Ferroelectric materials, which were discovered nearly a hundred years ago, and have led to a huge range of applications, including digital information storage and neuromorphic computing, are still not completely understood. One theory that describes them well is the Landau theory, but this also predicts that the materials could, peculiarly, have a negative capacitance. Researchers at the Nanoelectronics Materials Laboratory (NaMLab gGmbH), Dresden University of Technology and the National Institute of Materials Physics in Romania, have now confirmed this prediction for the first time by performing electrical measurements on ferroelectric hafnium zirconium oxide (Hf0.5Zr0.5O2). Negative capacitance could be exploited to improve the energy efficiency of electronics devices and since this material is already found in today’s computer chips, real-world applications might be possible relatively quickly.
Ferroelectric materials have permanent electric dipole moments – in the same way that their ferromagnetic counterparts have permanent magnetic dipole moments. Ferroelectrics can be used in a wide range of devices because their dipole moments can be oriented using electric fields, which are much easier to create than the magnetic fields used to manipulate ferromagnetic materials.
Since the 1940s, researchers have modelled these materials using the Landau theory of phase transitions. This theory was first applied to ferroelectrics by physicists Ginzburg and Devonshire, and models based on the Landau-Ginzburg-Devonshire (LGD) approach, as it is called, are important for understanding the basic physics of ferroelectricity.
Double-well shape
In this theory, a ferroelectric material can have a negative capacitance, which shows up in the double-well shape of the free energy in a ferroelectric as an electric field is applied. Although predicted over 70 years ago, most scientists believed that negative capacitance was impossible to observe in an experiment. Observing the signature of negative capacitance is important, however. This is because it can amplify an applied voltage and could thus be exploited to reduce the power dissipation of future electronics devices beyond conventional limits.
Researchers led by Michael Hoffmann have now measured the double-well energy landscape in a thin layer of ferroelectric Hf0.5Zr0.5O2 for the first time and so confirmed that the material indeed has negative capacitance. To do this, they first fabricated capacitors with a thin dielectric layer on top of the ferroelectric. They then applied very short voltage pulses to the electrodes of the capacitor, while measuring both the voltage and the charge on it with an oscilloscope.
“Since we already knew the capacitance of the dielectric layer from separate experiments, we were then able to calculate the polarization and electric field in the ferroelectric layer,” Hoffmann tells Physics World. “We then calculated the double-well energy landscape by integrating the electric field with respect to the polarization.”
The advantages of Hf0.5Zr0.5O2
The negative capacitance is a direct result of the intrinsic energy barrier between two stable polarization states of the ferroelectric layer, he says. In regular ferroelectric capacitors with two metal electrodes, this barrier region is not accessible because the ferroelectric polarization is screened by free electrons in the electrodes. In this work, the researchers were able to inhibit this screening by inserting another dielectric layer.
Organic ferroelectrics finally stick in the memory
“Most ferroelectric materials are very difficult to integrate into current semiconductor fabrication processes but Hf0.5Zr0.5O2 is already employed in modern electronics,” explains Hoffmann. “This means that future products utilizing this effect might not be far away. Another advantage of Hf0.5Zr0.5O2 is that it retains its ferroelectric properties even for films thinner than 10 nm, which is important for further miniaturization.”
The researchers say they also need to closely match the positive capacitance of the dielectric layer to the negative capacitance of the ferroelectric layer in their devices, which is much easier to do when using Hf0.5Zr0.5O2 compared to many other ferroelectrics. “This is thanks to the relatively high coercive field (that is, the electric field at which the ferroelectric switches from one polarization to another) of Hf0.5Zr0.5O2 compared to these other materials,” says Hoffmann.
Future applications
One of the most promising applications utilising negative capacitance are electronic circuits with much lower power dissipation that could be used to build more energy efficient devices than any that are possible today, he adds. “We are working on making such devices, but it will also be very important to design further experiments to probe the negative capacitance region in the structures we made so far to help improve our understanding of the fundamental physics of ferroelectrics.”
The research is detailed in Nature 10.1038/s41586-018-0854-z.
