Overcoming the Challenges Associated with the Direct Analysis of Trace Metals in Seawater Using ICP-MS

The direct analysis of seawater by inductively coupled plasma–mass spectrometry (ICP-MS) is notoriously difficult because of the high matrix content of the sample that leads to both spectroscopic (for example, polyatomic interferences) and nonspectroscopic interferences (for example, signal suppression). Additionally, the low target concentration levels demand a noncontaminating, robust sample introduction technique. The latest ICP-MS techniques provide high-throughput methods that are able to process large numbers of samples presented for analysis.

The oceans cover 71% of the planet and have a combined volume of 1.4 × 10²¹ L (3.3 × 10⁸ cubic miles). Any attempt to understand the chemical inputs to the planet's ecology must include a close look at seawater. Toxic substances and high concentrations of certain trace elements may present a health hazard for irrigation, swimming, fishing, boating, and industrial uses. These conditions also can affect wildlife, which use the water for drinking or a habitat.

The analysis of trace elements in seawater is one of the most challenging analytical tasks in the field of environmental monitoring, mainly because of matrix interferences from its high dissolved salt content. In recent years, methods of adequate sensitivity became available for true ultratrace-metal determinations in drinking water. However, in the case of seawater, only now has it become possible to both resolve the complex mixtures and achieve the very low detection limits required.

The most commonly employed method of analysis for trace metals in seawater is inductively coupled plasma–mass spectrometry (ICP-MS). Although the technique is very powerful by virtue of its sensitivity and selectivity, it is limited in terms of the amount of dissolved solids that can be introduced. When aspirating samples with high levels of dissolved solids, salt may deposit on the interface (cones) during the analytical run, degrading signal stability. Additionally, high levels of dissolved salts can cause spectral (polyatomic) interferences. Because these interferences originate from the seawater matrix itself, standard ICP-MS for seawater analysis can often lead to false positive results that are much higher than the actual concentrations in the solution.

To address these analytical challenges, a range of different sample preparation and introduction techniques have been coupled with ICP-MS. This article demonstrates the contamination-free, fully automated, direct, trace-metal analysis of seawater using a specialized sample introduction system in conjunction with ICP-MS.

Methods

The results in this study were obtained using an Elemental Scientific Inc. (Omaha, Nebraska) PC3 Fast sample introduction system with a Thermo Scientific (Bremen, Germany) XSERIES 2 ICP-MS system. The ICP-MS system was configured with the sample introduction system in combination with an autosampler. The schematic in Figure 1 outlines the principle of the sample introduction system, which utilizes a vacuum pump to load samples onto a PFA loop mounted across a six-port switching valve. In addition to providing direct, routine trace-metal analysis of high matrix samples including seawater, the system ensures high sample throughput and minimizes sample deposition on the ICP-MS interface for increased long-term stability. The system also ensures minimal contamination as a result of automated sample manipulation and short sample pathway consisting of high-purity components.

Figure 1: Schematic of the sample introduction system.

Experimental

The ICP-MS instrument parameters are outlined below.

Nebulizer: ESI PFA-ST nebulizer

Spray chamber: Quartz cyclonic spray chamber

Injector: Demountable torch with a 2.5-mm diameter

Interface: Ni Xs high sensitivity cone

Additional gas: 2% methane in argon at 100 mL/min

Nebulizer gas: 0.93 L/min

Forward power: 1500 W

Collision cell: 4.0 mL/min 7% hydrogen in helium at 2 V (kinetic energy discrimination [KED] mode)

Seawater Analysis

As part of the experiment, undiluted seawater samples were loaded onto the autosampler and the internal standard containing diluent was added online using a T-piece located directly after the exit port from the Fast valve before the nebulizer. The ratio of seawater to diluent was 1:7. Online dilution is recommended to minimize unnecessary contamination. The NASS-5 and CASS-4 seawater reference material standards for trace metals (National Research Council Canada, Ottawa, Ontario) were selected to evaluate accuracy. Locally sourced filtered and acidified seawater was spiked to 10 ppb with a range of elements and used for the long term stability evaluation. A three-point external calibration was used for quantification with standard concentrations chosen relative to the expected concentration in the samples. Gallium (Ga), yttrium (Y), indium (In), and bismuth (Bi) were used as internal standards. All elements were measured in KED mode using the standard single gas mixture of 7% hydrogen (H₂) in helium (He) for optimum sensitivity and interference suppression.

Results

Figure 2 shows the external calibration curves for the determination of trace metals in seawater by ICP-MS; the selection of elements includes iron, nickel, copper, and zinc. Figure 3 compares the raw count rates for the internal standards in both a 2% nitric acid (HNO₃) and an undiluted seawater sample measured. Signal suppression of <20% was found in the undiluted seawater.

Figure 2: Calibration curves for Fe, Ni, Cu and Zn.

Figure 3: Internal standard count rates (counts per second) from Thermo Scientific PlasmaLab software. Suppression in the 1:7 diluted seawater is less than 20%.

Quantitative results for the seawater samples measured are shown in Table I. The findings indicate a good agreement between measured and certified values. To assess the stability of the analysis method, 180 samples of 10 ppb spiked seawater were analyzed using the described system over a 6-h period. The stability of the recovery for the 10 ppb spike is shown in Figure 4.

Table I: Comparison of measured with certified values for the NASS-5 and CASS-4 reference materials. Detection limits (ng/mL) were calculated from 3 standard deviations of 100 blank runs.

Figure 4: Measured values in 10 ppb spiked Bremerhaven seawater for 180 samples over a6-h period.

Conclusion

To understand the quality of seawater, the analysis of its chemical properties is vital. The ability to effectively undertake such analyses is invaluable for seawater environmental protection.

The method described in this work provides a rapid, contamination free sampling technology with full automation for high sample throughput. Through implementation of the new method, previous limitations on total dissolved solids with ICP-MS are eliminated, providing a sensitive and interference-free instrumental solution for trace-elemental determination even in difficult matrices such as seawater.

Shona McSheehy-Ducos is an ICP-MS application specialist at Thermo Fisher Scientific in Bremen, Germany.

References

(1) http://www.epa.gov/regulations/laws/cwa.html