铁电材料的集成：下一代计算和存储设备的终极解决方案,ACS Applied Electronic Materials

当前位置： X-MOL 学术 › ACS Appl. Electron. Mater. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

铁电材料的集成：下一代计算和存储设备的终极解决方案
ACS Applied Electronic Materials ( IF 4.3 ) Pub Date : 2021-06-30 , DOI: 10.1021/acsaelm.0c00851
Robin Khosla _{1,

2} , Satinder K. Sharma ₃

Affiliation

自从铁电被发现以来的几十年里，铁电材料已经成为广泛的半导体技术和电子设备应用的基石，特别是在最先进的互补金属氧化物半导体 (CMOS) 逻辑电路和数字信息存储介质中. 权衡了集成铁电材料（尤其是用于电子设备的高 κ 电介质）最近前所未有的进步和未来前景。重点是（i）应用（逻辑和记忆）；(ii) 铁电材料（有机、无机和二维 (2D)）；(iii) 器件结构（金属/铁电/金属（MFM）、金属/铁电/半导体（MFS）、金属/铁电/绝缘体/半导体（MFIS）和金属/铁电/金属/绝缘体/半导体（MFMIS））；(iv) 下一代电子器件（负电容场效应晶体管 (NC-FET)、铁电 RAM (FeRAM)、铁电场效应晶体管 (FeFET) 和铁电隧道结 (FTJ)）。在 NCFET 中，铁电层用作负电容器，因此沟道表面电位可以放大到超过栅极电压。因此，器件可以克服“玻尔兹曼暴政”并以 < 60 mV/dec 的陡峭亚阈值摆幅和 < 0.5 V 的电源电压运行。因此，NC-FET 将更适合高速逻辑操作、可扩展性、低功耗和成本效益，针对 14T 型 CPU 寄存器和 6T 型高速缓存静态随机存取存储器 (SRAM) 等应用。由于独特的结构和电子特性，铁电体还为解决与技术缩放相关的问题开辟了道路。根据最佳性能、成本和最终用户要求，预计铁电存储器有不同的风格。在此，我们探讨了开发器件结构的令人兴奋的可能性，例如具有快速访问时间（<10 ns）、高耐用性（∼>10 ns）的单晶体管单电容器（1T-1C）型 FeRAM¹⁴个周期），适度的数据保留被认为是易失性动态随机存取存储器 (DRAM) 的有力竞争者，而对于非易失性存储器应用，1T 型铁电门控晶体管，称为 FeFET，具有无损读出、快速访问时间（∼< 100 ns)，中等耐力 (>10 ⁹周期）和高保留时间（>10 年）有可能与嵌入式固态驱动器 (SSD) 竞争。最后，具有三维交叉点架构的 FTJ 是高密度利基存储应用的有力竞争者，可与低成本的外部硬盘驱动器互换。最后，我们简要回顾了铁电体的最新进展以及下一代计算和存储设备应用的潜在未来比较，因此该领域可能会扩展，并为半导体技术的大批量制造铺平道路，直至 5 纳米以下节点未来几年。

"点击查看英文标题和摘要"

Integration of Ferroelectric Materials: An Ultimate Solution for Next-Generation Computing and Storage Devices

Over the decades since ferroelectricity was revealed, ferroelectric materials have emerged as a cornerstone for a wide spectrum of semiconductor technology and electronic device applications, particularly in state-of-the-art complementary metal oxide semiconductor (CMOS) logic circuits and digital information storage media. Recent unprecedented advancements and future perspectives on integrating ferroelectric materials, particularly with high-κ dielectrics for electronic devices, are weighed. The emphasis is on (i) application (logic and memory); (ii) ferroelectric materials (organic, inorganic, and two-dimensional (2D)); (iii) device structures (metal/ferroelectric/metal (MFM), metal/ferroelectric/semiconductor (MFS), metal/ferroelectric/insulator/semiconductor (MFIS), and metal/ferroelectric/metal/insulator/semiconductor (MFMIS)); and (iv) next-generation electronic devices (negative capacitance field effect transistors (NC-FETs), ferroelectric RAM (FeRAM), ferroelectric field effect transistors (FeFETs), and ferroelectric tunnel junctions (FTJs)). In NCFETs, the ferroelectric layer serves as a negative capacitor so that the channel surface potential can be amplified more than the gate voltage. Hence, devices can overcome the “Boltzmann tyranny” and operate with a steep subthreshold swing < 60 mV/dec and supply voltage < 0.5 V. Thus, NC-FETs would be more suitable for high-speed logic operations, scalability, low-power, and cost-effectiveness, targeting applications such as 14T-type CPU registers and 6T-type cache static random access memory (SRAM). Ferroelectrics also opens a path to solving the problems associated with technology scaling due to the unique structural and electronic properties. Ferroelectric memories are anticipated to be in different flavors based on optimum performance, cost, and end-user requirements. Herein, we deliberate on the exciting possibilities for the development of device structures such as one-transistor one-capacitor (1T-1C)-type FeRAM with fast access time (<10 ns), high endurance (∼>10¹⁴ cycles), and moderate data retention being considered as a strong contender for volatile dynamic random access memory (DRAM), while, for nonvolatile memory applications, 1T-type ferroelectric gated transistors, called FeFETs with nondestructive readout, fast access time (∼<100 ns), moderate endurance (>10⁹ cycles), and high retention time (>10 years) have the potential to compete with embedded solid-state drives (SSDs). Finally, the FTJs with three-dimensional cross-point architecture are strong contenders for high-density niche storage applications to interchange with low-cost per bit external hard disk drives. We conclude with a brief survey of recent ferroelectrics advances and potential futuristic comparisons for next-generation computing and storage device applications so the field may expand and pave the way for high-volume manufacturing of semiconductor technology down to the sub-5 nm node over the coming years.

更新日期：2021-07-27

点击分享查看原文

点击收藏

阅读更多本刊新发论文本刊介绍/投稿指南