gallery of project results (2019)

Compared to previous years, the 2019 edition of the project for this course was a little different. The project was not a mandatory part of the course any longer for for-credit students. As a result, only 1 team of volunteering students took up the challenge this year. This brave effort got stuck in some technical issues, but did end up nevertheless in a report that can be read here.

The peer feedback by other students on this project paper, is available here.

For whoever reads this page out of sync with the Fall 2019 delivery of this course, it might be useful to know the task that was given to the project team (copied underneath).

This was the task originally given:

Crystalline nickel has the fcc crystal structure. However, when grown as a thin film on a sapphire substrate, a fcc nickel film that is not too thick spontaneously develops an even thinner hcp nickel film on top of the fcc film. In this paper, researchers report on DFT calculations that reveal the mechanism behind this peculiar effect (well, to be honest, they don’t observe it for pure nickel, only for an alloy of 80% nickel and 20% iron). This is how they summarize their observation and explanation in the abstract of their paper: “Both the high-resolution X-ray diffraction and the cross-sectional high-resolution transmission electron microscopy observations revealed the phase separation of the Ni80Fe20 films into two parallel layers of the face-centered cubic (adjacent to the substrate) and hexagonal close-packed (on the top of the film) phases of similar compositions. Our density functional theory (DFT) calculations indicated that this phase separation is driven by the decrease of the film surface and interface energy, leading to the thermodynamically equilibrium thickness of the metastable hexagonal close-packed phase.”

Your task for this project is:

• Reproduce these reported nickel calculations with your preferred DFT code.
• Repeat this procedure for cobalt. What do you expect based on your calculations?
• Devise a rapid procedure that could identify other fcc or hcp materials that are likely to develop a surface layer of the other phase. Focus on ‘rapid’: which minimal information would you need to calculate in order to make it likely that a particular element would display this effect? If you have the time, use this procedure to hunt for other candidates.

Some more guidance for the first of these three steps:

• Read sections 2.3 and 4.4 of the paper. Fig. 10 of the paper and Fig. S7 of the supplementary information are useful as well. Try to understand all of this.
• Calculate the bulk energies of fcc Ni and hcp Ni, as a function of lattice parameter. For a common volume per atom (chosen as dictated by the sapphire substrate that will be introduced later), which phase has the lowest energy? (consider magnetism)
• Calculate the (111)-surface energies for fcc Ni and hcp Ni, at the same volume per atom. Converge with respect to numerical parameters, slab thickness and vacuum thickness.
• Calculate the interface energy of the fcc/hcp Ni interface, at the same volume per atom. Converge with respect to slab thickness.
• Calculate substrate + fcc Ni and substrate + fcc+hcp Ni, and judge on stability (see fig. S7). Compare with the inequality obtained from the data you calculated in previous steps. Estimate the maximal hcp-thickness.