Variable Fin Pitch Control via Curable Silicon Nitride Etch Masks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for manufacturing FinFETs face challenges in defining variable and controllable fin width and pitch due to inadequate lithography processing windows, resulting in line-width variability and line roughness, which limits the achievable fin-to-fin spacing.
Innovation Solution
A multi-patterning process using sidewall image transfer (SIT) technique with structured etch masks formed from curable silicon nitride material, allowing for precise control of fin dimensions by curing and uncuring portions of the silicon nitride layer to form distinct fin arrays with independent fin pitch and critical dimension control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional lithography is used to define fin width and pitch, then the process is simple, but line-width variability and line roughness increase due to inadequate lithography processing window
Solution Approach 1:
The fin patterning process is segmented into multiple steps: first forming mandrels at a relaxed pitch, then using spacer deposition and selective removal to create fins at the target pitch. This multi-step segmentation allows each step to be optimized independently, achieving precise fin width and pitch control while avoiding the limitations of single-step lithography.
Solution Approach 2:
Spacer layers are introduced as intermediary structures between the mandrels and the final fin pattern. These spacers serve as self-aligned masks that define the fin width with high precision, eliminating the need for direct lithographic definition of fin dimensions and thereby reducing line-width variability and roughness.
2Adaptability or versatility
If fins are cut from an initial array to achieve variable spacing, then fin-to-fin spacing can be adjusted, but the achievable spacing is limited
Solution Approach 1:
The methodology applies different spacer thicknesses and selective removal patterns to different regions of the substrate, enabling local control of fin pitch and width. This allows variable fin spacing across the wafer while maintaining precise control in each local region, overcoming the limitations of global fin cutting approaches.
Solution Approach 2:
The process uses dynamic control of spacer deposition thickness and selective etching to adaptively adjust fin pitch and width across different device regions. By varying the spacer thickness locally and selectively removing spacers in specific areas, the methodology dynamically optimizes fin geometry for different functional requirements (e.g., logic vs. memory) on the same substrate.
3Adaptability or versatility
If variable fin pitch and critical dimension are defined in situ, then distinct fin arrays for different functionalities can be formed, but lithography processing window becomes inadequate
Solution Approach 1:
The methodology transitions from two-dimensional lithographic patterning to three-dimensional spacer-based patterning. By depositing conformal spacers around mandrels and using vertical dimension control (spacer thickness) to define horizontal fin dimensions, the process achieves superior precision and eliminates lithography-induced line-width variability while enabling variable fin configurations across the substrate.
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
Enables the formation of semiconductor fin arrays with controlled fin geometry, facilitating disparate functionalities like logic and memory applications by allowing independent control of fin pitch and critical dimensions, thereby improving electrostatic control and device performance.
Implementation Method 1
A curing process and the attendant volume reduction can be used to tune the dimensions of silicon nitride structures
Data Source
AI summary
A multi-masking process is used to form semiconductor fin arrays having a controlled and variable fin pitch and fin critical dimension within different arrays. A layer of curable silicon nitride is incorporated into a patterning architecture, patterned to form an etch mask, and locally cured to further modify the etch mask geometry. The use of cured and uncured structures facilitate the tuning of the resultant fin geometry.