What Causes Hydrogen Embrittlement in Steel?

A high-strength fastener can meet its specified chemistry, hardness, and tensile strength, then fracture hours or weeks after installation under a load well below its expected capacity. In many cases, the initiating question is simple: what causes hydrogen embrittlement? The engineering answer is less simple. Failure requires an unfavorable combination of available hydrogen, a susceptible material, tensile stress, and a microstructure or environmental condition that allows cracking to develop.

Hydrogen embrittlement is not a single defect or one easily identified failure mode. It is a family of hydrogen-assisted damage mechanisms that can reduce ductility, fracture resistance, and load-bearing reliability. It is particularly consequential in high-strength steels, welded components, pressure-containing equipment, pipelines, fasteners, springs, and other safety-critical assets where delayed cracking can be difficult to detect before failure.

What Causes Hydrogen Embrittlement?

Hydrogen embrittlement occurs when atomic hydrogen enters a metal, migrates through its microstructure, and concentrates at locations where the material is already vulnerable to fracture. These locations can include grain boundaries, inclusions, dislocations, interfaces between phases, weld heat-affected zones, and highly stressed crack tips.

Molecular hydrogen gas does not readily diffuse into steel in the same way. The concern is atomic hydrogen, generated through chemical reactions, corrosion processes, welding, electrochemical treatment, or exposure to hydrogen-containing service environments. Once absorbed, hydrogen can move through the lattice and become trapped at microstructural features. Under sufficient tensile stress, its local concentration can lower resistance to crack initiation or crack growth.

The result may be a brittle-looking fracture in a material that would normally exhibit useful ductility. Delayed failure is a defining operational risk: the component may pass inspection and proof loading, only to crack after hydrogen redistributes under sustained service stress.

Hydrogen introduction during manufacturing

Many embrittlement events begin before a component reaches service. Acid pickling and descaling can introduce hydrogen when acids react with the steel surface. Electroplating, galvanizing, phosphating, electropolishing, cathodic cleaning, and some chemical stripping operations can create the same condition if process controls are inadequate.

Welding is another significant source. Moisture in electrodes, fluxes, shielding gases, or surface contamination can generate diffusible hydrogen in weld metal and adjacent heat-affected zones. The risk rises where welding procedures, consumable storage, preheat, interpass temperature, heat input, or post-weld hydrogen removal are not controlled for the material and restraint condition.

Manufacturing-related hydrogen does not automatically cause failure. A low-strength, lightly loaded component may tolerate it. However, higher-strength steels can become increasingly vulnerable as hardness and tensile strength increase, particularly when residual stress from forming, machining, grinding, threading, or heat treatment is present.

Hydrogen generated in service

Hydrogen can also enter steel during service. General corrosion, pitting corrosion, galvanic coupling, sour environments containing hydrogen sulfide, acid exposure, and cathodic protection can all produce atomic hydrogen at a metal surface.

Cathodic protection illustrates why conditions must be assessed rather than assumed. Properly designed cathodic protection is an essential corrosion-control measure for many buried, submerged, and marine structures. If protection potentials are excessively negative, however, hydrogen evolution may increase. Whether that creates embrittlement risk depends on the alloy, strength level, temperature, coating condition, stress state, exposure duration, and service environment.

Hydrogen service itself demands separate consideration. Components used in hydrogen production, storage, compression, transport, or processing may experience gaseous hydrogen embrittlement, hydrogen-assisted fatigue, or related mechanisms. Pressure cycling, permeation rate, gas purity, temperature, and localized stress concentrations can materially change performance.

The Conditions That Turn Hydrogen Into a Failure Risk

Hydrogen availability alone is not enough. A practical assessment considers four interacting factors: the hydrogen source, material susceptibility, tensile stress, and time or temperature conditions that support diffusion and crack development.

Material strength is often the first screening criterion. High-strength quenched and tempered steels, hardened fasteners, high-hardness weld heat-affected zones, and certain martensitic microstructures are generally more susceptible than low-strength ferritic steels. This does not mean low-strength steel is immune, nor that every high-strength steel will fail. Heat treatment condition, alloy composition, cleanliness, grain size, inclusion content, and local hardness all influence behavior.

Tensile stress is equally critical. Applied service load, assembly preload, residual stress, thermal stress, weld shrinkage stress, and stress concentration at threads, sharp radii, pits, weld toes, or machining marks can supply the driving force for cracking. Compressive surface stress may be beneficial in some cases, but it does not eliminate the need to control hydrogen entry and material condition.

Time matters because hydrogen diffusion and trapping are not instantaneous. A component may be most vulnerable after plating, pickling, welding, or exposure to a corrosive environment, when hydrogen has time to move from the surface toward highly stressed regions. Elevated temperature can accelerate diffusion, while hydrogen traps can delay release and complicate bake-out or dehydrogenation decisions.

Common Locations for Hydrogen-Assisted Cracking

Failure investigations frequently identify cracking at sites where stress and metallurgical susceptibility overlap. Thread roots are a classic example because they concentrate stress and may retain plating-related hydrogen. Other common locations include undercuts at weld toes, hard heat-affected zones, cold-formed bends, ground surfaces, corrosion pits, notches, and transitions between thick and thin sections.

In welded fabrication, hydrogen-assisted cold cracking may appear in the weld metal or heat-affected zone after the weld has cooled. Thick, highly restrained joints are more vulnerable because they combine high residual stress with slower opportunities for stress relaxation. For this reason, a welding procedure must address hydrogen control alongside strength, toughness, hardness, heat input, and joint restraint.

Fracture surfaces can provide important evidence, but appearance alone is rarely conclusive. A brittle or intergranular region, secondary cracking, and limited plastic deformation may support a hydrogen-assisted mechanism. They must still be evaluated against loading history, material properties, chemistry, hardness, microstructure, environmental exposure, and the timeline of failure.

How to Investigate Suspected Hydrogen Embrittlement

A defensible investigation starts with preserving evidence. Avoid destructive removal methods that erase fracture features, alter crack surfaces, or contaminate deposits. Record the component orientation, installation condition, load history, environmental exposure, manufacturing route, coatings, heat treatment, welding documentation, and the timing between processing, installation, and failure.

Laboratory work typically combines visual examination, dimensional assessment, fracture examination, metallography, hardness mapping, chemical verification, and scanning electron microscopy with energy-dispersive spectroscopy where appropriate. These methods help establish whether the failure initiated at a pit, inclusion, machining defect, weld feature, or other local discontinuity.

Hydrogen is difficult to measure and interpret after a failure because it can diffuse out of the component during handling and storage. Direct hydrogen analysis can be valuable when sampling is prompt and the test plan is tailored to the alloy and suspected mechanism. Even then, measured hydrogen content should not be treated as a standalone verdict. The relevant issue is often local diffusible hydrogen at the crack initiation site, not only bulk hydrogen concentration.

Mechanical testing may be needed to determine susceptibility under sustained load or slow strain-rate conditions. Depending on the component and governing requirements, this can include proof-load testing, notch tensile testing, fracture toughness assessment, or hydrogen embrittlement evaluations aligned with applicable industry methods. Testing should replicate the credible material condition, coating system, stress level, and environment as closely as practicable.

Reducing Risk Through Process and Design Control

The most effective controls are applied before hydrogen-assisted cracking becomes an investigation. Material selection should account for strength level, hardness limits, service medium, anticipated cathodic protection conditions, and fabrication route. Where high strength is necessary, designers should minimize sharp stress raisers and specify suitable surface preparation, coating, heat-treatment, and inspection requirements.

For plated or chemically processed steel, control begins with cleaning chemistry, inhibitor use, exposure time, current density, bath condition, and rinse practices. Post-process baking may reduce diffusible hydrogen, but its effectiveness depends on prompt commencement, temperature, duration, section geometry, coating requirements, and the material’s trapping behavior. Baking is not a substitute for limiting hydrogen introduction in the first place.

For welding, low-hydrogen consumables, controlled storage, clean joint preparation, appropriate preheat, qualified parameters, and restraint management work together. Hardness surveys of the heat-affected zone can help verify that cooling rates and material response have not created an unnecessarily susceptible condition.

In service, manage corrosion and cathodic protection with monitoring rather than fixed assumptions. Review protection potentials, coating damage, water chemistry, process upsets, and load cycles, particularly where high-strength components operate in wet, sour, marine, or hydrogen-containing environments.

AECTL can support this work through accredited materials testing, corrosion assessment, weld and coating inspection, failure analysis, and targeted engineering consultancy. The objective is not merely to identify a crack after the fact, but to establish a technically defensible cause and practical controls for preventing recurrence.

When delayed cracking affects a critical asset, the strongest next step is a focused evidence-based review of the component, its processing history, and its actual service conditions. That approach turns a suspected hydrogen issue into an actionable decision on repair, replacement, qualification, or process change.