The recent column on hydrogen embrittlement (June) prompted readers to send in a number of finishing-related questions. Here are some of the inquiries about that process, and several others:

Hydrogen Embrittlement

Can I eliminate the potential for hydrogen embrittlement if I use an electroplating process that has 100 percent cathode efficiency?

Hydrogen embrittlement becomes a possibility any time that high-strength steels are exposed to hydrogen. Using a 100 percent-efficient plating process does not necessarily eliminate all contact with hydrogen. Cathodic electrocleaning, stripping in an acidic solution, and acid pickling also subject parts to hydrogen contact. Furthermore, plating processes that are 100 percent cathode efficient are sometimes less than 100 percent efficient at high current densities, such as points and sharp edges. Lastly, parts that are selectively plated by shielding or masking may be exposed to hydrogen by contact with acidic plating solutions.

At what strength level can I eliminate the baking required to relieve hydrogen embrittlement?

There is no definite cut-off. Steels with tensile strength levels as low as 140,000 psi have been reported to be embrittled by hydrogen. A "rule of thumb" is that steels below Rockwell "C" 35 are far less susceptible, but if the steel has been subjected to "charging"óas would be the case if it were electrocleaned cathodically, or if it were cathodically treated in an acidic solutionóthere is no lower strength limit.

How long should I bake plated high strength steel parts to eliminate the chance of hydrogen embrittlement?

Very thick coatings (0.0005" or more) of dense deposits may require double bakingó once after about 0.0002" of plating is applied, and a second bake after the balance of the specified thickness is added.

The possibility of hydrogen embrittlement can never be eliminated. Baking at 375 +25,-0 F has been shown to be an effective method of reducing the tendency toward hydrogen embrittlement. If the part can withstand higher temperatures, baking at 400-450 F will drive the hydrogen out faster. The efficiency of baking is determined by a number of factors, including how long the part waited to be baked after processing; the thickness of the plating; its porosity level; the efficiency of the plating process; the duration of the baking; and other factors, including part geometry and the ultimate tensile strength of the steel.

My specification states that baking for embrittlement relief is to occur within 15 minutes of processing. Do you think this is unreasonable?

The higher the strength of the steel, the faster the baking must be done after processing. Many newer specifications require baking of steel with Rockwell "C" values above 50 to be baked within 15 minutes of being processed. There have been reports of spring steels self-destructing while waiting to be baked. Compliance with the specified maximum delay cannot be overemphasized. In some cases, a part cannot be removed from the plating tank, rinsed, dried, de-masked and baked within on hour. In such cases, the delay should be as short as physically feasible.

Why is the length of the delay before baking so critical?

Once the embrittlement mechanism has been completed, it is essentially impossible to "cure" the embrittlement by baking. If the parts are held prior to baking for so long a time that the embrittlement mechanism has had a chance to be completed, the part is usually lost. Baking drives out the absorbed hydrogen before the mechanism has a chance to proceed to completion. In effect, the finisher is in a race against time.

Must I specify whether or not my fasteners must be baked, or will the finisher perform this operation "automatically"?

Baking for hydrogen embrittlement relief must be specified, if the finisher is to be held responsible for embrittlement failure. Many plating companies are well informed on the subject, and will inquire as to whether or not they are to bake a part, if the customer does not specify baking. But dont count on it. You should take the precaution of specifying the temperature, duration of bake and the maximum delay before baking.

Is there some method that will verify that a fractured part failed because of hydrogen embrittlement?

There is no 100-percent-certain method that will verify that the cause of a fracture was hydrogen embrittlement. Brittle fractures, caused by hydrogen, have similar surface structures to other brittle fractures, including evidence of cleavage and microvoid coalescence. The investigator will review the evidence, including processing methods, baking steel structure and other factors, to determine the "most likely cause" of the fracture. If the part was Rockwell "C" 45 and was electroplated and not baked, the most likely cause (in the absence of other causes) would be hydrogen embrittlement.

We analyzed a fractured part and it contained 9ppm hydrogen. Isnt that evidence that the metal finisher did not adequately bake the part?

Baking does not totally remove all hydrogen from the steel. The mechanism by which baking reduces hydrogen embrittlement is not fully understood. No concentration levels of hydrogen that are either safe or harmful have been reported. There are too many other variables to be considered.

By what method can a zinc coating be applied to fasteners, to avoid the entire hydrogen embrittlement problem?

Mechanically applied zinc coatings are generally accepted as free of hydrogen embrittlement, as long as the parts do not require acidic pickling or other contact with hydrogen prior to application of the mechanically applied coating. Such coatings are difficult to apply to large parts, or parts that cannot tolerate tumbling against each other. Maintaining dimensional tolerance is also difficult with such a process.