Merged Spacer Fin Generation for Tapered IC Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The reduced space between adjacent elements in advanced IC devices, such as the 7 nanometer technology node, leads to increased challenges in fin formation and removal due to reduced error tolerance in lithography processes, making accurate alignment of masks difficult and prone to errors.
Innovation Solution
A method involving the use of mandrels and dummy spacers with a block mask to form and remove spacers, allowing for increased error tolerance in fin placement without additional masks, by creating a merged spacer and utilizing specific mask placements to enhance alignment accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the space between adjacent fins is reduced to increase integration density, then the device integration efficiency is improved, but the error tolerance for mask placement is reduced making alignment increasingly challenging and prone to errors
Solution Approach 1:
The patent introduces merged spacers as intermediary structures that are formed before the actual fin pattern. These merged spacers act as a buffer zone that absorbs placement errors, allowing the final fin pattern to achieve high precision even when mask alignment has some error. The merged spacer serves as a mediator between the imperfect lithography process and the required precise fin structure.
Solution Approach 2:
The methodology performs preliminary actions by forming merged spacers and dummy spacers before the actual fin formation. The merged spacer is created in advance as a reference structure that defines the eventual fin positions. This preliminary structure allows subsequent etching and material deposition to proceed with relaxed alignment tolerances while still achieving precise final fin placement.
2Adaptability or versatility
If traditional cut-masks are used to remove subset of fins, then the device configuration is achieved, but the alignment accuracy becomes increasingly challenging and prone to errors due to reduced spaces between adjacent fins
Solution Approach 1:
Instead of using masks to directly define and remove fins, the patent inverts the approach by first forming merged spacers that define the fin positions, then using these spacers as the actual pattern-defining structures. The traditional mask role is replaced by the self-aligned spacer structures, which are more tolerant to placement errors.
Solution Approach 2:
The methodology creates dummy spacers that are copies or replicas of the merged spacer structure. These dummy spacers serve as templates or references that can be selectively removed using block masks, allowing device configuration changes without requiring precise mask alignment on the actual fin structures. The dummy spacers absorb the alignment error.
Data Source
AI summary
Methods for processes to form and use merged spacers in fin generation and the resulting devices are disclosed. Embodiments include providing first and second mandrels separated from each other across adjacent cells on a Si layer; forming first and second dummy-spacers and third and fourth dummy-spacers on opposite sides of the first and second mandrels, respectively; removing, through a block-mask, the first and fourth dummy spacers and a portion of the second and third dummy-spacers; forming first spacers on each exposed side of the mandrels and in between the second and third dummy-spacers, forming a merged spacer; removing the mandrels; removing a section of the merged-spacer; forming second spacers on all exposed sides of the first spacers and the merged-spacer; removing the merged-spacer and the first spacers; removing exposed sections of the Si layer through the second spacers; and removing the second spacers to reveal Si fins.