There are three fundamental guiding principles of ICP-AES sample preparation that must be followed for every calibration and analysis, regardless of what is being analyzed (rocks, sediments, or interstitial waters).
1. Filtration: All solutions introduced to the instrument must be filtered. This is most efficiently performed through Acrodisc filters that fit over a syringe. Failure to filter every solution may lead to a partially or completely clogged nebulizer, resulting in poor analytical precision or analysis failure. Clogged nebulizers are also extremely difficult to clean (see "Basic Maintenance Suggestions").
2. Matrix Matching: The term "matrix" refers to the sum of all compositional characteristics of a solution, including its acid composition. Calibration standards and samples must be matrix-matched in terms of composition, total dissolved solids (TDS), and acid concentration (in percent) of the solution. For analysis of solid materials that have been dissolved, the use of LiBO2 flux effectively means that the standards and samples have been compositionally matrix matched (as will be detailed in "Preparation of Rock/Sediment Samples"), but it is also important to have the standards and samples be of the same acid concentration. TDS refers to the total mass (in milligrams) of powder (sample plus flux) dissolved in a given mass of solution, and is commonly quantified as a function of dilution factor (see "Dilution Factors"). There is a wide range of acceptable acid concentrations, any one of which is likely to suffice, but it is of critical importance that whatever acid concentration is used during a particular run, that it be the same for all standards and samples in that run. Typical concentrations for rock and sediment samples are 5% or 10% HNO3. For interstitial water analysis, the calibration standards must be prepared with the same background salt concentration.
3. Cleaning: Throughout the laboratory, great care and diligence must be exercised to keep sample bottles and ICP glassware used for the analysis of difference matrices separate. Because Li and B are of particular interest to interstitial water chemists, all sample bottles, autosampler vials and tubes, Teflon tubing, peristaltic pump tubing, and torch assembly glassware that are used during the analysis of rocks and sediments (and thus have been exposed to tremendous amounts of Li and B in the LiBO2 flux) must be kept separate from those items used in the preparation and analysis of interstitial waters. As importantly, bottles and glassware that have been used in the preparation and analysis of the LiBO2 flux solutions must be cleaned separately from interstitial water apparatus. Duplicate acid baths and other cleaning routines need to be maintained with the greatest rigor.
The procedure used here is similar to that used previously on board the Resolution for major element X-ray fluorescence (XRF) analysis. Whereas in major element XRF the fused glass disk serves as a solid target for the X-ray beam, in ICP-AES preparation the glass bead is dissolved in 10% HNO3, the solution is filtered and diluted, and the resultant aqueous solution is subsequently introduced into the ICP-AES for analysis.
In the following presentation, the analyst must decide whether to use ignited samples, depending on the analytical needs. It is most common to use ignited samples for igneous rocks, with the final analysis including a determination of loss on ignition (LOI). For sediments and sedimentary rocks, particularly carbonate-rich sediments, determination of LOI is often not performed, partly because LOI comprises a significant proportion of the total weight percent. Regardless, the following guidelines apply to the use of ignited or nonignited samples equally well. The most important consideration is that the proper sample and acid concentrations be used during calibration and analysis.
LiBO2 flux without La absorber, and indeed only ultrapure grade LiBO2, should be used. The heavy La absorber also contains other rare earth elements (REEs), and thus the standard and sample solutions will be compromised for future analyses, and the ICP-AES will be contaminated with background concentrations of REEs.
The most critical aspect of the preparation is the need to maintain a constant sample to flux ratio. This can be adjusted as needed on a case by case basis, but a ratio of 1:4 should suffice in most situations. ICP-AES analysis requires a far smaller sample mass than XRF. Typically, 0.1 g of sample powder mixed with 0.4 g of LiBO2 flux will result in adequate solution for a thorough major and trace elemental analysis. However, if only limited sample mass is available (e.g., for the analysis of volcanic glasses), a smaller sample mass may be used, but the same sample:flux ratio must be maintained for all samples and calibration standards (otherwise the matrix will not match). For example, 0.05 g of sample would require 0.2 g flux to maintain the suggested 1:4 ratio. Use of such smaller masses results in a smaller mass (volume) of analyte solution, but if a complete geochemical analysis is not required, very small sample masses (e.g., basaltic glass rims) can be analyzed for only a few target elements of high interest.
1. Weigh 0.1 g of powdered, dried sample. ODP standard practice is to weigh within a range of 0.0995-0.1005 g, with a precision of 0.5% of the measured value. If the sample weight falls within this range, weighing differences need not be accounted for during data reduction.
2. Mix the sample powder with 0.4 g of LiBO2 flux. In practice, this mass can be weighed out between 0.395 and 0.405 g. Transfer the powder mix into a Pt-Au crucible, add 10 µL of 0.172 mM LiBr wetting agent, and fuse at 1050° C for 10-12 min. The wetting agent is prepared by dissolving 0.15 mg of ultrapure LiBr powder into 10 g of DI.
3. Let the bead cool and solidify in the bottom of the crucible. After cooling, pop the bead off the crucible bottom. A sharp "whack" squarely on the hard desktop facilitates release.
4. Dissolve the bead in 50 mL of 10% HNO3 acid solution and shake. Dissolution is effective in a sonicator bath or a wrist-action shaker, and should take ~45 min to 1 hr in most cases. This solution is now at what is broadly termed a 100× dilution for TDS because the 0.5 g of flux + sample has been dissolved in ~50 g of HNO3 (50 g/0.5 g = 100). This is also referred to as a 1% solution because 1% of the total mass consists of TDS. It is important to note that the dilution of the sample is significantly greater (500×), because 0.1 g of powdered sample has been dissolved in 50 g of HNO3 (50 g/0.1 g = 500). The very small addition of LiBr does not significantly increase the TDS. The exact mass of the acid solution is not critical (i.e., 50 mL HNO3 is ~50 g), but it is extremely important that the unknown samples, blanks, replicates, etc., be prepared identically to the rock SRMs used for calibration (see "Dilution Factors").
5. After complete dissolution of the bead, the solution should be filtered. This is best achieved by aspirating ~20 mL of the solution into a syringe and filtering through an Acrodisc into a cleaned scintillation vial. The remaining unfiltered solution can be saved for additional analysis, replicate analysis, or transport to a shorebased laboratory.
6. The final analyte solution is prepared by pipetting a 5-mL aliquot of the filtered solution and diluting it with 35 mL of 10% HNO3 (to a total of 40 mL). This is an additional 8× dilution (40 mL/5 mL = 8) of the original 500× diluted solution described in Step 4 above (to arrive at a total 4000× dilution of the sample [not of the sample + flux]). This solution is dilute enough to enable use of a type-C Meinhard concentric nebulizer (see "Nebulizers").
7. An analytical procedural blank is prepared identically to the samples, with the exception that only 0.4 g of flux is fused and dissolved. An additional 0.1 g of flux is not added to mimic the TDS of the 0.5 g mix of sample + flux because this would provide an inaccurate quantitation of the impurities introduced by the amount of flux used in preparation of the unknowns.
1. After HNO3 acidification of an aliquot of interstitial water sample, the preparation of interstitial water for ICP-AES analysis primarily involves dilution. Because the pore-water solutions have already been filtered during squeezing, additional filtration is not required.
2. The most common dilution for interstitial water samples is 10×. Depending on the analysis speed and the number of elements analyzed, one can usually generate enough analyte with 0.5 mL of interstitial water diluted with 4.5 mL of DI (for a total analyte of 5.0 mL). A 1-mL sample + 9 mL DI solution allows for replicate analysis. In some cases, undiluted interstitial water may be analyzed, but care must be taken to not clog the nebulizer, as the likelihood of salt buildup at the nebulizer orifice is greatly increased.
3. An analytical blank is prepared identically by analyzing pure DI acidified to matrix match the samples. Be sure to use DI from the same bottle as used during dilution of the samples.