Reliable Measurements   • Introduction / Contents • About the Author  Foundations   • 1. Laying the Foundation  Sample Collection   • 2. Planning the Project • 3. Sampling and Sub-sampling  Sample Preparation   • 4. An Introduction to Sample Preparation • 5. Container Material Properties • 6. Container Transpiration • 7. Stability of Elements at ppb Concentration Levels  Contamination   • 8. Environmental Contamination 9. Contamination From Reagents • 10. Contamination From the Analyst and Aparatus  Preparation Techniques   • 11. Acid Digestions of Inorganic Samples • 12. Acid Digestions of Organic Samples • 13. Sample Preparation by Fusion • 14. Ashing  Sample Measurement   • 15. ICP-OES Measurement • 16. ICP-MS Measurement  Conclusions   • 17. Method Validation

Sample preparations using acid digestion, fusion, or ashing all typically use water as the primary reagent. Most water used in trace metal laboratories is produced by systems that use ion-exchange purification. This water is commonly referred to as "conductivity water" because its conductivity approaches the theoretical conductivity of water (0.055 microhm / cm {18.2 megohm water} at 25�C).

Under normal laboratory conditions, conductivity water never measures to be 18.2 megohm due to the presence of CO₂ (H₂CO₃HCO₃^- + H⁺). Furthermore, if you could find water giving a conductivity of 18.2 megohm, it is not necessarily free of trace elemental contaminants because only ionized compounds are detected by conductivity measurement.

Through carefully controlled experiments and measurements in clean room facilities, we have found that conductivity water will typically give readings closer to 16 megohm and that it is free of trace metallic impurities down to conventional ICP-MS and axial view ICP-OES detection limits. We have also found that sub-ppb level impurities that were once thought to be coming from the water, are in actuality from the atmosphere and the container materials (see earlier parts of this guide). We have found that the use of clean room facilities and high temperature nitric acid-leached LDPE bottles are necessary for reliably measuring common contaminant elements in water. Therefore, do not assume that your water has significant levels of elemental contaminants if it gives conductivity readings between 16 and 17 megohm and your ICP-MS or OES is detecting trace levels of the common environmental contaminants.

High purity water should be used ASAP. "Stored" high purity water may pick up impurities from the storage container. Popular storage containers are made from quartz, polyethylene (both high and low density), and fluoropolymers.

Quartz (fused quartz or vitreous silica) typically contains 98.8% SiO₂, and impurities consisting mainly of Na₂O, Al₂O₃, Fe₂O₃, MgO, and TiO₂. Quartz has a solubility in water of 11 ppm.¹ We have measured a solubility of quartz in conductivity water of 11.2 ppm as silicic acid (equilibration time is ~4 weeks using 400 mesh quartz powder). [ A Closer Look at Quartz ]

A significant amount of HDPE is manufactured using alumina / silica based catalysts. Long term storage in high density polyethylene (HDPE) can result in ppm levels of Ca, Mg, Si, Ti, Al and ppb levels of Cr, V and Fe. LDPE can be manufactured using an organic catalyst. Storage in HNO₃ leached LDPE is optimum. Through study, we've discovered that short term (1-5 days) storage in both 20 liter HDPE and LDPE cubi containers that have been leached with dilute HNO₃ do not leach any elements at ICP-MS / OES detection limits.

Fluoropolymers are not as clean as generally thought (see below figures). Studies performed in our own laboratories confirm these results. It is our recommendation that you save your money and use LDPE.

Figures 9.1 and 9.2 show total trace metals and major contributing ions for cut parts and from PFA 1 resin pellets and extruded tubing following extraction.²

Fig 9.1: Total Trace Metals & Contributing Ions for Cut Parts



- Extracted for 5 days in 10% ultrapure HNO₃ at 25�C -


Fig 9.2: Total Trace Metals & Major Contributing Ions from PFA 1 Resin Pellets and Tubing



- Extracted for 5 days in 2% ultrapure HNO₃ at 25�C -

Figure 9.3 shows the total extractable fluoride, chloride, and sulfate ions from cut fluoropolymer parts following extraction.²

Fig 9.3: Fluoride, Chloride, and Sulfate Ions from Cut Parts



- Extracted for 5 days in 18 MΩ Di H₂O at 85�C for 1 hour -

High purity acids have been commercially available for years with the major impurities typically less than 1 ppb. Distilling the acid yourself may offer some improvement in purity. If you are using more than 500 mL of mineral acid per month, you may want to consider the monetary savings of distilling the acid yourself. Diagrams 9.1 and 9.2 below show typical quartz and teflon stills.

Diagram 9.1: Pure Quartz Sub-Boiling Still




Diagram 9.2: All-Teflon Sub-Boiling Still

Tables 9.1 - 9.5 below show impurities in different Grades of the common mineral acids.

The "purity" situation for salts and other reagents is typically not as favorable as it is for water and acids. Typically, the highest purity solid that can be confirmed is 99.999% (5-9's or TMI 10ppm). This translates to 100 ppb total impurities for a 100 fold dilution (i.e. - 1 g of 5-9's pure reagent into 100 mL of solution). We recommend that you know your supplier's definition of "high purity". DO NOT use solid reagent grade materials when preparing samples for trace metals analysis whenever possible. If necessary, a blank should be performed to confirm the acceptability of reagents(s) or to identify problem elements / impurities. Purification of unacceptable reagents may be accomplished by mercury cathode electrolysis, extraction with dithiozone or cupferron, ion-exchange, or crystallization. 

 Part 8 • Part 10