Direct Analysis of Metallic Impurities in SiC and GaN Wafers by LA-GED-MSAG-ICP-MS/MS

Importance of the Topic

This application note addresses the growing need for precise impurity analysis in wide-bandgap semiconductor materials such as silicon carbide (SiC) and gallium nitride (GaN). These materials enable higher breakdown voltages and superior thermal performance in power electronics used for electric vehicles, renewable energy inverters, and high-frequency communications. Substrate purity directly impacts device reliability and efficiency, demanding analytical techniques capable of detecting trace metal contamination at ultratrace levels.

Objectives and Study Overview

The primary goal was to develop and demonstrate a fully automated, direct-analysis workflow for 12-inch semiconductor wafers using a combination of laser ablation, gas exchange, aerosol standard generation, and triple quadrupole ICP-MS. Key objectives included:

• Overcome the size and sensitivity limitations of traditional vapor phase decomposition (VPD) and conventional LA-ICP-MS.
• Establish quantitative calibration by the method of standard addition (MSA) without requiring matrix-matched solid standards.
• Validate the technique on spiked Si wafers, research and dummy SiC wafers, and bulk and film GaN wafers.
• Demonstrate spatially resolved impurity mapping and single nanoparticle detection capabilities.

Methodology and Instrumentation

The analysis platform integrates four key components:

• Femtosecond laser ablation system with galvo mirror and x-y-z-θ stage for high-speed, large-area sampling of whole 12" wafers.
• Gas Exchange Device (GED) to replace ambient air with argon carrier gas, ensuring efficient transport of ablated particles to the ICP-MS.
• Metal Standard Aerosol Generation with dual-syringe pumps (MSAG-DS) to deliver precise volumes of aqueous calibration standards and blanks, enabling dynamic standard addition at constant total flow.
• Agilent 8900 ICP-QQQ in MS/MS mode for effective removal of spectral interferences and enhanced sensitivity. Automated switching between ammonia and hydrogen reaction gases optimizes interference control for different analytes.

Main Results and Discussion

Spiked Si wafer experiments demonstrated sharp, transient signals for 30 elements, confirming the spatial resolution and high sensitivity of the LA-GED-MSAG-ICP-MS/MS approach. Quantitative analysis of SiC wafers revealed sub-ppb to tens of ppb levels of impurities such as Cu, Ag, Sn, and Fe, with clear differences between research and dummy grades. In GaN wafers, bulk single-crystal and thin films showed distinct impurity profiles, with matrix effects effectively compensated by standard addition calibration. Single nanoparticle mode enabled detection of lead particles down to an equivalent diameter of 10 nm, illustrating the system’s capability for particulate contamination assessment.

Benefits and Practical Applications

This technique offers several advantages for semiconductor fabrication and quality control:

• Direct whole-wafer analysis without the need for large ablation chambers or destructive wet-chemical preconcentration.
• Quantitative accuracy via liquid standard addition, eliminating dependence on matrix-matched solid standards.
• Spatially resolved impurity mapping and nanoparticle detection for defect and particle contamination studies.
• High throughput and automation using FOUP integration, wafer transfer robotics, and preset ablation patterns.
• Compatibility with a wide range of wafer materials beyond silicon, including SiC and GaN.

Future Trends and Applications

Emerging opportunities include:

• Integration of real-time data analytics and machine learning for automated defect recognition and process control.
• Extension of the workflow to other advanced substrates, heterostructures, and composite materials.
• Further miniaturization of particulate detection limits through optimized laser parameters and reaction gas chemistries.
• Seamless integration into semiconductor fab metrology lines for inline monitoring of wafer purity.

Conclusion

The LA-GED-MSAG-ICP-MS/MS methodology delivers a powerful, fully automated solution for direct impurity analysis of large-area semiconductor wafers. By combining femtosecond laser ablation, efficient gas exchange, dynamic aerosol standard addition, and triple-quadrupole ICP-MS, this approach achieves ultratrace detection limits, spatial mapping, and nanoparticle analysis without the shortcomings of traditional VPD and LA-ICP-MS techniques. It represents a valuable tool for quality assurance and process development in advanced semiconductor manufacturing.

References

• Ichinose T., Kawabata K., Sakai K. Automated Surface Analysis of Metal Contaminants in Silicon Wafers by Online VPD-ICP-MS/MS; Agilent Technologies Publication 5994-6135EN; 2021.
• Halicz L., Günther D. Quantitative Analysis of Silicates Using LA-ICP-MS with Liquid Calibration. Journal of Analytical Atomic Spectrometry 2004, 19, 1539–1545.
• Suzuki K., Nishiguchi K., Kawabata K., Yamanaka M. Analysis of Metallic Impurities in Specialty Semiconductor Gases Using GED-ICP-MS; Agilent Technologies Publication 5994-5321EN; 2020.