Thermal ionization mass spectrometers

Neder H, Heusser G, Laubenstein M (2000) Low-level y-ray germanium-spectrometer to measure veiy low primordial radionuclide concentrations. ApplRadiat Isot 53 191-195 Palacz ZA, Freedman PA, Walder AJ (1992) Thorium isotope ratio measurements at high abundance sensitivity using a VG 54-30, an energy-filtered thermal ionization mass spectrometer. Chem Geol 101 157-165... 

Personnel working in some programs at the Los Alamos National Laboratory (LANL) may handle radioactive materials that, under certain circumstances, could be taken into the body. Employees are monitored for such intakes through a series of routine and special bioassay measurements. One such measurement involves a thermal ionization mass spectrometer. In this technique, the metals in a sample are electroplated onto a rhenium filament. This filament is inserted into the ion source of the mass spectrometer and a current is passed through it. The ions of the plutonium isotopes are thus formed and then accelerated through the magnetic held. The number of ions of each isotope are counted and the amount of Pu-239 in the original sample calculated by comparison to a standard. 

The relatively small mass differences for most of the elements discussed in this volume requires very high-precision analytical methods, and these are reviewed in Chapter 4 by Albarede and Beard (2004), where it is shown that precisions of 0.05 to 0.2 per mil (%o) are attainable for many isotopic systems. Isotopic analysis may be done using a variety of mass spectrometers, including so-called gas source and solid source mass spectrometers (also referred to as isotope ratio and thermal ionization mass spectrometers, respectively), and, importantly, MC-ICP-MS. Future advancements in instrumentation will include improvement in in situ isotopic analyses using ion microprobes (secondary ion mass spectrometry). Even a small increase in precision is likely to be critical for isotopic analysis of the intermediate- to high-mass elements where, for example, an increase in precision from 0.2 to 0.05%o could result in an increase in signal to noise ratio from 10 to 40. 

Iron. Fe has 4 isotopes of which the heaviest Fe has a very small abimdance of about 0.3%. The precision of thermal ionization mass spectrometers is around 10 s on this isotope and there is only a hint in some normal inclusions for an excess in 5 Fe (VoUcening and Papanastassiou 1989). Recent ICPMS measurements at the 2 s precision level display normal isotopic compositions for Fe in planetary materials but no Allende inclusion was reported in this study (Kehm et al. 2003). If excesses of similar magnitude to Ca, Ti, Cr were present they would not be clearly resolved in agreement with the observations. When Fe and Fe are used to correct for instrumental mass fractionation, Fe exhibits normal abundances, suggesting all three isotopes are present in solar relative abundances. 

The observed range of natural variations of 5 Ca is about 4 to 5%o in terrestrial materials and up to 50%o in high temperature condensate minerals in carbonaceous chondrites. The typical reproducibility of measurements is about +0.15%o. Broader application of Ca isotope measurements in geochemistry may be possible, particularly if the reproducibility can be improved to 0.05%o to 0.03%o. There is hope that this can be achieved either with inductively coupled plasma source mass spectrometry (Halicz et al. 1999) or with a new generation of multi-collector thermal ionization mass spectrometers (Heuser et al. 2002). 

Thermal-ionization mass spectrometers use a hot filament to ionize the sample. The element of interest is first purified using wet chemistry and then is loaded onto a source filament, often along with another substance that makes ionization easier and a more stable function of temperature. The filament is heated and as the sample evaporates, it is ionized. Both positive and negative ions can be created by thermal ionization, depending on the electronegativity of the element to be measured. Thermal-ionization mass spectrometers are used to measure a wide variety of elements, including magnesium, calcium, titanium, iron, nickel, rubidium, strontium, neodymium, samarium, rhenium, osmium, uranium, lead, and many others. 

Thermal-ionization mass spectrometers (TIMS) combine a hot-filament source with a magnetic-sector mass spectrometer. The mass spectrometers are operated at low to moderate mass-resolving power. A large number of elements can be measured with thermal ionization mass spectrometry. Special care is taken to purify the samples using ion exchange columns. Samples are loaded onto the filaments along with an emitter, and a typical run may take several hours. Modem systems have multiple collectors so that several isotopes can be measured simultaneously. High-precision measurements are done with Faraday cup detectors, but low-abundance isotopes can be measured on electron multipliers. Modem machines are capable of precisions of 0.1 to 0.01 permit. 

A multiple ion collector device is required for the simultaneous determination of separated ion beams in precise and accurate isotope ratio measurements in order to study, for example, isotope fine variation in Nature or during tracer experiments using enriched stable isotope tracers. In thermal ionization mass spectrometers or in ICP-MS, mostly a system of several Faraday cups (up to 16) and/or ion counters (electron multipliers) is applied. In the photographs in Figures 4.7 and 4.8 examples of multiple ion collector systems are shown from the mass spectrometers MC-ICP-MS... 

Figure 5.30 Experimental setup of multiple collector thermal ionization mass spectrometer a) MC-TIMS Triton, Thermo Fisher Scientific, Bremen, Germany. (Reproduced by permission of Thermo Fisher Scientific Bremen.)...

The analysis of stable Isotopes requires extensive sample preparation and sophisticated, expensive Instruments. In our work we use a thermal ionization mass spectrometer. Thermal ionization mass spectrometers cost approximately 300,000. In contrast, analysis of the radioisotope requires little sample preparation and analysis can be done quickly and easily using a gamma counter. 

The mass spectrometer we now use for zinc analysis, in the laboratory of Maynard Michel of Lawrence Berkeley Laboratory, is a thermal ionization mass spectrometer, a single direction focusing instrument with a 12" radius magnetic sector, double filament, rhenium ionizing source and electron multiplier detector. In addition, have done some preliminary work for Fe and Cu analysis with an automated TI/MS which speeds analysis considerably with excellent precision. We hope to be able to develop methods to use this automated Instrument for zinc analysis as well. 

There are few pubhshed Lu-Hf isotope studies of mantle xenohths because of difficulties in efficient ionization of hafnium by thermal ionization mass spectrometers. Multicollector plasma mass spectrometers are a solution to this problem and data are emerging that promise to be a more revealing tool in mantle environments than neodymium isotopes. The variety of Lu/Hf fractionation displayed by mantle minerals (Figure 42) indicates that, as with other isotope systems, isotopic variation should be considerable and initial results are confirming this. Salters and Zindler (1995) found very radiogenic Hf/ Hf at relatively unradiogenic neodymium isotope compositions in spinel peridotites from Salt Lake Crater, Hawaii. Radiogenic Hf/ Hf also characterizes low-T circum-cratonic... 

Samples weighing 2-5 mg are then dissolved in 5-molar nitric acid. The strontium fraction is purified using ion-specific resin and eluted with nitric acid followed by water. This solution is loaded onto a titanium filament for placement in the instrument (Fig. 4.20). Isotopic compositions are obtained on the strontium fraction thermal ionization mass spectrometer (TIMS). This is a single focusing, magnetic sector instrument equipped with multiple Faraday collectors. Strontium is placed on a thin filament and measured. Sr/ Sr ratios are corrected for mass fractionation using an exponential mass fractionation law. Sr/ Sr ratios are reported relative to a value of 0.710250 for the NIST 987 standard (e.g., if the Sr/ Sr ratios for the standards analyzed with the samples average 0.710260, a value of 0.000010 is subtracted from the ratio for each sample). 

Fig. 4.20 Loading sample strontium solution on a filament for measurement in the thermal ionization mass spectrometer (TIMS)...

---