Metal Gate Selective Etching for FinFET Threshold Voltage Spread
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional metal-oxide-semiconductor (MOS) devices face challenges in creating a sufficient threshold voltage spread between transistors, particularly due to the poly-depletion effect and the difficulty in maintaining the thickness of work function tuning layers during etching processes, which affects the performance of Fin Field-Effect Transistors (FinFETs).
Innovation Solution
Doping aluminum into the titanium nitride work function tuning layer increases etching selectivity between the tantalum nitride barrier layer and the titanium nitride work function tuning layer, allowing for selective thinning of the barrier layer while minimizing the loss in work function tuning layer thickness, thus maintaining the threshold voltage spread between transistors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a metal gate is formed using conventional multi-layer deposition and CMP processes, then the poly-depletion effect is eliminated, but the threshold voltage spread between transistors is insufficient
Solution Approach 1:
The patent changes the material composition parameter of the work function tuning layer by doping aluminum into titanium nitride. This parameter change modifies the etching selectivity, allowing the barrier layer to be selectively thinned while preserving the work function tuning layer thickness, thereby achieving sufficient threshold voltage spread without increasing device complexity
Solution Approach 2:
The patent applies local quality by creating different aluminum doping concentrations in different regions of the work function tuning layer. By selectively doping certain areas with aluminum, the etching selectivity is locally enhanced, enabling precise control over threshold voltage spread in specific transistor regions while maintaining overall structure simplicity
2Manufacturing precision
If the barrier layer is thinned to adjust threshold voltage, then the threshold voltage spread increases, but the work function tuning layer thickness is also reduced due to low etching selectivity
Solution Approach 1:
The patent changes the chemical composition parameter of the work function tuning layer by incorporating aluminum-doped titanium nitride. This parameter change dramatically increases the etching selectivity between the barrier layer and work function tuning layer, allowing the barrier layer to be thinned with minimal loss to the work function tuning layer thickness
Solution Approach 2:
The patent uses composite materials by creating a work function tuning layer composed of titanium nitride doped with aluminum. This composite material structure provides both the desired work function characteristics and enhanced etching selectivity, enabling precise barrier layer thinning while preserving work function tuning layer integrity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively reduces the adverse reduction in threshold voltage spread, ensuring that FinFETs maintain desired performance characteristics by enhancing etching selectivity and minimizing the thickness loss of work function tuning layers, thereby supporting the requirements of different circuits.
Implementation Method 1
Doping aluminum into the titanium nitride work function tuning layer increases etching selectivity between the tantalum nitride barrier layer and the titanium nitride work function tuning layer
Implementation Method 2
allowing for selective thinning of the barrier layer while minimizing the loss in work function tuning layer thickness
Data Source
AI summary
A method includes forming a gate dielectric comprising a portion extending on a semiconductor region, forming a barrier layer comprising a portion extending over the portion of the gate dielectric, forming a work function tuning layer comprising a portion over the portion of the barrier layer, doping a doping element into the work function tuning layer, removing the portion of the work function tuning layer, thinning the portion of the barrier layer, and forming a work function layer over the portion of the barrier layer.