What is serial sectioning and what insights into a material can it provide?

DR. BRETSCHNEIDER: With the method known as serial sectioning, we gradually remove layers of material from the sample. After each cut, we take a high-resolution image of the exposed surface. From this, we create a complete 3D reconstruction of the microstructure. This is important for the complex microstructures found in almost all technically used materials. The more sections, the higher the resolution. We typically work with around 100 microsections. Depending on the complexity, only 30 or up to 300 planes may be required.

In contrast to many non-destructive methods, serial sectioning provides high-precision information on the smallest microstructures. For example, X-ray inspection or computer tomography often only offer limited resolution and accuracy. We achieve resolutions ranging from nanometers to millimeters. This enables us to identify the structures (e.g., grain boundaries in a metal) or the distribution of inhomogeneities (e.g., pores or non-metallic inclusions) in the selected material volume. Within the institute, we work closely with colleagues from other technology fields. We often analyze weld seams and use 3D analysis to provide information on process optimization. Another exciting field of application is additive manufacturing. Here, we also cooperate in investigating the material properties resulting from the process control.

How are the samples ablated, and what effort is required if an average of 100 cuts are necessary for a 3D structure?

DR. BRETSCHNEIDER: That depends on the required resolution. We use several metallographic preparation steps for the usual sizes, including grinding, polishing, and etching. We can reach the nanometer level with electron microscopy and FIB (Focused Ion Beam) technology. In this process, the material is removed layer by layer using an ion beam. The layer removal is correspondingly delicate, and fine details are visible. The imaging and material analysis are carried out using electron microscopy.

The metallographic process is very time-consuming when done manually. However, since 2021, we have been able to automate it because at Fraunhofer IWS, we use a metallography robot (RoboMet-3D) developed in the USA. Our technician installs the prepared sample, sets the parameters appropriately, and starts the process. Within the system, the robot processes the sample step by step and places it independently on the built-in light microscope using a robotic arm to generate images. The system repeats this process again and again, even without supervision. As a rule, we let the robot run for the entire working day. Even if this process sounds simple, we need experienced technicians to set the robot settings precisely. Without metallographic expertise, users cannot optimize the system effectively.

Which expertise do you mean exactly, and how widespread is the use of such robots?

DR. BRETSCHNEIDER: Factors such as grinding time, the selection of grinding wheels, and the correct quantities of polishing agents and chemicals are just a few examples. The metallographic expertise of the researchers in our team forms the indispensable basis. After all, the RoboMet-3D is not a plug-and-play system. The robot does the routine work, but the human has to understand the processes fully and set all the parameters correctly. There are around 20 of these robots worldwide, so far almost exclusively in North America. We were the first institution to use such a system in Europe and installed it in December 2021. We received extensive training, but it was still necessary to learn a lot in practical use in order to achieve expert handling of the system, as this requires extensive specialist knowledge and time. Our pioneering work in Europe has given us a certain head start. We now use the robot on a daily basis and know how to fine-tune the parameters.

How could this process be further developed?

DR. BRETSCHNEIDER: Every system has its limits. In microscopy, only some things have been fully exploited. The RoboMet-3D currently only uses a light microscope. We could obtain even finer images if an electron microscope were connected. Hardness measurements like the Vickers method are also not currently integrated. We are already considering combining such methods with the robot in the future.

Mr. Bretschneider, you are now ending your active work at Fraunhofer IWS. We would like to take this opportunity to thank you on behalf of the entire institute. How do you see the near future?

DR. BRETSCHNEIDER: I remain curious and think that the contact with the institute will continue. Today will certainly not be the last time we discuss the great potential of automated serial sectioning. My colleagues in the team will continue to work on this topic with great concentration and expertise. I am pleased to be able to hand the topic over to competent hands.