Fin Width Adjustment via Spacer Merging in FinFET Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The semiconductor industry faces challenges in scaling beyond the 100 nm process technology node, particularly with gate-oxide thickness, source and drain doping depths, and current density, which complicates the formation of high aspect-ratio semiconductor fins in FinFET devices, requiring improved process control for contact landing and multi-gate structures.
Innovation Solution
A method involving the growth of parallel mandrels on a semiconductor substrate, followed by the formation of spacers and adjusting mandrel width to close gaps between spacers, allowing for the creation of first and second type fins, where the second type fins are wider than the first, and the use of a spacer-merging process to form fins of varying widths without requiring new photomasks, maintaining fin density within existing transistor structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high aspect-ratio semiconductor fins are formed in FinFET devices to improve device performance, then transistor speed and reliability are improved, but contact landing precision and process control become more difficult
Solution Approach 1:
The patent segments the fin structure into multiple types (first type fins and second type fins with different widths) by using spacers of different widths between adjacent mandrels. This segmentation allows different regions to have optimized fin dimensions for their specific functions, improving overall device performance while maintaining manufacturability through a single photomask process.
Solution Approach 2:
The patent applies local quality by creating fins with different widths at different locations. The second type fins have a wider width than the first type fins, allowing local optimization for specific device requirements such as contact landing areas or emitter efficiency, while the overall high aspect-ratio fin structure maintains transistor reliability.
2Reliability
If fins of varying widths are created to improve contact landing and emitter efficiency, then device performance is improved, but new photomasks would be required increasing manufacturing complexity
Solution Approach 1:
The patent uses preliminary action by forming spacers around mandrels before removing the mandrels. The spacers are deposited in a conformal manner that naturally creates different widths between adjacent mandrels based on the mandrel spacing. This preliminary formation of spacers with varying widths allows subsequent fin formation without requiring additional photomasks or lithography steps.
Solution Approach 2:
The spacer formation process is self-service in that the conformal deposition automatically creates the desired varying fin widths based on the mandrel arrangement. The process uses the mandrel-spacer-merge sequence to self-generate the multi-width fin structure without external intervention or additional patterning steps, thereby avoiding increased photomask complexity.
3Manufacturing precision
If mandrel width is adjusted to close gaps between spacers, then fin width control is improved, but process control stringency increases
Solution Approach 1:
The patent applies parameter changes by systematically varying the mandrel width and spacing parameters during the spacer formation process. By controlling the mandrel pitch and width, the process naturally produces spacers with different widths that close gaps between adjacent structures. This parameter control achieves precise fin width control while the systematic approach manages process control complexity.
Data Source
AI summary
A method includes growing a plurality of parallel mandrels on a surface of a semiconductor substrate, each mandrel having at least two laterally opposite sidewalls and a predetermined width. The method further includes forming a first type of spacers on the sidewalls of the mandrels, wherein the first type of spacers between two adjacent mandrels are separated by a gap. The predetermined mandrel width is adjusted to close the gap between the adjacent first type of spacers to form a second type of spacers. The mandrels are removed to form a first type of fins from the first type of spacers, and to form a second type of fins from spacers between two adjacent mandrels. The second type of fins are wider than the first type of fins.