US Patent Application 18447453. Ferroelectric Semiconductor Device and Method simplified abstract
Contents
Ferroelectric Semiconductor Device and Method
Organization Name
Taiwan Semiconductor Manufacturing Company, Ltd.
Inventor(s)
Chia-Cheng Ho of Hsinchu City (TW)
Ming-Shiang Lin of Hsinchu City (TW)
Ferroelectric Semiconductor Device and Method - A simplified explanation of the abstract
This abstract first appeared for US patent application 18447453 titled 'Ferroelectric Semiconductor Device and Method
Simplified Explanation
The patent application describes a method for creating a ferroelectric semiconductor device.
- The method involves using a diffusion anneal process to drive dopant elements through a silicon layer and into a gate dielectric layer.
- This forms a doped gate dielectric layer with a gradient depth profile of dopant concentrations.
- The doped gate dielectric layer is then crystallized during a post-cap anneal process.
- This creates a gradient depth profile of ferroelectric properties within the crystallized gate dielectric layer.
- A metal gate electrode is then formed over the crystallized gate dielectric layer.
- This results in a ferroelectric transistor with multi-ferroelectric properties between the gate electrode and the channel.
- The ferroelectric transistor can be used in deep neural network applications.
Original Abstract Submitted
A ferroelectric semiconductor device and method are described herein. The method includes performing a diffusion anneal process to drive elements of a dopant film through an amorphous silicon layer and into a gate dielectric layer over a fin to form a doped gate dielectric layer with a gradient depth profile of dopant concentrations. The doped gate dielectric layer is crystallized during a post-cap anneal process to form a gradient depth profile of ferroelectric properties within the crystallized gate dielectric layer. A metal gate electrode is formed over the crystallized gate dielectric layer to obtain a ferroelectric transistor with multi-ferroelectric properties between the gate electrode and the channel. The ferroelectric transistor may be used in deep neural network (DNN) appl