New technologies have become more important than ever for developing efficient and clean energy supply systems. Hydrogen technology is one example that holds high potential both as an accumulator and as a fuel. However, cryogenic liquid hydrogen is typically stored in special steel tanks – a circumstance that presents its own safety challenges: Should the structure of the tank fail in any way, the hydrogen can escape uncontrollably and form inflammable mixtures with other elements like oxygen present in the air. Therefore, material testing is absolutely essential in any quality control process for tanks used to store liquid hydrogen.
Austenitic steel is often used as a base material for producing hydrogen storage tanks. However, the face-centred cubic (FCC) crystalline structure of the alloy is only metastable. Indeed, the manufacturing process itself (cold rolling or forming) can cause the FCC crystals to transform into the body-centred tetragonal (BCT) microstructure of martensite.
Also dropping below the MS (martensite start) temperature can be problematic. When allowed to cool slowly, the austenite transforms into a mixture of ferrite and cementite. But in a rapid cooling process (i.e. quenching, employed to harden the steel), there is no time for the carbon atoms to diffuse out of the crystalline structure in large enough quantities to form ferrite and cementite, resulting in martensite.
Too much martensite is undesirable in steel destined for hydrogen tanks, because hydrogen can settle at the grain boundaries of the martensite (hydrogen embrittlement or cold cracking), which can then lead to material failures. Therefore, testing the steel’s martensite content with precise measurements is required to determine its suitability for this purpose.
An easy-to-use technique is the magnetic induction measurement method. The FERITSCOPE® FMP30, originally designed by FISCHER to measure the ferrite content of steel, has now been further enhanced to additionally measure the martensite content. The switch-over to “martensite testing mode” entails just a few clicks in the software.
The calculation of the martensite content from the probe’s signal is based on the relations published by Talonen et al. (Comparison of different methods for measuring strain induced α-Martensite content in austenitic steels, Materials Science and Technology, Dec. 2004).
In order to assess the potential risk of material failure and avoid subsequent damage, it is quick, easy and cost-effective to determine the α-martensite content in hydrogen tanks using the FERITSCOPE® FMP30. Please contact your FISCHER representative for more information.