Bridge failure sparks concern over hydrogen embrittlement

Engineers know hydrogen embrittlement is a bad thing. However, ask many to explain what it is and how to avoid it, and you may be met by silence. The phenomenon occurs when hydrogen atoms penetrate the crystalline molecular structure of metals and weaken the bonds.

The effect is that elongation and ductility is severely reduced. Parts that fail because of hydrogen embrittlement have almost no plastic deformation, with the fractured cross-sections looking more like a ceramic.

The problem is well known by industrial fastener manufacturer, EJOT. Its sales manager for industrial products Steve Wynn, explains the problem: "There are two causes of hydrogen embrittlement; where hydrogen from the environment assists with the failure i.e. corrosion, and where hydrogen from the manufacturing process assists with the failure."

Hydrogen embrittlement is well known to fastener suppliers that know it can be induced during electroplating, heat treatment and also cleaning (pickling) processes during manufacture.
However, overlooked and misunderstood are the effects of the environment, especially on highly stressed large bolted joints. This is the realm of designers and specifiers. Uckfield based TR Fasteners has experienced various cases in the past where large bolted joints have unexpectedly failed because of poor specification, and not manufacturing errors.

"We are a manufacturer and supplier of components and we don't always get told where our fasteners end up," says Geoff Budd, managing director at TR Fastenings. "And that is why this is such an important topic. People think when they have this sort of failure it is caused by problems during manufacture, but actually it's often the specification that has brought on the phenomenon. That is why we want to educate customers, and get designers to fully understand it."

Perhaps the most high profile case at the moment is California's replacement San Francisco–Oakland Bay Bridge, currently under construction. The bridge uses shear keys, giant steel and concrete boxes sandwiched between the bridges' deck and supporting concrete pier, to counteract movement during an earthquake.

Anchoring the shear keys in place are a series of 3-inch threaded seismic safety rods around 5m long. In March, this year, when workers began to fully tighten the bolts, 32 of the 96 rods catastrophically failed completely snapping them in two.

As the finger of blame was pointed toward supplier, Dyson Corp, the company hit back at the state agency responsible for the bridge construction Caltrans saying, 'Dyson supplied exactly what was requested from us'.

Expert reports have since shown that the rods were indeed compliant with mechanical and chemical requirements, which the company cites as evidence that it is not responsible for the failures. It went on to say that it had no involvement in the application of the products, and that it was the specification that was ultimately at fault.

Though full details are yet to emerge, numerous errors have already come to light. The rods were forged, tempered and delivered in 2008. However, a 'paperwork error' meant that documents showing heat-treatment couldn't be produced. As a result Caltrans sent them back to raise the correct paperwork.

This meant the rods were heat treated twice making the steel significantly harder. Tests show the steel was rated at 170,000 pounds per square inch (1172MPa), well beyond the minimum acceptable level of 140,000 pounds per square inch (965MPa).

The rods were subsequently dipped in a molten zinc to prevent rusting, however this process on such hard steel is known to increase the likelihood of micro-cracking, potentially allowing hydrogen molecules to enter the metal. Additionally, under stress, these cracks begin to act like a sponge and actually absorb hydrogen at a much faster rate.

The rods and bolts were again delivered to the bridge and installed on top of a concrete pillar and, while under stress, were not fully tightened. This is how they stayed for the next five years during which time the holes that the rods sat within slowly, but surely, filled with rain water.

Experts say this could have activated a 'galvanic couple', causing a release of hydrogen which then found its way into the hard, and more venerable, steel via tiny flaws. When workers finally came to tighten the bolts, hydrogen embrittlement had well and truly taken hold resulting in the high percentage of failures. However, it is still unclear from official reports whether hydrogen got in to the steel during manufacture or during the five years subsequent because of environmental conditions.

The rods are not easily replaced. Since they have been installed the roadway has been constructed above and workers no longer have the headroom to take out the old bolts or put in new ones. A series cables is now likely to replace the broken rods.

US based consultant Louis Raymond is a leading expert on hydrogen embrittlement and has authored the standards for testing anchor bolts, which may have prevented the failure but were never used.

He says: "The standards focus on hydrogen embrittlement degradation from service exposure, not just from manufacturing, which is what gets blamed every time there is this type of bolt failure.

"I hope that we can make the bolt manufactures aware enough to prevent this from occurring in the future."

Manufacturers are often not aware of exactly where products are used or in what environment. This has opened up the potential for hydrogen embrittlement to occur after manufacture and installation if a specification doesn't take in to account environmental conditions.

"It is the responsibility of designers to select the most appropriate plating and fastener combination for a given application," adds Dr Raymond. "Currently, this area is severely neglected throughout the industry and needs immediate attention.

"The environmental effects of galvanic coupling in large bolted joints should be conducted as due diligence to ensure the service life."

Dr Raymond has developed a hydrogen embrittlement accelerated test that has lead to the commercialisation of the Rising Step Load (RSL) test system. RSL is similar to a slow strain rate tensile test, however, the load is incrementally increased and sustained long enough to detect hydrogen induced cracking.

Seeking the advice of fastener manufacturers and expert consultants is highly desirable in identifying the most appropriate requirements for large bolted joints, especially on big projects.
Sharing information with the supplier about applications is often vital so the fasteners supplied are fully fit for purpose.

Justin Cunningham

This material is protected by MA Business copyright
See Terms and Conditions.
One-off usage is permitted but bulk copying is not.
For multiple copies contact the sales team.


Do you have any comments about this article?
It is necessary to coat or plate this high strength bolts used in corrosive environment, however this process definitely introduce hydrogen. So is that nitrocarburization the better way to make the bolts anticorrosive?

Comment Charles Fu, 15/07/2013

Your comments/feedback may be edited prior to publishing. Not all entries will be published.
Please view our Terms and Conditions before leaving a comment.

© MA Business Ltd (a Mark Allen Group Company) 2022