Characterizing Nanoparticle Element Oxide Slurries in Chemical-Mechanical Planarization Using Single Particle ICP-MS

Topics Covered

Introduction
Analysis of Metallic Nanoparticles
Sample Preparation and Instrumentation
Instrument Validation
Experimental Results
     CeO2 Nanoparticles
     Al2O3 Nanoparticles
Conclusion

Introduction

In the semiconductor and nanoelectronics fabrication industry, the chemical-mechanical planarization (CMP) process is used for smoothing semiconductor surfaces for photolithography using chemical and mechanical forces. Nowadays, CMP slurries containing nanoparticles are used for miniaturization of electronic devices and improvement of productivity.

Determining the presence of larger particles and the size distribution of CMP slurry nanoparticles is a crucial QC process for photolithography as these factors can influence the eminence of silicon wafers. This article outlines the results of a study involving the quantitation and characterization of CMP slurry nanoparticles such as Al2O3 and CeO2.

Analysis of Metallic Nanoparticles

ICP-MS operating in single particle mode (SP-ICP-MS) is a proven analysis method for metallic nanoparticles as it is capable of measuring the dissolved concentration of both analytes and individual nanoparticles. SP-ICP-MS is increasingly used in the detection and measurement of inorganic-based nanoparticles, thanks to its sensitivity, analysis speed, and flexibility.

This technique involves complete ionization of the nanoparticles in an ICP, followed by detection of the ensuing ions by a mass spectrometer. Since the signal intensity depends on the particle size, SP-ICP-MS provides information about particle size, concentration and size distribution.

Measuring single particles using the SP-ICP-MS technique is carried out by diluting the sample to obtain temporal resolution between particles. The mass spectrometer must have the ability to perform extremely rapid measurements in the range of 10 to 50 µs to detect nanoparticles, owing to the 300-500 µs transient signal from a 50 nm nanoparticle (Figures 1 and 2).

Figure 1. Acquiring data faster than the transient signal allows surface area integration of nanoparticle signal.

Figure 2. Signal of one gold 60 nm particle acquired using the NexION 350 ICP-MS operating in single particle mode (50µs dwell time and no settling).

The number of data points that can be acquired by an instrument for a single mass per second is defined as the "transient data acquisition speed," which is an essential parameter for SP-ICP-MS. The PerkinElmer NexION® 350 ICP-MS running in single particle mode can acquire up to 6 million data points/min, without any electronic settling time.

Sample Preparation and Instrumentation

For SP-ICP-MS analysis, sample preparation involves ultrasonication of the sample slurries for five minutes, followed by dilution with laboratory pure Type I water to obtain the final sample for analysis.

Click here to see the NexION® 350 ICP-MS Spectrometer from PerkinElmer

This experiment used a PerkinElmer NexION® 350 ICP-MS. Table 1 summarizes instrumental parameters and conditions of analysis. PerkinElmer Pure commercially available NIST traceable standards were used to build the dissolved calibration curves.

Table 1. Instrumental parameters and conditions for SP-ICP-MS.

Parameter Value
Instrument NexION 350D ICP-MS
Nebulizer PFA Concentric
Spray Chamber Cyclonic
Torch and Injector Quartz Torch and Alumina 2.0 mm ID injector
Power (W) 1600
Plasma Gas (L/min) 15
Aux Gas (L/min) 1.2
Neb Gas (L/min) 1.02
Sample Uptake Rate (mL/min) 0.25
Sample Tubing Orange/Green
Dwell Time (µs) 100
Sampling Time (s) 60

The dissolved calibration curve consisted of one blank and four standards. Two gold nanoparticle standards (50 and 80 nm) from Ted Pella, Inc. were used to establish the particle calibration curve and determine the system transport efficiency, which was then evaluated with the SRM 8013 60 nm gold nanoparticles.

The NexION 350 allows for continuous and rapid data acquisition as it completely eliminates the quadrupole electronics settling time. As a result, the NexION 350 ICP-MS captures every single particle event, thereby providing accurate particle counting.

Instrument Validation

NIST SRM 8013 60 nm Gold Nanoparticles was used as a quality control to perform method validation. Figure 3 shows that the size distribution of this sample is centered around 60 nm, thus validating the accuracy of the methodology.

Figure 3. NIST SRM 8013 Gold 60 nm SP-ICP-MS size distribution graph.

Experimental Results

CeO2 Nanoparticles

Slurry #1 and Slurry #2 both consist of a broad distribution of CeO2 nanoparticles. Figure 4 shows the normalized frequency graph for Slurry #1, revealing a size range of 12-42 nm. The mean size distribution of CeO2 nanoparticles was 22.3 nm. The cumulative graph reveals that nanoparticles with a size of less than 30 nm represent 80% of the sample.

Figure 4. Slurry #1 normalized frequency particle size distribution graph.

Figure 5 presents the normalized frequency graph for Slurry #2, showing a size range of 22-76 nm and a mean size distribution of 47.8 nm. The nanoparticles with a size of less than 62 nm represent 80% of the sample. The observation of an elevated background signal revealed the dissolution of some amount of Ce into the solution. Table 2 summarizes the results for both slurry samples.

Figure 5. Slurry #2 normalized frequency particle size distribution graph.

Table 2. Results for CeO2 Slurry Samples

Sample ID Mean (nm) Size Median (nm) Size Particle Conc. (particles/mL) Dissolved Conc. (ppb)
CeO2 Slurry #1 22.3 21.4 196821 <0.01
CeO2 Slurry #2 47.8 47.3 267029 0.13

Al2O3 Nanoparticles

Figure 6 shows the normalized frequency graph for Slurry #3 containing Al2O3 nanoparticles, showing a size range of 28-58 nm and a mean size distribution of 44.4 nm. The cumulative graph reveals that the nanoparticles with a size below 54 nm represent 80% of the sample, of which half of the nanoparticles are less than 48 nm. The two dominant sizes are 28-38 nm and 50 nm.

Figure 6. Slurry #3 normalized frequency particle size distribution graph.

The normalized frequency graph for an Al2Onanoparticle slurry appears in Figure 7. The results show a size range of 22-46 nm and a mean size distribution of 32.5 nm. The cumulative graph reveals that nanoparticles smaller than 38 nm represent 80% of the slurry sample. Table 3 summarizes the results for both Al2O3 slurries.

Figure 7. Slurry #4 normalized frequency particle size distribution graph.

Table 3. Results for Al2O3 Slurry Samples

Sample ID Mean (nm) Size Median (nm) Size Particle Conc. (particles/mL) Dissolved Conc. (ppb)
Al2O3 Slurry #3 44.4 47.2 122536 <0.02
Al2O3 Slurry #4 32.5 31.8 148960 <0.02

Conclusion

The results of this study clearly demonstrate the ability of the NexION 350D ICP-MS to accurately characterize nanoparticles when operating in single particle mode. Method validation was performed with NIST-certified nanoparticles, and the methodology was then applied to CeO2 and Al2O3 slurries.

Despite not being pure metallic particles, SP-ICP-MS is easily able to measure and characterize metal oxide particles, yielding information about particle size, size distribution, and concentration.

This information has been sourced, reviewed and adapted from materials provided by PerkinElmer Inc.

For more information on this source, please visit PerkinElmer Inc.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Submit