The frequency-dependent ferroelectric properties of 45 nm (Al,Sc)N films sputter deposited on complementary metal-oxide-semiconductor (CMOS)-compatible Al metal electrodes are measured and compared. Low in-plane compressive stress (À10 AE 20 MPa) is observed in (Al,Sc)N thin films deposited on Al electrodes. The (Al,Sc)N films exhibit an imprint in the measured coercive fields (E c ) of À4.3/þ5.3 MV cm À1 at 10 kHz. Utilizing positive-up negative-down (PUND) measurements, ferroelectric switching is observed within %200 ns of an applied voltage pulse, which demonstrates the ability of ferroelectric (Al,Sc)N to achieve the fast read/write speeds desired in memory devices.

No abstract

The deluge of sensors and data generating devices has driven a paradigm shift in modern computing from arithmeticlogic centric to data-centric processing. Data-centric processing require innovations at the device level to enable novel compute-inmemory (CIM) operations. A key challenge in the construction of CIM architectures is the conflicting trade-off between the performance and their flexibility for various essential data operations. Here, we present a transistor-free CIM architecture that permits storage, search, and neural network operations on sub-50 nm thick Aluminum Scandium Nitride ferroelectric diodes (FeDs). Our circuit designs and devices can be directly integrated on top of Silicon microprocessors in a scalable process. By leveraging the field-programmability, nonvolatility, and nonlinearity of FeDs, search operations are demonstrated with a cell footprint <0.12 μm 2 when projected onto 45 nm node technology. We further demonstrate neural network operations with 4-bit operation using FeDs. Our results highlight FeDs as candidates for efficient and multifunctional CIM platforms.

In this study, we report the effects of a multilayer architecture on the electrical breakdown strengths and ferroelectric characteristics of 45 nm thick aluminum scandium nitride (AlScN) films. Multilayered films (three-layer, five-layer, and seven-layer) are deposited via sequential deposition of Al0.72Sc0.28N and Al0.64Sc0.36N while maintaining constant volume ratios in all three samples. The effect of the increased number of interfaces is compared to 45 nm single layer Al0.72Sc0.28N and single layer Al0.64Sc0.36N films. The Weibull analysis shows an increase in the characteristic breakdown field from 5.99 and 5.86 MV/cm for single layer Al0.72Sc0.28N and Al0.64Sc0.36N to as high as 7.20 MV/cm in the seven-layered sample. The breakdown field to coercive field (EBD/Ec) ratios also increase from 1.37 and 1.26 in single layer Al0.72Sc0.28N and Al0.64Sc0.36N to up to 1.44 in the seven-layered sample with no significant change in remanent polarization. The enhancement of the characteristic breakdown field can be understood as the propagation of the electrical tree being deflected by multilayer interfaces and/or being slowed by the relative compressive stress in the alternating layers.

The deluge of sensors and data generating devices has driven a paradigm shift in modern computing from arithmetic-logic centric to data centric processing. At a hardware level, this presents an urgent need to integrate dense, high-performance and low-power memory units with Si logic-processor units. However, data-heavy problems such as search and pattern matching also require paradigm changing innovations at the circuit and architecture level to enable compute in memory (CIM) operations. CIM architectures that combine data storage yet concurrently offer low-delay and small footprint are highly sought after but have not been realized. Here, we present Aluminum Scandium Nitride (AlScN) ferroelectric diode (FeD) memristor devices that allow for storage, search and neural network-based pattern recognition in a transistor-free architecture. Our devices can be directly integrated on top of Si processors in a scalable, back-end-of-line process. We leverage the field-programmability, non-volatility and non-linearity of FeDs to demonstrated circuit blocks that can support search operations in-situ memory with search delay times < 0.1 ns and a cell footprint < 0.12 µm2. In addition, we demonstrate matrix multiplication operations with 4-bit operation of the FeDs. Our results highlight FeDs as promising candidates for fast, efficient, and multifunctional CIM platforms.