ICP-MS for Elemental, Particle Analysis in Semiconductor Testing

 ICP-MS for Elemental, Particle Analysis in Semiconductor Testing

Inductively coupled plasma mass spectrometry (ICP-MS) is an elemental analysis technology.  It can detect most of the periodic table of elements at milligram to nanogram per liter concentrations, making it one of the most important, powerful, and versatile spectrometric techniques.

ICP-MS is widely used because it provides extremely low detection limits for nearly all the elements it can measure—almost the entire periodic table. The only elements ICP-MS can’t measure are hydrogen, helium, argon, nitrogen, oxygen, neon, and fluorine. However, even fluorine can be analyzed indirectly using a triple quadrupole ICP-MS, showcasing the comprehensive elemental coverage of ICP-MS.

ICP-MS can detect many naturally occurring elements below 0.1 part per trillion—equivalent to one drop of water (50 µL) in 200 Olympic-sized swimming pools (500 million liters).1 However, ICP-MS can also measure elements at concentrations of up to 100s or even 1000s of parts per million. One thousand ppm is 0.1%, and the concentration range from 0.1 ppt to 0.1% covers 10 orders of magnitude.1 This wide range of detection allows ICP-MS to be used in various industries, from environmental monitoring to semiconductor testing, making it a versatile and powerful tool.

From routine environmental monitoring, consumer product testing, and food and pharmaceutical safety applications to life science and clinical research, mining and metals analysis, geochemistry, nuclear and petrochemicals, ICP-MS analysis is critical. Its role becomes even more pronounced in semiconductor testing and manufacturing, where measuring ultra-trace impurities and controlling particle contaminants is paramount.

Semiconductor testing

Semiconductors are at the core of modern electronics and are used in everything from smartphones to automobiles. As the semiconductor industry evolves, devices are becoming smaller, faster, more reliable, and more powerful. However, as components shrink to the single-nanometer scale, contamination monitoring and control becomes ever more crucial because even ultra-trace contaminants can reduce manufacturing yields, cause product reliability issues, or lead to product failures. The semiconductor industry estimates suggest that contamination accounts for around 50% of yield losses.2

Analysis of impurities in semiconductors and electronics must be done at each stage of the manufacturing process and on everything from testing wafers, raw materials, and process chemicals to QA/QC of final products.

Impurities analysis

Monitoring and controlling trace element contamination begins with the high-purity wafer substrate. The substrate is usually silicon, but other materials such as silicon carbide, silicon nitride, and gallium arsenide can also be used.

ICP-MS can measure trace metallic contamination in bulk silicon after dissolving the silicon in hydrofluoric acid. Trace metals in the sliced wafer are measured using a surface analysis technique—such as vapor phase decomposition (VPD)—where the metals are extracted from the silicon substrate into a droplet that ICP-MS then analyzes.

In addition to the high-purity wafer substrate, the purity of chemicals used throughout the wafer fabrication process must be controlled to avoid the introduction of contaminants. Metallic contaminants are of concern because they can affect the electrical properties of the finished device by, for example, reducing dielectric breakdown voltage.

The concentration of trace metals on the wafer surface must be determined to ensure that metal contaminants do not adversely affect integrated circuits. The bare silicon layer on the surface of the wafer quickly oxidizes to silicon dioxide (SiO2) when exposed to atmospheric oxygen and water. This naturally oxidized layer is ~0.25 nm thick. If the integrated circuit (IC) design requires an insulating film, a much thicker oxide layer is formed on the wafer surface by heating the wafer between 900 to 1200°C in the presence of O2 or water vapor. This thermally oxidized layer may be up to 100 nm thick. For both native and thermally oxidized SiO2, the trace metals in the oxide layer can be measured at extremely low concentrations using VPD coupled with ICP-MS.2

VPD-ICP-MS is a proven method of measuring trace metal contamination in silicon wafers. The VPD wafer sampling approach has good sensitivity because it concentrates the metals in the oxide layer from a large wafer surface area into a single droplet of solution for measurement. VPD can be performed manually, although it takes an experienced operator to consistently recover the dissolved metals in the SiO2 layer.2 Automating the VPD process ensures consistency and reduces the potential for contamination.

For example, the Agilent 7900 and 8900 ICP-MS instruments can be integrated with VPD systems for the automated analysis of metallic impurities in silicon wafers. Both systems provide the good matrix tolerance required for analysis of thermally oxidized SiO2, where the SiO2 matrix concentration can be up to 5000 ppm in the extraction.2 The 8900 has the added benefit of MS/MS operation, providing the most effective interference removal of any ICP-MS, delivering lower detection limits and improved accuracy.2

Subscribe to our e-Newsletters!
Stay up to date with the latest news, articles, and events. Plus, get special offers from Labcompare – all delivered right to your inbox! Sign up now!

Particle analysis

Monitoring particulate contamination is also fast becoming a priority for the semiconductor industry. Contamination from metallic particles such as iron, chromium, and nickel, which can be released from IC processing equipment, is of particular concern. Single-particle ICP-MS (spICP-MS) can determine the number, size, size distribution, and elemental composition of multiple elemental nanoparticles in process chemicals and baths. The technique provides more comprehensive particle characterization than traditional particle counting methods while also determining the dissolved element concentrations in the same rapid acquisition.3

There is growing awareness that metallic nanoparticles (NPs), especially iron NPs, can lead to defects on the surface of wafers. Even small amounts of impurity in processing reagents can affect product yield and reliability. spICP-MS is increasingly being used to characterize these metallic nanoparticles.

In a recent application3, Agilent researchers and collaborators in Taiwan used the spICP-MS method to measure isopropyl alcohol, propylene glycol methyl ether acetate, and propylene glycol monomethyl ether spiked with 15 nm Fe2O3 NPs. The data showed the NP signals were clearly separated from the background signals, and the ratio of signal intensities of NPs was accurate and acceptable. Additionally, both the detected particle number and the size were constant over a 10-hour period.3

After a 6-month period the samples were remeasured ans the signals from 30 nm Fe NPs were clearly observed. The shape of the size distribution was recorded as very similar to the initial data in a fresh solution. The results show that Fe NPs are stable in isopropyl alcohol, suggesting they don’t dissolve into or precipitate out of the solvent. Ultimately, the authors concluded that the 8900 spICP-MS method satisfies the emerging needs of the semiconductor industry to monitor low concentrations of small-sized particles in high-purity solvents.3

“By combining particle counting, sizing, and elemental identification in a single measurement, spICP-MS solutions enable semiconductor manufacturers to implement more effective contamination control ensure quality and support the industry's continued evolution toward smaller, more complex devices,” says Yuri Tanaka, ICP-MS Product Marketing Manager for Atomic Spectroscopy Division, Agilent Technologies.

References

  1. Agilent Technologies. (2025) ICP-MS Overview https://www.agilent.com/en/product/atomic-spectroscopy/inductively-coupled-plasma-mass-spectrometry-icp-ms/what-is-icp-ms-icp-ms-faqs
  2. Agilent Technologies (2022). Measuring Inorganic Impurities in Semiconductor Manufacturing. Retrieved from Agilent Technologies website: https://www.agilent.com/cs/library/applications/appcompendium-semiconductor-icp-ms-5991-9495EN-us-agilent.pdf
  3. Hsu, D., Shimamura, Y., Mizobuchi, K., Liao, B., Wang, K., Su, C., Hsu, C.H. (2020). Analysis of 15 nm Iron Nanoparticles in Organic Solvents by spICP-MS. Retrieved from Agilent Technologies website: https://www.agilent.com/cs/library/applications/application_fe_nanoparticles_8900_icp-ms-5994-1747en_us_agilent.pdf
  • <<
  • >>