24th February 2023 | Author: Pat Trimby
Recently I was reading a fascinating book – “Bounce: The Myth of Talent and the Power of Practice” by the English journalist and author, Matthew Syed. Now Syed was, prior to his career as a wordsmith, a top-level international table tennis player. He grew up in the English town of Reading, a place not particularly well-known for producing international level sports stars yet, remarkably, in the 1980s, the single street on which Syed lived produced “more outstanding table tennis players than the rest of the nation combined”.
The cause of this bizarre statistic? Practice. Or, more specifically, a dedicated teacher with a passion for the sport, unlimited access to a single table tennis table in a local club that was open 24 hours a day and a group of teenagers who loved playing against each other. And therein lies the essence of the book – people who reach the top level in what they do, whether that is sport, music, business or even chess, do so because of experience rather than latent talent. In fact, to reach the very top, Syed provides evidence that you need about 10,000 hours of high-quality practice.
This is all very interesting (and I do recommend the book), but what has this to do with microscopy and, more specifically, electron backscatter diffraction (EBSD)? I am just coming up to my 30th year working with electron microscopes and, during that time, I have encountered and had the privilege of working alongside many extremely “talented” microscopists. I am sure you know the type of person I mean – those who, when you watch them working the controls of their particular instrument, are seemingly brilliant at what they do. I’ve had this feeling watching TEM experts generate atomic resolution images of Al alloys, FIB-SEM specialists making in-situ lift-out look easy, electron microprobe operators delivering consistently reliable quantitative analyses or AFM experts working magic with the instrument controls to deliver astonishing data, seemingly out of nothing. However, when I think about all of these talented scientists, the common thread is that they have all had a lot of experience, working in their fields for years, honing their skills almost on a daily basis. We think of them as “talented”, but perhaps the “Talent Myth” is just that?
Now, this revelation is a problem for a company like Oxford Instruments and, in particular, for a product manager like me. We want to sell our equipment to more and more scientists, and we want to broaden the applications of techniques like EBSD, so that they can be used in diverse areas of science such as dentistry, battery technology, nanoscience, semiconductors and so on. But EBSD, as with many of these high-end techniques, is not a simple technique, and it can take many 100s of hours of practice to develop the skills to be able to analyse, effectively, more challenging sample types. Unfortunately, time is at a premium in the modern age; researchers and technicians are expected to generate quality results from their new instruments almost immediately, regardless of whether they have 10 or 1000 hours of experience with the technique in question.
How can we accommodate these widely differing levels of expertise in our end customers? Well, that has been a focus of considerable development here at Oxford Instruments, and we are now starting to make some real progress in this field. In the latest release of AZtec, version 6.1, we are introducing a new workflow in our EBSD software that we call “Optimize Experiment”. The idea behind this is to develop a system that helps our users to find the best settings for their sample and their desired analysis. This will include detector settings, indexing parameters, phase settings and so on. In addition, more informative timings data will help users estimate, with a degree of certainty, how long their analyses will take – a particularly important metric in any multi-user laboratory.
In the annotated video shown below, you will see Optimize Experiment in action: the principle is quite simple – at the start you decide the “experimental priority” (i.e. are you looking for high speed, high precision, some all-round balance etc.?) and define an approximate grain size - then the system will prompt you to collect a short test map. If you subsequently tweak any of the settings, the test map will be reprocessed automatically (the EBSD patterns are stored in memory) and at the end of each map, you will see a detailed summary of the data quality. This includes some new data quality metrics that will help users to evaluate how well the system is separating phases, how reliable the orientation measurements are, whether the angular precision of the data is improving or worsening and so on. In this way, it is easy to evaluate, statistically, whether the changes you make to any of the settings are benefiting your analyses, so that you can be confident of making the best use of your time at the microscope.
In this example, the user is setting up the system to characterise a heat-treated duplex stainless steel that includes 2 intermetallic phases (Sigma and Chi), so the emphasis is on good quality data (including phase differentiation) rather than the speed of the analysis. If you watch the video through to the end, you will see how the final data collection had an acquisition speed of 226.0 Hz with an indexing rate of 99.1%. The prediction from the test map analysis was 225.1 Hz and 99.2% - almost spot on!
Of course, this will not yet choose all of the settings for you – a degree of knowledge about your sample, as well as an understanding of what you might need to do to improve the data quality or speed of analysis, will still be required. And for those “experts” who still want the workflow that they are used to, that will still be available. However, we want to continue to develop in this direction, improving the intelligence of our EBSD software and making EBSD an easier and more accessible technique.
In the future, we don’t want our users to require 10,000 hours of practice to become masters of microstructural characterisation, or even 1,000 hours. But perhaps we can settle on 10?
Dr Pat Trimby,
EBSD Product Manager, Oxford Instruments
Dr. Pat Trimby graduated with a degree in Geology from Oxford University and a PhD from Liverpool University. Following more than 20 years working in both commercial and academic electron microscopy, Pat joined Oxford Instruments in 2017. He has always maintained a strong focus on SEM-based microanalysis, with a particular interest in developing new electron backscatter diffraction (EBSD) techniques and applications. He is currently working as the EBSD product manager within the Nanoanalysis marketing team.
We send out monthly newsletters keeping you up to date with our latest developments such as webinars, new application notes and product updates.
© Oxford Instruments 2025