Quantification of Additives in a Commercial CMP Sample Using HPLC With Photodiode Array and Mass Detection

Importance of the Topic

Chemical mechanical polishing slurries are critical in semiconductor manufacturing to achieve uniform wafer surfaces. Precise control of slurry composition influences yield, device performance, and process reliability. Routine monitoring of additives ensures formulation consistency and prevents slurry degradation under mechanical and thermal stress.

Objectives and Study Overview

This study demonstrates a rapid, fifteen-minute analytical workflow for quantifying key chemical additives in a commercial CMP slurry. The method combines high performance liquid chromatography with photodiode array and single quadrupole mass detection to achieve quantitative analysis of UV-active components and qualitative confirmation of challenging analytes.

Methodology and Instrumentation

The workflow uses minimal sample preparation: standards dissolved in ammonium formate buffer; CMP slurry centrifuged at 4 °C, 2200 rpm, and supernatant diluted. Additive standards include citric acid, benzotriazole, polyethylene glycol and phosphoric acid. Analytical conditions are:

• System: Waters Arc HPLC with quaternary pump and column oven at 50 °C
• Column: XBridge BEH 5 μm 4.6×150 mm
• Mobile phase: 10 mM ammonium formate in water (A) and acetonitrile (B) with gradient elution at 0.6 mL/min
• Detection: PDA at 254 nm and ACQUITY QDa mass detector in ESI+ and ESI– modes over 30–1200 Da
• Data processing: Empower 3 chromatography data system

Results and Discussion

All target additives were baseline separated in under fifteen minutes. Retention time repeatability showed RSD < 0.1 % and peak area RSD values of 0.3–10 % for consecutive injections. Calibration curves of individual standards and mixed additives yielded coefficients of determination R2 > 0.995 (with PEG slightly lower at 0.98 due to mixture interactions). Recovery studies in the slurry matrix demonstrated benzotriazole recoveries of 99–101 %, citric acid 78–83 % (matrix suppression), and PEG 81–111 %. Mass detection confirmed identities via characteristic m/z values, improving confidence for low-UV-activity analytes.

Benefits and Practical Applications

• Rapid routine analysis with minimal sample prep
• Combined UV and mass detection for quantitative and qualitative assurance
• Robust column performance compatible with UHPLC without revalidation
• Integrated data acquisition and processing in a single software platform

Future Trends and Opportunities

Opportunities include scaling to sub-2 μm columns for faster separations, automation of sample handling for high throughput, integration with laboratory information management systems, and extension to emerging slurry chemistries. Advances in mass detection sensitivity and data analytics will further enhance quality control in semiconductor manufacturing.

Conclusion

The presented Arc HPLC-PDA-QDa method offers a robust, high-confidence approach for routine quantification of CMP slurry additives. Excellent precision, linearity, and recovery demonstrate its suitability for formulation development and quality control in semiconductor processing.

References

1. Raghavan S, Keswani M, Jia R. Particulate Science and Technology in the Engineering of Slurries for Chemical Mechanical Planarization. Kona Powder and Particle Journal. 2008;26:94–105.
2. Nagendra Prasad Y, Ramanathan S. Role of Amino Acid Adsorption on Silica and Silicon Nitride Surfaces during STI CMP. Electrochemical and Solid-State Letters. 2006;9(12):G337.
3. Toray Research Center Inc. Analysis of Organic Additives in CMP Slurry. Technical Data S00194E.
4. Chempoint. Ludox TM-50 Colloidal Silica Technical Data Sheet.