Litcius/Paper detail

High‐<i>κ</i> Dielectric (HfO<sub>2</sub>)/2D Semiconductor (HfSe<sub>2</sub>) Gate Stack for Low‐Power Steep‐Switching Computing Devices

Taeho Kang, Joonho Park, Hanggyo Jung, Haeju Choi, Sang‐Min Lee, Sang‐Min Lee, Nayeong Lee, Ryong‐Gyu Lee, Gahye Kim, Seung‐Hwan Kim, Hyung‐jun Kim, Cheol‐Woong Yang, Jongwook Jeon, Yong‐Hoon Kim, Sungjoo Lee, Sungjoo Lee

2024Advanced Materials35 citationsDOIOpen Access PDF

Abstract

Abstract Herein, a high‐quality gate stack (native HfO 2 formed on 2D HfSe 2 ) fabricated via plasma oxidation is reported, realizing an atomically sharp interface with a suppressed interface trap density ( D it ≈ 5 × 10 10 cm −2 eV −1 ). The chemically converted HfO 2 exhibits dielectric constant, κ ≈ 23, resulting in low gate leakage current (≈10 −3 A cm −2 ) at equivalent oxide thickness ≈0.5 nm. Density functional calculations indicate that the atomistic mechanism for achieving a high‐quality interface is the possibility of O atoms replacing the Se atoms of the interfacial HfSe 2 layer without a substitution energy barrier, allowing layer‐by‐layer oxidation to proceed. The field‐effect‐transistor‐fabricated HfO 2 /HfSe 2 gate stack demonstrates an almost ideal subthreshold slope (SS) of ≈61 mV dec −1 (over four orders of I DS ) at room temperature (300 K), along with a high I on / I off ratio of ≈10 8 and a small hysteresis of ≈10 mV. Furthermore, by utilizing a device architecture with separately controlled HfO 2 /HfSe 2 gate stack and channel structures, an impact ionization field‐effect transistor is fabricated that exhibits n‐type steep‐switching characteristics with a SS value of 3.43 mV dec −1 at room temperature, overcoming the Boltzmann limit. These results provide a significant step toward the realization of post‐Si semiconducting devices for future energy‐efficient data‐centric computing electronics.

Topics & Concepts

Materials scienceStack (abstract data type)OptoelectronicsDielectricSemiconductorPower (physics)High-κ dielectricElectrical engineeringComputer scienceQuantum mechanicsProgramming languageEngineeringPhysicsSemiconductor materials and devicesFerroelectric and Negative Capacitance DevicesAdvanced Memory and Neural Computing