Matrix-matched calibration in LA-ICP-MS of silicate, phosphate and carbonate minerals: application of G-Probe samples

Laser ablation (LA) sampling provides fast microelement ICP-MS analysis of a wide range of solid materials without their dissolution, thus decreasing contamination from water and reagents as well as reducing polyatomic isobaric interferences from acid solutions. However, the issue of matrix-matched calibration becomes crucial for LA-ICP-MS due to differences in behaviour during laser interaction and evaporation of solid samples. There are several approaches to LA calibration: simultaneous supply of standard solutions into a spray chamber; calibration using a set of NIST 61х synthetic glasses and glasses prepared from natural rocks and minerals (basalt, nephelinite, etc.) or pressed synthetic samples (calcium carbonates, phosphates and sulphides produced by USGS). A set of natural glasses for microanalysis is available from the International Association of Geoanalysts (IAG) in co-operation with the USGS. The G-Probe proficiency testing programme has been operating since 2008 and deals with solid samples for microanalysis (LA-ICP-MS, EPMA, EDS-SEM). A number of samples of different compositions were distributed: BBM-1G and BSWIR-1G natural basaltic glasses, GSM-1 gabbro; NIST SRM-based basaltic and diabase glasses; GP-MACS synthetic pressed calcium carbonate, GP-MAPS phosphate and some others. The aim of the present work was to estimate the LA-ICP-MS analysis quality using matrix-matched calibration with G-probe samples of various composition. All G-Probe samples were analysed using an ELAN 9000 Q-ICP-MS combined with a LSX-500 (Nd:YAG, 266 nm) laser ablation system. For silicate rocks, TB-1 basaltic glass was used for calibration; the remaining samples were analysed as unknowns. MAPS-4 calibration material were used for phosphate rock analysis. A combination of external matrix-matched calibration and internal normalisation was used for calculating element concentrations. LA-ICP-MS analysis quality was estimated using z-scores. Most of the results obtained were in a good agreement with assigned values.


Introduction
Laser ablation (LA) sampling in inductively coupled plasma mass spectrometric (ICP-MS) analysis allows the rapid analysis of the trace element composition of solids to be carried out without their dissolution. At the same time, the contamination from the reagents used is minimised, and polyatomic isobaric interferences arising due to the presence of acid solutions are significantly reduced [1]. LA-ICP-MS is widely applied in the studies of microobjects with high spatial resolution, individual mineral grains, and spatial distributions of trace elements.
However, the issue of matrix-matched calibration becomes crucial for LA-ICP-MS of solids due to differences in behaviour during laser interaction and evaporation of solid samples, especially when using 193 nm excimer and 213 and 266 nm Nd:YAG lasers [2].
To date, there are several widely practiced approaches to constructing calibration curves in the analysis of solid samples of various compositions using laser ablation sampling: some include the simultaneous supply of aqueous calibration solutions to the spray chamber of a mass spectrometer [3,4]; others use synthetic glasses of the NIST SRM 61x series (manufactured by the National Institute of Standards and Technology, USA), as well as glasses made by fusion of natural rocks and minerals (basalt, nephelinite, etc.) or pressed synthetic non-silicate samples (calcium carbonates and phosphates, sulphides, US Geological Survey, USA).
The first approach leads to a complication of design features (for example, the manufacturing, often in-house, of chambers with additional inputs for solutions, the connection of an additional gas for spraying), as well as the formation of polyatomic isobaric interferences from the solvents (water and acids). The use of synthetic glasses for calibration provides a certain unification of the results obtained, but can be justified only in some cases (for example, for the analysis of silicate samples), while their composition does not reflect the natural sample composition with their wide range of trace element content [5]. For example, about 33 microelements with concentrations from 15 to 80 μg/g are certified in the most widespread NIST SRM 612 glass, which is insufficient when analysing the entire variety of rocks and minerals.
Taking into account the above-mentioned, the most acceptable way to calibrate a mass spectrometer when analysing solid samples is to use matrix-matched calibration samples, and NIST SRM 612 glasses to optimise the analytic parameters of the device (adjusting the interface, monitoring the level of oxides, etc.).
The range of solid reference materials (RM) of natural composition has expanded significantly at the moment: The US Geological Survey produces 4 natural basalt glasses and 1 nephelinitic glass, which are the melts of powdered BCR-2, BHVO-2, BIR-1, TB-1 and NKT-1 certified reference materials (SRM) [5], respectively, as well as synthetic pressed MACS-3 calcium carbonate and MAPS-4 phosphate and MASS-1 polymetallic sulphide.
The glass making procedure [5] involves the fusion of the powdered rock material in an oven in a platinum crucible at 1325 -1645°C for 2-6 hours with several stirrings with a platinum rod and subsequent rapid cooling in deionised water. The pieces of glass are then dried, fixed in epoxy resin and distributed to analytical laboratories. For pressed powdered samples [6], a specially developed procedure of trace element co-precipitation with pure calcium carbonate or phosphate in a reaction vessel is used. The resulting suspension is powdered to less than 40 μm, dried at 110 °С and pressed into pellets with a diameter of 19 mm. The homogeneity and composition of the samples obtained is confirmed by a number of studies in USGS laboratories (XRF, LA-ICP-MS, EPMA, etc.). Thus, the samples described above are microanalytical reference materials (MRM), that is, they have passed the certification procedure and issued certificates with certified concentrations of major and trace elements.
An expanded set of natural glasses for microanalysis is presented as part of the G-Probe

Experimental
The following geological glasses and microanalytical reference materials were studied (Table   1). The following LA operational parameters (Table 2) were used when analysing the samples of various compositions based on the previously obtained data taking into account the specific features of rock/mineral laser evaporation [8,9].  In this case, finding the analyte content in the sample is carried out according to the formula: Thus, the silicate samples were analysed using a TB-1G basalt glass for calibration with internal silicon standardisation. Phosphate sample was analysed using MAPS-4 synthetic phosphate RM for calibration and internal calcium standardisation.

Results and discussion
Scoring and statistical analysis in GeoPT is undertaken according to the ISO 13528 Standard relating to statistical methods used in proficiency testing [11] based on the earlier recommendations of the IUPAC International Harmonised Protocol [12].
According to the GeoPT Protocol [13], the results of the analysis are evaluated using z-scores in the form: where x i is the result of the analysis of a particular laboratory, x pt is the assigned value of the element content in the test sample, σ pt is the corresponding standard deviation for proficiency testing (SDPT), or target precision, based on a GeoPT fitness for purpose criterion.
In the G-Probe programme, the values of element concentrations obtained during analysis in USGS laboratories, as well as the data from NIST SRM certificates (in case glass was fused from a standard sample), and results of analysis of bulk powder samples from previous rounds are taken as x pt assigned elemental content.
In the Protocol of the G-Probe programme, a model of the standard deviation dependency on concentration is adopted as an σ H estimate of the target precision in the form of the Horwitz function [14]: where the values of concentrations and precision should be expressed in mass fractions (for example, 1 ppm = 10 -6 , 1% = 0.01); the coefficient k = 0.02 corresponds to the results of the second category of results -applied geochemistry.
Accordingly, a z-score outside the range ±3 implies that an unacceptable source of bias may be present in the participant's analytical system and that the need for remedial action should be considered. Z-scores more extreme than ±2 carry the same message to a lesser degree [13].
However, an assumption has been made by [15] that The results of LA-ICP-MS analysis and z-scores are given in Supplementary Table S1. The measurement result is the mean value of two measurements performed on two different sample fragments each consisting of two parallel measurements.
For a number of samples studied (Table 1), the values of the z-scores were calculated in accordance with the Protocol of the G-Probe programme, as well as the z'-scores using the dependency of the standard deviation for low concentrations S = 0.035 • C 0.8495 (C≤0.1%) since almost all the concentrations of determined elements in studied samples were less than 0.1% [15]. When the quality assessment is performed using the z'-scores, a greater number of results are recognised as satisfactory, and only some elements still fall outside the range ±3.
Most often, during LA-ICP-MS analysis of geological samples, a number of elements are of the greatest interest -these are the rare-earth elements (REE), yttrium, uranium, thorium and lead.
To illustrate the quality of their analysis, z (z') plots were constructed (Fig. 2).

Conclusions
This study describes the matrix-matched calibration approach to the analysis of geological