Analyzing Drinking Water by ICP-MS for Heavy Metal Detection

Syringe filter units for sample preparation

The Minisart syringe filter holders used in our study are filtration units designed for a single use and contain a microporous filter membrane between two housing parts made of highly pure plastic. This membrane is thermally sealed inside the housing without any addition of glue. Filter membranes of different materials, such as polyethersulfone (PES), cellulose acetate (CA) and regenerated cellulose (RC) can be used for analysis of metals in drinking water. Common international standards recommend a pore size of 0.45 µm, which is used to remove undissolved solids [4, 6]. In addition to PES and CA with a pore size of 0.45 µm, RC filter material (pore size of 0.2 µm) was also tested in this study. For the housing material, either polypropylene or an acrylic-based multi-polymer mixture is used. During manufacture in compliance with DIN EN ISO 9001, all syringe filter units are automatically integrity-tested. Subsequently, the quality assurance department tests the pressure-hold characteristics and flow rate performance, among other properties, of each lot manufactured.

Test method for extraction of metals

To extract metals, 0.5 mL of ultrapure Type 1 water/cm² of filter area was filtered through each group of three interconnected units of the same Minisart filter type by pressing in the syringe plunger. A 10 mL disposable B. Braun syringe was used per filtration run. Table 1 lists the volumes and eluate quantities used.

ICP-MS technology was employed to analyze the eluate obtained from the syringe filter holders for a total of 38 chemical elements [7]. The instrument used was an Agilent ICP-MS system 7500. The results section of this paper lists only the results of the elements determined according to the requirements of the WHO, as such elements are of special significance for testing drinking water.

Each eluate was collected in a 30 mL Sarstedt vial and acidified using nitric acid and hydrochloric acid for complete mineralization. For the blank, 5 mL of ultrapure water were drawn into a 10 mL single-use syringe and transferred into a Sarstedt vial. Nitric acid and hydrochloric acid were also added to this vial. To prepare for analysis using ICP-MS, rhenium was used as an internal standard at a concentration of 10 µg/L, and the eluate was filled up to a volume of 10 mL.

Calibration of the ICP-MS measurements was performed for all elements examined. Standard solutions of the respective elements together with the instrument blanks (zero values) were injected into the Agilent ICP-MS 7500 system, and the calibration curves were recorded. Figure 3 shows examples a, b and c of the calibration slopes of the elements Ni, Cu and 207Pb as a function of the ratio (corresponds to the signal of the particular element to the signal of rhenium 10 µg/L used as an internal standard) to the concentration of the element expressed in µg/L. The concentration of each element of the tested samples was calculated from the respective calibration curves and is listed in Table 2.

Results and discussion of the drinking water samples

The relevant parameters for drinking water analysis are shown in Table 2. The detection limit is 0.1 µg/L, and the recovery of the internal standard was 100% (10 µg/L). For seven out of twelve elements, the values were below this limit, and metal ions could not be detected in the solutions. For five out of twelve elements, metal quantities above the detection limit were found. This applies to the elements Al, Cu, Fe, Ni and Na. Sodium, in particular, consistently showed the highest values (16 µg/L to 69 µg/L) for all samples including the blank. For iron, a blank of 0.58 µg/L was determined, and for Minisart NML, a value of 1.2 µg/L was obtained. Aluminum was detected only in the blank (0.69 µg/L). Copper and nickel were detected for Minisart RC, with values of 0.13 µg/L and 0.33 µg/L, respectively.

The test results clearly show that the concentrations in µg/L (ppt) of the elements examined in the eluate of the various Minisart types are considerably below the limit values required by the WHO, and are also below the detection limit of 0.1 µg/L for many elements. For individual samples, particularly sodium, values above the detection limit were measured. However, these values are several powers of ten below the limits for drinking water and do not add any significant quantities of metal.

Contamination by syringe filter holders negligible

The results obtained significantly underscore that the Sartorius Minisart syringe filters used do not release any significant quantities of metal ions into samples and are excellently suited for preparing samples, especially for removing undissolved constituents, for the detection of metals in drinking water by ICP-MS and ICP-AES.

Acknowledgements: Special thanks go to Dr. E. Herbig for reviewing the manuscript and for his constructive discussions on this subject.n

Literature

[1] Forth/Henschler/Rummel, Allgemeine und spezielle Pharmakologie und Toxikologie, 8. Auflage [8th edition], 2001, p. 1043

[2] WHO, Guidelines for Drinking-water Quality, 3rd edition, Vol. 1 Recommendations, 2004

[3] Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption

[4] ISO 17294-2:2003, Application of inductively coupled plasma mass spectrometry (ICP-MS)

[5] Worley, J., and Kvech, S. (version dated 23 August 2011, 11:31) “How the Agilent 7500cx ICP-MS Works” (http://wiki.manchester.ac.uk/wrc/index.php/ICP-MS)

[6] Standard Methods for Examination of Water and Wastewater 22nd edition, 2012

[7] Prüfbericht [German-language Test Report], Currenta GmbH u. Co. OHG, Leverkusen, 2015

* K. Schöne: Sartorius AG, 37075 Göttingen/Germany

(ID:44811131)