Paving the way for new and challenging matrix reference materials-particle suspensions at the core of material processing providing RMs for method development and method validation.

Sufficient homogeneity of the certified parameter(s) over the whole fill series of a matrix reference material (RM) is a fundamental quality criterion. In practice, the heterogeneity of the target parameter is evaluated, whereby a relative value can be calculated of how much the target parameter is varying over the RM-batch. A high degree of homogeneity (low heterogeneity) is an inherent quality mark of a good RM. Here, we report how challenging matrix RMs were produced by using particle suspensions at the core of the material processing step. The examples of matrix RMs produced span from whole water reference materials for persistent organic pollutants, PM2.5-like atmospheric dust certified for specific ions to microplastic RMs. Most of these RMs were subsequently used in different phases of analytical method development or for method validation. Common to all these matrices is that they cannot be easily mixed, handled, or dosed to prepare larger sample batches. In all cases, a continuously stirred suspension of particles was used during material processing. In general, relative between-bottle heterogeneities from 1.6 to 6% were achieved for the target parameters in these matrix presentations. Concerning developments of new CRMs in emerging fields, the co-dependence between the availability of validated analytical methods with good repeatability and testing materials with a known and high homogeneity of the target parameter(s) becomes particularly challenging. This situation is an RM/Method causality dilemma. To overcome that hurdle, strategies are proposed for stepwise processes where RM producers and a network of analytical method developers could work hand in hand. In addition, development of a portfolio of inexpensive and well-homogenised common samples coupled with a reporting interface is suggested. This would benefit method developers and RM producers alike. As more and more data is compiled for a specific matrix, it paves the way for new and challenging RMs that can later be used by a wider community.

Cadmium determination in natural waters at the limit imposed by European legislation by isotope dilution and TiO2 solid-phase extraction.

The cadmium content in surface water is regulated by the last European Water Framework Directive to a maximum between 0.08 and 0.25 µg L(-1) depending on the water type and hardness. Direct measurement of cadmium at this low level is not straightforward in real samples, and we hereby propose a validated method capable of addressing cadmium content below µg L(-1) level in natural water. It is based on solid-phase extraction using TiO(2) nanoparticles as solid sorbent (0.05 g packed in mini-columns) to allow the separation and preconcentration of cadmium from the sample, combined to direct isotope dilution and detection by inductively coupled plasma mass spectrometry (ID-ICP-MS). The extraction setup is miniaturised and semi-automated to reduce risks of sample contamination and improve reproducibility. Procedural blanks for the whole measurement process were 5.3 ± 2.8 ng kg(-1) (1 s) for 50 g of ultrapure water preconcentrated ten times. Experimental conditions influencing the separation (including loading pH, sample flow rates, and acid concentration in the eluent) were evaluated. With isotope dilution the Cd recovery rate does not have to be evaluated carefully. Moreover, the mathematical model associated to IDMS is known, and provides transparency for the uncertainty propagation. Our validation protocol was in agreement with guidelines of the ISO/IEC 17025 standard (chapter 5.4.5). Firstly, we assessed the experimental factors influencing the final result. Secondly, we compared the isotope ratios measured after our separation procedure to the reference values obtained with a different protocol for the digested test material IMEP-111 (mineral feed). Thirdly, we analysed the certified reference material BCR-609 (groundwater). Finally, combined uncertainties associated to our results were estimated according to ISO-GUM guidelines (typically, 3-4% k = 2 for a cadmium content of around 100 ng kg(-1)). We applied the developed method to the groundwater and wastewater samples ERM-CA615 and BCR-713, respectively, and results agreed with certificate values within uncertainty statements.

Large volume injection in ion chromatography Separation of rubidium and strontium for on-line inductively coupled plasma mass spectrometry determination of strontium isotope ratios.

Large volume injection, up to 5 mL, was evaluated and optimised for the on-line ion chromatographic separation of Rb and Sr before ICP-MS measurement of Sr isotope ratios. Flat-topped chromatographic peaks, ideally suited for multicollector ICP-MS isotope ratio measurements, could be obtained when the composition of the mobile phase (nitric acid and 18-crown-6 ether) was identical to the matrix of the sample. Under those conditions rubidium eluted at the dead volume of the column while strontium produced a flat-topped transient signal with several minutes of stable plateau. On-line data acquisition during several minutes at the plateau of Sr signal allowed high precision Sr isotope ratio measurement. The developed procedure was evaluated for Sr isotope ratio measurements on different types of samples, including cider, apples, apple leaves, and soil extracts, in the frame of a long-term project aiming at origin authentication using strontium isotope ratio measurements. It was observed that sample matrix caused broadening of the strontium chromatographic peak and loss of flat-topped peak profile. Under those circumstances the addition of the complexing crown-ether 18-crown-6 both to samples and chromatographic eluent provided two distinct advantages. First, a drastic increase in the retention of strontium was observed which could be modulated by increasing the concentration of nitric acid in the eluent up to 900 mM. This increase in the eluent HNO(3) concentration allowed the application of the method to acid soil digests and other high acidity samples. Second, the matrix of the sample did not affect any more the chromatographic peak profile and similar chromatographic separations could be obtained for samples and standards maintaining the flat-topped Sr peak profile. Sample preparation consisted of a simple 1:10 dilution of the cider or pre-treated solid samples by adding HNO(3) (900 mM) and 18-crown-6 ether (5mM) to obtain similar composition in the sample solution and the HPLC eluent.

Evaluation of strontium isotope abundance ratios in combination with multi-elemental analysis as a possible tool to study the geographical origin of ciders.

In order to evaluate alternative analytical methodologies to study the geographical origin of ciders, both multi-elemental analysis and Sr isotope abundance ratios in combination with multivariate statistical analysis were estimated in 67 samples from England, Switzerland, France and two Spanish regions (Asturias and the Basque Country). A methodology for the precise and accurate determination of the (87)Sr/(86)Sr isotope abundance ratio in ciders by multicollector inductively coupled plasma mass spectrometry (MC-ICP-MS) was developed. Major elements (Na, K, Ca and Mg) were measured by ICP-AES and minor and trace elements (Li, Be, B, Al, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Rb, Sr, Y, Mo, Cd, Sn, Sb, Cs, Ba, La, Ce, W, Tl, Pb, Bi, Th and U) were measured by ICP-MS using a collision cell instrument operated in multitune mode. An analysis of variance (ANOVA test) indicated that group means for B, Cr, Fe, Ni, Cu, Se, Cd, Cs, Ce, W, Pb, Bi and U did not show any significant differences at the 95% confidence level, so these elements were rejected for further statistical analysis. Another group of elements (Li, Be, Sc, Co, Ga, Y, Sn, Sb, La, Tl, Th) was removed from the data set because concentrations were close to the limits of detection for many samples. Therefore, the remaining elements (Na, Mg, Al, K, Ca, Ti, V, Mn, Zn, As, Rb, Sr, Mo, Ba) together with (87)Sr/(86)Sr isotope abundance ratio were considered for principal component analysis (PCA) and linear discriminant analysis (LDA). Finally, LDA was able to classify correctly 100% of cider samples coming from different Spanish regions, France, England and Switzerland when considering Na, Mg, Al, K, Ca, Ti, V, Mn, Zn, As, Rb, Sr, Mo, Ba and (87)Sr/(86)Sr isotope abundance ratio as original variables.