Ferroelectric Orthorhombic ZrO₂ Thin Films Achieved Through Nanosecond Laser Annealing

Abstract

A new approach for the stabilization of the ferroelectric orthorhombic ZrO2 films is demonstrated through nanosecond laser annealing (NLA) of as‐deposited Si/SiOx/W(14 nm)/ZrO2(8 nm)/W(22 nm), grown by ion beam sputtering at low temperatures. The NLA process optimization is guided by COMSOL multiphysics simulations. The films annealed under the optimized conditions reveal the presence of the orthorhombic phase, as confirmed by X‐ray diffraction, electron backscatter diffraction, and transmission electron microscopy. Macroscopic polarization‐electric field hysteresis loops show ferroelectric behavior, with saturation polarization of 12.8 µC cm−2, remnant polarization of 12.7 µC cm−2 and coercive field of 1.2 MV cm−1. The films exhibit a wake‐up effect that is attributed to the migration of point defects, such as oxygen vacancies, and/or a transition from nonferroelectric (monoclinic and tetragonal phase) to the ferroelectric orthorhombic phase. The capacitors demonstrate a stable polarization with an endurance of 6.0 × 105 cycles, demonstrating the potential of the NLA process for the fabrication of ferroelectric memory devices with high polarization, low coercive field, and high cycling stability.

Pure ZrO2 Ferroelectric Thin Film for Nonvolatile Memory and Neural Network Computing

“…Further investigations revealed that the ferroelectricity of ZrO 2 films can be influenced by many factors, such as doping, − film thickness, , epitaxial strain, interfacial layers, electrode material, and oxygen source dose time . It was found that the macroscopic ferroelectric responses can be ascribed to the formation of the orthorhombic or rhombohedral (r) phase in the ZrO 2 films. ,− …”

“…To obtain a ferroelectric ZrO 2 film, one method is to introduce an in-plane tensile strain to stabilize the ferroelectric o phase . It has been reported that the in-plane tensile strain after rapid thermal processing (RTP) can be introduced by utilizing an electrode material with a lower thermal expansion coefficient (TEC) than that of the ferroelectric layer (the TEC of ZrO 2 is ∼10.5 × 10 –6 K –1 ). , For example, by utilizing a low-TEC tungsten electrode (∼4.5 × 10 –6 K –1 ), a remnant polarization ( P r ) of ∼21.5 μC/cm 2 can be achieved in the W/ZrO 2 (10 nm)/W structure, but the leakage current density is high (∼5 × 10 –7 A/cm 2 at 1 MV/cm) as a result of the oxygen vacancy (V O ) generated during the interfacial oxidation of W. ,− In comparison to ZrO 2 , RuO 2 also has a low TEC of ∼6.4 × 10 –6 K –1 and may contribute to stabilizing the ferroelectric o phase in ZrO 2 . Moreover, the interfacial oxygen diffusion from RuO 2 to ZrO 2 can suppress the V O concentration in the ZrO 2 film and reduce the leakage current. − Therefore, investigating the ferroelectricity of ZrO 2 with RuO 2 electrodes and ferroelectric-ZrO 2 -based ferroelectric field-effect transistors (FeFETs) can be innovative.…”

Influence of Biaxial Strain and Interfacial Layer Growth on Ferroelectric Wake-Up and Phase Transition Fields in ZrO2

“…In the past decade, several nonvolatile memory devices have been investigated based on fluorite-structured ferroelectric materials. For example, a large number of investigations have focused on capacitors for ferroelectric random-access memory (FeRAM), , ferroelectric field-effect transistors (FeFETs), − and ferroelectric tunnel junctions (FTJs). , Significant progress has been made in demonstrating improvements in the ferroelectric properties and reliability of fluorite-structured ferroelectric capacitors by incorporating different dopants, , varying film growth process parameters, , and using different electrodes . However, a critical issue in bringing capacitors from the laboratory to the industrial scale is that the ferroelectric behavior is strongly influenced by small differences in the parameters used during film growth.…”

Enhanced polarization and endurance properties of ZrO2-based ferroelectric capacitor using HfO2 interfacial layer

“…For example, Suraj et al investigated the thickness-dependent transition from antiferroelectric to ferroelectric by changing the thickness of ZrO 2 thin films from 10 nm to 5 Å, and they successfully discovered an approach to convert the conventionally antiferroelectric tetragonal phase of ZrO 2 to the ferroelectric orthorhombic phase through reduced dimensionality [26]. Crema et al presented a new approach, using nanosecond laser annealing, for the stabilization of ferroelectric orthorhombic phase in W/ZrO 2 /W structure, and demonstrated an optimal remnant polarization of 12.7 µC cm −2 and endurance up to 6 × 10 5 cycles [27]. The effect of oxygen vacancies and epitaxially strain etc on the ferroelectric properties of rhombohedral ZrO 2 thin films was also reported [28,29].…”

“…A high annealing temperature is not compatible with the back-end-of-line process of advanced semiconductor technology nodes, where the thermal budget temperature is typically limited below 400 • C. It was recently reported that the deposition temperature and the annealing temperature for ferroelectric ZrO 2 thin films can be as low as 300 • C and 400 • C, respectively [24], suggesting a lower thermal budget in the crystallization process of zirconia than hafnia. Furthermore, ZrO 2 is much more abundant in nature than HfO 2 [33], thus it could be a more appealing candidate for low-cost and large-scale use in the next-generation memory [27,28].…”