In this second part, we conclude the discussion of hydrogen embrittlement.

If the part is electroplated, there is a greater chance that the plating itself will form a barrier that prevents the hydrogen from escaping. That is why most critical hardened-steel components must be baked if they are electroplated and will be subject to sustained tensile stresses.

Threaded fasteners and spring clips are two types of components that are generally used at high sustained-stress levels in the hardened and plated condition. As previously stated, in general, the baking process does not remove the hydrogen from the part. You don’t “bake it out.” There are a lot of websites that state that the post-plate bake is a “bake out.” However, it is very difficult for hydrogen to diffuse through a zinc layer. The openings through which the hydrogen atom must squeeze itself to get through the zinc layer are much smaller than the openings within iron crystals. So, again, most of the hydrogen is not baked out.

What is thought to happen in baking is that many of the hydrogen atoms are given just a bit more thermal energy that lets them jump around more. By random motion, many of the hydrogen atoms then end up in what are called “deep thermodynamic traps.” This means that the hydrogen ends up preferentially sitting at dislocation tangles, inclusions and second-phase particle boundaries rather than in the grain boundaries. The atoms are very comfortable in these places and not prone to leave them once there. Then, even in the presence of a sustained tensile stress, the hydrogen atoms can’t cause a problem by “greasing” the grain boundaries.

Some of the required baking programs are quite long. These may well be necessary, especially if low temper temperatures have been used to maintain a hardness close to the as-quenched value. However, even long bake cycles may not cause hydrogen atoms that have made themselves comfortable in grain boundaries to go where we want them. Again, the basic good practice of proper chemical maintenance in cleaning and plating operations to minimize the amount of hydrogen that gets into the parts in the first place, and strict adherence to proper baking procedures, will minimize hydrogen-embrittlement problems.

Before anyone “blames” hydrogen for causing an embrittlement condition (this usually means that the plater will be held accountable), a metallographic cross section should be prepared to examine the microstructure. Just because the part is hardened steel and plated and the crack is intergranular, it doesn’t mean that the hydrogen is the cause. There are other causes of intergranular cracks that should be ruled out before the hydrogen is blamed.

Quench cracks are often intergranular, for example. Also, the history of the crack must be studied. In general, a crack precipitated by an impact load will not be hydrogen related. Hydrogen embrittlement takes time. Usually, at least several hours are needed for the hydrogen to diffuse to the grain boundaries where they can cause a problem. The steel might also have a temper embrittlement or other problem that is intergranular but not due to hydrogen. Furthermore, some hydrogen-embrittled parts have DUCTILE features visible on the fracture surface when viewed in a scanning electron microscope.

The situation is complicated enough to get interesting!