed since start-up of the tank operation. It was noted, however, that some operational problems with the rectifier may have interrupted its protection. Investigation (visual inspection, on-site examination and testing, EDS analysis of scale deposits, and MIC testing of the soil) supported the conclusion that corrosion may have been caused by an interruption in cathodic protection. The effectiveness of cathodic protection on established microbial deposits is questionable. Recommendations included ultrasonically testing the tank floor and replacing portions based on the remaining wall thickness. Doubling the wall thickness of the floor plates was also recommended.

Welded steel storage vessels used to hold mildly alkaline solution were produced in exactly the same manner from deep-drawn aluminum-killed SAE 1006 low-carbon steel sheet. After the cylindrical shell was drawn, a top low-carbon steel closure was welded to the inside diameter. The containers were then filled with the slightly alkaline solution, pressurized, and allowed to stand under ambient conditions. A small number, less than 1%, were returned because they began to leak in service. Inspection revealed general corrosion and pitting on the inner surfaces. However, other tanks that experienced the same service conditions developed no corrosion. Corrosion was linked to forming defects that provided sites for localized corrosion, and to lack of steam drying after cleaning, which increased susceptibility to general corrosion.

This paper describes the investigation of a corrosion failure of bottom plates on an aboveground tank used for the storage of potable water. The tank was internally inspected for the first time after six years of service. Paint blisters and rust spots were observed on the bottom plates and first to third course shell plates. Sand blasting and repainting of the bottom plates and first course shell plates was to be used as a remedial measure. However, during the sand blasting, holes and deep pitting were observed on the bottom plates. On-site visual inspection, magnetic flux leakage (MFL) inspection, ultrasonic testing (UT), and evaluation of the external cathodic protection (CP) system were used in the failure analysis. The corrosion products were analyzed using energy-dispersive X-ray analysis (EDAX). The failure is attributed to the ingress of water and its impoundment under the tank bottom along the periphery inside the ring wall and failure of water side epoxy coating. Various measures to prevent such failures in the future are recommended.

A storage tank had been in service at a petrochemical plant for 13 years when inspectors discovered cracks adjacent to weld joints and in the base plate near the foundation. The tank was made from AISI 304 stainless steel and held styrene monomer, a derivative of benzene. The cracks were subsequently welded over with 308 stainless steel filler wire and the base plate was replaced with new material. Soon after, the tank began leaking along the weld bead, triggering a full-scale investigation; spectroscopy, optical and scanning electron microscopy, fractography, SEM-EDS analysis, and microhardness, tensile, and impact testing. The results revealed transgranular cracks in the HAZ and base plate, likely initiated by stresses developed during welding and the presence of chloride from seawater used in the plant. It was also found that the repair weld was improperly done, nor did it include a postweld heat treatment to remove weld sensitization and minimize residual stresses.

Two type 420 martensitic stainless steel load cell bodies, which had been installed under two of the four legs of a milk storage tank failed in service. The failure occurred near a change in section and involved fracture of the entire cross section. Examination showed a brittle fracture that was preceded by a small fatigue region. Pitting corrosion was evident at the fracture origin. The areas around the load cells had been subjected to regular washdowns using high-pressure hot water, and the pitting was attributed to crevice corrosion between the load cell and the holddown bolts. Prevention of such corrosion by the use of a flexible sealant to eliminate the crevice was recommended.

A 22 m (72 ft) diameter filled grain storage bin made from a 0.2% carbon steel collapsed at a temperature of −1 to 4 deg C (30 to 40 deg F). Failure analysis indicated that fracture occurred in a two-step process: first downward, by ductile failure of small ligament from a bolt hole near the bottom of the tank to create a crack 25 mm (1 in.) long, and then upward, by brittle fracture through successive 1.2 m (4ft) wide sheets of ASTM A446 material. Site investigation showed that the concrete base pad was not level. Chemical analysis indicated that the material had a high nitrogen content (0.020%). The allowable stress based on yield was estimated using four different design criteria. Correlation among those results was poor. The different criteria indicated that the material was loaded from the maximum allowable to approximately 30% less than allowable. Nevertheless, at this stress level, fracture mechanics indicated that the 25 mm (1 in.) starter crack exceeded or was very near the critical crack length for the material. Additional factors not taken into account in the design equations included cold work from a hole punching operation, thread imprinting in bolt holes, and an additional hoop stress created by forcing an incorrectly formed panel to fit the pad base radius. These factors increased the nominal design stress to a sufficiently large value to cause the critical crack length to be exceeded.

Several type 304L stainless steel dished ends used in the fabrication of cylindrical vessels developed extensive cracking during storage. All of the dished ends had been procured from a single manufacturer and belonged to the same batch. When examined visually, several rust marks were observed, indicating contamination by rusted carbon steel particles. Liquid penetrant testing was used to determine the extent of the cracks, and in situ metallographic analysis was performed over the cracked region. The morphology of the cracks was indicative of transgranular stress-corrosion cracking (TGSCC). Conditions promoting the occurrence of the TGSCC included significant tensile stresses on the inside of the dished ends, the presence of surface contamination by iron due to poor handling practice using carbon steel implements, and storage in a coastal environment with an average temperature of 25 to 32 deg C (77 to 90 deg F), an average humidity ranging from 70 to 80%, and an atmospheric NaCl content ranging from 8 to 45 mg/m2 /day. Recommendations preventing further occurrence of the situation were strict avoidance of the use of carbon steel handling implements, strict avoidance of cleaning practices that cause long-term exposure to chlorine-containing cleaning fluid, and solution annealing of the dished ends at 1050 deg C (1920 deg F) for 1 h followed by water quenching to relieve residual stresses.

The dished ends of a heavy water/helium storage tank manufactured from 8 mm (0.3 in.) thick type 304 stainless plate leaked during hydrotesting. Repeated attempts at repair welding did not alleviate the problem. Examination of samples from one dished end revealed that the cracking was confined to the heat affected zone (HAZ) surrounding circumferential welds and, to a lesser extent, radial welds that were part of the original construction. Most of the cracks initiated and propagated from the inside surface of the dished ends. Microstructures of the base metal, HAZ, and weld metal indicated severe sensitization in the HAZ due to high heat input during welding. An intergranular corrosion test confirmed the observations. The severe sensitization was coupled with residual stresses and exposure of the assembly to a coastal atmosphere during storage prior to installation. This combination of factors resulted in failure by stress-corrosion cracking. Implementation of a new repair procedure was recommended. Repairs were successfully made using the new procedure, and all cracks in the weld repair zones were eliminated.

Large quantities of aluminum-clad spent nuclear materials have been in interim storage in the fuel storage basins at The Savannah River Site while awaiting processing since 1989. This extended storage as a result of a moratorium on processing resulted in corrosion of the aluminum clad. Examinations of this fuel and other data from a corrosion surveillance program in the water basins have provided basic insight into the corrosion process and have resulted in improvements in the storage facilities and basin operations. Since these improvements were implemented, there has been no new initiation of pitting observed since 1993. This paper describes the corrosion of spent fuel and the metallographic examination of Mark 31A target slugs removed from the K-basin storage pool after 5 years of storage. It discusses the SRS Corrosion Surveillance Program and the improvements made to the storage facilities which have mitigated new corrosion in the basins.

This paper describes the remote ultrasonic (UT) examinations of a high-level radioactive waste storage tank at the Savannah River Site in South Carolina. The inspections, carried out by E.R. Holland, R.W. Vande Kamp, and J.B. Elder, were performed from the contaminated, annular space of the 46 year old, inactive, 1.03 million gallon waste storage tank. A steerable, magnetic wheel wall crawler was inserted into the annular space through small (6 in., or 150 mm, diam) holes/risers in the tank top. The crawler carried the equipment used to simultaneously collect data with up to four UT transducers and two cameras. The purpose of this inspection was to verify corrosion models and to investigate the possibility of previously unidentified corrosion sites or mechanisms. The inspections included evaluation of previously identified leak sites, thickness mapping, and crack detection scans on specified areas of the tank. No indications of reportable wall loss or pitting were detected. All thickness readings were above minimum design tank-wall thickness, although several small indications of thinning were noted. The crack detection and sizing examinations revealed five previously undetected indications, four of which were only partially through-wall. The cracks that were examined were found to be slightly longer than expected but still well within the flaw size criteria used to evaluate tank structural integrity.

Aluminum-clad spent nuclear fuel is stored in water filled basins at the Savannah River Site awaiting processing or other disposition. After more than 35 years of service underwater, the aluminum storage racks that position the fuel bundles in the basin were replaced. During the removal of the racks from the basin, a failure occurred in one of the racks and the Savannah River Technology Center was asked to investigate. This paper presents the results of the failure analysis and provides a discussion of the effects of corrosion on the structural integrity of the storage racks.

When a large LPG low-carbon steel storage tank was put into service for the first time and filled beyond the proof testing level, a brittle fracture crack initiated at a fillet weld between a stiffener ring and the wall. The crack propagated to a length of 5.5 m and arrested. Analysis showed that the plates satisfied the criteria of BS 4741. It was concluded that the cause of crack initiation was the lack of a mouse hole at the junction between the stiffening ring and the wall of the tank. The tank was repaired and put back in service. When it was filled beyond the proof test level, again a brittle crack was initiated at a horizontal weld defect and propagated vertically, destroying the tank and the liquefaction plant. The initiation site was a thumbnail elliptical crack in a horizontal weld, having a depth of 1.5 mm, and a length of 4.5 mm. This showed that as late the mid-1970s, misunderstanding of brittle fracture led to the wrong design and construction of an LPG storage tank. The best design specification is to use a correlation between LAST, the Lowest Anticipated Service Temperature, and the DBTT measured by either Charpy tests or DTT.

A four-million gallon capacity (15,142 cu m) oil storage tank ruptured upon filling after re-erection near West Elizabeth, PA on 2 Jan 1988. The tank shell split vertically with failure originating at a flaw existing prior to the reconstruction. Brittle fracture occurred both up and down from the defect when the stress induced by filling reached a critical value for the steel, which had poor toughness properties. This steel had been used in the original construction of a tank in Ohio more than 40 years previously. The defect at which brittle fracture originated in the tank shell showed evidence of burning from a torch. This tank failure was the catalyst for the introduction of new rules concerning the inspection and assessment of older storage tanks.

A 100,000 barrel crude oil storage tank rupture caused extensive property damage in Dec 1980, in Moose Jaw, Saskatchewan. Failure was attributed to a brittle fracture that originated at a weld between a reinforcing pad and a manway nozzle. Factors that contributed to the brittle fracture included incomplete penetration in a single-bevel groove weld, poor impact properties of the hot rolled ASTM A283 low-carbon steel base material, and air temperature down to 27 C on the day of failure. Details of the analysis and results of impact testing are discussed.

A substantial number of copper alloy C27000 (yellow brass, 65Cu-35Zn) ferrules for electrical fuses cracked while in storage and while in service in paper mills and other chemical processing plants. The ferrules, made by three different manufacturers, were of several sizes. One commonly used ferrule was 3.5 cm long by 7.5 cm in diam and was drawn from 0.5 mm (0.020 in.) thick strip. Investigation (visual inspection, metallographic examination, and a mercurous nitrate test, which is an accelerated test used to detect residual stress in copper and copper alloys) of both ferrules from fuses in service and storage in different types of plants, and ferrules from newly manufactured fuses, supported the conclusion that the ferrules failed by SCC resulting from residual stresses induced during forming and the ambient atmospheres in the chemical plants. The atmosphere in the paper mills was the most detrimental, and the higher incidence of cracking of ferrules there was apparently related to a higher concentration of ammonia in conjunction with high humidity. Recommendations included specifying that the fuses meet the requirements of ASTM B 154.

This paper deals with disk drive failures that occur in the interface area between the head and disk. The failures often lead to the loss of stored data and are characterized by circumferential microscratches that are usually visible to the unaided eye. The recording media in disk drives consists of a metal, glass, ceramic, or plastic substrate coated with a magnetic material. Data errors are classified as ‘soft’ or ‘hard’ depending on their correctability. Examination has shown that hard errors are the result of an abrasive wear process that begins with contact between head and disk asperities. The contact generates debris that, as it accumulates, increases contact pressure between the read-write head and the surface of the disk. Under sufficient pressure, the magnetic coating material begins wearing away, resulting in data loss.

The crosslinked epoxy resin encapsulant protecting an electromagnetic valve coil failed during long-term storage and was examined to determine the cause. The investigation included fault-tree analysis, FTIR and EDX spectroscopy, and differential scanning calorimetry with thermogravimetric analysis. Based on test data, the epoxy resin had not been properly cured and was hydrolyzed in its compromised state because of humidity. Hence, the depolymerized material gradually softened to the point where the effect of creep caused it to flow, ultimately causing the failure.

A riveted 0.25% carbon steel oil-storage tank in Oklahoma was dismantled and reassembled in Minnesota by welding to form a storage tank for soybean oil. An opening was cut in the side of the tank to admit a front-end loader. A frame of heavy angle iron was welded to the tank and drilled for bolting on a heavy steel plate. The tank was filled to a record height. In mid-Jan the temperature dropped to -31 deg C (-23 deg F), with high winds. The tank split open and collapsed. The welding used the shielded metal arc process with E6010 electrodes, which could lead to weld porosity, hydrogen embrittlement, or both. At subzero temperatures, the steel was below its ductile-to-brittle transition temperature. These circumstances suggest a brittle condition. Steps to avoid this type of failure: For cold conditions, the steel plate should have a low carbon content and a high manganese-to-sulfur ratio and be in a normalized condition, low-hydrogen electrodes and welding practices should be used, all corners should be generously radiused, the welds should be inspected and ground or dressed to minimize stress concentrations, postweld heating is advisable, and radiographic and penetrant inspection tests should be performed.