Hydrogen embrittlement

Hydrogen attacks the iron at elevated temperatures. This phenomenon is clear while welding of steel where the moisture (located at surface of steel or from filler metal or gas used in welding, .. etc.) dissolved into O₂ and H₂. Most of hydrogen inside the weld bead escapes through the weld to air. The adjacent portion to weld or what we called the HAZ is transformed to austentitic when heated by welding, hydrogen is soluble in this region. At HAZ the hydrogen remains as it is, and upon cooling the hydrogen builds up pressure that can cause underbead cracking or what we called hydrogen embrittlement. The same phenomenon occurs for iron at elevated temperatures.

In welding of steel, we can minimize underbead cracking by: preheating surface of steels, using low hydrogen electrodes, and allow for a low cooling rate for weld bead to give a chance for hydrogen to escape.

In addition to slow cooling it is desirable to perform an Hydrogen bake-out usually at 600 degrees F for two hours or so. Some steels (P91) require an hydrogen bakeout and a PWHT (stress relieve) within a few days.

Any hydrogen that remains after the hydrogen bakeout, is dissipated during PWHT.

I don't believe that exposure to steam will elevate the hydrogen in the metal, weld or in the heat affected zone. There is a large variety of steels used in the steam powered generating plants that use a large range of temperatures and pressures. The creep failures that occur, are due to improper welding technique including preheat and PWHT.

Hi Guest,

It is well established that H (nascent hydrogen) when trapped / dissolved in molten steel (or steel at high temperatures for that matter; but exposed for longer in this case) would unite to form H₂ within 24 hrs, assume enormous volume and exert pressure on its container walls proportionately. In 48 hrs it begins to manifest itself definitely by underbead cracking in the case of welds (HAZ is most affected) or transgranular cracking in the other cases. The whole of this reaction completes in about 3 weeks when all stresses are released fully via cracking, if left unattended. This I learned during my training in India for AWS's CWI.

Hope it answers your question.

Regards,

Nachi

You don't need high temperatures to cause hydrogen embrittlement. When I worked at a plating shop, we zinc electroplated seat belt bolts. If we didn't properly bake them after plating, the heads would pop off the bolt about 2-6 hr after tightening.

The bake cycle, IIRC, was about 30 min at 160 F.