Elsevier

Chemical Geology

Volume 390, 18 December 2014, Pages 61-73
Chemical Geology

Accurate analysis of shallowly implanted solar wind ions by SIMS backside depth profiling

https://doi.org/10.1016/j.chemgeo.2014.10.003Get rights and content

Highlights

  • We report a method to analyze solar wind (SW)captured during the GENESIS mission

  • Backside depth-profiling by secondary ion mass spectrometry (SIMS) was used.

  • Nearly complete depth profiles of implanted SW were obtained.

  • We present quantitative analyses of SW fluences for several elements.

Abstract

A method to quantitatively determine the fluences of shallowly-implanted solar wind ions returned to Earth by the Genesis Discovery mission is described. Through backside depth-profiling, we recover nearly complete depth profiles of implanted solar wind for several nonvolatile elements, including Mg, Al, Ca, Cr, and, to a lesser extent, Na, in silicon targets that collected bulk solar wind and solar wind from specific velocity regimes. We also determine the fluences of the volatile elements C, N, and O in silicon targets that collected bulk solar wind. By the use of appropriately calibrated ion implanted standards, fluences as low as 2 × 1010 atoms cm 2 can be determined with precision and accuracy typically in the few percent range. Specific approaches to sample preparation, sputtering artifacts during depth profiling by secondary ion mass spectrometry, and quantification including the production of ion implant standards are discussed.

Introduction

NASA Discovery Mission V, Genesis, collected solar wind in space, 1.5 million km sunward from Earth, for a period of 2.3 years and returned this sample for laboratory analysis (Burnett et al., 2003, Burnett and Team, 2011, Reisenfeld et al., 2013). The solar wind can serve as a proxy for the composition of the Sun and, thus, average of composition of the solar nebula (Marty et al., 2011, McKeegan et al., 2011), provided that fractionation processes occurring during the formation of the solar wind (e.g., Heber et al., 2012a) can be understood and quantitatively modeled. The objective of the Genesis mission is to obtain elemental and isotopic abundances of elements heavier than helium in the solar wind with precisions and accuracies better than those available from spacecraft data, which have uncertainties on the order of up to several tens of percent (e.g., von Steiger et al., 2000, Giammanco et al., 2008).

The solar wind at 1 AU (astronomical unit) is comprised of plasma with ion speeds of about 400 km s 1, equivalent to a mean particle energy of ~ 1 keV per amu (atomic mass unit). This enables solar wind ions to penetrate, i.e. implant, into exposed collectors that are high purity semiconductor materials (e.g., silicon wafer, diamond-like carbon, sapphire; (Jurewicz et al., 2003)). The implantation results in a depth distribution of the solar wind ions in the collectors that peaks at ~ 40 nm and a tail up to a few hundreds of nanometers. The solar wind comprised ~ 96 at.% hydrogen, ~ 4 at.% helium, with all heavier elements occurring as minor or trace species in roughly solar abundances, e.g., 0.05 at.% oxygen and 3 ppma (parts per million atomic) aluminum (Asplund et al., 2009). Thus, the analytical challenge for Genesis is to accurately measure these low-abundance species with precision useful for planetary science (Burnett and Team, 2011), a challenge made more difficult by the shallow depth of the implanted solar wind. Despite the excellent purity of the collector materials themselves, the terrestrial contamination on their surfaces is significant. Terrestrial contamination varies among the collectors sampled and can include: a natural oxide layer (complicating oxygen analyses), a molecular film deposited in space by outgassing of organic compounds and subsequent UV polymerization (Allton et al., 2006), and particulates (e.g., soil, salt, pulverized collectors) deposited during the hard landing of the science canister in the Utah desert in September 2004.

For elements other than noble gases, the most widely used method to extract and analyze the implanted solar wind is depth profiling by secondary ion mass spectrometry (SIMS) (e.g., Marty et al., 2011, McKeegan et al., 2011, Huss et al., 2012, Heber et al., 2013, Heber et al., 2014, Rieck et al., 2014). The SIMS method is extensively employed in the semiconductor industry to analyze in-depth distributions of dopant elements (e.g., Zinner, 1983, Wilson et al., 1989). The high depth resolution achieved by SIMS, in principle, enables the separation of surface contamination from the ion implant. The ion microscope capability of the Cameca IMS 1270 (UCLA) and 7f-Geo (CalTech) SIMS instruments employed in this work allows the insertion of a field aperture in an image plane thus restricting the collection of secondary ions to only the center of a sputtered pit and minimizing contributions from the crater walls. In addition, the ion imaging capability allows the detection of micron-sized contaminant particles on the sample surface so that they can then be avoided.

Despite the excellent depth resolution inherent in the secondary ion formation by sputtering, and even considering the restriction of accepted ions from the instantaneous crater bottom enabled by the ion microscope, the extremely low solar wind fluences of elements implanted in the Genesis targets (e.g., ~ 9.7 × 1010 atoms cm 2 for sodium in the bulk solar wind collector, see Fig. 10) coupled with the shallow penetration depths and ubiquitous surface contamination make quantitative analysis of the solar wind profiles highly problematic by conventional SIMS depth-profiling. For example, atomic mixing and knock-on of ions from a high-abundance surface contaminant to the low concentration region of solar wind can partially or entirely mask the signal of interest; this problem is especially acute for the volatile elements carbon, nitrogen, and oxygen that comprise residual vacuum species and which are often the components of surface films on Genesis collectors (see for nitrogen Huss et al. (2012)). The degradation of depth resolution by the addition of these knocked-on ions may result in erroneously high solar wind fluences or incorrect isotopic compositions. A further problem for quantitation of ion signals in the very near surface layers results from preferential sputtering effects, a problem long recognized in the semiconductor industry (e.g., see review by Zinner (1983)). At the beginning of the sputter process (or at any sharp interface of changing target composition), secondary ion yields vary markedly. The effect is particularly strong when sputtering with reactive ion beams (e.g., cesium or oxygen) that are used to enhance secondary ion formation (and, thus, sensitivity). The nonequilibrium sputtering effects scale with implantation depth of the primary ion beam (Wilson et al., 1989), and even under favorable conditions (e.g., low impact energy) correspond to depths of typically up to ~ 10 nm in the ion microscope instruments considered here. For some elements, certain “tricks”, such as low impact energy primary ion beams and/or gaseous oxygen flooding of the sample analysis area, can be used to minimize ion beam mixing of surface contamination and transient effects on SIMS ion yields due to preferential sputtering. However, as the peak intensity of the solar wind implant is at a depth of only ~ 40 nm in the Genesis collectors, even with exceptional measures, the intrinsic limitations of ion beam mixing result in a sputtering region that changes with depth until steady-state is achieved; this transient region at least partially affects the solar wind depth profile when sputtering is initiated from the front-side of a collector surface.

Given these constraints, to obtain accurate depth profiles we adopt the method of sputtering from the backside of a Si-collector target (backside depth profiling) with a low impact energy primary ion beam. This technique has been previously applied in the semiconductor industry for analysis of ultrashallow ion implants or impurities (e.g., Jackman et al., 1990, Yeo et al., 2002, Yeo et al., 2003, Fujiyama et al., 2011). We developed this technique further in our laboratory for the application to solar wind analysis. In this contribution, we show that our backside depth profiling technique is capable of analyzing nearly complete depth distributions of many elements in the solar wind, even in the presence of high levels of surface contamination. We have successfully applied the method to the analysis of bulk solar wind with fluence as low as 2 × 1010 atoms cm 2 and, for some more abundant elements, to the analysis of all three solar wind velocity regimes collected by Genesis (Reisenfeld et al., 2013). The regime collectors were only exposed to solar wind during specific conditions: slow (fluence-weighted mean proton speed: 384 km s 1), fast (578 km s 1) and coronal mass ejection events (CME) (452 km s 1) that include higher energetic ions (up to 1600 km s 1). These collections have only about one third each of the fluence of the bulk solar wind sample (Reisenfeld et al., 2013) and are characterized by different depth distributions of the implanted solar wind.

In the following, we describe sample preparation, point out potential pitfalls, as well as generally explain the SIMS analytical conditions, standardization of measured signal, and data reduction procedures. Possible systematic errors are considered and evaluated. We show that these procedures allow us to obtain precise and accurate absolute fluences of the implanted solar wind in both bulk and regime-specific targets. The solar wind fluences of the analyzed elements as well as their implications on solar wind formation processes will be published elsewhere. Preliminary data are published in several abstracts (e.g. Heber et al., 2013, Heber et al., 2014).

Section snippets

Sample preparation

Fragments of Float-Zone (FZ) silicon wafers with implanted solar wind, on the order of 8 × 8 mm2, were cleaned at the NASA Johnson Space Center curatorial laboratory by megasonically activated ultrapure water (5 min at 40 °C) that efficiently removed particles ≥ 5 μm (Allton et al., 2007) followed by an exposure to UV radiation in an ozone atmosphere (30 min) that removed the molecular film contamination deposited in space (Calaway et al., 2007). The final steps involved cleaning in a sequence of

SIMS analytical conditions for backside depth profiling

Depth profiling analyses for C, N, and O were performed with the Cameca 7f-Geo at the California Institute of Technology and for Na, Mg, Al, Ca, and Cr mainly with the Cameca IMS-1270 at the University of California, Los Angeles. The IMS-1270 was chosen for analysis of the latter elements because high mass resolving power (m/Δm ~ 3000) is needed to separate the desired atomic ions from molecular ion interferences, and this can be achieved in that large-format instrument without sacrificing

General analytical conditions

Backside depth profiles of implanted solar wind were generally analyzed in three to four steps (Fig. 5). At first, a (125 × 125) μm2 raster (step 1) removed the surface contamination of the backside sample until the measured intensity of the isotope of interest reached background level. This was followed by a (100 × 100) μm2 raster for analysis (steps 2 to 4). During steps 1 and 2, both the isotope of interest and the matrix element were measured. During step 3 only the isotope of interest was

Data reduction in backside depth profiles

The Si-normalized relative sensitivity factor, RSF, of each element analyzed in this work was determined from respective reference implants measured prior to and after the sample profile. In SIMS depth profiling, by convention, the RSF is a conversion factor from secondary ion intensity to atom density (see Section 3.1 in Wilson et al., 1989):di=ni/nSi×RSF,where di represents the atom density and ni the ion intensity in counts per second of the element of interest, and nSi the ion intensity in

Standardization and absolute calibration of the fluences of the reference implants

Sample fluences were calibrated against reference implants via the relative sensitivity factor (RSF), see Section 5. The reference implants were produced based on the following criteria: the isotope fluence was chosen to be high enough to result in a secondary ion signal well above background but low enough to be measured on the electron multiplier with, if possible, only insignificant or low deadtime corrections. The use of a minor isotope of the element of interest increased the range of

Summary

We have documented in detail our approach of backside depth-profiling for quantitatively analyzing the fluences of shallowly implanted solar wind ions returned to Earth by the Genesis Discovery mission. The main analytical challenges overcome by this approach are: (1) unacceptable levels of mixing of residual surface contaminants into the solar wind signal and (2) the inability to quantify relative SIMS ion yields in the nonequilibrium sputtering regime which overlaps the implant profiles in

Acknowledgments

We appreciate the expertise and cooperation of the curatorial team at NASA Johnson Space Center in cleaning Genesis collector fragments prior to preparation for backside profiling. Detailed reviews by Andrew Davis and Ulrich Ott are gratefully acknowledged. This work was supported by grants from the NASA Laboratory Analysis of Returned Samples (LARS) program. V. Heber thanks the NASA for financial support and Rainer Wieler (ETH Zurich, Switzerland) for the discussion and correction of the paper

References (36)

  • D.S. Burnett

    The Genesis discovery mission: return of solar matter to earth

    Space Sci. Rev.

    (2003)
  • D.S. Burnett

    Ion implants as matrix-appropriate calibrators for geochemical ion probe analyses

    Geostand. Geoanal. Res.

    (2014)
  • M.J. Calaway

    Decontamination of Genesis array materials by UV ozone cleaning

    (2007)
  • Cameca

    Cameca IMS 1270/1280 User's Guide

    (2004)
  • S.M. Eggins et al.

    Compositional heterogeneity in NIST SRM 610-617 glasses

    Geostand. Geoanal. Res.

    (2002)
  • N. Fujiyama

    A beneficial application of backside SIMS for the depth profiling characterization of implanted silicon

    Surf. Interface Anal.

    (2011)
  • C. Giammanco et al.

    Minor ion abundances in the slow solar wind

    ApJ

    (2008)
  • V.S. Heber

    Isotopic mass fractionation of solar wind: evidence from fast and slow solar wind collected by the Genesis mission

    ApJ

    (2012)
  • Cited by (13)

    • Matrix effects on the relative sensitivity factors for manganese and chromium during ion microprobe analysis of carbonate: Implications for early Solar System chronology

      2017, Geochimica et Cosmochimica Acta
      Citation Excerpt :

      This method will achieve the best results when applied to materials with very low abundances of the element of interest. Secondly, it is also important to independently calibrate the implant fluence because the ion implanting instruments have low mass resolution so minor interferences can affect the dose measured during the implant such that the nominal fluence can only be considered 10–20% accurate (Heber et al., 2014; Burnett et al., 2015). The fluence of the implant can be calibrated by comparison with a standard of known concentration and ideally ought to exceed this concentration by a factor of 2 to 10 to enable good precision during SIMS analysis.

    • Discrimination and quantification of Fe and Ni abundances in Genesis solar wind implanted collectors using X-ray standing wave fluorescence yield depth profiling with internal referencing

      2016, Chemical Geology
      Citation Excerpt :

      A formidable challenge for the surface cleaning applications is to remove the contamination from the surfaces without affecting the implanted solar wind ions residing at depths on the order of 100 nm. For quantification of implanted solar wind, SIMS and RIMS are the two surface sensitive techniques that have been applied successfully (e.g., Burnett, 2011, 2013; Heber et al., 2014; Veryovkin et al., 2011). Although these techniques are destructive in that the implanted species plus the collector substrate are sputtered away with the energetic primary ions, the analysis footprint (sputtered crater) is relatively small (~ hundreds of microns) compared to the available size of the collector fragments (~ a few mm) thus permitting many independent analyses on the same collectors at different locations.

    View all citing articles on Scopus
    1

    Present address: Paul Scherrer Institute, Division for Radiation Safety and Security, 5232 Villigen PSI, Switzerland.

    View full text