Launch of High-Sensitivity Analysis Service for Impurity Nanoparticles in Chemical Solutions to Support Expanding Semiconductor Demand in the AI Era

【Abstract】 Toray Research Center, Inc. (Headquarters: 1-7-2 Nihonbashi Honcho, Chuo-ku, Tokyo; President: Yoshiki Makabe; hereinafter "TRC") has developed a technology capable of analyzing the particle concentration and size distribution of impurity metal nanoparticles—a significant issue in semiconductor manufacturing processes—with the highest level of sensitivity in Japan. This was achieved by highly optimizing the sample preparation methods and measurement conditions for Single Particle Inductively Coupled Plasma Mass Spectrometry (spICP-MS*1, *2). Leveraging this technology, TRC is launching a contract analysis service targeting nanoparticles in chemicals used for semiconductor manufacturing.

As semiconductor miniaturization progresses, nanometer-sized*3 particles (nanoparticles) can cause defects such as open circuits or short circuits, even in extremely small quantities. Consequently, managing the purity of chemicals used in manufacturing processes has become increasingly critical for ensuring production yields.

spICP-MS is a high-sensitivity analysis method that detects metal nanoparticles in liquids one by one, allowing for the simultaneous measurement of particle size and number concentration. By constructing a proprietary analysis system that maximizes the features of this method, TRC has enabled the detection and quantitative evaluation of trace impurity nanoparticles in chemical solutions. This service is expected to contribute to the further purification of chemicals and the improvement of yields in advanced semiconductor manufacturing.

【Background】 As semiconductors become higher-performing and more energy-efficient, device miniaturization is accelerating. Meanwhile, contamination by metal nanoparticles is viewed as a major factor causing open and short circuits in integrated circuits. Various chemicals, including photoresists*4, acids, and organic solvents, are used in semiconductor manufacturing processes. The impact of impurity metal nanoparticles contained in these chemicals on device reliability and yield cannot be ignored, creating a strong demand for technology that can evaluate trace metal nanoparticles in chemicals with high sensitivity and quantitative accuracy.

spICP-MS is an analytical method capable of detecting metal nanoparticles dispersed in liquids at concentrations as low as a few ppt*5 (one part per 100 billion), making it effective for evaluating impurity metal nanoparticles in chemicals. However, challenges have included noise signals originating from reagents used in sample preparation, metal ions or contaminants in the sample, and contamination within the measurement equipment, as well as sensitivity fluctuations due to minor differences in measurement conditions, which have made the detection of trace, minute particles difficult.

【Technology and Analysis Example】 To address these challenges, TRC has constructed an analysis system that maximizes the features of spICP-MS by combining the use of high-purity solvents for sample preparation, operational management to keep the equipment in a clean state, and the optimization of measurement conditions according to the specific chemicals and particles being analyzed. As a result, it has become possible to detect and evaluate extremely trace amounts of metal nanoparticles present in semiconductor manufacturing chemicals with high sensitivity and stability.

Figure 1 shows an example demonstrating the effectiveness of this technology: the confirmation of impurity metal nanoparticle levels in Propylene Glycol Monomethyl Ether Acetate (PGMEA*6), which is used as a solvent for photoresists and as a cleaning agent in semiconductor manufacturing processes, using spICP-MS.

When commercially available high-purity PGMEA was measured as-is, numerous metal nanoparticles such as aluminum (Al) and iron (Fe) were detected. This result indicates that metal nanoparticles exist as impurities even in chemicals generally considered "high-purity." When sample preparation is performed using this PGMEA, it is difficult to determine whether the detected particles originated from the sample or the PGMEA used for preparation. On the other hand, when the same analysis was performed after purifying this solvent, metal nanoparticles were almost no longer detected, demonstrating that the use of appropriately purified reagents enables more sensitive and reliable impurity evaluation. The Fe particles detected in the PGMEA before purification corresponded to a concentration of 0.04 ppt in the solvent, achieving extremely high detection sensitivity.

Furthermore, this method allows for the evaluation of metal nanoparticles contained in resins by dissolving them in an appropriate solvent, which is also effective for estimating contamination sources in chemicals derived from solid resins. The analysis of trace metal nanoparticles in resins became possible only by using a solvent that had been purified to a high degree for dissolving the resin. Moving forward, TRC will expand the scope of application to various solvents and materials to meet an even wider range of analytical needs.

[Figure 1: Analysis results of metal impurity nanoparticles in PGMEA]

【Future Outlook】 Against the backdrop of expanding demand for products such as semiconductors for AI and data centers, the sophistication of semiconductor manufacturing technology is expected to continue. This technology is expected to be utilized in a wide range of scenarios, from research and technical development to mass production processes, as a means to evaluate impurity nanoparticles in chemicals with high sensitivity.

Based on the fundamental philosophy of "contributing to society through advanced technology," TRC will continue to advance its micro-particle detection technology and expand the range of supported samples, thereby supporting the resolution of customer challenges by meeting diverse analytical needs, including those in the semiconductor industry.

【Glossary】 *1 Inductively Coupled Plasma Mass Spectrometry (ICP-MS) A method that ionizes a sample solution with plasma and measures the types and amounts of metal elements contained therein using a mass spectrometer. It possesses the highest level of sensitivity among modern analytical methods and excels at detecting and measuring metal impurities at extremely low concentrations.

*2 spICP-MS Abbreviation for Single Particle ICP-MS. It uses an ICP-MS device to detect the types and amounts of metal elements contained in each individual particle in a liquid. The volume of each particle is determined from the amount of metal elements, and the particle size is calculated assuming a spherical shape. Additionally, the particle concentration in the sample liquid is calculated from the frequency at which particles are detected.

*3 Nanometer A unit of length (nm). 1 nm is equivalent to one-millionth of a millimeter.

*4 Photoresist A liquid chemical consisting of polymer resin, photosensitive material, and solvent. It has the property of changing chemically in response to light and is used to form circuit patterns in semiconductor manufacturing processes.

*5 ppt A unit of concentration. Abbreviation for Parts Per Trillion, meaning one part per trillion. 1 ppt is equivalent to one part per 100 billion of 1%.

*6 PGMEA Propylene Glycol Methyl Ether Acetate. A colorless, transparent solvent also used in inks and dyes in fields other than semiconductor manufacturing.