Hydrogen inclusions mapped in high-resolution and in 3D
Using a nano-tomography method, material scientists have succeeded in localizing hydrogen inclusions within a structural metal in 3D for the first time. This was made possible by a uniquely modified measurement method developed at ETH Zurich.
Hydrogen is a problem for many metals. If hydrogen atoms are included in a metal, the material properties can be severely affected, causing it to become brittle or cracked. To improve materials, it is thus important to quantify the hydrogen inclusions, which means that these deposits need to be located. However, until now it has not been possible to three-dimensionally locate individual hydrogen atoms in a material sample.
Material scientists at the University of Oxford, together with microscopy experts from ETH Zurich, have now three-dimensionally mapped individual volatile hydrogen atoms in a solid material for the first time, and have presented their findings in the latest issue of the journal Science. They made the measurements in a sample of high-strength steel.
Evaporating atom by atom
The researchers used atom probe tomography for the mapping. In an analysis chamber, a small, needle shaped material sample is placed under high vacuum, and analysed atom by atom: each time, during a short voltage pulse, a few atoms from the sample’s pointed end are evaporated and measured. This then continues until the atoms in the three-dimensional region being investigated have evaporated. The process controllably destroys the relevant sample area. From the trajectories of the evaporated and detected atoms, a computer can then generate a three-dimensional model of the sample’s atomic structure in 3 dimensions.
Atom probe tomography is generally suited to the investigation of alloys, minerals or semiconductors. Using it to locate hydrogen is incredibly difficult, because the hydrogen atoms are mobile within the sample (volatile).
Analysis at very low temperatures
However, atoms of hydrogen belonging to the sample can be located using atom probe tomography if the sample is cooled to a very low temperature. In this state, the hydrogen included in the sample is not volatile. ETH Zurich possesses the only atom probe microscope that is currently able to keep samples constantly below minus 140 degrees Celsius, and can thus be used for such measurements – which is why the Oxford material scientists conducted their measurements in Zurich.
“It is a good challenge to introduce a cold sample into a high vacuum system so that it stays cold and doesn’t condense, which would interfere with the measurements,” explains Stephan Gerstl, senior scientist at ScopeM, the Scientific Center for Optical and Electron Microscopy at ETH Zurich. Together with senior scientist Roger Wepf and other colleagues at ScopeM, he developed a system for transporting and transferring samples, allowing the researchers to keep the sample at minus 140 degrees Celsius even in a vacuum, and cool it to minus 250 degrees Celsius before measurement.
The trick with heavy hydrogen
As even the smallest hydrogen impurities in the measuring chamber could have falsified the measurements, the scientists had to use a particular trick: to establish the procedure, they produced a metal sample containing inclusions from a heavier hydrogen isotope (deuterium), which is extremely rare in nature. This made it possible for them to distinguish between the deuterium in the metal inclusions and the contaminating inherent hydrogen from outside the sample, and thus clearly verify the hydrogen/deuterium in the inclusions.
The now established low-temperature atom probe tomograph could also be of interest for many other materials; for example, soft samples such as rubber or polymers, and even liquids. The ETH scientists have a lot planned for their unique measuring capability.
Reference
Chen YS, Haley D, Gerstl SSA, London AJ, Sweeney F, Wepf RA, Rainforth WM, Bagot PAJ, Moody MP: Direct Observation of Individual Hydrogen Atoms at Trapping Sites in a Ferritic Steel. Science, 17 March 2017, doi: external page 10.1126/science.aal2418