Q&A: Direct observation and impact of co-segregated atoms in magnesium having multiple alloying elements

Prof. Jian-Feng Nie at the Department of Materials Science and Engineering, Monash University, is the corresponding author on a Nature Communications publication on atomic imaging in metals. Kristina Kareh, the editor who handled the paper, asks him some behind-the-scenes questions about this work.

Like Comment
Read the paper

1. What are the key findings of your paper?

There are three key findings, each with a different significance.

First, we report the direct observation and identification of atomic elements that are segregated at the atomic scale in a metal – this is the first time we see segregated atoms using X-ray mapping in metallic materials.

The second novelty is that with this new imaging, we detect a new pattern of solute segregation in twin boundaries in a magnesium alloy. This is important because once segregation patterns are recognised, we can come up with a plan to manipulate them via alloy composition and thermomechanical processes in order to tailor the mechanical response (for example, to make twinning either easier or more difficult).

Third, we report a new boundary migration mechanism in magnesium alloys. It is usually said that for twin boundaries to migrate, there needs to be the formation of specific defects (disconnections) with a minimum height of two atomic layers. However, we find that once there is segregation at the boundary, depending on the segregation element(s), twin boundaries can migrate via a single atomic layer.  

For us, the most important finding is the first one, which is the X-ray mapping for atomic element resolution. This can be applied to different kind of boundaries in different materials. Some functional materials that would be of interest would be silicon, germanium, and other semiconductors, because their function is tied to impurity elements. In the past, these were investigated using Z-contrast imaging, but this new X-ray atomic mapping method should work. Specifically, what we have learned is that for any type of segregation, operation of the microscope at a 300 keV voltage i.e. normal operation voltage, is too high for X-ray mapping! 

2. What was your role in this work?

I designed the entire experiment. In the old days, I used phase contrast to look at atomic features before I realised it wasn’t very good. I then moved to Z-contrast STEM to look at the structure of precipitates and defects in metals and alloys. But 6 or 7 years ago, I also realised Z-contrast imaging was not very good. This is what led me to want to figure out how to do X-ray mapping. But in order to do that, you have to figure out what is the inherent problem with the imaging. I therefore had to make sure that we identified the problem before tackling it. In terms of materials, I picked magnesium and magnesium alloys because this is my main research effort. Today, we have also quickly expanded this method to aluminium alloys.

Essentially, I designed the experiment and did a lot of analysis. We also spent a lot of time at night trying to minimise drifting of the sample!

3. Everyone has a story: how did you come to do this particular piece of research?

Twenty years ago, it was very difficult to deduce whether there was phase boundary segregation, as we only had phase contrast imaging. In 2005, variations in Z-contrast enabled us to see where the atoms were, but only in special cases. In addition, X-ray mapping was already possible for ceramics, but the ceramics are thermally stable and have a high melting temperature (the golden standard is strontium titanate). But I am a metallurgist, and metals are not so! And light metals like magnesium and aluminium are even worse.

Initially, I thought adapting this X-ray mapping method to metals would be easy before I realised it definitely was not going to be. Atomic resolution element mapping like what we did requires the right facilities, the right sort of imaging, and the right sort of operating system to obtain the right images. At Monash, we have a double aberration-corrected Titan but it does not have this X-ray capability. This is why this paper is the result of a collaboration with Chongqing University in China, as only a small fraction of current microscopes can do this. Usually, microscopes are operated at voltages of 200 or 300 keV, but in this case, when we tried to visualise the atoms, the radiation damage was too severe and the cold stage setup didn’t work. We finally lowered the voltage to 120 keV and used trial and error to figure out if we could make the imaging work, which was challenging because the resolution of the instrument is proportional to the electron voltage.   

4. What was the most critical moment during the study? Any anecdotes you’d like to share with us about challenges with obtaining these results and preparing the manuscript?

We spent several years working out the successful conditions for X-ray mapping, and at one stage we wanted to give up! We tried all sorts of different things before we finally tried a lower voltage. But it must be said that mapping a single layer of atomic resolution is very difficult: the pictures in the manuscript are the best we could select. As an example of a challenge, sample drift led to a lot of failures. The completion of this work is therefore the result of patience. Without patience this wouldn’t have happened.

One thing I find funny is that the project this manuscript came out of is not actually an imaging project but a project on magnesium alloys.


5. What do you hope will be the impact of your research?

This is very difficult to answer… My own view is that down the line, there will be more and more results about X-ray imaging at atomic resolution because once people know how to use this mapping technique, they can apply it to their own problems. In terms of material insights, the new segregation pattern and the new boundary mechanism we report may hopefully be useful for modellers. There are therefore two different fronts in terms of impact! I hope that this work will encourage more publications on segregation and precipitation, as we can now visualise a single layer of added alloying elements, which couldn’t be seen before. I also feel that another way forward would be quantitative X-ray mapping which is equivalent to a combination of TEM and atom probe tomography.

6. Which is the development that you would really like to see in the next 10 years?

In terms of the imaging, I can imagine an evolution from traditional chemical information-based atomic imaging (EDX and EELS) to the point where a single image or map combines both structural and chemical information. This will take at least 5 years if not more, but I can see it happening: some of the modern microscopes have windows that are getting larger and larger, which makes X-ray mapping easier.

In terms of materials, solute segregation is a potential ‘hot topic’. In the past, we couldn’t quantify it easily and couldn’t do very much. But now that we can see where we are at the atomic level, we can start thinking about things that we are doing. For example, we can look at pinning of the boundaries, solutes around precipitates, we can start to understand heterogeneous nucleation or precipitate stability via interfacial energy change… Once you can see where solutes are, you can think about what sort of function they play. This is true not only for how solutes affect mechanical behaviour, but also what parts they play in phase transformations.

7. What would your dream conference be like?

Nowadays, I really appreciate smaller-scale workshops having well-defined topics or themes. Ideally, 20-30 people who are true experts in their fields. This allows people to spend time chatting and talking, allows for crossovers between the experimental and the computational level, and results in more in-depth discussions. This is now very attractive for me. And of course the conference should be in a nice place! Somewhere where it is quiet and people can be relaxed.

8. What is your favourite material?

That is obviously magnesium! I have spent 25 years working on it, and I still think there is a long way to go. It is a beautiful metal: it is light and with many unique features, and we still do not understand it well. We need a much deeper understanding of it at the fundamental level. In a way, it is like a growing baby compared to steel and nickel-based superalloys.

9. If you weren’t a material scientist, what would you like to be (and why)?

I must say I would like to be a materials engineer. Science is all well and good, but ideas have to be turned into a real technology and a real product. I would spend more time with people working on the industrial side to make sure we can fabricate more products out of magnesium for people to use, for example, biomedical implants. The aim is really to enable people’s lives to become more and more comfortable. It must also be said that engineering is very different from pure research, and it has its own sets of challenges.

10. Where do you go from here?

We will now spend more time and make more effort on X-ray imaging at atomic resolution, including atomic resolutions EELS. We would like to make sure that, using this new technique, we can dig out as much information as possible out of the materials we deal with. I would also like to extend this technique to other materials (other metals and alloys, ceramics, semiconductors….).

Header image by Daniel Grayson from Pixabay

Kristina Maria Kareh

Senior Editor, Springer Nature

Kristina joined Nature Reviews Methods Primers in August 2020. Prior to that, she was an editor at Nature Communications from March 2017, where she handled manuscripts spanning all areas of metallurgy and structural material science. She completed her MEng and PhD at Imperial College London, where she investigated real-time deformation of semi-solid aluminium-copper alloys using synchrotron X-ray tomography. She continued on at Imperial College as a postdoctoral scientist imaging semi-solid steels and solid oxide fuel cells. At Nature Reviews Methods Primers, Kristina handles primers covering methods across the physical and the social sciences. Kristina is based in the London office.

No comments yet.