Electronic Synapses Enabled by an Epitaxial SrTiO_3‐δ/ Hf_0.5Z_r0.5O₂Ferroelectric Field‐Effect Memristor Integrated on Silicon

Abstract

Synapses play a vital role in information processing, learning, and memory formation in the brain. By emulating the behavior of biological synapses, electronic synaptic devices hold the promise of enabling high‐performance, energy‐efficient, and scalable neuromorphic computing. Ferroelectric memristive devices integrate the characteristics of both ferroelectric and memristive materials and present a far‐reaching potential as artificial synapses. Here, it is reported on a new ferroelectric device on silicon, a field‐effect memristor, consisting of an epitaxial ultrathin ferroelectric Hf0.5Zr0.5O2 film sandwiched between an epitaxial highly doped oxide semiconductor SrTiO3‐δ and a top metal. Upon a low voltage of less than 2 V, the field‐effect modulation in the semiconductor enables to access multiple states. The device works in a large time domain ranging from milliseconds down to tens of nanoseconds. By gradually switching the polarization by identical pulses, the ferroelectric diode devices can dynamically adjust the synaptic strength to mimic short‐ and long‐term memory plasticity. Ionic contributions due to redox processes in the oxide semiconductor beneficially influence the device operation and retention.

Growth of emergent simple pseudo-binary ferroelectrics and their potential in neuromorphic computing devices

“…Most importantly, HfO 2 - and ZrO 2 -based non-volatile FeFETs have already been established as promising contenders for synaptic elements in neuromorphic devices due to their established deposition methods, high ferroelectric polarization, scalability, high endurance, defect-free operation mechanism, and CMOS compatibility. 63–68 The impressive performance of HfO 2 - and ZrO 2 -based thin film FeFETs is getting significant attention due to high scalability, good endurance, and CMOS compatibility. 4…”

“…In addition, these group-IV-based oxide materials exhibit important performance metrics, including good linearity/symmetry, multilevel states, high on/off ratios and low write energy characteristics, which are desirable for neuromorphic applications. 4,63–65,67–69 However, HfO 2 - and ZrO 2 -based thin films typically struggle with parameters such as imprint, wake-up, fatigue, and consequently with uniform cycling endurance. Also, reducing the film thickness, as is required in some cases, may increase E c and therefore necessitate higher operating voltages, leading to higher leakage currents during program/erase operations.…”

“…Recently, researchers have reported analog synaptic transistors based on HfO 2 - and ZrO 2 -based ferroelectric materials, which employ a variety of channel materials, including WO x , indium gallium zinc oxide (IGZO), indium zinc tin oxide (IZTO), poly-GeSn, and WS 2 . 63–69 HfO 2 - and ZrO 2 -based ferroelectric materials are attractive due to their ferroelectricity in ultra-thin films (from ∼1 up to 10 nm) with improved reliability, low programming voltage and CMOS compatibility. 63–66 For instance, Halter et al studied an Hf 0.57 Zr 0.43 O 2 thin-film-based (thickness ∼9.6 nm) FeFET with WO x as the channel material for synaptic elements.…”

“…63–69 HfO 2 - and ZrO 2 -based ferroelectric materials are attractive due to their ferroelectricity in ultra-thin films (from ∼1 up to 10 nm) with improved reliability, low programming voltage and CMOS compatibility. 63–66 For instance, Halter et al studied an Hf 0.57 Zr 0.43 O 2 thin-film-based (thickness ∼9.6 nm) FeFET with WO x as the channel material for synaptic elements. 63 This work showcased an Hf 0.57 Zr 0.43 O 2 /WO x stack as an FeFET with good voltage control and strong linearity and symmetry in potentiation and depression at different voltage amplitudes with identical pulse trains, which facilitate facile programming of learning algorithms.…”