Single crystal magnesium oxide is commonly used as a substrate for the deposition of a wide range of thin film materials. But, researchers at Linköping University, Sweden, have tested samples of commercially available MgO and found the crystal quality varies so widely as to potentially cast doubt on the validity of countless experiments using the poorer quality substrates.
In work to be published in the June issue of the Journal of Crystal Growth, Jeremy Schroeder and colleagues explain the importance of substrate selection for thin film deposition, the substrate lattice acting as a template for growth, after all. [Schroeder et al, 2015, J Cryst Growth; DOI: http://dx.doi.org/10.1016/j.jcrysgro.2015.03.010] They point out that single crystal substrates are the ideal choice because they lack defects that might interfere with the crystal quality of the thin film. Moreover, matching the substrate lattice to the proposed thin film structure can also avoid mismatches and generate the optimal product.
Magnesium oxide produces minimal mismatches in growing films of titanium nitride (TiN) , for instance, as both have the rock salt crystal structure. Similarly, other transition metal nitrides - HfN, ZrN, and ScN - benefit from the support of MgO. As such, researchers buy commercially available MgO substrates anticipating that they will have perfect single crystals for their experiments with superconducting and magnetic thin films, their epitaxial metallic superlattices, piezoelectrics, ferroelectrics, thermoelectrics, hard coatings, and more.
Unfortunately, Schroeder and colleagues are the bearers of bad news. They obtained thirty MgO substrates from six different vendors and tested them for their claimed single crystal characteristics. They were disappointed to find that many of the samples contained multiple crystal domains, rather than just one. Fundamentally, a poor quality substrate means a poor quality thin film, as demonstrated by one to one correlation with the team's titanium nitride films. Indeed, more than 70 percent of the MgO substrates were demonstrated to be of less than the claimed single crystal quality.
"We think that an important aspect of our study is the inconsistent quality of MgO substrates," Schroeder told Materials Today. "We can purchase substrates from an individual vendor one month and get excellent quality substrates and then the next month we can purchase substrates from the same vendor and get poor quality substrates."
The researchers suggest that there is an urgent need for researchers to do in-house quality control on these substrates to ensure consistency in their experiments. Moreover, it is also time that vendors worked more closely with the scientific community to raise standards in this area. The team demonstrates how vendors can carry out quality control by acquiring X-ray diffraction omega–phi maps in batch processes, as can the researchers themselves. They also propose a simple quality grading scheme to allow MgO substrates of varying quality to be more readily differentiated.
"We have contacted all the single-crystal MgO substrate vendors of which we are aware, which totals eleven vendors," Schroeder told us. "We have already received responses from three vendors. We are actively seeking to work together with vendors and MgO crystal growers to resolve the issue of poor quality MgO substrates," he adds.
David Bradley blogs at Sciencebase Science Blog and tweets @sciencebase, he is author of the popular science book "Deceived Wisdom".