Precise measurement of tellurium isotope ratios in terrestrial standards using a multiple collector inductively coupled plasma mass spectrometry

Five commercially available terrestrial Te standards have been used for the measurement of Te isotope ratios. Atomic absorption spectrometry standard solutions were obtained from Kanto Chemical (Lot no.103G9089, Kanto Chemical Co. Inc., Tokyo, Japan), Wako Chemical (Lot. No. JCF9875 Wako pure chemical industries, Ltd., Osaka, Japan), Aldrich Chemical (Lot no.12329LR, Aldrich Chemical Company, Inc., Milwaukee, WI, USA), Spex CertiPerp (Lot No. 6-250TE, Spex CertiPerp Company, Inc., Metuchen, NJ, USA), and Johnson Matthey Chemicals (Lot No. 801141G, Royston, UK). All standard solutions were concentration at 1000 µg·mL⁻¹. The dilution was carried out in a class 100 laminar flow hood using deionized water (>18 MΩ·cm⁻¹) produced with a Milli-Q system (Merck Millipore, Burlington, MA, USA) and Tamapure AA-100 ultrapure (Tama Chemicals, Kawasaki, Japan) HNO₃.

Te isotope ratio measurements were performed using a Nu Plasma 3D (Nu Instruments Ltd., Wrexham, UK) MC–ICP–MS at National Institute for Quantum and Radiological Sciences (QST), in low-resolution mode using nickel sampler cone and nickel wide-angle skimmer cone. MC–ICP–MS consisted of fixed 21 collectors; among these, 16 were Faraday cups and 5 were Daly collectors (Table 5). The L6 and L7 Faraday cups were equipped with a fixed 10¹² Ω resistor. Ax, H2, H3, H4, H5, H6, H7, and H8 Faraday cups were equipped with fixed 10¹¹ Ω resistors. H1, L1, L2, L3, L4, and L5 Faraday cups were equipped with a switchable dual resistors setup of 10¹¹ Ω and 10¹² Ω, respectively. The switching of the resistor for Faraday cups was performed using the Nu Plasma software. Gain calibration of each Faraday cup resistors was performed every day using the software operated standard procedure by supplying 4 V. It is consecutively applied for all the Faraday cups with 10¹¹ Ω to 10¹² Ω resistors. The gain values were within 20 ppm error range.

Recently, researchers have used Faraday detection systems with amplifiers equipped with 10¹² Ω and 10¹³ Ω resistors in the feedback loop for the measurements of low ion intensities [19,20]. The 10¹² Ω resistor provides 10 times higher voltage compared to 10¹¹ Ω resistor for a given ion beam, whereas the Johnson–Nyquist (JN) noise level of the resistor increases by a factor of 10−−√10. Therefore theoretical 3-fold improvement in the signal to noise ratio is expected but, in practice, this ratio improves only by a factor of two.

The response time of the 10¹² Ω resistors takes time for the signal on the resistors to reach their baseline value, which is slower compared to the 10¹¹ Ω resistors. The curve fitting decay time for each resistor was determined by measuring the signal after closing the second line of sight valve using the Nu Plasma software. The tau corrections were carried out for the 10¹² Ω resistor. Therefore, a relatively slow response does not affect the data quality of MC–ICP–MS analytical performance.

The operating conditions of MC–ICP–MS for Te isotope ratio measurements are given in Table 4. Prior to each measurement session, the instrument was carefully tuned to maximum Te signal intensity by adjusting the torch position, Ar gas flow, lens voltages, and deflector settings. ¹²⁶Te isotope mass is selected as monitoring isotope for peak centering before the measurement of each block. On-peak background subtraction was performed using beam intensities that were measured by introducing 2% HNO₃ before sample measurement.

Kanto chemical Te standard was used as laboratory standard and Te isotope ratio measurements were carried out in both wet plasma and dry plasma mode. In wet plasma mode, the measurement was performed using a 200 ng·mL⁻¹ Te concentration solution. Sample solutions were introduced into the plasma through a micromist nebulizer with an aspiration rate of 200 µL·min⁻¹. After each measurement, a washout was performed with 2% HNO₃ for 15 min. The isotope ratio measurement comprises 10 blocks of 20 cycles with an 8 s integration time. One block consists of 20 cycles (isotope ratios).

In dry plasma mode, Aridus-3 desolvating nebulizer (Teledyne CETAC Technologies, USA) was used where 10 ng·mL⁻¹ Te concentration solution was aspirated at a rate of 100 µL·min⁻¹. The Ar sweep gas flow rate was typically 4.25 L·min⁻¹ with a nebulizer gas flow rate of 0.9 L·min⁻¹. No additional N₂ gas was used for the measurement. The other instrumental operating settings are identical to wet plasma measurements. The washout was approximately 30 min using a 2% HNO₃. Te standard with a concentration of 10 ng·mL⁻¹ and 200 ng·mL⁻¹ was analyzed before and after every terrestrial Te standard to confirm the absence of drift in dry and wet plasma mode of measurement, respectively. The amount of Te consumed in one measurement was approximately 27 ng and 1067 ng for dry and wet plasma mode, respectively.

There were 11 Faraday cups used for the simultaneous collection of ion beams. The L5 Faraday cup resistor setup was changed from 10¹¹ Ω to 10¹² Ω resistor. The measurement of Te isotope ratios was carried out using Faraday cups with a mixed resistor of 10¹¹ Ω and 10¹² Ω in a static cup configuration mode. The ¹¹⁸Sn, ¹²⁰Te, were collected using L6 and L5 10¹² Ω resistor Faraday cups, and ¹²¹Sb, ¹²²Te, ¹²³Te, ¹²⁴Te, ¹²⁵Te, ¹²⁶Te, ¹²⁸Te, ¹²⁹Xe, and ¹³⁰Te were detected using L4, L3, L2, L1, Ax, H1, H3, H4, and H5, 10¹¹ Ω resistor Faraday cups, respectively (Table 5).

The Te isotope ratios (all ratios) obtained from the Faraday cups were corrected for mass fractionation by normalizing with ¹²⁴Te/¹²⁸Te = 0.14853 [11] and ¹²⁵Te/¹²⁸Te = 0.22204 [1] using exponential fractional law. However, the previously published isotope results have used different normalization ratios to correct for internal normalization. ¹²⁴Te/¹²⁸Te normalization was carried out for TIMS and N–TIMS measurements. Due to ionization potential, isobaric interference of Sn and Xe could be controlled [6,10]. The direct comparison of data with the previously published data is possible [10,11,12,15,23]. Since there is no isobaric interference of Xe or Sn with ¹²⁵Te, it was preferred in MC–ICP–MS studies. The internal normalization of ¹²⁵Te/¹²⁸Te was selected for mass bias correction, because the mass-bias-corrected ¹²⁶Te/¹²⁸Te and ¹³⁰Te/¹²⁸Te isotope data do not require any Sn correction, which is very much important for the sample containing a significant amount of Sn.

The possible isotopic interference for Te isotopes was listed and the interference corrections for Te isotopes were carried out by the earlier workers [11,12,23]. A similar correction procedure was followed for the Te isobaric correction in this study. The major isobaric interferences on Te isotopes can be generated from Sn, Sb, and Xe. Therefore, the ion currents of ¹¹⁸Sn⁺, ¹²¹Sb⁺, and ¹²⁹Xe⁺ were measured during the measurement to do interference corrections. Xenon is present in the Ar plasma gas and the typically reported yield of Xe/Te ratios are 4–9 × 10⁻⁴ [12]. The correction of ¹¹⁸Sn on ¹²⁰Te, ¹²²Te, ¹²¹Sb on ¹²³Te, ¹²⁴Te, and ¹²⁹Xe on ¹²⁴Te, ¹²⁶Te, ¹²⁸Te, and ¹³⁰Te isotopes were applied online during the measurement using exponential law. The intensities of ¹¹⁸Sn, ¹²¹Sb, and ¹²⁹Xe were 2.3 × 10⁻¹⁶ A, 2.05 × 10⁻¹⁵ A, and 6.28 × 10⁻¹⁶ A for wet plasma condition, and 1.5 × 10⁻¹⁶ A, 8.8 × 10⁻¹⁶ A, and 9.6 × 10⁻¹⁵ A for dry plasma, respectively. The typical production rate of tellurium hydride (TeH) was 2 × 10⁻⁴ and <1 × 10⁻⁵ for wet and dry plasma, respectively. The TeH correction was not necessary for the dry plasma condition because of its low production rate [15]. As it is already mentioned by the earlier studies that Te standard solutions typically displayed Sn/Te concentration ratios of about 1–3 × 10⁻⁴ and Sn/Te ratios of up to 1.5 × 10⁻³ could be tolerated for samples without compromising the accuracy of the analytical results [12].