Mark Kelinske

Gas chromatography (GC) coupled to inductively coupled plasma–mass spectrometry (ICP-MS) is a powerful and flexible analytical solution that is well-established for measuring organometallic compounds such as organotin, organomercury, and organolead in environmental samples, foodstuffs, and consumer products. GC–ICP-MS is also used for industrial applications, such as monitoring catalyst poisons and environmental contaminant elements in petrochemical processing and plastics manufacturing. The semiconductor industry uses a range of high-purity specialty gases and volatile liquids as precursors in wafer substrate production, and for processes such as plasma etching and deposition of thin films by chemical vapor deposition (CVD). Semiconductor manufacturers are constantly developing new integrated circuit (IC) chips with smaller sizes, higher speeds, lower power consumptions, and greater transistor densities. This trend means that contaminants must be controlled at even lower levels in process chemicals and precursors. In this article, we show how GC–ICP-MS, particularly using triple quadrupole or tandem ICP-MS (ICP-MS/MS), enables determination of the lowest levels of contaminants in the specialty gases and volatile liquids used to make the most advanced electronic devices.

Integrated circuit (IC) chips play an increasingly important role in modern life, supporting technologies from consumer electronics and self-driving vehicles to the high-speed communications and servers used in accelerated computing and artificial intelligence (AI). The manufacturing of advanced semiconductor chips requires high-purity materials, metals, and a range of process chemicals, such as ultrapure acids and solvents. Wafer manufacturers and chip fabrication plants (FABs) also use a variety of bulk and electronic specialty gases (ESGs) and volatile liquids throughout the manufacturing process (1,2). For example, silane (SiH

) and chlorosilanes are used as feedstocks for gas purification of silicon to the nine to eleven nines purity used to make semiconductor wafers. Nine nines (9N) Si is 99.9999999% pure, meaning there is only 1 part per billion (ppb) of total impurity elements in the solid silicon crystal. Alternative, non-silicon wafers and films are often made from “III-V compound” semiconductors, such as gallium arsenide (GaAs) and indium GaAs (InGaAs), which may use arsine gas (AsH3) as a precursor together with trimethyl gallium (TMG).

Fabricating an advanced IC chip involves upwards of a thousand steps where thin layers of conducting and insulating material are alternately deposited and etched to create the millions of transistors and other features that make up the circuit. Dopant elements are added to modify the conductivity of the silicon substrate or surface film. High-purity gases are used extensively in these “front end” IC fabrication processes. For example, phosphine (PH

) is used to dope the required level of phosphorus into thin-film transistor (TFT) displays. Diborane (B

) is used for boron doping and ion implantation, for growing metal-boron thin films, and as a component of chemical vapor deposition (CVD) of boron-phosphate-silicate-glass (BPSG) films. Fluorocarbon (FC) gases, such as 2-fluorobutane and hexafluoro butadiene, are also used in the electronics industry (for example, in dry etching and for plasma enhanced chemical vapor deposition [PECVD]). The high reactivity and low toxicity of some FC gases means they are also used to clean reaction chambers between fabrication processes. In a typical semiconductor FAB, high-purity gases can make up the second largest material cost after high-purity silicon.

Controlling the level of contaminants in specialty gases is critical to ensuring acceptable chip yield and the performance and reliability of the finished semiconductor device. To understand why contamination is so important, it is useful to consider that advances in semiconductor device technology are driven by reductions in scale. Packing more transistors into a smaller space improves the speed and performance of the chip. The latest chips have circuit features and line widths measured in single nanometers (nm, one millionth of a millimeter), and the presence and position of individual atoms in the silicon crystal lattice can affect the electrical properties of the device. This makes the chips highly susceptible to contamination, so controlling the levels of impurities in the high-purity gases used during fabrication is fundamental to the production process.

Controlling the levels of dopant elements and metallic contaminants in substrate materials, process chemicals, and gases, is a particular industry concern. Dopant elements are deliberately added to the silicon substrate to make it “semiconducting” and control the electrical properties, as illustrated in Figure 1. Silicon has four valence electrons, and pure silicon forms a crystal lattice in which all these electrons are shared with the adjacent silicon atoms. As a result, pure silicon is a poor conductor of electricity. N-type (negative) dopants are Group V elements such as phosphorus, arsenic, antimony, and bismuth, which have five valence electrons, so they add one free electron to the silicon lattice. P-type (positive) dopants are Group III elements, which have three valence electrons, so they contribute a “missing” electron or a positive “hole” in the lattice. The level of doping varies depending on the intended properties, but low-level doping can add as few as one dopant atom per 100 million silicon atoms, so controlling contamination for these elements is critical.

FIGURE 1: Schematic showing: (a) Pure silicon crystal lattice; (b) silicon doped with phosphorus, an 

-type dopant element; (c) silicon doped with boron, a 

ICP-MS instruments are already widely used in the semiconductor industry to monitor the levels of ultratrace contaminants in high-purity materials and process chemicals. Contaminants in high-purity gases often need to be controlled at lower levels than can be achieved using traditional gas chromatography (GC) detectors, which may not have sufficient selectivity for the required analysis to begin with. Also, for some matrices and compounds, standards are unstable or not available, making calibrating and quantifying contaminants difficult.

As a detector for GC, ICP-MS, particularly with a triple quadrupole or tandem mass spectrometer (MS/MS), can overcome many of the limitations of other GC detectors. GC–ICP-MS can detect and quantify almost every element at low levels in gases and volatile liquids. The technique gives element and isotope-specific analysis, and is highly selective. As a result, matrix components and other compounds have little to no impact on the signal for the target analyte. GC–ICP-MS also gives practically the same elemental signal regardless of the compound, so non-specific or “compound independent” calibration (CIC) can be used, simplifying the quantification of compounds that are unstable or for which standards are not available. GC–ICP-MS is a versatile and sensitive technique that enables semiconductor manufacturers to reliably quantify trace-level contaminants in the wide range of specialty gases and volatile liquids used in semiconductor manufacturing.

Examples of different electronic gas analysis applications are presented in this work. The details of the specific GC columns and methods used for the different sample types and compounds measured are provided in the discussion. Specialty electronic gas analysis using GC–ICP-MS often requires a special approach to sample introduction and analysis, as many of the gases are toxic, flammable, and pyrophoric (spontaneously combust on contact with air or moisture). Silicon compounds such as silanes, siloxanes, and chlorosilanes decompose to form silica, which would deposit in a standard GC injector. Flammable and pyrophoric gases can be analyzed directly from the cylinder, using a switch- ing valve and loop to deliver the required sample volume directly to the column in the GC injector port. A 10-port sampling valve provides the option to automatically add a second gas as an online internal standard, as illustrated in Figure 2. In a similar way, volatile liquids such as trichlorosilane (TCS) can be sampled and injected directly into the GC–ICP-MS system using a high-pressure switching valve (3).

FIGURE 2: Direct sampling of flammable and pyrophoric gases for analysis by GC–ICP-MS, using a 10-port valve to allow online addition of an internal standard to the sample gas. Showing the (a) load and (b) inject sequences.

A further consideration when GC– ICP-MS is used to analyze samples such as silanes, chlorosilanes, AsH

) is that these compounds are inorganic. This means the matrix is not easily decomposed and volatilized in the ICP instrument, unlike with hydrocarbon matrices. To successfully analyze these inorganic matrices, a low-volume Deans Switch microfluidic flow control device was used to divert the matrix to vent, preventing excessive matrix load from reaching the ICP. The Deans Switch was positioned in the GC oven in the flow path between the column exit and the ICP-MS transfer line. Once the retention time of the matrix peaks were established using a low volume injection, the Deans Switch was used to divert the matrix to vent during the GC separation.

Both single and triple quadrupole (QQQ) Agilent ICP-MS systems were used for elemental detection. ICP-QQQ is a tandem MS configuration (MS/MS); as a result, it supports the use of reactive cell gases to give more effective control of spectral overlaps. In many cases, analysis can be performed on either type of mass spectrometer, with ICP-MS/MS typically giving detection limits (DLs) that are approximately an order of magnitude lower or more, depending on the element. Because the ICP-MS plasma is “dry” (no aqueous solution), many spectral backgrounds are much lower with GC–ICP-MS than with the conventional liquid sample introduction. As a result, non-typical ICP-MS elements, such as silicon and sulfur, can be measured at their major isotopes using single quadrupole ICP-MS. However, lower DLs are possible if an ICP-MS/MS method is used with a reactive cell gas to further reduce spectral overlaps. For more typical ICP-MS elements, such as germanium and arsenic, part per trillion (ppt) level DLs are routinely achievable. As a multi-element detector, GC–ICP-MS can be used to analyze hydride gas contaminants such as SiH

 at trace level (single ppb down to 10 ppt DLs) in other ultrapure hydride gases (4).

The sulfur compounds, hydrogen sulfide (H

S) and carbonyl sulfide (COS), were determined as ppb-level contaminants in PH

 using a 60-m (2 x 30 m) x 0.53 mm i.d. x 20 μm film mega-bore Plot U GC column at 50 

C with helium carrier gas at 17.5 psig. Sulfur was measured as the 

P matrix peak. The chromatogram is illustrated in Figure 3, demonstrating the complete separation of the sulfur compounds from the PH

 gas. The shaded region of the time plot shows when the PH

 matrix was diverted to vent using the Deans Switch.

FIGURE 3: GC–ICP-MS/MS analysis of trace sulfur compounds in phosphine (PH

 matrix to vent, avoiding excessive matrix loading to the ICP. Sulfur compounds at 25 ppb each. ICP-MS/MS compound DLs were approximately 1.5 ppb.

 matrix is difficult for organic MS as the two compounds both have their large molecular ion peak at 

 34 and their fragmentation patterns are similar. As illustrated in Figure 4, the only major difference in the GC–TOF-MS spectra is the 

 31, which is not useful for quantifying sulfur. Although H2S is separated from the PH

 matrix peak chromatographically, the tail of the matrix peak causes a raised background for H

S. By contrast, low-level determination of H

 matrix is straightforward using ICP-MS, which is a mass-specific elemental detector that determines 

 when measured by GC–TOF-MS. Mass-specific elemental detection of 

P, enabling interference-free, trace-level analysis of H

 is a critical high-purity process gas, in which sulfur compounds and other hydride contaminants must be monitored at trace levels. However, PH

 also needs to be measured as a trace-level contaminant in other gases and volatile liquids used in the semiconductor industry, to avoid unwanted contamination by phosphorus, an 

-type dopant element. Dichlorosilane (DCS) is used as a precursor for high-purity silicon crystal growth, and in combination with ammonia (NH

) to grow silicon nitride (SiN) thin films by CVD. The purity of the silicon crystal or SiN film depends on the purity of the precursor gases, so contaminants including PH

 must be monitored at ppb levels in the DCS. PH

 was determined in DCS by GC–ICP-MS/MS using a 60 m x 0.32 mm x 5.0 μm film DB-1 column using helium carrier gas at 20 psig. The oven program had a starting temperature of 35 

C/min ramp and a final temperature of 220 

Figure 5 shows overlaid chromatograms for a PH

 being determined as a trace contaminant in DCS. The PH

 contaminant was quantified at 4.8 ppb. The use of ICP-MS/MS with O

 47]) enabled low-level phosphorus analysis with minimal contribution from the silicon matrix.

FIGURE 5: Overlaid GC–ICP-QQQ chromatograms showing trace PH

 contaminant quantified at 4.8 ppb in DCS matrix, overlaid with PH

GC–ICP-MS was used to measure higher boranes as trace impurities in high-purity diborane (B

) using a 60 m x 0.32 mm x 1.5 m Rtx-Volatiles column. The oven program used a 35 

C/min ramp and a final temperature of 240 

C. Tolerances for diborane purity are extremely tight (<1% variation), so accurate measurement of higher boranes is essential for production control. Standards are not available or stable for higher boranes, but GC–ICP-MS provides a solution, as illustrated in Figure 6. The chromatogram shows the separation and quantification of tetraborane (B

 matrix. The higher boranes were quantified at single ppm levels in 2% B

 using CIC against a diborane standard. Quantification by CIC is based on the consistent response for elemental boron and a simple mathematical correction for the different amounts of boron in the different compounds. ICP-MS also offers isotope specific measurement, which is beneficial in this application, as some uses of borane require B

 that is isotopically enriched up to 99.99% B-11. The 

B isotopic abundances can easily be measured using GC–ICP-MS.

FIGURE 6: Tetra and pentaborane quantified at single ppm levels in 2% B

 sample using compound independent calibration with B

ICP-MS and ICP-QQQ are well-established techniques that are widely used in the semiconductor industry for measuring ultratrace levels of elemental contaminants in wafer substrate materials, metals, and process chemicals. Combining ICP-MS with GC extends the use of the technique to monitoring contaminants in bulk, specialty gases, and volatile liquids. These sample types require special GC sampling approaches, as many of the samples are toxic, flammable, or pyrophoric. Such samples can be handled safely using an appropriate switching valve or loop injection gas sampling system.

As a sensitive, multi-elemental detector for GC, ICP-MS gives low detection limits, mass-specific measurements, and excellent selectivity. GC coupled to a triple quadrupole ICP-MS system enables trace-level analysis of compounds containing elements traditionally considered difficult to determine by ICP-MS, such as silicon and phosphorus. The high temperature of the ICP ion source means that ICP-MS gives nearly equal elemental response regardless of the compound’s composition. This enables the use of compound independent calibration, which is useful when 

compound-specific calibration standards are unstable or unavailable.

Trace Analysis of Specialty and Electronic Gases

(2) Linde. Electronic Special Gases Index. 

High-Pressure Liquid Injection Device for Agilent Gas Chromatographs

(4) Geiger, W. M.; McElmurry, B.; Anguiano, J.; Kelinske, M. 

Determination of Trace Impurities in Electronic Grade Arsine by GC-ICP-QQQ

 is with Consolidated Sciences (CONSCI) Ltd., in Pasadena, Texas. 

 are with Agilent Technologies, Inc. Direct correspondence to 

Articles

Manufacturing advanced electronic devices requires the production of high-quality semiconductors and integrated circuit chips. In this article, the authors explain how GC, when coupled with ICP-MS, enables the detection of elements that are essential in semiconductor production.

GC-ICP-MS for Process and Quality Control in Semiconductor Manufacturing

There are well recognized difficulties in measuring nanoparticles (NPs) accurately by conventional single-quadrupole inductively coupled plasma–mass spectrometry (ICP-MS), because many natural and engineered NPs are based on elements that are challenging to measure at trace level by ICP-MS, such as iron, titanium, and silicon. This month's "Atomic Perspectives" column discusses how complex isobaric and polyatomic spectral interferences can be mitigated using triple-quadrupole ICP-MS (ICP-MS/MS) utilizing a collision-reaction cell (CRC), allowing for the multielement characterization and detection of smaller nanoparticle sizes.

	Metal-based engineered nanoparticles (NPs) are widely used in industrial and technological processes, medical applications, foods, and personal care products. NPs exhibit unique physical and chemical properties compared to bulk chemicals, because of the large surface area of particles with one or more external dimensions ranging in size from 1 to 100 nm. While scientists and specialty chemical suppliers are beginning to harness these novel properties to develop new applications, much remains unknown about the impact of NPs on biological and aquatic systems. Given the potential for NP release during production, use, or disposal, reliable measurement methods are required to assess the concentration of NPs in a range of samples.

	Single-particle inductively coupled plasma-mass spectrometry (SP–ICP-MS) is fast-becoming an accepted technique for the characterization of NPs in liquid samples (1). The technique provides valuable information on the mean size, size distribution, and number of nanoparticles in the sample, as well as the concentration of the element dissolved in the solution. A number of early studies used conventional single-quadrupole ICP-MS instruments to characterize simple, spherical nanoparticles composed of a single element, such as silver, gold, or zirconium (2,3). However, large quantities of NPs are manufactured from various oxides of iron, zinc, titanium, and silicon. Because these NPs are composed of elements that are much more difficult to measure at trace levels by conventional ICP-MS, and because of overlaps from polyatomic interferences, SP-ICP-MS methods have been developed using triple quadrupole ICP-MS (ICP-MS/MS) (4).

	In ICP-MS/MS, an additional mass filter (Q1) is located before the instrument's collision–reaction cell (CRC), enabling a double mass selection capability known as 

. Q1 controls the ions that can enter the cell, ensuring that the reaction processes in the cell are controlled and consistent. This extra selection step enables the more predictable and reliable removal of interferences on elements such as aluminum (Al), sulfur (S), and iron (Fe), using reactive cell gases, compared to conventional ICP-MS (5).

	When using reactive gases, ICP-MS/MS avoids isobaric and polyatomic interferences by utilizing either on-mass or mass-shift mode. In on-mass mode, the target isotope is measured at its original mass, and the interference or interferences are shifted to a different mass. In mass-shift mode, the target isotope is measured as a reaction product ion at a different mass, free from the interference. This is exemplified in Figure 1, which shows how ICP-MS/MS on-mass mode avoids polyatomic interferences from argon ion (ArO

56 via proton transfer with ammonia in the CRC. Figure 1 also demonstrates how ICP-MS/MS mass-shift mode for titanium (

Ti) avoids an isobaric interference from 

 48 to pass to the cell, while all other ions are rejected. 


	Figure 1: A schematic to show how (a) the on-mass mode, and (b) the mass-shift mode are used in triple quadrupole ICP-MS (ICP-MS/MS) CRC to remove interferences. 

Multiple Element Nanoparticle Characterization

	To date, most SP-ICP-MS applications have been used to determine samples with a known NP content, but many natural samples contain mixtures of NPs of unknown or variable composition. Also, some NP samples contain more than one metal. In this study, SP-ICP-MS has been further developed to determine multiple element NPs in samples containing a mix of NPs using ICP-MS/MS to provide high sensitivity while removing interferences that can be challenging to eliminate using single-quadrupole ICP-MS. The method has the potential to more fully characterize the NP content of products, chemicals, and environmental samples, as part of a routine monitoring program.

	With appropriate sample preparation and dilution, SP-ICP-MS enables NPs to be detected from the elemental signals generated from individual particles as they pass through the plasma. The signal generated by a single NP lasts for about 1 ms, so rapid data analysis software is needed to capture the transient signal. This is done using fast time resolved analysis (TRA) acquisition, which allows single element acquisition at a sampling rate of 100 µs (10,000 measurements per second). With the option of using no settling time between measurements of a particular element, the software can integrate multiple points across the ion plume for each single-particle event, providing sufficient points to define the nanoparticle peak shape (~5 to 10 points/peak). For the SP-ICP-MS analysis of multiple elements, additional rapid acquisition software has been developed that can collect data for up to 16 elements in a single sample, and also optimize the method parameters, thus simplifying the analysis.

	Using a rapid multielement NP analysis program, data are collected sequentially for multiple elements, using optimum conditions for the measurement of each individual element. The results for each NP are automatically collated into a single table of results, speeding up the analysis compared to measuring each element individually. Sampling in this manner also prevents agglomeration, or settling of the sample, so a more representative aliquot is analyzed for each analyte under the same analytical conditions. This is demonstrated in Figure 2, which shows a schematic representation of conventional (one element at a time, requiring sample uptake and rinse steps for each element) and rapid multielement NP acquisitions (requiring just one sample uptake and rinse time for all analytes).


	Figure 2: Example of (a) the conventional single element SP-ICP-MS method (top three batches) as compared to (b) a multi-element SP-ICP-MS method (bottom).

Reference Materials, Samples, and Sample Preparation

	Two platinum (Pt) reference materials from nanoComposix (San Diego, CA, US) were used to confirm the accuracy of the method. The reference materials contained 30 and 50 nm sized particles, and were diluted with deionized water (DIW) to a particle concentration that was calculated to give approximately 200–500 particle counts per min. The solutions were sonicated for ~3 min before analysis, to ensure sample homogeneity.

	A stainless steel nanopowder containing multiple nanoparticles was prepared for analysis by placing 41.6 mg of the nanopowder into 10 mL of 0.1% ethanol, and making up to 50 mL with DIW. The solution was sonicated, and then diluted an additional 10,000 times with DIW in a 50 mL vial and 50 µL of ethanol was added to aid suspension.

	A sample of moat water and bottled drinking water were sonicated as received for about 3 min. Two separate tap water samples were diluted 100x using DIW before sonication. Various natural waters were collected, including a river estuary sample, a sample collected upstream of a drain, water from a runoff drain, water downstream of the drain, and a seawater sample. These water samples were diluted 1000 times, using DIW before sonication. The reagents as 1% tetramethylammonium hydroxide (TMAH), 1% sulfuric acid (H

), 1% hydrochloric acid (HCl), 1% nitric acid (HNO

OH) were prepared using 0.5 mL of the reagent into 50 mL of DIW. Then, 50 µL of ethanol was added to all samples before analysis, to aid suspension of insoluble NPs in solution.

	An 8900 Triple Quadrupole ICP-MS instrument (Agilent Technologies) was used for the analysis of NPs in the reference materials, stainless steel nanopowder, environmental samples, and laboratory reagents. The instrument's ion lens system was configured for maximum sensitivity, together with a PFA microflow nebulizer. The samples were delivered via the SPS4 autosampler (Agilent Technologies) through a peristaltic pump. Hot plasma conditions were used initially to confirm the sample preparation method, and to assess the accuracy of the method for the determination of the Pt NP reference materials. Following a successful initial evaluation, the system was further optimized based on the evaluation of a gold (Au) NP reference materials from nanoComposix (San Diego, California). Warm plasma conditions were used as a compromise between high sensitivity and low background for the measurement of Cr, Al, Fe, Ni, Cu, and Au NPs. Instrument operating parameters are given in Table I.

	To convert the signals measured using SP-ICP-MS to the particle content of the original sample, it is necessary to calculate the nebulization efficiency, which is the ratio of the amount of analyte entering the plasma to the amount of analyte delivered to the nebulizer. Here, the nebulization efficiency was calculated by measuring a reference material of known particle size. A 60 nm Au NP reference material dispersed in 0.1% ethanol in deionized water and a 1.0 ppb ionic gold solution in (1% nitric acid in 0.5% hydrochloric acid) were measured. The single NP software automatically calculated the nebulization efficiency of 4.9% and 3.2% for the hot and warm plasma conditions, respectively. The time-resolved signal and size distribution graphs of Au NPs in 0.1% ethanol in deionized water are shown in Figure 3.


	Figure 3: Time-resolved signal and size distribution of (a) 60 nm, and (b) 30 nm gold nanoparticles (Au NPs) in 0.1% ethanol in deionized water solution.

	The 50 nm and 30 nm Pt NP reference materials were measured in duplicate, as shown in Figure 4. The respective measured mean size of 48.5 nm and 31 nm and median size of 48.5 nm and 31 nm agreed well with the certificate values of 46 + 5 nm and 30.6 + 3.1 nm obtained by transmission electron microscopy (TEM)-ultraviolet-visible (UV-vis), and dynamic light scattering (DLS).


	Figure 4: Size distribution results in duplicate of (a, b) 50 nm, and (c, d) 30 nm platinum nanoparticles (Pt NPs) as reference materials.

	The 60 and 30 nm Au NP reference materials were also measured in duplicate, as shown in Figure 5. The respective measured 60 nm and 27.5 nm for both the mean and median size agreed well with the certificate values of 60 + 6 nm and 31 + 3 nm obtained by TEM, UV-vis, and DLS.


	Figure 5: Size distribution results in duplicate of (a, b) 60 nm, and (c, d) 30 nm gold nanoparticles (Au NPs) as reference materials.

Analysis of Multiple Element NPs in Stainless Steel

	The multielement SP-ICP-MS method was used to analyze a stainless steel nanopowder, to demonstrate that the method could determine multiple NPs in a complex material. The time resolved signal and size distribution for Cr, Al, Fe, and Cu NPs are shown in Figure 6. The results in Table II show that good reproducibility was obtained for the mean and median particle sizes, although some elements showed size variability as the NP's were not all uniform in diameter.


	Figure 6: Time-resolved data (left) and particle size distribution (right) for chromium (Cr), aluminum (Al), iron (Fe), and copper (Cu)-containing nanoparticles (NPs) in a stainless steel nanopowder.

Application of the Method to the Analysis of Natural Waters and Chemicals

	Once a multiple element NP method has been established, it can easily be applied to the analysis of a wide range of samples. As a proof-of-concept, a series of environmental samples including a bottled drinking water, a sample taken from a large pool of water located outside of an office complex, and various river and seawaters, were analyzed using the method, together with six common laboratory reagents. The results given in Tables III and IV show that relatively high levels of Al and Fe-based NPs were present in the environmental samples, especially the moat, river, and sea waters. Relative to this group of samples, the concentration of Cr NPs was highest in the moat water and in tetramethylammonium hydroxide.

	Because Fe is a common component of the process equipment used in general manufacturing of chemicals and products, there is an increasing demand to measure Fe-based NPs. The concentration of Fe (and Cr) NPs in tetramethylammonium hydroxide and NH

OH was relatively high compared to the other reagents.

	The mean size of the particles ranged from as small as 10 nm up to 95 nm. However, for the analysis of the smallest-sized particles, ICP-MS instrumentation with very high sensitivity and low background is essential to support detection.

	The determination of multiple element nanoparticles by SP-ICP-MS is a relatively new application. This study has shown that a triple quadrupole ICP-MS with a double mass selection capability (MS/MS mode) has delivered the high sensitivity, low background, and interference control necessary for the characterization of Cr, Al, Fe, Ni, Cu, Pt, and Au nanoparticles.

	The study has also showed that environmental and industrial chemical samples containing nanoparticles composed of different or multiple elements can be determined quickly using SP-ICP-MS/MS. Compared to acquiring data separately for each NP, specially developed rapid acquisition software was used to simplify the analytical method and significantly shorten the sample run times.

	Given that regulations are likely to be introduced for nanoparticles across different industry-sectors, ICP-MS is expected to play a central role in the enforcement and development of any new guidelines. Multiple element SP-ICP-MS methods have the potential to more fully characterize the NP content of bulk chemicals, foods, consumer products, medicines, and environmental samples, as well as supporting research projects.

	(1) F. Laborda, E. Bolea, and J. Jiménez-Lamana, 

	(2) J. Tuoriniemi, G. Cornelis, and M. Hassellöv, 

	(3) C. Degueldre, P.-Y. Favarger, and C. Bitea, 

	(4) E. Bolea-Fernandez, D. Leite, A. Rua-Ibarz, L. Balcaen, M . Aramendía, M. Resano, and F. Vanhaecke, 

	(5) E. Bolea-Fernandez, L. Balcaen, M. Resano and F. Vanhaecke, 

 is an Application Scientist at Agilent, in Santa Clara, California. Craig has been with Agilent since 2004, and has been involved with multiple type of applications for ICP-MS, including environmental, pharmaceutical, nutraceutical, semiconductor, geologic, and clinical analyses, to name a few. Craig obtained a bachelor of science degree in chemistry from Fort Lewis College, in Durango, Colorado.

 is a Product Specialist at Agilent in Elm Mott, Texas, specializing in advanced ICP-MS and ICP-MS/MS techniques. He received his undergraduate and graduate degrees from Texas A&M University, in College Station, Texas. Prior to Agilent, Mark was a senior research scientist and research group manager with Southern Research Institute, in Birmingham, Alabama, where he focused on low-level analytical chemistry, method development, and research program management.

 is an ICP-MS Product Specialist at Agilent, in Seattle, Washington. He received his undergraduate degree from Cornell University. Prior to Agilent, he worked at Brookhaven National Labs studying acid rain in the Adirondack Mountains of northern New York, and then went on to work at a number of commercial analytical labs operating GC and GC/MS. Direct correspondence to: 

 column editor of "Atomic Perspectives," is the principal of Scientific Solutions, a consulting company that serves the application and writing needs of the trace element user community.

Complex isobaric and polyatomic spectral interferences can be mitigated using triple quadrupole ICP-MS (ICP-MS/MS) with a  collision–reaction cell (CRC). This configuration allows for the multielement characterization and detection of smaller nanoparticle sizes.

Mark Kelinske

Articles

GC-ICP-MS for Process and Quality Control in Semiconductor Manufacturing

Rapid Multielement Nanoparticle Analysis Using Single-Particle ICP-MS/MS