An on-line solid-phase extraction method linked to inductively coupled mass spectrometry (ICP-MS) has been developed for the determination of low levels of uranium and thorium in environmental samples. The hyphenation of lab-on-valve (LOV) and multisyringe flow injection analysis (MSFIA), coupled to an ICP-MS, allows the simultaneous determination of thorium and uranium in different types of environmental sample matrices achieving high selectivity and sensitivity levels. On-line separation and preconcentration of thorium and uranium are carried out by means of UTEVA resin. The potential of the LOV-MSFIA makes possible the full automation of the system by the on-line regeneration of the column. The limits of detection reached are 0.4 ng L⁻¹ of uranium and 2.8 ng L⁻¹ of thorium. The reproducibility of the LOV-MSFIA-ICP-MS is 1.7% of RSD. Moreover, a high sensitivity, a wide working range (0–200 μg L⁻¹ for uranium and thorium) and an injection frequency up to 9 h⁻¹ (depending on the sample volume) should be highlighted. Different water sample matrices (seawater, well water, freshwater, tap water and mineral water), a phosphogypsum sample with natural uranium and thorium content and a channel sediment reference material were satisfactorily analyzed with the proposed method.

A multicommuted system has been developed for lead (Pb) isolation, pre-concentration and determination using an atomic fluorescence spectrometer as the detector. The lab-on-valve (LOV) technique allows automatic separation and pre-concentration of lead by means of solid phase extraction. The Pb resin was conditioned with nitric acid (1 mol L⁻¹) and the analyte was eluted with 1.3 mol L⁻¹ glycine solution. Chemical and flow variables affecting the extraction/pre-concentration of Pb²⁺ were investigated by a multivariate approach. First, screening of the independent variables and their interactions was carried out by a two level full factorial design (2^k). Then, a face centered central composite design was performed in order to find the optimal values of the significant variables. Because a wide range of sample volumes can be loaded on the microcolumn (1–100 mL), a mass calibration curve was obtained in order to quantify Pb in sample batches with a wide variability in lead concentration. Under optimal conditions, the limit of detection and the quantification limit were 0.004 and 0.014 ng of Pb²⁺, respectively. The linear response range was 0.014–20 ng of Pb²⁺. The repeatability, expressed as relative standard deviation (RSD), is 2.4% for 4 μg L⁻¹ Pb²⁺ (n = 10). A small amount (0.6 g) of Pb resin is enough to fill the microcolumn of the LOV device. The durability of the resin reached 154 injections. The injection frequency permitted between a sample frequency of 1–10 h⁻¹ depending on the sample pre-concentration volume. This method was validated using certified reference materials such as SRM-1400 (bone ash), SRM-2976 (mussel tissue) and BCR-610 (groundwater), and the results obtained showed good agreement with the certified values. The proposed LOV system was successfully applied to different samples of environmental and biological interest (water, fish and human teeth) with satisfactory recoveries, between 90–110%. The proposed system has some advantages such as minimization of sample handling, sensitivity, good precision, reduction of reagent volumes, high durability of the column and the versatility achieved by a variety of managed sample volumes.