Introduction of the Pre-calibration Package GEO-TRACE-PAK: An Analytical Package for Trace Elements Including Rare Earth Elements in Rocks

GEO-TRACE-PAK is a pre-calibration package for quantitative WDXRF analysis of trace elements in geological materials. It was developed using approximately 110 standard materials, primarily rocks, soils, ores, minerals, and related oxide materials.

The package contains pre-established measurement conditions, calibration parameters, and correction factors, eliminating the need for users to develop calibration curves from their own large collection of standards. It supports 37 trace elements, including rare earth elements, as well as eight major components used as references for matrix corrections. It is designed for use with the Rigaku ZSX Primus IV, ZSX Primus IVi, and ZSX Primus III NEXT.

The package supports quantitative analysis of F, Sc, V, Cr, Mn, Co, Ni, Cu, Zn, Ga, Ge, As, Br, Rb, Sr, Y, Zr, Nb, Mo, Ag, Cd, Sn, Sb, Ba, La, Ce, Pr, Nd, Sm, Eu, Yb, Hf, W, Pb, Bi, Th, and U.

This range includes several rare earth elements—La, Ce, Pr, Nd, Sm, Eu, and Yb—as well as geologically important trace elements such as Rb, Sr, Y, Zr, Nb, Th, and U. Na₂O, MgO, P₂O₅, S, K₂O, CaO, TiO₂, and Fe₂O₃ are also included as reference components for matrix corrections.

Geological samples contain complex combinations of major and trace elements, producing numerous overlapping X-ray lines. Trace analytes may occur at ppm concentrations while interfering elements are present at concentrations ranging from fractions of a percent to tens of percent. This makes accurate separation of analytical peaks and background signals particularly important.

Matrix effects add another complication. Major constituents such as Fe₂O₃, TiO₂, CaO, and K₂O can substantially alter the absorption or excitation of X-rays from trace elements. Reliable quantitative analysis therefore requires careful selection of analytical lines, crystals, detectors, background positions, primary X-ray filters, and matrix-correction methods.

Analytical conditions are selected according to the interference environment surrounding each elemental line. In the heavy-element region, the high angular resolution of a LiF(220) analyzing crystal is frequently combined with a scintillation counter and high-resolution slit to improve peak separation. LiF(200) is used where the wavelength lies outside the accessible scanning range of LiF(220) or where other sensitivity and interference considerations apply.

Alternative analytical lines can also avoid problematic overlaps. As-Kβ1 is used rather than the more intense As-Kα because As-Kα nearly coincides with Pb-Lα. Sm-Lα avoids difficulties caused by Fe interference near Sm-Lβ1, Ce-Lβ1 avoids Ba interference affecting Ce-Lα, and Pr-Lβ1 avoids overlap between Pr-Lα and La-Lβ1.

Primary X-ray filters selectively remove characteristic and continuous X-rays emitted by the X-ray tube before they reach the sample. Proper filter selection can reduce background and tube-related interference, improving the peak-to-background ratio and lowering the detection limit.

Different filters are optimized for different energy regions. For example, a Ni40 filter increased the Sr-Kα peak-to-background ratio from 2.03 to 5.83 and improved the lower limit of detection from 0.93 to 0.77 ppm. An Al125 filter increased the Ce-Lβ1 peak-to-background ratio from 1.87 to 4.42 and improved its detection limit from 11 to 8.4 ppm. A Ni400 filter can also eliminate Rh tube radiation that interferes with Cd-Kα.

Two principal approaches are used depending on the analytical line. For higher-energy lines, the Rh-Kα Compton scattering intensity can serve as an internal standard. Because Compton scattering varies with the mass absorption characteristics of the sample, normalizing the analyte intensity to the scattering intensity compensates for variations in the geological matrix.

For other wavelength regions, absorption corrections based on major components such as TiO₂, CaO, and K₂O are used. The appropriate method depends on the relationship between the analytical line and the absorption edges of major matrix elements. For Sr-Kα, the scattering internal-standard method reduced calibration accuracy error from 58 ppm without correction to 12 ppm. For Ba-Lα, absorption correction was preferable, reducing the error from 65 ppm without correction to 37 ppm.

Yes. A binder can be mixed with powders that do not form mechanically stable pressed pellets on their own. GEO-TRACE-PAK includes a correction that compensates for dilution and changes in scattering intensity caused by binder addition. The binder ratio is entered into the software rather than requiring every sample to be prepared at exactly the same ratio.

Tests using Sr and Ba showed that corrected results for samples containing binder were generally within a few percent of binder-free reference results. However, binder effects become substantially greater for low-energy X-ray lines because the binder absorbs more of the emitted radiation. This limitation is particularly important for fluorine.

Fluorine is particularly sensitive to binder addition because its low-energy F-Kα radiation is strongly absorbed by the binder coating the powder particles. This produces an intensity reduction greater than would be expected from simple dilution, and conventional binder corrections cannot completely compensate for the effect.

In testing, a sample measured at 1262 ppm F without binder produced a result of 1101 ppm with 10% binder, a relative error of about −13%. At 20% binder, the result fell to 848 ppm, an error of approximately −33%. Binder addition should therefore be limited to 10% for GEO-TRACE-PAK samples when fluorine and other light components are important, with the understanding that fluorine values may still be roughly 10% lower than the true concentration.