Prepare all nitric (HNO₃) and hydrofluoric (HF) acid solutions from double distilled acids (GFS Chemicals Inc. Columbus, OH). Use water deionized to ≥18 mΩ·cm for all solutions (e.g., as produced by an Aqua Solutions Ultrapure Water System - Aqua Solutions, Inc., Jasper, GA). Collect contributions for base urine pools from healthy volunteers with measured total urine uranium concentrations of < 5 ng /L, and acidify them to 1% v/v HNO₃. Prepare natural uranium QC solutions by spiking base urine with dilutions of a uranium standard, (SPEX Industries, Inc., Edison, NJ) traceable to the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA). Prepare aqueous CRM solutions by dissolution of uranium oxide CRMs in powder or pellet form.²² Prepare low uranium ratio QC solutions by spiking base urine with dilutions of CRM NBL U005-A, NBL 115 (U.S. Department of Energy, New Brunswick Laboratory, Argonne, IL) and a mixture of natural uranium and NBL U005-A. Prepare high uranium ratio QC solutions by spiking base urine with dilutions of CRM NBL U015 (U.S. Department of Energy, New Brunswick Laboratory, Argonne, IL) and mixture of natural uranium and NBL U015.

The optimum urine sample volume is 2 mL unless the total uranium concentration is greater than 300 ng/L or less than 100 ng/L. For samples with uranium concentrations higher than 300 ng/L, use 1 mL (or less, depending on the uranium concentration) of the sample, diluted to ~ 200 ng/L for analysis. Use 4 mL or more of the sample, as appropriate, if the total uranium concentration is less than 100 ng/L. Acidify urine sample by adding 375 µL concentrated HNO₃ per mL of (diluted, if appropriate) urine sample using 4 mL polystyrene sample cups (VMR, Suwanee, GA) prewashed with 5% v/v HNO₃. Pour the contents into individual 0.22g TRU resin (Eichrom, Darien, IL) solid phase extraction columns, previously washed extensively with ≥18 mΩ·cm water , 10% v/v HNO₃ (30 mL) and 5% v/v HF (30 mL × 2) and equilibrated with 10% v/v HNO₃ (5 – 6 mL). Next wash each column with two to five mL 10% v/v HNO₃, pouring each wash through the column, followed by two 5 mL 10% v/v HNO₃ column washes to force unretained ions through each column. Elute the uranium from the samples into acid-cleaned 4 mL polystyrene conical bottom sample cups (Thermo-Fisher Scientific, Suwanee, GA) with 2 mL 5% v/v HF, or proportionately higher volumes for dilution purposes. The reagent blank for this method is 5% v/v HF (Figure 1).

This method measures Uranium isotope ratios using an extended dynamic range high resolution ICP-MS model Element XR (Thermo Fisher Scientific, Bremen, Germany), which is a double focusing magnetic sector field inductively-coupled-plasma mass spectrometer with a high performance, discrete dynode, dual mode secondary electron multiplier detector and a high current (for percent level content) faraday detector (Mascom, Bremen, Germany). It uses the ICP-MS, equipped with nickel sampler and skimmer cones and a CD-2 guard electrode, in triple mode. The sample introduction system consists of a computer controlled ASX-112 (Cetac, Omaha, NE) autosampler and Aridus II™ (Cetac, Omaha, NE) desolvation unit. Samples self-aspirate from the autosampler into the desolvation unit through an Apex perfluoroalkoxy (PFA) 100 µL/minute nebulizer (ESI, Omaha, NE, or equivalent). Sample desolvation occurs within the AridusII™ unit in a PFA spray chamber (Cetac, Omaha, NE) set at 110 °C. With the aid of argon sweep gas and nitrogen gas for sensitivity enhancement, the sample passes through a semi-permeable membrane coil set at 160°C. Optimize flow rates as needed, with argon sweep gas at ~ 2–7 L/min and nitrogen gas at ~ 3–9 mL/min. The desolvated sample exits the desolvating unit into a 1.8 mm I.D. sapphire injector and a standard quartz torch, and then into the mass spectrometer.

Some method parameters are crucial for the precision of uranium isotope ratios determination by SF-ICP-MS. Table 1 reflects optimized method parameters, including numbers of method passes, runs and samples based on the precision of the measurement of uranium isotope ratios. Further, the method requires the performance of individual run experimental parameter optimizations with respect to maximum ion intensity of ²³⁸U and minimum uranium oxide formation rate using a tuning solution containing natural uranium. Table 1 also contains summary examples of these optimized operating conditions.

The sample separation protocol uses solid phase chelation extraction (SPE), with TRU columns for separation of the actinides from the urine matrix, which is a modification of previously published methods for determination of uranium isotope ratios in urine.^{20, 21, 23} Because the Aridus II™ demonstrates exceptional signal stability, and in order to avoid introduction of interference or contamination into the samples, this method uses no tracer (or internal standard). SPE separation recovery is evaluated by analyzing the spiked internal quality control material both before and after running the sample material through the columns. The recoveries are acceptable, at approximately 90% to 104%. We use a 5% v/v HF solution as the reagent and instrument blank for this method since it is the sample matrix introduced to the ICP/MS, and the results obtained for blank samples (2 mL of water), which underwent exactly the same chemical separation procedure as the urine samples, showed very similar cps to 5% v/v HF.

Sensitivity requirements for ²³⁴U and ²³⁶U are extremely stringent, and this limits the utility of SF-ICP-MS measurements. In order to determine possible improvement of instrumental sensitivity by use of an Aridus II™, we analyzed a 5 ng/L tuning solution which contained natural uranium. The Aridus II™ sensitivity for uranium is approximately 10 times higher than that of a Meinhard® quartz nebulizer (Type TQ-30–43) and the quartz cyclonic spray chamber (~3.0 × 10⁵ cps/ppt versus 3.0 × 10⁴ cps/(ng/L). The acid trap bottle (1M NaOH), connected to the sweep gas output line of the Aridus II™, affects the back pressure of the instrument’s introduction system, and has a significant effect on instrument sensitivity. The position of this gas line under the surface of the NaOH solution is very important and must be checked/optimized for best instrument performance.

Previous CDC methods, ^{20, 21} in which the target ²³⁸U eluent concentration from the TRU column is ~ 40 ng/L, use one detection mode (Secondary Electron Multiplier, or SEM, in counting mode) for the ²³⁵U/²³⁸U isotope ratio measurement.^{20, 21} For this method, we increased the target ²³⁸U eluent concentration from the TRU column to ~200 ng/L because of the extremely different isotopic abundances among ²³⁴U, ²³⁵U, ²³⁶U, and ²³⁸U. Thus, the “counting mode only” method, with its limited dynamic range, is no longer practical. This method uses the “Triple” mode, where both SEM Counting and Analog detection modes are used, to perform the analysis. Counting mode is used for the low abundance ²³⁴U, ²³⁵U and ²³⁶U isotopes and analog mode is used for ²³⁸U. It is important to run a short, separate method before each batch of analysis with ²³⁸U at about 2 to 4 million cps in “Triple” mode for >10 scans, which updates the analog conversion factor (ACF), i.e., the factor between SEM counting and analog detection modes (for cross calibration between modes). Optimal samples per peak for this method is 600, and 1% of the Mass Window (see Table1) is used to avoid continuous automatic ACF updates for all analyses.

The abundance sensitivity of the instrument is critically important for the ²³⁶U/²³⁸U measurement since ²³⁶U is present only in very small proportion relative to ²³⁸U. The abundance sensitivity during the experiments was 2.3 × 10⁻⁶, which was calculated by averaging the measured ²³⁶U/²³⁸U ratio of the spiked natural uranium internal quality control material during the QC characterization of this method. ²³⁶U/²³⁸U ratio results for all samples in this method are mathematically corrected for abundance sensitivity.

The Aridus II™ desolvating system reduces the contribution of ²³⁵U¹H to the ²³⁶U signal. We used the signal at mass M = 239 (due to ²³⁸U¹H⁺) to estimate the significance of any remaining ²³⁵U¹H⁺ contribution to the ²³⁶U signal at mass M = 236. Our experiments showed that this hydride contribution to the ²³⁹U/²³⁸U signal, with an average value of 1.2 × 10⁻⁵, is equivalent to < 1 × 10⁻⁷ for the analytically more important ²³⁶U/²³⁸U ratio, and can therefore be ignored for this method.

CDC’s procedures require investigation of the uranium isotope ratios in patient urine samples when the study protocol calls for it, or if a urine specimen is determined to have an ‘elevated’ total uranium concentration. For this method, based on the known ²³⁸U concentration in the urine sample determined by the other available methods, the volume of urine sample loading on the TRU column is determined to target the uranium concentration of the column eluent at approximately 200 ng/L (as is, diluted or pre-concentrated). Thus, the linearity range requirement is not critical as long as the total U concentration is < 300 ng/L. However, we tested the linearity range using aqueous solutions of CRM NBL U005-A, with total uranium concentrations ranging from 5 ng/L to 400 ng/L, and calculated the corresponding concentrations of ²³⁴U, ²³⁵U, ²³⁶U and ²³⁸U based on the certified uranium isotope ratios of NBL U005-A for comparison. The results are summarized in Figure 2.

We performed analyses of aqueous CRM solutions from the U. S. Department of Energy’s New Brunswick Laboratory, and the observed uranium isotope ratios were in good agreement with the certified values. Table 2 shows this, along with the typical precision observed at low, medium and high ratios of daily quality control materials that were analyzed at the beginning, in the middle and at the end of each analytical run.

The method was best optimized for a total uranium concentration between 100 ng/L and 300 ng/L. Limit of detection (LOD) determination is not critical for this method since it is generally intended for ratio determination in samples with relatively high (> ~200 ng/L) total uranium content, and because ratios present are of concern across a defined, detectable range (~0.2–95% for ²³⁵U/²³⁸U). However, in order to confirm the concentration range for which reliable uranium isotope ratios may be determined, we prepared three urine pools with lower concentrations (spiked with CRM NBL U005-A, at 50 ng/L, 100 ng/L and 150 ng/L of total uranium) than that of the target total uranium concentration (200 ng/L) and analyzed these samples during the internal quality control materials’ characterization procedure. Analyses at all of these concentration levels showed good precision. However, samples with a total uranium concentration of < 100 ng/L (~3.54 pg/L of ²³⁴U, and ~1.18 pg/L of ²³⁶U) had a trend of increased ratio RSD and might therefore require more volume for pre-concentration. Samples with uranium concentrations > 300 ng/L may be pre-diluted with 1 mL of 5% v/v HNO₃ or less to 200 ng/L, depending on uranium concentration (Figure 3).