EXPloring Antiferroelectric oxides as a new technological brick for futur Nanoelectronic Devices – EXPAND

Ferroelectric (FE) materials that are characterized by a switchable polarization under application of an electric field have already attracted considerable interest for future electronic devices like memories because of their potential for fast switching capabilities, long retention time and high integration density. The ability of the FE polarization to tune the carrier density at the interface of an adjacent material has been recently used in FE-based tunnel junction (TJ) memories allowing a non-destructive readout of the information. Beyond that, the ability to tune continuously the polarization state and therefore the resistance through the FE barrier permitted to realize FE-based memristors mimicking biological synapses. This should therefore lead to a radically enhance of the computational power and energy efficiency of electronic devices. Alternatively, it has been also demonstrated very recently that using negative capacitance of a FE gate insulator in a FE-based Field Effect Transistor (FET), the so-called subthreshold swing which is limited to 60 mV/decade in classical FET can be subsequently reduced which in turn diminishes the power supply voltage and energy dissipation in the FET.
It is worth to realize that the physical mechanism at the origin of the above mentioned effects is the FE instability/metastability (which is static in FTJs or transient in the negative capacitance response of FE-FETs). Interestingly, antiferroelectrics (AFEs), which are less studied materials especially as thin or ultrathin films, show also polar instability/metastability. Indeed, AFEs are very close in energy to FEs e.g. electric/elastic field can induce FE state from AFE one. However, AFEs have not, or scarcely, been considered in any memory or logic devices.
Here in the framework of EXPAND project, we would like to explore the potential of AFEs in both TJs and FETs. In a AFE-based TJ, we aim at realizing a relaxation oscillator that emulates the behavior of a spiking neuron. Integrating this device into a RC circuit should thus allow building a spiking oscillator, whose spiking frequency can be adjusted by the voltage applied to the device. In AFE-based FETs, the energy landscape of the polarization which is at the origin of the negative capacitance is richer and therefore may not only significantly decrease the subthreshold swing but also generate novel logic functionalities such as dynamic hysteresis control. Reaching proof-of-concepts of such devices requires a better understanding of the fundamental and stability conditions of antiferroelectricity in thin films and a better knowledge of the interplay of AFEs with various interfaces including metal, dielectric, FE or another AFE layer. These heterostructures will then serve as building blocks for the design of a new nanoelectronic with expected boosted properties as well as innovative memory and logic devices.
EXPAND is a multidisciplinary project requiring tight articulation between AFE-based film and heterostructure growth, precise structural and physical characterization with continuous feedback with modeling work to finally integrate them into prototype devices with improved performances and novel functionalities. Using AFE materials in innovative electronic devices will therefore lead to an improvement of Information Communication Technologies device performances, especially in term of power consumption and efficiency with storage and ultimate logic beyond C-MOS and more than Moore technologies.