Making the invisible measurable and translating physical phenomena into practical technological applications: This is the aim of the research of Rebecca Petrich, a doctoral candidate at the Institute of Micro and Nanotechnologies (IMN) MacroNano and the Institute of Physics at TU Ilmenau. In UNIonline she reports on her doctoral journey - with wafers and microchips in her hand luggage from Ilmenau to the Black Forest via Scotland and Berlin to the border triangle near Basel and back to Ilmenau.
To develop a particularly energy-efficient and highly sensitive measuring system, a so-called MEMS magnetometer, tailored to specific industrial requirements - from concept to finished prototype: this is the research assignment given to Rebecca Petrich by Endress+Hauser as part of the XMEN project. It uses the magnetoelectric effect to measure the smallest magnetic field changes and is to be used to monitor process and characteristic variables.
Ms. Petrich, why did you choose this university and this subject?
I did my Nawitur here at the university when I was still at school. We took student courses and did internships during the vacations. That was a great experience - we programmed, milled, wired and got to know many different areas. I found the interface between natural sciences and engineering particularly appealing. That’s why I decided to study biomedical engineering. Over the course of my studies, I became increasingly fascinated by the question: How can the invisible be made measurable?
And this question brought you to microand nanotechnology?
Exactly. At some point, the specific application in medicine faded into the background - instead, I was fascinated by the fundamental question. That’s why I studied micro and nanotechnologies in my Master’s degree and specialized in microsensor technology. At the time, the Master’s course lasted four semesters and was very broadly based - from materials-oriented design to supramolecular chemistry. It was the ideal scientific breeding ground for me. In my final thesis, I then developed my first sensor based on my own concept. In the process, I learned my first soft skills to bring an idea to the chip in time.
What makes the type of sensors that you develop in your work so special?
The MEMS magnetometers I develop are suitable for fast, accurate, energy-efficient and budget-friendly measurements - even in environments with comparatively high offset or interference fields. In such situations, many highly sensitive magnetometers reach their limits, as they become saturated by the strong ambient magnetization and are "blind", so to speak. Our sensors, on the other hand, remain functional. They are also CMOS-compatible and can be efficiently manufactured in large quantities using established microtechnology processes and integrated into industrial processes.
What exactly will the measuring device be used for by your client?
At Endress+Hauser, the sensors are to be used to monitor the condition in a tank from the outside. Until now, this has often been done invasively, which has a number of disadvantages. One example is the more complex maintenance.
In your opinion, how does this type of industry-funded doctorate differ from others?
If you do your doctorate as part of an industry project, you are given a clear research assignment - and you work in the interests of the company. This means that you have to focus your own research on practical applicability right from the start and take numerous industrial framework conditions into account. This is usually different for publicly funded or university-financed doctorates. Here too, later use in industry can play a role, but it is often seen more as a long-term perspective. Research is conducted primarily in the public interest, but also out of academic curiosity. Both paths - industrial or academic - bring their own challenges and opportunities.
What were the particular challenges?
One of the biggest challenges in chip development in research is process stability. In large industrial production facilities, the same processes are usually used and these are stabilized in such a way that they run almost flawlessly. This is different for us: the parameters often change on the same system due to different materials and processes, which makes stability more difficult. In addition, we are an educational institution in which students and doctoral students, who often have little experience, operate the systems alongside those responsible for the systems. This is part of the learning process, but it can also lead to a wafer breaking or a system failing, for example. That’s why it’s important to plan well in advance and always have alternatives ready. For example, I ended up outsourcing two processes to industrial partners to ensure stability.
What were these processes?
The two riskiest steps take place at the end of production: The entire wafer, i.e. the carrier material for my sensor chip, is etched through, the individual chips are separated from each other and the sensor structures are exposed at the same time - all without damaging the sensitive structures. My sensors only work if these structures can vibrate freely, which is a major technical challenge. To achieve this, I use a combination of an SF6 Bosch etching process and a XeF2 etching process. The first process is quite aggressive and involves some risks. In order to identify suitability and potential problems at an early stage, I carried out preliminary tests at the Center for Micro and Nanotechnologies (ZMN) and then transferred the process to Hahn-Schickard in the Black Forest.
What happened from there?
The final process step took place in Scotland, directly at the plant manufacturer memsstar. In the XeF2etching process, the last remaining silicon is removed. A major advantage of this process is that it is very gentle, as it requires almost no energy input. This makes it particularly suitable for sensitive structures. However, these structures are under residual stress due to the manufacturing process. Only at the end can it be determined whether they remain intact or are damaged by this tension - for example through cracks. In addition, the sensor then had to survive transportation back to Germany and integration into the production facility at Endress+Hauser near Berlin. So this was a really challenging phase - not for the faint-hearted.
That sounds like a grueling time...
Yes, it definitely was. Every time I completed a step, I was relieved and happy - but at the same time I also knew that the likelihood of everything really working out in the end was very low. You can’t calculate all the details right up to the end, there’s always a risk. And even if the chips survive the entire manufacturing process and transportation, I didn’t know whether they would actually work as planned for the customer. As I was working with new materials, it was not possible to predict the sensor behavior precisely beforehand - the data was simply too thin. It was only after about three years that I was able to test the finished sensors at the customer’s premises and see whether everything was right. It often takes just a few seconds to decide whether That’s why it was classified as a high-risk project.
And did you win?
Yes, and really well. And that was anything but a matter of course. Because my client’s measuring principle requires a very special sensor. Of course, there are already a wide variety of magnetometers - but none that meet exactly these requirements. This means that if you don’t hit the right frequency range, for example, the sensor simply won’t work for the intended application.
Were you there when your sensor was tested and what was the moment like for you?
Yes, I was there - and it was an incredible moment. I had developed over 100 different chip variants and handed them over to the customer. You can’t calculate in advance which variant will meet the parameters exactly, so I deliberately implemented many different designs. In the end, I had to trust my instincts. Then we connected the sensor - and it worked straight away. At that moment, it felt like we had landed on the moon. This success was only possible thanks to the great cooperation of everyone involved.
Based on this experience, what is the most important piece of advice you would give to young people interested in a career in science?
Get out of your comfort zone, be brave and trust in your abilities.
A portrait of doctoral candidates
A doctorate is often a springboard for an academic career and opens doors in industry and business. But what does the path to a doctorate really feel like? What opportunities, challenges and experiences does a doctorate at TU Ilmenau entail? In our new series "Doctoral candidate of the month", young scientists give personal insights into their doctorate, their motivation and their plans for the future.Rebecca Petrich completed a Bachelor’s degree in Biomedical Engineering and a Master’s degree in Microand Nanotechnologies at TU Ilmenau and is currently a research assistant at the university’s Institute for Microand Nanotechnologies (IMN) MacroNano , where she is writing her dissertation on a research contract from Endress + Hauser SE + Co. KG, the development of a new type of magnetometer in the "XMEN" project. As spokesperson for the IMN graduate college and deputy chairwoman of the board of the Förderverein für Mikround Nanotechnologien Ilmenau e.V. (Ilmenau Association for the Promotion of Micro and Nanotechnologies), her main aim is to promote exchange between science and industry. Among other things, she co-initiated a job and project exchange with companies and co-organizes the annual PhD Day at the TU Ilmenau, a further education and networking event to promote doctoral students.
