Our everyday lives are so riddled with electronics that we hardly notice them anymore. When we casually reach for our smartphone, we hardly think about how complex such a device actually is. Hundreds of tiny components work together in it - each of them a high-precision masterpiece of engineering. These barely noticeable components include frequency filters. They ensure that a device only receives the right signals, whether via WiFi or mobile networks. Every device that communicates wirelessly contains such filters. They are often based on so-called piezoelectric thin films. Piezoelectric materials have a special feature: they generate an electrical voltage when they are deformed and deform in return when an electrical voltage is applied.
In addition to frequency filters, piezoelectric thin films are used for many other components in microelectronics, whether as sensors, actuators or tiny energy converters. Additional applications, such as for quantum technologies, are the subject of ongoing research. However, one thing is clear: for such thin films to do their job, they must be of a high quality. Depending on the composition and function of the thin film, different manufacturing processes are required. Empa researchers from the "Surface Science & Coating Technologies" department have developed a new coating process for piezoelectric thin films. The special thing about it is that their method allows the high-tech layers to be produced in very high quality on insulating substrates and at a relatively low temperature - a first. The researchers have published their results in the journal "Nature Communications" and applied for a patent for the process. There are applications for the new method in the semiconductor industry as well as in future quantum and photonics technologies.
HiPIMS magnetron sputtering as a basis
The researchers used a common coating process called HiPIMS - short for "high power impulse magnetron sputtering" - as a basis. Magnetron sputtering is a coating process in which material is deposited from a starting material - the target - onto a component to be coated - the substrate. For this purpose, a process gas plasma is ignited at the target. The process gas ions - usually argon - are shot at the target, from which they then knock out atoms that subsequently land on the substrate and form the desired thin film. Many materials can be used as targets. For piezoelectric applications, it is often metals that can be used to produce nitrides, e.g. aluminum nitride, by adding nitrogen. The HiPIMS process basically works in exactly the same way - except that the process does not take place continuously, but in short but particularly high-energy pulses. This not only means that the ejected target atoms travel faster. Many of them are also ionized on their way through the plasma. This makes the process exciting for research. In contrast to neutral atoms, ions can be accelerated, for example by applying a negative voltage to the substrate. The process has been used to produce hard coatings for around 20 years. Here, the high energies ensure particularly dense and resistant layers.
However, the process has not yet been used for piezoelectric thin films. This is because when a voltage is applied to the substrate, not only are the layer-forming target ions accelerated onto the substrate, but also the argon ions from the process gas. This argon bombardment must be avoided. "Several percent argon can sometimes be trapped in hard material layers," says Empa researcher Sebastian Siol. "High voltages often have to be applied to a piezoelectric thin film. This would lead to a catastrophic electrical breakdown." Nevertheless, the researchers working with Siol believed in the potential of HiPIMS for piezoelectric thin films. The high energy with which the ions fly towards the substrate is extremely advantageous. If the ion hits the substrate with sufficient energy, it remains mobile on the substrate for a short time and can find an optimal position in the growing crystal lattice. But what can be done about the argon inclusions?
Jyotish Patidar developed a clever solution as part of his doctoral thesis. Not all ions arrive at their destination at the same time. The majority of argon ions are located in the plasma in front of the target. This means that they often reach the substrate faster than the target ions, which first have to be knocked out of the target and also have to cross the entire distance. Patidar's trick was the timing: "If we apply the voltage to the substrate at exactly the right moment, we only accelerate the desired ions," explains Siol. The argon ions have already flown past at this point - without the additional acceleration, they have too little energy to gain a foothold on the substrate.
"Electronic shower" as an air traffic controller
This trick enabled the researchers to produce high-quality piezoelectric thin films using the HiPIMS process for the first time - equivalent to or even better than conventional methods. Now came the next challenge: depending on the application, the thin film needs to be produced on an insulating substrate, such as glass or sapphire. However, if the substrate is non-conductive, no voltage can be applied to it. Although there is a way in industry to accelerate the ions anyway, this often leads to argon inclusions in the layer.
This is where the Empa researchers achieved a breakthrough. To accelerate the ions onto the insulating substrate, they use the magnetron pulse itself - the short impulse that shoots the process gas ions onto the target. The plasma in the chamber contains not only ions, but also electrons. Each pulse from the magnetron automatically accelerates these negatively charged elementary particles onto the substrate. The tiny electrons reach their target much faster than the ions. Normally, this "electron shower" is not relevant for the HiPIMS process. However, when the electrons arrive at the substrate, they give it a negative charge for a fraction of a second - enough to accelerate ions. If the researchers trigger a subsequent magnetron pulse at exactly the right time interval, the electron shower accelerates the target ions that "flew off" during the previous pulse. And of course, the timing can also be adjusted here so that only the right ions end up in the thin film.
Producing piezoelectric thin films with SFP-HiPIMS
The results are impressive: "With our method, we were able to produce piezoelectric thin films on insulating substrates just as well as on conductive ones," summarizes Siol. The researchers call the process "Synchronized Floating Potential HiPIMS", or SFP-HiPIMS for short. The big advantage: with SFP-HiPIMS, piezoelectric thin films can be produced in very high quality at low temperatures. This opens up new possibilities for the production of chips and electronic components, which often cannot withstand temperature extremes. The process for insulating substrates is particularly important for the semiconductor industry: "The processes in semiconductor production are designed in such a way that there is often no possibility of applying an electrical voltage to the substrate," says Siol.
Together with his research group, he will next focus on the production of ferroelectric thin films - another key technology in today's and tomorrow's electronics. Based on this success, the Empa researchers are also launching several projects with other research institutions to bring their thin films into applications ranging from photonics to quantum technologies. And finally, they want to further optimize the innovative process with the help of machine learning and high-throughput experiments (OM-8/25).
Literature
J Patidar, O Pshyk, K Thorwarth, L Sommerhäuser, S Siol: Low temperature deposition of functional thin films on insulating substrates enabled by selective ion acceleration using synchronized floating potential HiPIMS; Nature Communications (2025); doi: 10.1038/s41467-025-59911-y
Contact
Empa
Coating Technologies
Überlandstrasse 129
8600 Dübendorf (Switzerland)
Phone +41 58 765 11 11
www.empa.ch/web/s208/coating-technologies
About Empa
The Swiss Federal Laboratories for Materials Science and Technology (Empa) is the ETH Domain's interdisciplinary research institute for materials science and technology. As a bridge between research and practical application, it develops solutions for the primary challenges facing industry and creates the scientific basis for the sustainable development of our society.