In this interview, we had the pleasure of speaking with Dr. Burak Ozdol, a TEM expert at Western Digital. With his background in semiconductor material and knowledge of TEM sample preparation, Dr. Ozdol shared his insights on FIB preparation for in situ microscopy, and electrical device studies.
You have done a lot of research in the past, on a variety of topics including semiconductors, magnetic nanoparticles and TEM sample preparation. Could you describe your background and how you got into your current field of research?
I began my studies and career in Turkey, earning a Bachelor’s degree in Metallurgy and Materials Engineering from METU in Ankara. From there, I moved to Germany to pursue a Master’s at Kiel University, where I encountered TEM for the first time while studying functional materials. The process of characterizing these materials sparked my deep interest in TEM, leading me to pursue a PhD at the Stuttgart Center for Electron Microscopy within Max Planck Institute. There, I explored all aspects of TEM, including advanced analytical techniques.
At MPI, my research focused on the characterization and methodology development for strain-engineered semiconductor devices. A key challenge was overcoming sample preparation difficulties, which I found particularly exciting. What made this work especially meaningful was the opportunity to analyze real industrial samples—transistors and LEDs straight from the shelf—allowing me to bridge fundamental research with real-world applications.
After my PhD, I moved to Berkeley, California, to join the National Center for Electron Microscopy (NCEM) as a PostDoc. Here, I expanded my TEM expertise into in situ techniques, focusing on nano-deformation studies. I conducted heating, biasing, and strain testing on semiconductor samples, which was fascinating because we could observe devices evolving in real time under operating conditions. My work included studying InAs/In0.6Ga0.4 core-shell nanowires for semiconductor research as well as deformation dynamics and structural changes in metallic glasses.
My passion for working with real materials in real applications ultimately led me to industry. Today, I’m leading the TEM FIB group at Western Digital in San José, California, where I contribute to world-class analytical science in a fast-paced, cutting-edge environment.
Image from: Nano Lett. 2018, 18, 8, 4949–4956
You have used TEM research both in academia and in industry; do you feel your approach to experimental design for TEM is different in the two instances?
The approach to TEM research in industry is definitely different from academia, mainly because of priorities. At Western Digital, we focus heavily on failure analysis and supporting future device development. R&D is important, but everything has to be efficient and results-driven. That means we don’t always have the flexibility to explore fundamental questions the way academia does. Even when exciting new techniques emerge, it can be tough to implement them because the workflow needs to stay streamlined.
We have the tools for in situ TEM in-house, and we have talented engineers who can run these experiments. To make the process smoother, automation is key—especially in sample prep, which is a huge focus for us. FIB automation has already made a big impact, and the next step is applying that same efficiency to TEM and data analysis. If we can fully automate these processes, it will open the door to exploring more fundamental research in the future, allowing us to push the boundaries of what’s possible in materials characterization.
In academia, there’s a lot more room for fundamental research. It’s easier to experiment with new or niche techniques simply because researchers have the time to do so. That flexibility is a big difference between the two environments.
Rendering of the FIB preparation process, snapshot from this instructional video.
From your perspective, what benefits does in situ TEM offer for the development of semiconductors or magnetic nanomaterials that might be challenging to achieve through other methods?
In situ TEM is a powerful and sophisticated technique, and while it comes with challenges, the insights it provides can be incredibly valuable. With a well-structured workflow, it can actually be more efficient than traditional methods by allowing us to observe changes in real time rather than relying on multiple separate experiments.
In hard disk drives, device structures consist of nanoscale layers, each serving a specific function—whether for magnetic, thermal, optical or structural properties. These layers must work in perfect harmony for the device to function properly. The goal of using in situ microscopy with these samples is to understand how high temperatures affect interfacial diffusion, grain evolution, and size changes that could ultimately lead to poor performance or device failure. One of the key advantages of in situ TEM is its ability to reveal these transformations in real time, helping us pinpoint the exact temperatures at which material changes begin—crucial information since these changes may directly impact the properties of each functional layers.
This approach also eliminates the need for extensive ex situ imaging across multiple devices. Instead of preparing multiple FIB lamellae and capturing hundreds of images to piece together what’s happening, we can gather the same insights from a single experiment on a single device—improving both efficiency and accuracy.
The gap between nanoscale and bulk-scale results is often a challenge in research. How do you address this in your work?
Bridging the gap between nanoscale and bulk-scale results is always a challenge, but we address it by carefully designing our experiments to be as representative as possible. That said, in our case, “bulk” is relative—our devices are already on the order of hundreds of nanometers.
One of the biggest challenges we face is sample preparation. To achieve atomic resolution in TEM, we have to thin our materials using FIB, but this process can alter how the material behaves, especially in the Z direction. To account for this, we sometimes prepare thicker FIB lamellae as a comparison. While this may slightly compromise resolution, it provides a more accurate link between our in situ nanoscale observations and larger-scale behaviors.
Ultimately, keeping the full picture in mind is key—understanding both the broader material properties and how they relate to what we observe under the microscope. Thoughtful experimental design ensures that our nanoscale findings translate to real-world applications.
For someone first starting electrical biasing in situ experiments, are there any considerations when designing experiments to ensure the nanoscale results are relevant?
Biasing experiments are extremely interesting for us, as we can actively investigate hard drives at the nanoscale. Hard drives have two primary components that are particularly interesting to investigate: the spinning platter (disk) and the head, which acts as both a reader and a writer. The head moves over the platter to write 0’s and 1’s by changing the magnetic orientation of the nano-granular film on the disk surface, and it can later move back to read the same data. The head itself is especially important to study, and if we can prepare it onto an E-chip at the nanoscale, we can investigate its properties using in situ TEM with biasing. This is an extremely challenging goal, but one we are working towards.
One of the major challenges with this process is how to prepare the head on an E-chip without introducing interference from the preparation method. From my experience, preparing the electrical contacts on the E-chip can significantly impact how the sample interacts compared to bulk studies. The way these contacts are fabricated can directly affect the results, so it’s essential to carefully design the process. Although there have been many studies on preparing FIB lamella for biasing experiments, there are still some uncertainties in this area. For our purposes, we would need to separate the contact fabrication from the in situ experiments to make the process more time-efficient. This can be achieved by working closely with our R&D team to ensure that the contacts are fabricated further away from the critical areas, thus minimizing overspray or unintended interference during biasing studies.
Additionally, electrical measurements on thin membranes and nanoscale devices bring their own set of challenges, especially in maintaining stable and reproducible conditions. Careful planning and close collaboration across teams help us address these challenges and ensure that our in situ experiments provide meaningful, real-world insights.
Looking ahead, what advancements or capabilities would you most like to see in (in situ) TEM technology, especially in its application to semiconductors and magnetic nanomaterials?
From a practical point of view, the progress in TEM technology over the past 10 years has been truly remarkable. The microscopes, detectors, and in situ equipment are all advancing rapidly, with TEMs becoming more capable, stable, and achieving higher spatial, analytical and temporal resolution. There are also more detectors available, such as EDS, EELS, and 4D-STEM, which significantly enhance the data we can collect. Additionally, in situ companies are working hard to improve user-friendliness, which makes these technologies more accessible. However, the next step for us is to leverage the full capabilities of these hardware advancements. We want to focus on making in situ experiments more repeatable, accurate, and productive. Currently, we’re seeing progress with drift correction algorithms, which have already made a difference, but the next major step will be improving automated focus adjustments. The tools and technology are improving at an impressive rate, and there’s much potential to use these advancements to push in situ work forward.
From a scientific perspective, one area that’s particularly exciting is the exploration of magnetic fields in situ. Anything related to magnetism, especially in the context of in situ TEM, is something we’re eager to see developed further. Newer microscopes with advanced Lorentz TEM capabilities are opening up more opportunities for magnetic measurements, which is a field we’re really interested in. As these microscopes become more sophisticated, the potential to explore magnetism in real time in the nanoscale environment grows.
Example of 4D-STEM on FIB lamella. Image taken from Nano Lett, (2022) 22, 9544–9550.
How do you stay up to date with the latest developments in the field?
To stay up to date with the latest developments in the field, attending conferences like the M&M conference in the US is really beneficial for us. Since we focus on TEM and FIB analysis, these microscopy conferences are a great way to learn about the latest advancements in electron microscopy. We get to explore new detectors, cameras, TEM technologies, and emerging systems. These events also offer a chance to connect with experts and stay informed about the tools and techniques shaping the field.
In addition, I’ve developed an increased interest in following webinars put on by commercial companies. This has become a valuable resource for keeping up with developments without having to dig through a lot of papers. Webinars allow us to easily access information and catch up on the latest trends at our convenience, even after the session has ended. It’s an efficient way to stay informed without the time commitment of reading extensive literature.
How do you foster collaboration across different research disciplines, and what opportunities are there for interdisciplinary work?
At Western Digital, we actively collaborate with several universities, building long-lasting relationships that have been cultivated over many years. These collaborations allow us to combine the expertise and resources of both industry and academia, fostering a mutually beneficial exchange of knowledge. Additionally, we collaborate with research groups that can conduct studies on our behalf, which further strengthens our connections to the academic community.
One of our main goals is to bridge the gap between academia and R&D. By working together with PhD students and academic researchers, we can address fundamental questions that drive innovation. While we do have the talent in-house to conduct fundamental research, the time constraints in industry make it more difficult to prioritize this type of work. Many of the PhDs and postdocs who specialize in fundamental TEM or in situ TEM in their academic studies often bring valuable expertise to the industry. However, only a small percentage of them continue this exact type of research once they transition to industry roles. By fostering strong collaborations, we can complement our in-house capabilities with the deep expertise from academia, ensuring that we can still address the fundamental questions that drive technological advancements efficiently.
Thank you so much Dr. Ozdol for sharing your insights and giving us a really interesting perspective on research in industry!
