Sample Preparation for Quantitative Metal Impurity Testing

When analyzing for trace metals using inductively coupled plasma (ICP), sample preparation is crucial. For sample preparation in general, a key question is what is the best method to get the test article in solution. The ideal situation would be to dilute the test article in water or dilute acid. The time and effort are minimized, thereby allowing for quicker result generation. Using this simplistic approach, however, may not be appropriate for most test materials when using ICP methods.

Sample preparation methods

The United States Pharmacopeia (USP) proposed General Chapter <233> Elemental Impurities—Procedures suggests four sample-preparation methods (1). These solutions may be analyzed using an inductively coupled plasma–optical emission spectrometer (ICP–OES) or an inductively coupled plasma–mass spectrometer (ICP–MS). Liquid samples may be analyzed neat, and solid or liquid samples may be analyzed in aqueous or organic solutions or digested in a closed-vessel apparatus. It is important to ensure the blank and standard solutions are prepared in the same matrix as the sample by using the same solvent, acid concentration, and stabilizers.

PHOTODISC / GETTY

Neat. A few test articles can be analyzed without dilution or digestion. This technique is appropriate when the sample is not too viscous for aspiration into the ICP–OES or ICP-MS. An example would be water.

Direct aqueous solution. Test articles that are water-soluble can be prepared in a dilute acid and analyzed directly. It is important to ensure complete dissolution of the test article and that no precipitate or turbidity is present.

Direct organic solution. Test articles that are not water-soluble may be prepared in an organic solvent. To run organic solutions on the ICP–OES or ICP-MS, a cooled spray chamber, as well as a special torch, may be required. A separate oxygen hookup may be needed as well. The ICP parameters need to be optimized to run organic solutions and may differ significantly from parameters used for analyzing aqueous solutions.

Indirect solution. Test articles requiring concentrated acid for dissolution can be prepared using a closed-vessel microwave apparatus. This method minimizes the loss of volatiles but may require multiple microwave cycles and acid additions to complete the digestion. The analyst must consider the test articles being digested when using the microwave digestion vessels. The vessels are expensive, and contamination of the Teflon sleeves can occur from the test article. Additional precautions are needed when using hydrofluoric acid. In addition to the safety concerns of working with the acid, special microwave digestion vessels, a special ICP nebulizer, and torch are required when using hydrofluoric acid.

Other sample-preparation methods may be used that are not outlined in the USP proposed method. A few examples include refluxing with dilute acid and performing acid digestion according to EPA method 3050B (2).

Other considerations

Glassware used for trace-metal testing must be scrupulously clean before use. To prepare glassware for analysis, rinse the clean glassware with a dilute acid solution that is prepared from a trace-metal-grade acid and rinse with USP purified water (3). Glassware can be used immediately if being used for direct aqueous solution preparation. Let glassware dry for all other sample preparation methods.

Standards are prepared by diluting commercially available NIST traceable standard solutions. Precautions must be taken when making mixed standard preparations to ensure all of the elements are compatible in solution and that no precipitation occurs. It is important to prepare standards and blanks in the same matrix as that of the sample to eliminate interference.

Additionally, when screening for mercury, typically a stabilizer, such as gold (III) chloride (AuCl₃), is added to maintain mercury as the mercuric ion and to prevent reduction to elemental mercury.

Materials and methods

To illustrate how important method selection is in obtaining accurate results, several examples are provided. An analysis of arsenic, cadmium, lead, and mercury was performed to support oral dosage products using the limits within the USP proposed General Chapter <232> Elemental Impurities—Limits (4). Testing was performed at 1.5 μg/g arsenic, 0.5 μg/g cadmium, 1.0 μg/g lead, and 1.5 μg/g mercury, corresponding to the 100% specification level. One unspiked test article and triplicate spiked test articles at 100% of the specification were prepared. Standards were made at 50%, 100%, and 150% of the specification.

To determine if the sample preparation method is acceptable, the USP proposed general chapter provides validation criteria for elemental impurity testing. Two sets of criteria are included: one for a limit test and the second for a quantitative procedure. The work performed in this study followed the quantitative-procedure criteria. Accuracy and specificity were performed, but repeatability and ruggedness were not included in this study. The acceptance criterion for accuracy per the USP proposed General Chapter <233> Elemental Impurities—Procedures is 70–150% recovery for the mean value at each concentration (1). This study was performed using spiked test articles prepared at the 100% concentration only.

Test articles used were reagent-grade sodium chloride (EMD Chemicals) and polysorbate 80 (Tween 80, Fisher Scientific). All concentrated acids were trace-metal grade. Sodium chloride is freely soluble in water, and polysorbate 80 is very soluble in water (5). Sample-preparation methods used included direct aqueous solution, indirect solution using closed-vessel microwave digestion, refluxing, and an acid-digestion method under EPA 3050B. Instrumental analysis was performed using a ICP–OES spectrometer (Thermo Electron IRIS Intrepid II XDL ICP–OES) with a high-solids nebulizer. A mercury stabilizer was not added to the solutions because the ICP analysis was performed immediately after standard and sample-solution preparation.

Direct aqueous solution preparation method. 0.5 g of each test article were dissolved and diluted to 10.0 mL with 1% hydrochloric acid. Spiked test articles were prepared by adding 1.0 mL of spiking standard to a mixture of 0.5 g of the test article and 9.0mL of 1% hydrochloric acid. A blank and standards were prepared in 1% hydrochloric acid.

Indirect solution-preparation method. 0.5 g of each test article were transferred to a Teflon microwave vessel. 1.0 mL of USP purified water or spiking standard and 4 mL of nitric acid were added to the vessel. The solution was left to stand for 5 min before microwaving. The microwave program consisted of ramping the temperature to 120 °C in 2 min, and the temperature was maintained at 120 °C for 20 min. Once the Teflon vessels had cooled to room temperature (20–25 °C), the contents were quantitatively transferred to 10-mL volumetric flasks with USP purified water and diluted to volume with USP purified water. A blank and standards were prepared in 1% nitric acid.

Refluxed solution-preparation method. 0.5 g of each test article were transferred to a 50-mL round-bottom flask. 5.0 mL of 2% nitric acid or a mixture of 1.0-mL spiking standard and 4.0 mL of 2% nitric acid were added to the appropriate flask and refluxed for 30 min using water-cooled condensers. Once the flasks had cooled to room temperature, the contents were quantitatively transferred to 10-mL volumetric flasks with USP purified water and diluted to volume with USP purified water. A blank and standards were prepared in 2% nitric acid.

EPA 3050B solution-preparation method. 0.5 g of each test article were transferred to a 150-mL glass beaker, and 0.4 mL of nitric acid and 2 mL of hydrochloric acid were added. 1.0 mL of spiking solution was added for spiked samples. The beaker was covered with a ribbed watch glass and refluxed at 95 °C for 15 min on a hot plate. The samples were filtered through Whatman Grade No. 41 filter paper into a 10-mL volumetric flask and washed through the filter paper with 0.4-mL of hot hydrochloric acid and 2-mL of hot USP purified water. The filter paper was placed in the original beaker used for refluxing and 2 mL of hydrochloric acid were added. The beakers were heated to 95 °C on a hot plate until the filter paper dissolved. The resulting solution was quantitatively transferred through fresh Whatman Grade No. 41 filter paper into the 10-mL volumetric flask. All flasks were diluted to volume with USP purified water. A blank and standards were prepared in 2% nitric acid.

All solutions were analyzed by ICP-OES for arsenic, cadmium, lead, and mercury.

Results

The results are summarized in Tables I and II. Evaluation of the sodium chloride results showed that the indirect solution preparation using closed-vessel microwave digestion gave the best recovery values of the four methods. All four elements met the USP proposed accuracy criterion. The worst of the sample-preparation methods was refluxing. Although sodium chloride is freely soluble in water, the direct aqueous solution method was not the most appropriate for the metals that were analyzed. The cadmium recovery was below 70%.

Table I: Sample preparation methods using a test article of sodium chloride for testing for arsenic, cadmium, lead, and mercury with an inductively coupled plasma–optical emission spectrometer.

The polysorbate 80 results showed that no one method gave acceptable recovery for all four metals. The direct aqueous solution was the best method, but had mercury recovery above 150%. The indirect solution preparation using closed-vessel microwave digestion was the second best method, but had mercury recovery slightly below 70%. Mercury recovery may be improved using a stabilizer. The worst of the sample preparation methods was refluxing.

Table II: Sample preparation methods using a test article of polysorbate 80 for testing for arsenic, cadmium, lead, and mercury with an inductively coupled plasma–optical emission spectrometer.

Conclusion

These examples illustrate that solubility of the test article may not be the primary factor when choosing a sample-preparation method. Just because a test article is water soluble, does not mean that the direct aqueous solution method is the best choice. Validation studies need to be performed as described in the USP proposed General Chapter <233> to ensure the method is accurate, specific, precise, and rugged when used as a quantitative method (1).

Gayla Velez is director of analytical laboratory services, SGS Life Science Services, 616 Heathrow Drive, Lincolnshire, IL 60069, tel. 847.821.8900, gayla.velez@sgs.com.

References

1. USP Proposed General Chapter <233> "Elemental Impurities— Procedures," Pharmacopeial Forum 37 (3), Rev. May 26, 2011, www.usppf.com/pf/pub/data/v373/CHA_IPR_373_c233.html#CHA_IPR_373_c23, accessed Aug. 22, 2011.

2. EPA, Method 3050B, "Acid Digestion of Sediments, Sludges, and Soils" (Washington, DC), www.epa.gov/osw/hazard/testmethods/sw846/pdfs/3050b.pdf, accessed Aug. 1, 2011.

3. USP 34–NF 29 "USP Purified Water Monograph Water for Pharmaceutical Purposes" (USP, Rockville, MD, 2010), p. 4598.

4. USP, Proposed General Chapter <232>"Elemental Impurities— Limits," Pharmacopeial Forum 37 (3), Rev. May 26, 2011, www.usppf.com/pf/pub/data/v373/CHA_IPR_373_c232.html, accessed Aug. 22, 2011.

5. USP 34–NF 29 "Reference Tables: Description and Solubility" (USP, Rockville, MD, 2010), pp. 1035 and 1040.