W. M. Keck Cosmochemistry Laboratory - Facility Description

The Cameca ims 1280 ion microprobe

The Cameca ims 1280 microprobe in the W. M. Keck Cosmochemistry Laboratory.


Shown above is a schematic diagram of the Cameca ims 1280 [click for enlargement]. The primary ion column, shown in yellow on the left, provides a highly focused ion beam produced in either a duoplasmatron source (for O, Ar, and other gases) or in a microbeam Cs+ source. The sample is located where the primary ion column and the secondary-ion mass spectrometer join (lower left of the the image). The secondary-ion mass spectrometer, shown in light green, is a double focusing mass spectrometer with both electrostatic and magnetic sectors. Ions are detected with three separate systems. There is a single collector, which uses either an electron multiplier or Faraday cup. This is used for the applications requiring the highest mass resolving power. There is also a multi-collector (light blue on the right), which has five detectors and can be configured with any combination of electron multipliers or Faraday cups. And the machine can be used in imaging mode with a 2-dimensional channel plate and fluorescent screen. The new SCAPS detector is mounted in place of the channel plate detector.


The Cameca ims 1280 ion microprobe is a modern incarnation of Cameca's large geometry ion microprobe (Conty, 1990). Our ims 1280 (photo on right) was commissioned in March, 2006. Since that time we have added a SCAPS solid-state imaging detector (see below).

An ion microprobe (or Secondary Ion Mass Spectrometer; SIMS) works by generating a beam of charged particles (ions) to sputters atoms from the sample. Some of the sputtered atoms are ionized during sputtering. These secondary ions are extracted from the sample chamber and transferred into the mass spectrometer and measured. The secondary ion mass spectrometer must have high mass-resolving power because the sputtering process generates ions of every element in the sample and a wide variety of molecular compounds. In many types of mass spectrometry, samples are purified before being introduced into the mass spectrometer. The mass spectrometer only has to separate the isotopes of the elements of interest and perhaps a few interfering molecules. In the SIMS technique, the mass spectrometer must do all of the work, so the mass spectrometer has very high performance. A major advantage of the SIMS technique is that samples can be analyzed while retaining the petrographic context, allowing more information to be gathered from the sample. In the past, there was a big trade-off because the SIMS technique could not produce highly precise data. This changed with the ims 1280. New electronics permit much tighter control of instrument conditions. Run setup parameters can be saved, improving day-to-day reproducibility. It is now relatively routine to obtain 18O/16O ratios with a precision and accuracy of ±0.5 ‰ and data for 17O/16O and 18O/16O with a precision and accuracy of ~1 ‰. Magnesium isotopes in high-magnesium samples can be measured with precision and accuracy of better than 100 ppm. This level of precision permits investigation of a wide variety of problems using SIMS. ICPMS and Thermal Ionization Mass Spectrometry can obtain data with smaller uncertainties, but at the cost of petrographic context.

The ims 1280 consists of a primary ion column (highlighted in yellow in schematic diagram on right), a double focusing, magnetic sector mass spectrometer (green in the schematic), and a variety of detectors (the region highlighted in blue). The sample is introduced into the sample chamber, at the intersection of the primary column and the mass spectrometer (lower left in schematic) using an airlock system (purple in schematic).

Primary Ion Column: The primary ion column has two ion sources, a duoplasmatron and a microbeam cesium source. Two sources are used to measure different elements.

We measure elements that preferentially make positive ions using the duoplasmatron and 16O- primary ions. High-purity oxygen is introduced into the duoplasmatron where it is ionized to make a plasma. Negative oxygen ions are extracted from the source and are accelerated into the primary column at 13 KeV. 16O- ions are separated from 17O- and 18O- at the Primary Beam Mass Filter (an electromagnet) and are focused by several electrostatic lenses. The minimum spot size that can be obtained for 16O- is about 1 micron. The duoplasmatron can also be operated to make 16O+, 16O2+, and 16O2- beams. Argon and other gases can also be run in the duoplasmatron. A beam of 16O- ions (or other charged particles) hitting an insulating sample results in a charge buildup on the sample. This charge can deflect the primary beam and change the effective acceleration voltage of the mass spectrometer. To avoid this, samples are coating with carbon or gold to conduct away the accumulating negative charge.

We measure elements that make negative ions using the microbeam cesium source. The cesium is loaded into the source reservoir as CsCO3. Gentle heating of the reservoir during the initial conditioning of the source degasses the carbonate from the cesium. To operate the source, the cesium in the reservoir is heated with an electron-impact heater, and the resulting vapor moves down short a tube and comes into contact with a tungsten plate. The tungsten plate is heated to ~1100 C by a second electron-impact heater. The hot plate ionizes the cesium and the resulting Cs+ ions are extracted from the source and accelerated into the primary column 10 KeV. The Primary Beam Mass Filter directs the beam down the primary column, where it is focused by the lenses. The minimum spot size for a cesium primary beam is about 0.5 microns. When using a Cs+ primary beam, charge compensation must be provided by the normal incidence electron gun.

The primary beam can either be focused by the lenses to a small spot (critical illumination) or shaped by an aperture in the primary column to make a larger, homogenous beam (aperture illumination). Critical illumination is used for most measurements and is required to measure the smallest target grains. Aperture illumination produces a flat-bottomed crater, which can facilitate obtaining the highest-precision data. In aperture illumination, spot size depends mainly on aperture diameter and is independent of beam current, whereas in critical illumination, spot size depends on beam current.

The beam current can be monitored by means of a primary beam Faraday cup.

Sample Chamber and Air Lock: The sample inlet system on our ims 1280 has two positions, permitting us to have a standard block and a sample (or two samples) in the vacuum at the same time. The sample chamber routinely operates with a vacuum of ~1 x 10-9 torr. If the project requires it, we can reduce the pressure to 3-5 x 10-10 torr. The main reason that we do not always have such a good vacuum in the sample chamber is the samples themselves. Thick sections and samples with too much epoxy, for example, degas into the sample chamber, raising the pressure (see sample preparation).

Secondary Ion Mass Spectrometer: The high-sensitivity mass spectrometer on the Cameca ims 1280 is designed to have essentially full transmission at a mass resolving power of 6000 (M/ΔM, 10% definition). Even at a mass resolving power of ~15,000, enough counts reach the detector for high-precision measurements. The mass spectrometer on the ims 1280 can be operated as an ion microprobe or as an ion microscope. In microprobe mode, the secondary ion beam arrives at the detector as a focused beam. In microscope mode, the ions arrive at the detector preserving the relative positions from which they were generated on the sample, producing an ion image.

Most isotopic and abundance measurements are made in microprobe mode. The primary ion beam is focused to a spot and the beam is directed onto the mineral grain or other target that the user wishes to measure. In general, this is how one would obtain the highest-precision data.

Ion imaging is useful to investigate the details of the distribution of isotopes and elements in a sample. It also provides a means of excluding signals from unwanted materials near the spots of interest. The ims 1280 can gather ion images in two ways. Scanning ion imaging rasters a focused primary ion beam over a region of the sample (e.g., a 50 x 50 micron square). The signal is measured in an electron multiplier and the image is reconstructed using information about where the primary beam was at any given time. This is the imaging mode for the NanoSIMS.

The ims 1280 can also use the microscope mode to produce a direct ion image. In early incarnations ims-series Cameca ion probes, the ion imaging mode was used primarily for tuning. The image was detected with a micro-channel plate and phosphorescent screen assembly. (The micro-channel plate coverts an ion into a number of electrons, the phosphorescent screen converts the electrons to light, and the light is imaged by a camera.) We still use this system for tuning. Attempts to make this system quantitative so that data could be gathered in imaging mode were only partially successful. Gain variations across the micro-channel plate make calibration very difficult. The phosphorescent screen changes detection efficiency as well. These problems led to several attempts to make an ion-imaging detector. The most successful such detector to date (SCAPS, see below) was developed in Professor Hisayoshi Yurimoto's laboratory in Japan. Our laboratory manager, Kazuhide Nagashima, helped to develop the SCAPS detector as part of his Ph. D. thesis. He has installed this detector on our Cameca ims 1280.

Detectors: The ims 1280 has a wide variety of ion detectors. The original ion detectors on Cameca SIMS instruments were a single Faraday Cup for measuring large signals, and an electron multiplier for measuring small signals. Our ims 1280 still has these detectors and they are used for a variety of application. Because these detectors use the same ion path, only one can be used at a time; we call this monocollection mode. In monocollection mode, the magnetic field in the mass spectrometer is switched to put the different masses of interest into the detector sequentially. Because the magnetic field "jumps" from one mass to the next, this is called "jump scanning." In the ims 1280, there are not two monocollection Faraday cups, a low-precision Faraday cup (FC1) that is used mostly for tuning, and a high-precision Faraday cup (FC2) that is used for precise isotope measurements. FC1 and FC2 are equipped with with a 1010- and 1011-ohm resistors, respectively. The monocollection electron multiplier is a relatively large ETP multiplier that can measure signals from ~0.1 cps to ~106 cps. The background count rate of ~0.008 cps is the limiting factor at the low end, and the high end is governed by the uncertainty of the deadtime at >106 cps and increasing rapid degradation in gain over time with higher count rates.

The ims 1280 is also equipped with a multicollector, which consists of 5 separate detectors that can be moved to any required positions along a track (see the schematic drawing above right). The maximum dispersion between masses that can be measured simultaneously is ~16% (7Li and 6Li can be detected simultaneously). Each of the five detectors can be either a Faraday cup on an electron multiplier. The Faraday cups are high-precision cups with the same performance characteristics as the monocollector Faraday cup, FC2. They can be operated with either 1010- or 1011-ohm resistors, depending on the count rates to be measured. For high-precision oxygen isotopic analysis, we use a 1010-ohm resistor for the cup with the 16O signal. The multicollection electron multipliers are much smaller than the monocollection multiplier. The noise specification is essentially the same as for the monocollection multiplier. But the maximum count rate that can be reliably counted is quite a bit lower and the deadtime for the counting system used on these multipliers is significantly higher (~66 nanoseconds compared to ~30 nanoseconds for the monocollector EM). Signals above a few tens of thousands of counts per second can cause decreases in gains over a few days. Multicollection using electron multipliers is good for elements that are sample limited. Multicollection of elements that are not sample-limited (i.e., using Faraday cups) can eliminate the effects of time-dependent variations in ion signal(all isotopes experience the same time-dependent effects) and can give higher-precision data.

As mentioned above, there are two types of detectors for direct ion imaging. The micro-channel plate and phosphorescent screen provide a real-time image of the sample or of the beam at the entrance slit of the mass spectrometer. This system is used to tune the instrument. Those who have attempted to tune a mass spectrometer without access to an image of the entrance slit are often astonished at how much easier it is to tune when you have an image. This is a requirement for ion probe work because each project typically required a different setup and a different tuning. You cannot just tune the mass spectrometer once and forget about it. Direct ion imaging with the channel plate assembly is also useful for navigation on a sample (e.g., a grain mount). The other imaging detector is the SCAPS detector, which permits quantitative ion imaging and production of isotope ratio maps (see below).



The SCAPS Detector

The SCAPS vacuum housing equipped to the UH ims-1280 ion microprobe is shown in the left image. The top-right image shows inside of the SCAPS vacuum housing. The SCAPS device (bottom-right image) is attached to a cold finger and liquid nitrogen dewar, in order to cool the SCAPS down to ~80K. The central gold part of the SCAPS is imaging area composed of 608×576 pixels. [Click for enlargement.]

The University of Hawai'i Cameca ims 1280 has a new solid state imaging detector called SCAPS (Nagashima et al., 2009). A stacked CMOS-type active pixel sensor for charged particles, SCAPS was developed at the Tokyo Institute of Technology (Nagashima et al., 2001, Yurimoto et al., 2003). The SCAPS detector will permit direct ion imaging of fine-grained samples and will permit identification of isotopically or chemically anomalous grains at a spatial resolution of a few tenths of a micron (Nagashima et al., 2004; Kunihiro et al., 2005).

SCAPS is composed of a rectangular array of 608x576 independent micro-detectors or "pixels." When placed in the position where the channel plate normally sits, SCAPS can collect two-dimensional ion images. The SCAPS detector has sufficiently well understood and reproducible characteristics to permit quantitative isotope analysis in two dimensions using a stigmatic SIMS such as the Cameca ims 1280. SCAPS has several advantages over conventional systems, including two-dimensional detection, wide dynamic range, no insensitive period, direct detection of charged particles, constant ion sensitivities among nuclides, and a high degree of robustness. The SCAPS can measure high ion flux with an accuracy of within twice the statistical error and with a detection limit corresponding to 3 ions.





Above left: 27Al+ ion image obtained with SCAPS. Sample was a Cu-grid with 25 μm pitch on an Al-substrate. SCAPS can detect ions directly and avoid degrading spatial resolution associated with signal conversions in a conventional imaging system composed of micro-channel plate, fluorescent screen, and CCD camera. (right): An intensity profile across a boundary between Al-substrate and Cu-grid shows lateral resolution of SCAPS+ims-1280 is high as ~0.5 μm. [Click for enlargement.]
Above right: Comparison of scanning electron image (BSE) and SCAPS isotope images (Si-/O- and δ18O) of an area within a Ca-Al rich inclusion in Efremovka meteorite. The BSE and SCAPS Si-/O- images show that the area is composed of anorthite, melilite, fassaite, and spinel. Variation in oxygen isotopic compositions is clearly visible. δ18O gives the shift in isotope ratio in parts per thousand relative to standard mean ocean water (SMOW), i.e., bright gray regions have oxygen-isotope composition similar to SMOW, while dark gray regions have anomalous oxygen-isotope composition. The image demonstrates that our imaging technique has a capability of quantitative O-isotope mapping with permil-level precision and ~1 μm spatial resolution. [Click for enlargement.]


For further information :

  • Conty C. (1990) The IMS1270—A new high transmission SIMS. SIMS VII, 831-834.

  • Kunihiro T., Nagashima K., and Yurimoto H. (2005) Microscopic oxygen isotopic homogeneity/heterogeneity in the matrix of the Vigarano CV3 chondrite. Geochim. Cosmochim. Acta, 69, 763-773. [ abstract ]

  • Nagashima K., Huss G. R., Kosaka K., Kunihiro T., Keil K., Krot A. N., Taylor G. J., and Yurimoto H. (2009) Development of isotope imaging system with two-dimensional ion detector SCAPS for ims-1280 secondary ion mass spectrometer. Lunar Planet. Sci. Conf. 40, [ abstract 2066 ].

  • Nagashima K., Krot A. N., Huss G. R., and Yurimoto H. (2010) Micron scale oxygen isotope heterogeneity in anorthite of a forsterite-bearing type B CAI E60 from Efremovka. Lunar Planet. Sci. Conf. 41, [ abstract 2255 ].

  • Nagashima K., Krot A. N., and Yurimoto H. (2004) Stardust silicates from primitive meteorites. Nature, 428, 921-924. [ abstract ]

  • Yurimoto L., Nagashima K., and Kunihiro T. (2003) High precision isotope micro-imaging of materials. Applied Surface Science, 203, 793-797. [ abstract ]



21 April 2017