Contact-Area-Changeable CMP Conditioning for Enhancing Pad Lifetime

At a Glance

Metadata	Details
Publication Date	2021-04-14
Journal	Applied Sciences
Authors	Jungyu Son, Hyunseop Lee
Institutions	Tongmyong University, Dong-A University
Citations	12
Analysis	Full AI Review Included

Executive Summary

This study introduces a novel Contact-Area-Changeable (split) conditioning system for Chemical-Mechanical Polishing (CMP) pads, specifically designed to mitigate uneven wear and extend pad lifetime in semiconductor manufacturing.

Core Problem Addressed: Conventional swing-arm conditioning (Case I) causes locally excessive pad wear, leading to a rapid decrease in Material Removal Rate (MRR) and increased Within-Wafer Non-Uniformity (WIWNU) after approximately 12 hours.
Proposed Solution (Case II): Implementation of a split conditioner (inner disk and outer ring) where only the smaller inner disk is used for conditioning in the critical wafer-pad contact area (Zones 2-5).
Lifetime Enhancement: The proposed Case II system extended the effective pad lifetime to over 20 hours, compared to the 16-hour limit observed in Case I, where pad grooves wore out and wafers began breaking.
Performance Stability: Case II maintained exceptional process stability over 20 hours, showing only a 7.4% reduction in MRR and keeping WIWNU below 3%. Case I experienced a 44.9% MRR drop over 16 hours.
Wear Profile Control: Simulation and experimental results confirmed that Case II achieved a significantly flatter pad cut rate profile in the wafer-pad contact area (150-350 mm radial distance), preventing the non-uniform stress distribution associated with excessive local wear.
Mechanism Validation: Scanning Electron Microscope (SEM) images confirmed that pad grooves remained intact after 20 hours of conditioning in Case II, whereas they had completely disappeared after 16 hours in Case I.

Technical Specifications

Parameter	Value	Unit	Context
Wafer Diameter	200	mm	SiO₂ film (1.5 µm thick)
Platen Rotational Speed	93	rpm	Conditioning & CMP
Conditioner Rotational Speed	101	rpm	Conditioning
Conditioning Load	4	kgf	Total load
Swing Start Radius	370	mm	From pad center
Swing End Radius	50	mm	From pad center
Wafer Pressure	34.3	kPa	CMP process
Retaining-Ring Pressure	44.1	kPa	CMP process
Slurry Flow Rate	150	mL/min	TSO-12 silica slurry
Pad Type	KONI	N/A	Hard polyurethane pad
Case I Avg Pad Cut Rate	43.4	µm/h	Experimental result
Case II Avg Pad Cut Rate	22.2	µm/h	Experimental result
Case I MRR Stability	44.9	% reduction	Over 16 h conditioning time
Case II MRR Stability	7.4	% reduction	Over 20 h conditioning time
Case II WIWNU	<3	%	Maintained up to 20 h
Inner Conditioner Diameter	64	mm	Split conditioner design
Outer Conditioner Width	31	mm	Split conditioner design

Key Methodologies

The study utilized a combination of numerical simulation and marathon CMP experiments to validate the effectiveness of the contact-area-changeable conditioning system.

System Modification:
- A conventional swing-arm conditioning system was adapted to incorporate a split diamond conditioner (inner disk and outer ring).
- The loads applied to the inner and outer conditioners were independently controlled via a pneumatic cylinder and rotary joint.
Pad Wear Simulation:
- The numerical analysis method involved creating virtual grids (10 mm x 10 mm) on the polishing pad.
- The pad wear profile (pad cut rate distribution) was calculated by counting the number of times diamond grits passed over the virtual grid area during a simulated 1 hour of conditioning.
- Four conditioning cases were simulated, focusing on the use of full contact (Case I) versus selective contact (Case II, III, IV).
Conditioning Strategy (Case II Focus):
- The swing motion was partitioned into five radial zones (50-370 mm).
- In Case I (Reference), the full-contact conditioner was used across all zones.
- In Case II (Proposed), the full contact (inner + outer) was used only in Zone 1 (50-114 mm), while only the inner conditioner was engaged in Zones 2-5 (114-370 mm), which encompasses the critical wafer-pad contact area (150-350 mm).
Marathon CMP Experiment:
- Polishing pads were conditioned continuously, followed immediately by 1-minute CMP tests using 200 mm SiO₂ wafers.
- Pad Profile Measurement: A Pad Measurement System (PMS) was used to measure pad thickness profiles before and after conditioning intervals (1, 2, 3 hours, and subsequently hourly).
- Performance Metrics: MRR and WIWNU were calculated using standard deviation (σ) and average MRR (MRR_avg).
- Failure Criterion: The experiment was stopped when pad grooves wore out in the wafer-pad contact area or when wafer breakage occurred due to excessive pad wear (Case I stopped at 16 h; Case II ran to 20 h).

Commercial Applications

This technology offers significant improvements in process stability and operational costs for industries reliant on high-precision planarization.

Semiconductor Fabrication (Foundries): Directly applicable to critical planarization steps, including:
- Global planarization of thin films.
- Formation of Interlayer Dielectrics (ILD).
- Shallow Trench Isolation (STI).
- Manufacturing of advanced metal wiring (Cu, W) and contacts.
Manufacturing Efficiency and Cost Reduction:
- Extends the usable life of expensive polishing pads by over 50%, reducing consumable costs and replacement frequency.
- Increases tool uptime by minimizing the need for pad replacement and re-qualification.
Quality Control and Yield Improvement:
- Ensures stable CMP performance (low WIWNU and consistent MRR) over extended periods, leading to higher wafer yield, especially for highly integrated devices sensitive to non-uniformity.
Advanced CMP Tool Design: Provides a foundation for next-generation conditioning systems that incorporate dynamic, localized wear control based on real-time pad profile feedback.

View Original Abstract

Chemical-mechanical polishing (CMP) is a process that planarizes semiconductor surfaces and is essential for the manufacture of highly integrated devices. In CMP, pad conditioning using a disk with diamond grit is adopted to maintain the surface roughness of the polishing pad and remove polishing debris. However, uneven pad wear by conditioning is unavoidable in CMP. In this study, we propose a contact-area-changeable conditioning system and utilize it to conduct a preliminary study for improving pad lifetime. Using the conventional conditioning method (Case I), the material removal rate (MRR) decreased rapidly after 12 h of conditioning and the within-wafer non-uniformity (WIWNU) increased. However, the results of conditioning experiments show that when using a contact-area-changeable conditioning system, uniform pad wear can be obtained in the wafer-pad contact area and the pad lifetime can be extended to more than 20 h. Finally, the newly proposed conditioning system in this study may improve the CMP pad lifetime.

Tech Support

Original Source

DOI: https://doi.org/10.3390/app11083521

References

1995 - Observations on Polishing and Ultraprecision Machining of Semiconductor Substrate Materials [Crossref]
2020 - Review on modeling and application of chemical mechanical polishing [Crossref]
2004 - Chemical mechanical planarization for microelectronics applications [Crossref]
2010 - Chemical Mechanical Planarization: Slurry chemistry, Materials, and Mechanism [Crossref]
2010 - Effect of CMP Pad and Slurry to STI and ILD Polishing [Crossref]
2008 - Study on Adhesion Removal Model in CMP SiO2 ILD [Crossref]
2005 - A dishing model for STI CMP process [Crossref]
2004 - Characterizing STI CMP Processes with an STI Test Mask Having Realistic Geometric Shapes [Crossref]