tof-simsincosmochemistrygeosci.uchicago.edu/~tstephan/publications/tof... ·...

Planetary and Space Science 49 (2001) 859–906www.elsevier.com/locate/planspasci

TOF-SIMS in cosmochemistryThomas Stephan ∗

Institut f�ur Planetologie=Interdisciplinary Center for Electron Microscopy and Microanalysis (ICEM), Westf�alische Wilhelms-Universit�at,Wilhelm-Klemm-Str. 10, 48149 M�unster, Germany

Received 21 June 2000; received in revised form 27 February 2001; accepted 27 February 2001

Abstract

Time-of-.ight secondary ion mass spectrometry (TOF-SIMS) was introduced into cosmochemistry about a decade ago. Major advantagesof TOF-SIMS compared to other ion microprobe techniques are (a) parallel detection of all secondary ions with one polarity in a singlemeasurement—both polarities in subsequent analyses, (b) high lateral resolution, (c) su4cient mass resolution for separation of majormass interferences, and (d) little sample destruction. This combination makes TOF-SIMS highly suitable for the analysis especially ofsmall samples, like interplanetary and presolar dust grains, as well as tiny inclusions within meteorites. Limitations of this technique aremainly referring to isotopic measurements and quanti7cation. The possibility to measure molecular and atomic ion species simultaneouslyextends the applications of TOF-SIMS to the investigation of indigenous hydrocarbons in extraterrestrial material, which might have beenessential for the formation of life. The present work gives an overview of TOF-SIMS in cosmochemistry, technical aspects as well asapplications, principles of data evaluation and various results. c© 2001 Elsevier Science Ltd. All rights reserved.

Table of contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 860

2. Principles of secondary ion mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8612.1. Generation of secondary ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8612.2. Separation of secondary ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8622.3. Quanti7cation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8622.4. Static and dynamic SIMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8632.5. Charge compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8632.6. Mass interferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8632.7. Mass resolution in time-of-.ight mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . 865

3. Time-of-�ight secondary ion mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8653.1. Secondary ion mass spectrometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8653.2. Description of the TOF-SIMS instrument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8663.3. TOF-SIMS mass spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8693.4. Quanti7cation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8703.5. Element analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8723.6. SIMS sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8763.7. Isotope analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8773.8. Organic molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8783.9. Data processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 878

∗ Tel.: +49-251-8339050; fax: +49-251-8336301.E-mail address: [email protected] (T. Stephan).

0032-0633/01/$ - see front matter c© 2001 Elsevier Science Ltd. All rights reserved.PII: S 0032 -0633(01)00037 -X

860 T. Stephan / Planetary and Space Science 49 (2001) 859–906

4. Meteorites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8794.1. Ca,Al-rich inclusions in carbonaceous chondrites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8794.2. Organic molecules in Allan Hills 84001, Murchison, and Orgueil . . . . . . . . . . . . . . 8804.3. Further meteorite studies with TOF-SIMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 884

5. Interplanetary dust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8855.1. Volatile elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8875.2. Mineral identi7cation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8875.3. The collector project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8885.4. Particle impact residues from space probes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8905.5. Further TOF-SIMS studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 890

6. Presolar grains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8916.1. Isotope measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8916.2. Element mapping with TOF-SIMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 892

7. TOF-SIMS in comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 893

8. TOF-SIMS in space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 893

9. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 895

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 895

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 895

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 900

1. Introduction

The analysis of micrometer-sized samples like, e.g., in-terplanetary dust, presolar grains, and small inclusions inmeteorites has become more and more important in cosmo-chemistry. Recent technical developments lead towards theultimate analytical goal, the localization and identi7cation ofevery single atom and its chemical bonding state in a sample.The transmission electron microscope (TEM) with an ex-tremely high lateral resolution allows even seeing individualcrystallographic layers. Therefore, this technique has beenestablished widely into cosmochemistry (e.g., Tomeoka andBuseck, 1984, 1985, 1988; Bradley, 1994a). New develop-ments in mass spectrometry like resonant ionization massspectrometry (RIMS) enable to analyze the isotopic com-position of an element that is present with just a few atoms(e.g., Hurst et al., 1979; Pellin et al., 1987; Ma et al., 1995;Nicolussi et al., 1996, 1998a).Secondary ion mass spectrometry (SIMS) has also be-

come an increasingly important tool in geo- and cosmo-chemistry during the last two decades. After 7rst develop-ments in the 1960s, a generation of instruments with highmass resolutions like the Cameca IMS 3f and the SHRIMP(sensitive high mass resolution ion microprobe) allowed tointroduce SIMS into this scienti7c 7eld mainly during the1980s. Since then, this technique has been used for traceelement analysis and isotope measurements in terrestrial aswell as extraterrestrial geological samples. The applications

range from the analysis of rare earth elements in var-ious samples (e.g., Zinner and Crozaz, 1986; Bottazziet al., 1994; Floss and JolliK, 1998; Fahey, 1998) to iso-tope measurements (Zinner, 1989) in meteorites (e.g.,McKeegan et al., 1998; Leshin et al., 1998; Chaussidonet al., 1998; Eiler et al., 1998), interplanetary dust (e.g.,McKeegan et al., 1985; Zinner, 1991), presolar grains(e.g., Zinner et al., 1989; Zinner, 1991, 1998), and alsoterrestrial (e.g., Froude et al., 1983; Chaussidon and Li-bourel, 1993; Didenko and Koval, 1994; Schuhmacheret al., 1994) and lunar (e.g., Compston et al., 1984) minerals.Technical developments of conventional SIMS instrumentsduring the last years led to increased sensitivities at highmass resolution and high spatial resolution. This culminatedin the development of the Cameca IMS 1270, optimizedfor high mass resolution at high transmission (de Chambostet al., 1992, 1997), and the Cameca Nanosims 50 witha lateral resolution of 50 nm (Hillion et al., 1994, 1997;Stadermann et al., 1999a, b).Finally, since the early 1990s, also time-of-.ight (TOF-)

SIMS has been introduced to the analysis of extraterrestrialsamples (Stephan et al., 1991; Radicati di Brozolo et al.,1991).Common to all diKerent SIMS techniques is that sec-

ondary ions are released from a sample surface by sputteringwith a primary ion beam. In conventional ion microprobes,using mainly double-focusing magnetic mass spectro-meters (DF-SIMS), diKerent secondary ions are usually

T. Stephan / Planetary and Space Science 49 (2001) 859–906 861

measured sequentially during continuous sputtering. Sincethis process is destructive, the number of analyzed sec-ondary ion species is limited depending on the availableamount of sample material or, because the analyzed ma-terial may change during sputtering, on the heterogeneityof the sample. In TOF-SIMS, the secondary ions are usedmore e4ciently. All secondary ion species with one po-larity, either positive or negative, are here detected quasi-simultaneously. Destruction of the sample is minimized bypulsing of the primary ion beam. The high transmissionand in most cases su4cient mass resolution in combinationwith a high lateral resolution (∼0:2 �m) makes TOF-SIMShighly suitable for the analysis of small samples.The aim of the present work is to summarize 10 years

of analytical experience with TOF-SIMS in cosmochem-istry. After a more general introduction into the principlesof secondary ion mass spectrometry, TOF-SIMS will beexplained in detail with special emphasis on applicationsin cosmochemistry. The principles of TOF-SIMS data ac-quisition are illustrated as well as evaluation proceduresto achieve useful information on the elemental, isotopic,and molecular composition of the respective samples.The most intriguing results of TOF-SIMS studies in cos-mochemistry are reviewed in subsequent sections of themanuscript.

2. Principles of secondary ion mass spectrometry

Secondary ion mass spectrometry in general is a tech-nique used for surface analysis. Samples are bombardedwith primary ions at typical energies of 10–30 keV.Secondary particles are released in a so-called sputteringprocess from surface-near layers. Besides electrons, thesesecondary particles are atoms and molecules, which are inpart (approximately 1%) positively or negatively charged.The secondary atomic and molecular ions are extractedwith electric 7elds and then separated and detected in amass analyzer (Fig. 1). Depending on the type of mass ana-lyzer, one discerns between either conventional SIMS, withmagnetic or quadrupole mass spectrometers, or TOF-SIMS,with time-of-.ight mass spectrometers.

2.1. Generation of secondary ions

2.1.1. The sputtering processThe sputtering process can be described qualitatively, at

least for amorphous and polycrystalline samples, by cas-cades of atomic collisions (Sigmund, 1969). An impingingprimary ion experiences a series of collisions in the targetmaterial (Fig. 2). Recoiling atoms with su4cient energy gothrough secondary collisions and create further generationsof recoiling atoms. Both, primary ions and recoil atoms, havea chance to leave the target material as backscattered ionsor secondary atoms. The majority of sputtered particles re-sult from clouds of high-order recoil atoms. They have very

Fig. 1. The principle of secondary ion mass spectrometry: On the samplesurface, an energy-rich primary ion beam generates secondary ions, whichare separated and detected with a mass spectrometer.

Fig. 2. Primary ions transfer energy in a collision cascade to the targetatoms. Ion implantation (A) or backscattering (B) may occur. A third al-ternative, for thin samples only, is forward scattering (C). Atoms from thetarget material can leave the sample after several collisions as secondaryparticles (backward sputtering: a and b; transmission sputtering: c).

low energies (several electron volts) and originate from theuppermost atomic layers of the target.The collision cascade has a characteristic dimension of

10 nm in SIMS. Crystalline targets allow typically muchhigher penetration depths along open crystal directions in aso-called channeling process. The collision cascade model isnot applicable here, because binary collisions are relativelyrare. The charging state of the sputtered particles also cannotbe derived from this simpli7ed model, and multi-elementtargets are not considered.

2.1.2. Emission of secondary ionsThe ionization energy of an element is decisive for the

generation of positively charged secondary ions. Conse-quently, ions from alkali and alkaline–earth metals areformed most e4ciently during the sputtering process. Onthe other hand, for negative secondary ions, the electrona4nity plays the major role, and halogens have the highestion yields. Besides this, the ionization probability dependsstrongly on the charging state of the respective particlewithin the sample, i.e., its chemical environment. This char-acteristic, known as “matrix eKect”, complicates the properquanti7cation of SIMS results (see below).


2.1.3. Post-ionization of sputtered neutralsAbout 99% of all secondary particles escape the sample

electrically neutral and are not detectable in classical SIMS.However, by subsequent ionization with electron or laserbombardment (Oechsner, 1994; Wittmaack, 1994; KQotterand Benninghoven, 1997; Arlinghaus et al., 1997; Arling-haus and Guo, 1998), the sputtered neutrals can be measuredin a mass spectrometer. The mass spectrometry of secondaryneutrals (SNMS) has two major advantages: (1) The ion-ization e4ciency is increased compared to the SIMS pro-cess. (2) Matrix eKects are reduced because the secondaryparticles are ionized after they have left the solid body,and when the chemical environment has lost most of itsin.uence.In general, multiphoton processes are necessary to exceed

the 7rst ionization potential for atoms (4–17 eV for elementsaside from noble gases) and molecules. Individual photonenergies of most laser systems are typically too low to ionizethe atom or molecule directly. Two major techniques forpost-ionization with laser light are established, resonant andnon-resonant ionization.In the non-resonant multiphoton ionization (Kampwerth

et al., 1990; Terhorst et al., 1992, 1994; Niehuis, 1994;Schnieders et al., 1996, 1998; Schnieders, 1999) an atomor molecule has to transit through a series of intermediatestates to reach its ionization energy. These “virtual” statesare not quantum-mechanical excitation states and have there-fore very short lifetimes, of the order of 10−15 s. This pro-cess can be described as simultaneous absorption of severalphotons. Crucial for the ionization probability is the totalphoton .ux density. Practically all atomic and molecularspecies are aKected in this non-resonant laser ionization.The resonant case uses quantum-mechanical transition

states for ionization that are speci7c for an atom or amolecule (Hurst et al., 1979; Pellin et al., 1987; Ma et al.,1995). Therefore, the photon energies of one or a few lasershave to be tuned carefully to match exactly the requiredexcitation energies. The results are extremely high ioniza-tion yields for speci7c atoms or molecules. The ionizationof other neutral species at the same time is eKectively sup-pressed, if single photons cannot ionize these species andif the photon .ux density does not reach values necessaryfor non-resonant multiphoton ionization. Because of that,isobaric interferences at same nominal masses are avoided(e.g., interferences of 92Mo, 94Mo, and 96Mo with 92Zr,94Zr, and 96Zr) even with a relatively low mass resolutionof the spectrometer.In cosmochemistry, resonant laser post-ionization is up

to now mainly used together with laser ablation instead ofprimary ion bombardment. Applications have been the anal-ysis of titanium, zirconium, and molybdenum isotopic com-positions of individual presolar silicon carbide or graphitegrains (Ma et al., 1995; Nicolussi et al., 1997a, b, 1998a, b).Microprobe two-step laser desorption=laser ionization massspectrometry (�L2MS) has been used for the analysis of or-ganic molecules in various kinds of extraterrestrial material

(Kovalenko et al., 1991, 1992; Clemett et al., 1992, 1993,1996, 1998; Thomas et al., 1995).

2.2. Separation of secondary ions

2.2.1. Conventional SIMSIn conventional SIMS instruments, secondary ions are

generated continuously by use of a static primary ion beam,and the diKerent ion species are detected sequentially,e.g., in a double-focusing magnetic mass spectrometer ora quadrupole spectrometer. At a given instant, only sec-ondary ions of a previously selected mass-to-charge ratioreach the detector. Thus, most secondary ions are not mea-sured. Even in multi-detector instruments, only a few ionspecies with similar mass-to-charge ratios can be detectedsimultaneously. Most secondary ions are therefore lostduring the analysis.

2.2.2. TOF-SIMSContrary to this, time-of-.ight mass spectrometry al-

lows to measure all secondary ions with one polar-ity quasi-simultaneously. The sample is bombarded inTOF-SIMS with a short-time primary ion pulse with typicallengths between several hundred picoseconds and a fewnanoseconds. This is repeated every few hundred microsec-onds, resulting in a duty cycle of typically 10−5.All secondary ions with one polarity are accelerated in

an electric 7eld with potential U and mass separated in a7eld-free drift tube. Particles with diKerent mass-to-chargeratios m=q reach the detector after the time-of-.ight t:

m=q=2eUs2t2: (1)

Here, s is the eKective .ight path, corresponding to the lengthof the drift tube.

2.3. Quanti5cation

2.3.1. Element ratiosQuanti7cation, i.e., to determine useful element ratios, is

a problem in SIMS due to the above-described matrix eKect.Since the emission of secondary ions from multi-elementsamples is theoretically not well understood, only an em-piric approach is possible (Meyer, 1979). Standards witha chemical composition as similar as possible to the re-spective sample have to be used. These standards have tohave well-de7ned element abundances. Since usually onlya small volume of the standard is analyzed, this volumehas to be representative for the entire standard. This leadsto high demands on purity and homogeneity of the stan-dard. Amorphous or polycrystalline standards and samplesare desired, because crystalline targets lead to unwanted ef-fects like channeling. Only relative abundances can be deter-mined with SIMS, even if all these conditions are satis7ed.The secondary ion sensitivity for each element is calculatedfrom the standard relative to a reference element E0, often


silicon. Using these relative sensitivities, the respectiveelement ratios E=E0 in the analyzed sample can be calcu-lated from the determined secondary ion intensities SI :

(E=E0)Sample =[SI(E)=SI(E0)]Sample[SI(E)=SI(E0)]Standard

(E=E0)Standard : (2)

The relative sensitivity

RS(E; E0) =SI(E)=SI(E0)

E=E0(3)

depends on the respective ion species, the polarity, thesample material (matrix), as well as the analytical condi-tions, like primary ion type, energy, and angle of incidence,together with the detection e4ciency for the secondaryion species.

2.3.2. Isotopic ratiosThe matrix eKect plays a much less pronounced role for

isotopic ratios. Nevertheless, eKects of mass fractionationoccurring in the SIMS process or during subsequent massseparation and detection of the secondary ions have to betaken into account if high accuracy of isotope measurementsis required (Slodzian et al., 1980, 1998). Again, standardswith known isotopic ratios can be used to monitor theseeKects.

2.4. Static and dynamic SIMS

Secondary ion mass spectrometry in general is a destruc-tive technique. However, one often discerns between staticand dynamic SIMS due to the diKerence in sputtering in-tensity. Static SIMS describes the case where the samplesurface is not altered during the SIMS analysis. DynamicSIMS, on the other hand, is much more destructive butfacilitates quanti7cation. Interaction of the primary ionbeam—often O− for positive and Cs+ for negative secon-dary ions—with the sample generates more reproducibleconditions in dynamic SIMS. The high abundance of oxy-gen facilitates the formation of positive secondary ions andthe actual (pre-sputtering) oxidation state within the sam-ples loses its in.uence. Cesium, the element with the lowestionization potential, supports the generation of negativesecondary ions. With this strategy, dynamic SIMS becomesmuch less susceptible to matrix eKects than static SIMS.Static SIMS conditions cannot be achieved strictly speak-

ing, because primary ion bombardment even at lowest in-tensity alters the sample surface. However, TOF-SIMS is agood approach to static SIMS conditions especially with aless-focused primary ion beam. The probability for a pri-mary ion to hit an area on the sample already altered by sput-tering can be neglected, and sputter times of several hoursare necessary to consume one monolayer (Benninghoven,1973, 1975, 1985, 1994).

2.5. Charge compensation

Since most samples of geo- or cosmochemical interestare electrical insulators, sample charging during primary ionbombardment has to be considered (Werner and Morgan,1976; Werner and Warmoltz, 1984; Blaise, 1994). ChargingaKects the positioning of the primary ion beam, which isde.ected by the electric surface potential, and, on the otherhand, it aKects the energy of the secondary ions, becausethey are not extracted from a well-de7ned electric potential.For conventional SIMS analyses, samples are therefore

often coated with 20–100 nm of carbon or gold (Zinner,1989). The primary ion beam then sputters a small (up to20 �m) hole into this coating, and the underlying materialcan be analyzed. Small samples, single grains up to 30 �min size, can be pressed into gold foil (McKeegan et al.,1985; Zinner, 1989). Although, strictly speaking, the actu-ally investigated sample area is still insolating, charging isno longer critical in both cases. Electrons for charge com-pensation are picked up from the surrounding conductiveneighborhood, which also takes up surplus electrons whenthe sample is bombarded with negative primary ions.Surface coating has to be avoided in TOF-SIMS, because

only the uppermost atomic monolayers are analyzed. Even ifcoating of a single spot can be removed by sputtering, largerareas would not be accessible. However, a charge com-pensation system has been developed especially for pulsedprimary ion sources used in TOF-SIMS (HagenhoK et al.,1989). Electrons from a pulsed low-energy electron sourcereach the target between two primary ion shots. This processis self-regulative due to the low energy (10–20 eV) of theelectrons. The secondary ion extraction 7eld is switched oKduring the electron pulse. For the analysis of negative sec-ondary ions, a slight negative retarding 7eld at the extractorprevents the electrons from reaching the detector.

2.6. Mass interferences

A major problem in SIMS in general and TOF-SIMS inparticular is the separation of mass interferences. There-fore, this subject is discussed in detail here, especially forTOF-SIMS. The ability to separate two neighboring masspeaks in mass spectrometry is described as the mass resolu-tion or the mass resolving power (m=Vm). This mass reso-lution can be adjusted in DF-SIMS over a wide range at theexpense of the ion transmission of the instrument. The massresolution in TOF-SIMS mainly depends on the accuracy ofthe .ight time measurement as will be discussed later.The term “required mass resolution for separation of two

peaks” is used in the following for theoretical values result-ing solely from the mass diKerence of the respective peaks.It does not include considerations on peak shape, requiredaccuracy, or varying peak intensities, which often lead tomore strict limitations for a proper peak separation. Forthe experimentally determined mass resolutions of an actual


Fig. 3. Mass resolutions required for the separation of isobaric interfer-ences resulting from isotopes of other elements (resolving mA from mB).

mass spectrometer, the width (Vm) of a mass peak has tobe determined at a certain height of the peak. The full-widthhalf-maximum (FWHM) de7nition is used throughout thiswork if not speci7ed otherwise. The actual mass resolutionfor a su4cient separation of two peaks might be requiredat a lower peak height, depending on the relative intensi-ties of the respective ion signals and the required accuracy.This has especially to be considered in time-of-.ight massspectrometry, where peak integrals and not peak heights aredetermined.There are four major groups of mass interferences with

typical characteristic features: (1) Isobaric interferencesfrom nuclides of diKerent elements, (2) hydrides, (3) oxidesand hydroxides, (4) simple hydrocarbons. These groups arenow discussed in greater detail.

2.6.1. Isobaric interferences from other nuclidesIsotopes from diKerent elements at the same nominal mass

are one group of isobaric interferences. Two stable nuclidesexist for 52 isobars and even three naturally occurring nu-clides can be found for 7ve isobars. Mass resolutions be-tween m=Vm = 104 (48Ca and 48Ti) and 6 × 107 (187Reand 187Os) are required to separate their corresponding masspeaks from each other (Fig. 3). This is not achievable withthe available TOF-SIMS instruments, as will be shown later.However, for the analysis of element abundances, these in-terferences have to be considered but are not crucial. Theycan be corrected in most cases by using other isotopes fromthe respective element (see below).

2.6.2. HydridesDue to adsorbed hydrogen atoms or hydrocarbons, hy-

drides are important in surface analysis and are thereforeoften observed, especially in TOF-SIMS mass spectra.Sometimes they can reach the same intensity as the ele-ment peak itself. Hydrides from one isotope may interfere

Fig. 4. Mass resolution necessary for separating hydrides (m−1A1H) fromelement peaks (mB).

with an isotope of the same or a neighboring element. Therequired mass resolutions are typically lower here and canbe achieved in TOF-SIMS, at least for low masses, by usingvery short primary ion pulses. At masses above ∼90 amu(amu=atomic mass unit =1=12 of the mass of 12C) the re-quired mass resolution exceeds m=Vm=104 (Fig. 4), whichis approximately the maximum mass resolution reachablewith present TOF-SIMS instruments.

2.6.3. Oxides and hydroxidesOxides and hydroxides account for similar problems as

hydrides. Since 16O (15:995 amu) almost exactly lies on thenominal mass, any nuclide and its oxide have the same de-viation in mass from the nominal mass. This mass diKerencetypically increases for low-mass nuclides with increasingmass number. Between 88 and 140 amu it is practically con-stant, before it decreases again for higher masses. Therefore,in the middle mass range, around 110 amu for oxides and130 amu for hydroxides, their separation from atomic ionsrequires non-achievable high mass resolutions, well abovem=Vm= 104 (Fig. 5).

2.6.4. HydrocarbonsHydrocarbons occur almost throughout the entire mass

spectrum. Nevertheless, their interferences with atomic ionsare rather easy to separate, especially when several hydro-gen atoms are present. Each additional H-atom shifts thepeak 7:98× 10−3 amu to the right of the nominal mass (to-wards higher masses). Except for the light elements up tonitrogen and the very heavy elements thorium and uranium,elemental mass peaks lie left (towards lower masses) ofthe respective nominal mass. Upto 56Fe the mass-to-nucleonratio decreases (56Fe is the nuclide with the highest bind-ing energy). Consequently, the required mass resolution forseparation of hydrocarbons has a minimum at masses around56 amu (Fig. 6).


Fig. 5. Mass resolution needed for the separation of oxides (m−16A16O,solid symbols) and hydroxides (m−17B16O1H, open symbols) interferingwith nuclides at mass m.

Fig. 6. Mass resolution required for the separation of hydrocarbons12C1nHm−12n from element peaks at mass m.

2.7. Mass resolution in time-of-;ight mass spectrometry

In general, mass resolution in TOF mass spectrometrydepends on time resolution of the .ight time measurement.According to Eq. (1) follows

mVm

=t

2Vt: (4)

This resolution depends on the resolution of the mass ana-lyzer Ran (Mamyrin et al., 1973), the time resolution Vtregof detector and registration electronics, and the duration Vtpof the ionization event that in SIMS is the pulse width ofthe primary ion shot (JQurgens et al., 1992; JQurgens, 1993)

mVm

=

(1R2an

+ 2Vt2reg + Vt2p

t2

)−1=2

: (5)

From the linear correlation between t2 and m (Eq. (1))follows

mVm

=

(1R2an

+Vt2reg + Vt2p

c1m

)−1=2

; (6)

where the constant c depends on instrument parameters,length of the .ight path and extraction voltage.Actual values of mass resolutions achievable in

TOF-SIMS will be discussed later, when the TOF-SIMSinstrument used in this study is described.

3. Time-of-�ight secondary ion mass spectrometry

The following section will concentrate on practical as-pects of TOF-SIMS. First, for a better understanding, theprinciples of conventional DF-SIMS and TOF-SIMS instru-ments are compared.

3.1. Secondary ion mass spectrometers

3.1.1. DF-SIMSDouble-focusing magnetic mass spectrometers in SIMS

are separating secondary ions by electrostatic and magneticsector 7elds (Fig. 7). In the electrostatic 7eld, the secondaryions are dispersed according to their energy. Centrifugal andelectrostatic forces are in equilibrium here:

mv2

r= qE: (7)

The radius r depends in a given electrostatic 7eld E on thekinetic energy (mv2=2) per charge q. With the aid of theenergy slit (cf. Fig. 7), selection of secondary ion energiesis possible, and the system can be used as an energy 7lter.In the following magnetic 7eld B, the secondary ions are

dispersed according to their mass-to-charge ratio and theirenergy. It can be regarded as a 7lter for momentum (mv).Here, centrifugal and magnetic forces are in equilibrium

mv2

r= qvB: (8)

Fig. 7. Schematic view of a DF-SIMS instrument. Not shown areion-optical elements like de.ectors or electrostatic lenses.


The combination of electrostatic and magnetic 7elds al-lows compensation of energy dispersions from the electro-static and magnetic analyzers. The result is separation ofsecondary ions according to their mass-to-charge ratio re-gardless of their energy, the principle of double-focusingmass spectrometry.Nevertheless, the energy slit can be used for narrowing the

energy bandwidth of the secondary ions. The width of thesecondary ion energy distribution decreases with increasingcomplexity of the molecular ion and depends on the relativemass diKerence between the constituents of the molecule(Crozaz and Zinner, 1985). Consequently, energy 7lteringis not e4cient for the removal of hydrides and most monox-ides, but complex molecular mass interferences can be eKec-tively suppressed (Crozaz and Zinner, 1985). This is doneby setting the sample potential and the width of the energyslit to values that only secondary ions within a certain energywindow above a speci7c energy reach the detector. This en-ergy 7ltering technique has been successfully applied, e.g.,to the measurement of rare earth elements (Crozaz and Zin-ner, 1985; Zinner and Crozaz, 1986).A special feature of the experimental set-up shown in

Fig. 7 is that it can transport ion images, showing the lateraldistribution of the respective ion species. Secondary ionimages can be generated by using a defocused primary ionbeam. Direct imaging through ion optical projection is oftencalled ion microscope technique in contrast to SIMS instru-ments where secondary ion images are generated by scan-ning the primary ion beam over the sample (ion microprobetechnique).Energy 7ltering in DF-SIMS also reduces chromatic aber-

rations of secondary ion images. Analogous to light optics,the focal length of an ion-optical lens varies with the energyof the ions (equivalent to wavelength or color of light).

3.1.2. TOF-SIMSAll types of TOF-SIMS instruments are based on the same

principle: The velocity of an ion with a given kinetic en-ergy depends on its mass. One discerns between two majorgroups of TOF-SIMS instruments. Besides analyzers withthree hemispherical electrostatic analyzers (Schueler et al.,1990; Schueler, 1992), most TOF-SIMS instruments use amore simple design with an ion re.ector (Mamyrin et al.,1973; Niehuis et al., 1987; Niehuis, 1990). Only the lattertype is explained in detail here, because most analyses dis-cussed in this study were performed with this type of in-strument. However, the major diKerence between the twoset-ups is that the triple sector spectrometer allows directimaging through ion optical projection (ion microscope).Spatial information gets lost in the re.ector instrument andthe ion microprobe mode is used for imaging.A schematic view of a re.ector TOF-SIMS instrument

is given in Fig. 8. Secondary ions are extracted from thesample surface by acceleration in an electric 7eld. In thesubsequent drift tube, the ions are separated according to

Fig. 8. A schematic TOF-SIMS instrument.

their velocity before they reach the detector. Since slightlydiKerent starting energies result in slightly diKerent .ighttimes, even for ions of the same species, a re.ector is usedfor energy focusing (Mamyrin et al., 1973). Ions with higherenergies penetrate its electric 7eld deeper and are thereforedelayed compared to lower-energy ions. By adjusting theelectric 7elds carefully, all secondary ions of one speciescan reach the detector practically simultaneously.

3.2. Description of the TOF-SIMS instrument

The Cameca=ION-TOF TOF-SIMS IV instrument usedin this study is a re.ector-based instrument as describedin the previous section and shown in Fig. 8. The presentTOF-SIMS IV instrument at the Institut f�ur Planetologie inM�unster is equipped with two primary ion sources for anal-ysis, an electron impact ion gun mainly for argon primaryions (Niehuis et al., 1987) and a gallium liquid metal iongun (Ga-LMIS; Niehuis, 1997). In addition, the instrumenthas a sputter source, a combination of a cesium thermal ion-ization source and an electron impact source for argon oroxygen sputtering ions (Terhorst et al., 1997). It can be usedfor cleaning of the sample surface as well as for depth pro-7ling. All ion beams reach the sample with an angle of 45◦

and the secondary ions are extracted perpendicular to thesample surface with an extraction potential of 2 kV (Fig. 8).The sample is at ground potential.After the re.ection, the secondary ions are post-accelerated

to energies of 8–10 keV before they reach the detector. Thedetector consists of a microchannel plate for ion-to-electronconversion, a scintillator for electron-to-photon conversion,and a photomultipier, which is placed outside the vacuumchamber. All three steps lead to an ampli7cation of the sig-nal, but the major advantage of this set-up is the separa-tion of high voltage at the detector and low voltage of theelectronics. A constant fraction discriminator converts theincoming signal to a normalized signal, which is further


Fig. 9. The pulsing and bunching system of the electron impact primaryion gun. Focusing elements like lenses and de.ection units for beamalignments were omitted. Primary ions, mainly Ar+, are generated in aconventional electron impact ion source. A 90

◦-de.ection unit is used

to cut ion packages with pulse lengths of 10–15 ns out of a continuousion beam. These pulses are compressed to pulse lengths of ∼1:2 ns in abunching system.

processed in a time-to-digital converter (TDC). The timeresolution of the TDC is adjustable between 50 and 1600 ps,but it is mostly operated at 200 ps.A second detector system is used for detection of sec-

ondary electrons. With this, a secondary electron image sim-ilar to those generated in a scanning electron microscopecan be obtained by scanning the primary ion beam over thesample. A pulsed low-energy (∼18 eV) electron source isused for charge compensation.

3.2.1. The argon primary ion sourceA schematic view of the pulsing system of the electron

impact primary ion gun is given in Fig. 9. This primary ionsource is mainly used with 40Ar+ as primary ion species. Ionpulses with typically lengths of 10–15 ns are generated froma continuous ion beam with a 90◦ de.ection unit. Each ionpackage is compressed along its .ight direction to typically1:2 ns pulse length in an electrodynamic bunching system.Ions from the rear of the package are accelerated to catchup with those from the front when they reach the target.(Niehuis et al., 1987).The argon electron impact ion source has its main ad-

vantage in producing short primary ion pulses with rela-tively high intensities (cf. Table 1). The major drawbackof this type of ion source is the limited lateral resolution

Table 1Comparison of diKerent analytical modes for argon and gallium ion sourcesa

Primary Pulsing mode Pulse Spatial Primary Secondaryion species width resolution ions=cycle ions=cycle

40Ar+ Blanked+bunched 1:2 ns 10 �m 500 2.569Ga+ Blanked 5 ns 200 nm 50 0.2569Ga+ Blanked+chopped 1:5 ns 200 nm 10 0.0569Ga+ Blanked+bunched 600 ps 5 �m 50 0.2569Ga+ Burst mode n× 1:5 ns 200 nm n× 10 n× 0:05

aA value of one secondary ion per 200 primary ions is assumed as a rough estimate that is based on analyses of glass standards. For these conditions,one sputtered particle per primary ion can be assumed. The actual values strongly depend on sample material and surface conditions. The typical repetitionrates of 5–10 kHz result in several thousand secondary ion counts per second spread over the entire mass spectrum.

Fig. 10. The gallium liquid metal ion source. Focusing elements likelenses and de.ection units for beam alignments were omitted. Primaryion packages with pulse lengths of ∼5 ns are produced from a continuousbeam with a blanking system. These packages can be chopped a secondtime or compressed along their .ight direction in a bunching system tofurther reduce their pulse lengths.

(∼10 �m), due to the relatively large volume where ionformation takes place. Projection of this ionization volumeonto the sample surface leads to inevitable dispersion of theion beam. Acceleration of the ions to diKerent energies inthe bunching unit also reduces the lateral resolution. Theseenergy variations increase chromatic aberrations of the targetlens.

3.2.2. The gallium primary ion sourceThe gallium primary ion source is shown in Fig. 10. A

tungsten needle with a thin liquid gallium 7lm is exposedto a high electric 7eld. At typical extraction potentials of5–10 keV the liquid gallium forms a so-called Taylor-coneat the tungsten tip and Ga+ ions are emitted. These ions stemfrom a very small region with a typical size of 10 nm. Thissmall spot size is responsible for the achievable high lateralresolution of the Ga-LMIS (beam diameter ∼200 nm).The Ga-LMIS can be operated in diKerent modes, which

mainly diKer in length of the primary ion pulse, primaryion intensity, and achievable spatial resolution (cf. Table 1).Typical pulsing systems produce single ion packages with


Fig. 11. Timing schemes for blanker and chopper. The blanker is usedto select one or a series (burst mode, dotted line) of 1:5 ns ion pulsescontinuously produced by the high-frequency (40 MHz) chopper. Blankerand chopper are both pulsing units which are applied in diKerent ways.Sometimes, these names are used in the literature the other way round.

lengths of ∼5 ns by blanking a continuous ion beam. Thereare two possible techniques to further reduce the pulselength. Each ion package can be chopped a second time(Fig. 11) or compressed along its .ight direction in abunching system analogous to the argon source (Niehuis etal., 1987; Schwieters et al., 1991, 1992).In the 7rst case, the primary and therefore the secondary

ion intensity decrease with the pulse length. Extremely shortpulses result in extremely small signals, often not su4cientfor trace element studies. On the other hand, the spatial res-olution (beam diameter) remains unchanged in this case.Typical pulse lengths achieved with high-frequency chop-pers are 1:5 ns.The electrodynamic bunching system allows an axial

compression of the ∼5 ns ion package to ∼600 ps pulselength. Since no primary ion gets lost, secondary ion intensi-ties are usually by a factor of 7ve higher than in the choppedmode. This is mostly su4cient for trace element studies.On the other hand, chromatic aberrations of the target lenshere limit the achievable spatial resolution to ∼5 �m.

3.2.3. Mass resolution — spatial resolution — signalintensityThe three parameters mass resolution, spatial resolution,

and intensity of the secondary ion signals are strongly re-lated to each other in TOF-SIMS. The mass resolution canbe adjusted in DF-SIMS over a wide range by changing thewidths of the diKerent slits (Fig. 7) at the expense of theion transmission of the instrument. This is not the case forTOF-SIMS. Here, the mass resolution depends on the dura-tion of the primary ion pulse, the length of the .ight path,the energy focusing in the re.ector, the time resolution ofthe detection system, and, 7nally, on the mass itself. The de-pendence on time resolution of the .ight time measurementhas been discussed above.The mass dependence of the mass resolution at full-width

half-maximum (FWHM) is demonstrated in Fig. 12 for the

Fig. 12. Mass resolution (FWHM) for typical measurements with galliumand argon primary ion sources. Data points show selected peaks with noapparent mass interference. The gallium ion source was operated in threediKerent modes. The mass resolution of the analyzer is m=Vm = 10500in all cases.

argon primary ion source as well as for diKerent analyti-cal modes of the gallium source. From best 7ts accordingto Eq. (6) for the data points of undisturbed mass peaks,a resolution Ran for the analyzer of about 10,500 has beeninferred. This resolution describes the asymptotic limit athigh masses, whereas the accuracy of the time measurement,mainly from Vtp, shapes the 7t curve at low masses. Thetime resolution of the registration system Vtreg was 200 ps.Primary pulse lengths Vtp of 1:3 ns for the argon sourceand 5:2; 1:6, and 0:6 ns for blanked, blanked+chopped andblanked + bunched modes of the gallium source, respec-tively, have been determined.A simultaneous increase of all three parameters, mass res-

olution, spatial resolution, and intensity of the secondary ionsignals, is restricted, because the number of charged particleswithin a small volume is also limited due to space-chargeeKects. Extension of the ion package in .ight direction re-sults in low mass resolution. Extension perpendicular to the.ight direction results in low spatial resolution. The diKer-ent pulsing modes of the gallium source in Table 1 showthe mutual in.uence of these parameters.Nevertheless, if high lateral resolution and high mass res-

olution are required, the secondary ion signal can be in-creased by applying a series of 1:5 ns primary ion pulses,spaced 25 ns apart from each other, in every shot cycle.This so-called “burst” mode (Fig. 11) results in a series ofoverlapping mass spectra, which can be summed up by ap-propriate computer programs. For primary ion bursts of 7vesingle pulses, approximately the same intensity is reachedas in the “blanked” mode (cf. Table 1). This leads to use-ful results for masses up to ∼200 amu, where the last peakfrom one nominal mass begins to overlap with the 7rst peakof the next nominal mass. By varying the number of pri-mary ion shots within one burst, the usable mass range canbe adjusted. An increasing number of ion shots within one


Fig. 13. The dual beam sputter source. The special ion source combines acesium thermal ionization source with an electron impact source for argonsputtering ions. As in previous 7gures, focusing elements like lenses andde.ection units for beam alignments are not shown.

burst leads to a decreasing upper limit for the usable massrange (e.g., ∼50 amu for bursts of 10 shots). Therefore, theburst mode is not applicable to the investigation of largemolecules.

3.2.4. SputteringSince TOF-SIMS analyzes the uppermost surface of a

sample, contamination can be a severe problem. Therefore,the surface often has to be cleaned. To avoid adsorptionof molecules present in the ambient air after the cleaningprocedures, this is preferentially performed directly underultra-high vacuum conditions in the TOF-SIMS instrumentprior to the analysis. It can be done by sputtering with theprimary ion gun. The TOF-SIMS IV instrument is equippedwith an additional ion source exclusively for sputtering. Be-sides for surface cleaning, the sputter gun can also be usedfor depth pro7ling (Williams, 1998). The TOF-SIMS IVsputter source (Fig. 13) is a combination of a cesium ther-mal ionization source and an electron impact source for ar-gon or oxygen sputtering ions (Terhorst et al., 1997). Thebeam diameter is ∼50 �m.Secondary ion yields can be in.uenced and drastically

increased by selecting speci7c ion species for sputtering.Bombardment with oxygen ions, e.g., enhances the positivesecondary ion signal (Wittmaack, 1998); cesium ions in-crease negative ion abundances. Although such an increaseis desirable, the occurrence of molecular ions like oxides,stimulated by oxygen sputter ions, may complicate the inter-pretation of the mass spectrum. Therefore, sputtering withinert ion species like argon is more advantageous and hasbeen performed in this study.After removing adsorbed hydrocarbons through sput-

tering, even under ultra-high vacuum conditions (∼2 ×10−10 mbar), the sample surface can regain atoms andmolecules like hydrogen or hydrocarbons present in

Fig. 14. Timing scheme for simultaneous use of two ion sources forinsulating samples. A gallium primary ion source pulsed with 10 kHzperforms the actual TOF-SIMS analysis. Sputtering with a pulsed argonsource occurs between two gallium shots. Secondary ions are only ex-tracted directly after the gallium pulse. During the sputter pulse, a slightretarding 7eld at the extractor prevents secondary ions from reaching thedetector.

the residual gas. The primary ion source is not able toprevent this adsorption su4ciently with its low duty cy-cle, especially when a large area is scanned with the ionbeam. Therefore, an analytical mode has been developed,where the sample is analyzed with a gallium primary ionsource and sputtered simultaneously with an argon sputter-ing source (Fig. 14). The sputter pulse is applied betweentwo gallium pulses during charge compensation. The inten-sity of the sputter pulse is limited to ∼100 pA at 10 kHzrepetition rate (∼6 × 104 ions=shot) due to charging ofthe sample. Although the electron current for charge com-pensation easily exceeds 1 �A, the e4ciency of chargecompensation mainly depends on the number of electronsreaching the area of actual ion bombardment, ∼50 �m indiameter for the sputter source. However, the sputter currentof 100 pA is su4cient to suppress adsorption eKectivelyduring the analysis. The eKects of sputtering on various ionspecies are demonstrated in Fig. 15. Since the abundance ofhydride ions directly depends on the presence of hydrogenon the surface, hydride secondary ion intensities behaveanalogously to the hydrogen signal.

3.3. TOF-SIMS mass spectra

TOF-SIMS mass spectra contain a wealth of informationdue to parallel detection of an in principle unlimited massrange. Before the rules of data evaluation are explained, ex-amples for positive and negative secondary ion mass spectra(mass range 1–100 amu) obtained from a glass standard af-ter extensive argon sputtering are given in Figs. 16 and 17.Negative ion spectra are disturbed by electrons, which

partially result from charge compensation. They are alsogenerated as secondary or tertiary electrons from interactionof primary as well as secondary ions with various surfaceswithin the instrument. Some of these electrons appear as


Fig. 15. EKects of argon sputtering on the secondary ion intensities. A50 × 50 �m2 area of glass StHs6=80-G (cf. appendix) was bombardedwith 5 ns gallium shots. After 580 scans (256× 256 pixels, one shot perpixel) the sample was simultaneously sputtered with an argon beam. Thisleads to an increase of most secondary ion signals by factors of 50–200.Surface adsorbed hydrocarbons are e4ciently suppressed: The H+ andC+ signals decrease by factors of 2 and 10, respectively. Stable signalsare reached after 700–800 scans (∼80 min) with simultaneous sputtering.After switching oK the sputter source (after scan 2210) hydrocarbonsignals increase again to values similar to those before sputtering. Signalsfor other elements decrease slightly, but still remain clearly above thevalues before sputtering.

distinct lines or broad peaks in the respective spectrum, buta general baseline increase is also observed throughout theentire mass range.The spectra in Figs. 16 and 17 are highly compressed,

and peaks at identical nominal masses cannot be sepa-rated in these diagrams. The actual 7ne structures at 29and 43 amu, respectively, are shown in Fig. 18 for low-and high-resolution mass spectra of positive secondaryions. 29Si+ and 28Si1H+ can only be distinguished in thehigh-resolution spectrum at 29 amu. At low mass resolu-tion, only hydrocarbon peaks can be separated. This is alsoobvious at nominal mass 43 amu. The unambiguous iden-ti7cation of every single peak is even complicated for thehigh-resolution spectrum. Especially for some peaks (peaks3, 4, and 7–11 in Fig. 18) that lie between the element andthe pure hydrocarbon peak, assignments made in the 7gurecaption are equivocal.

Fig. 16. Highly compressed mass spectrum of positive secondary ionsobtained from glass standard KL2-G (cf. appendix) after extensive argonsputtering. Only the mass range from 1 to 100 amu is shown.

3.4. Quanti5cation

Quantitative evaluation of the mass spectra is one ofthe major objectives in TOF-SIMS analysis. From eachsignal in the spectrum, the abundance of the secondaryion species and ultimately the concentration of the respec-tive element, isotope, or molecule in the target have tobe deduced. However, only elemental or isotopic ratioscan be determined in SIMS. Absolute concentrations can-not be obtained due to insu4cient knowledge of sputtervolumes, absolute ion yields, transmission of the instru-ment, and detection e4ciencies. Prerequisites to determinethese ratios are, besides knowledge of the respective SIMSsensitivities, linearity of the detection and registration sys-tem, as well as control of instrumental mass fractionationeKects.

3.4.1. Linearity of the detectorLinearity of the detection system over a wide dynamic

range is indispensable for a quantitative evaluation of massspectra in TOF-SIMS, where the entire spectrum is measuredsimultaneously. Isotopic and element ratios up to several or-ders of magnitude occur, and secondary ion sensitivities span


Fig. 17. Mass spectrum (1–100 amu) of negative secondary ions fromglass standard KL2-G after argon sputtering. Broad peaks are caused byelectrons from various origins. They are also responsible for the higheroverall background in negative mass spectra compared to positive spectra(cf. Fig. 16).

a vast range. Therefore, high count rates for major peaks areoften compulsory to achieve su4cient ion intensities for mi-nor isotopes or trace elements. For single-ion counting, thetypical detection technique in TOF-SIMS, dead time eKectsfor high-intensity mass peaks have to be considered. Sincesignal intensities in TOF-SIMS .uctuate on an extremelyshort time scale (nanoseconds), these eKects become morecomplex than for constant or exponentially decaying signalintensities (SchiK, 1936; Esposito et al., 1991). However,dead time eKects have to be considered also in conventionalSIMS. Especially in isotope studies, they contribute signif-icantly to measurement uncertainties (Storms and Peterson,1994).The actual dead time of the TOF-SIMS IV instrument is

∼40 ns, which mainly results from the exponentially decay-ing scintillator signal. Therefore, only one secondary ion pernominal mass can be detected for each primary ion shot.However, dead time correction of TOF-SIMS spectra maybe performed using Poisson statistics. If only one mass peakis present at a given nominal mass, the dead time corrected

Fig. 18. Comparison of diKerent mass resolution settings near masses 29and 43 amu (top: 5 ns pulse length; bottom: 0:6 ns pulse length), respec-tively. At 29 amu the 7ve peaks in the high-resolution spectrum can beidenti7ed as (1) 29Si+, (2) 28Si1H+, (3) 12C1H16O+, (4) 12C1H314N+,and (5) 12C21H+

5 , respectively. At 43 amu (6) 43Ca+ with a contribu-tion from 42Ca1H+, (7) 27Al16O+, (8) 26Mg16O1H+, (9) 28Si12C1H+

3(10) 12C21H316O+, (11) 12C21H514N+, and (12) 12C31H+

7 emerge inthe high-resolution spectrum.

peak integral Icor becomes

Icor =−N ln(1− Iexp

N

); (9)

where Iexp is the measured integrated peak intensity and Nthe number of primary ion shots. The statistical one-sigmaerror of the corrected signal is

VIcor =(N IexpN − Iexp

)1=2: (10)

The correction of multiple peaks at one nominal mass be-comes more complicated, but even entire mass spectra canbe corrected for dead time eKects by applying similar formu-lae. This has been demonstrated in greater detail by Stephanet al. (Stephan, 1992; Stephan et al., 1994a). Limitations ofthese corrections result from two de7ciencies. First, the ac-tual dead time cannot be described by a single numericalvalue. A probability function has to be used instead, whichis only roughly known. Second, constant peak intensitiesduring the measurement are indispensable for an exact deadtime correction. However, simpli7ed correction techniquesdeliver results with su4cient precision (∼10%) for elementanalysis even for correction factors up to three. Restrictionsare more severe for isotope studies, where higher accuracy(few per mill) is required.

3.4.2. Mass fractionationInstrumental mass fractionation eKects play a less pro-

nounced role in TOF-SIMS. Here, general mass fractiona-tion eKects have not been observed so far. If there are anyeKects, they only have to be considered in isotope analy-sis, where high accuracy is required. But even there, statis-tical errors dominate at least for less abundant isotopes the


analytical uncertainties. For high-intensity ion signals, deadtime eKects often exceed mass fractionation eKects. How-ever, instrumental mass fractionation has always to be con-sidered, before deviations from “normal” isotopic ratios canbe claimed.

3.5. Element analysis

Peak assignments for quantitative analyses follow thesame principles as described by Eugster et al. (1985). Foreach isotope of a given element, at most the whole inten-sity at the respective mass window can be assigned. Thismass window is de7ned by the expected center of mass andthe prevailing mass resolution. A value for the expected to-tal elemental ion intensity can be calculated from each as-sumed isotope peak. Since mass interferences may aKect oneor more of these peak intensities, the minimum value de-duced from the diKerent isotopes is the upper limit for theintensity of the respective element. A proper analysis of therespective element can only be inferred, if the calculated to-tal elemental ion intensities for major isotopes agree withinstatistical error limits. If this is not the case, at least one iso-tope is aKected by interference and the others might also bein.uenced by similar interferences. Consequently, one hasto interpret the calculated elemental ion abundance as anupper limit. Strictly speaking, all calculated ion abundancesare upper limits, if interferences cannot be unequivocallyexcluded.Figs. 19 and 20 comprise peak assignments and the prin-

ciples of quantitative evaluation for positive and negativesecondary ions, respectively. Moderate mass resolution,su4cient for a complete separation of most interferingmolecules like hydrocarbons, was assumed (e.g., 5 ns pri-mary ion pulse length with the above-described galliumsource). In the following, the evaluation is discussed fordiKerent elements. Prerequisite for the here-described prin-ciples are “normal” (=cosmic) isotopic ratios of mostelements. —Throughout this study, isotopic abundances areused according to Anders and Grevesse (1989). —Cosmicisotope ratios can be assumed within statistical error limitsin most cases. Nevertheless, one has to be aware of thepossibility of non-solar isotopic ratios during the analysisof extraterrestrial material. Especially for the analysis ofpresolar grains, this has to be taken into account and will bediscussed in more detail below. Some of the interferenceslike hydrides, considered as non-separable in the presentsection, can be resolved by using very short primary ionpulses or mathematical peak deconvolution techniques. Thelatter will be explained in an extra section. Isotopic compo-sitions for some elements, especially the light ones, can bedetermined even at low mass resolution. This is discussednow, together with the principles of element analysis.

3.5.1. HydrogenAlthough hydrogen forms positive as well as negative

secondary ions during the SIMS process, the negative polar-

ity dominates. For positive secondary ions, a mass interfer-ence is observed for deuterium (2H+) from 1H+

2 molecularions. In addition, 1H+

3 -ions are often present in secondaryion spectra. The analysis of hydrogen with TOF-SIMS is af-fected by the residual gas in the vacuum apparatus, becauseit is typically dominated by hydrogen. Even at a pressureof several 10−10 mbar in the analysis chamber, hydrogen isstill adsorbed onto the sample surface during primary ionbombardment and can be observed in every mass spectrum.Therefore, the analysis of indigenous hydrogen isotopic ra-tios was not possible so far, although H2 forms practicallyno negative secondary ions and consequently 2H− is not in-.uenced by any mass interference.

3.5.2. Lithium, beryllium, and boronLithium, beryllium, and boron all form positive secondary

ions. Molecular interferences are almost negligible duringtheir analysis. Like all alkali metals, lithium is not observedas hydride in the positive spectrum. It would be unequiv-ocally identi7able as 7Li1H+ at nominal mass 8 amu. Anycontribution of 6Li1H+ at mass 7 amu to the 7Li+ mass peakwould be easy to subtract, particularly since 6Li has an iso-topic abundance of only 7.5% in “normal” lithium.Beryllium has only one stable isotope at nominal mass

9 amu. Here no mass interference occurs except 27Al3+,which can easily be separated (required mass resolutionm=Vm = 490). For boron, the convenient isotopic ratio ofabout four for 11B=10B allows a de7nite determination of theB+-intensity. BeH+ in most cases plays a minor role and, ifat all, only for the less-abundant isotope 10B+.

3.5.3. CarbonCarbon forms positive as well as negative secondary

ions during primary ion bombardment. Like observed forhydrogen, negative ions form the majority. Especially thepolarity of the carbon secondary ions depends on theirchemical environment. Hydrocarbons generate C+ as wellas C− secondary ions, whereas carbonates and in particularcarbides mainly form C−-ions. CH, CH2, and CH3 are alsoobserved in both polarities. Consequently, separation of the12C1H-signal from the 13C peak is essential for the mea-surement of 12C=13C isotopic ratios in both polarities. Aminimum in mass resolution of m=Vm= 2910 is necessaryfor this separation. Since the intensity of the 12C1H-ionsis often comparable with the 12C-signal, this mass resolu-tion is required at ∼1% of the peak height in case of solarcarbon (12C=13C = 89:9).For a proper measurement of isotopic ratios, preferentially

the amount of the hydride CH has to be minimized. Thiscan be done by sputtering of the sample, since formation ofhydrides is often due to the residual gas in the vacuumsystem (see above). To separate the two peaks at 13 amu, of-ten mathematical peak separation techniques have to be ap-plied (cf. section on presolar grains). A comparison betweenTOF-SIMS and DF-SIMS spectra for masses 12 and 13 amu,


Fig. 19. Flow chart showing the principles of peak assignments and determination of positive secondary ion abundances for elements from hydrogen tonickel.

respectively, is shown in Fig. 21. Although the TOF-SIMSspectrum (Fig. 21 top) was obtained under nearly optimalconditions—.at Te.on sample, only little surface adsorp-tion of hydrocarbons (low CH-signal), and short primary ionpulses (∼1:2 ns)—the two peaks at 13 amu were not sep-arated su4ciently. In contrary to this, these two peaks areseparated completely with DF-SIMS (Fig. 21 bottom).

3.5.4. NitrogenNitrogen has only a small ionization yield in the SIMS

process. Nevertheless, nitrogen can be measured as CN− ifcarbon is present. For the measurement of 14N=15N isotopicratios, 7rst, the 12C=13C isotopic ratio has to be calculated.Then, the mass peaks at 26 and 27 amu are determined. The

former can be identi7ed as 12C14N−, whereas the latter isformed by 13C14N− and 12C15N−. These two ion speciescannot be separated in TOF-SIMS (required mass resolutionm=Vm = 4270). The 14N=15N-ratio can be calculated nowwith the known 12C=13C-ratio according to

14N=15N =12C14N−12C15N− =

I26I27I26 −13C=12C

(11)

or

15N=14N =I27I26

−13C=12C; (12)

where Ix is the peak integral at x amu.


Fig. 20. Flow chart showing the principles of peak assignments and determination of negative secondary ion abundances for elements from hydrogen toiodine.

Fig. 21. Comparison of TOF-SIMS (top) with DF-SIMS (bottom, courtesyof F.J. Stadermann) spectra at masses 12 (12C) and 13 amu (13C and12C1H). With TOF-SIMS, positive secondary ions were retrieved from aTe.on sample by Ar+ primary ions. The DF-SIMS analysis was performedon a terrestrial silicon carbide grain. Mass resolutions (m=Vm) are givenat 12 amu for full width at 50, 10, and 1% peak height.

3.5.5. OxygenOxygen is the classical element providing negative

secondary ions in most SIMS applications. In oxidizedcompounds, it takes up the electrons released during theformation of positive metallic ions. Nevertheless, becauseof its high abundance in rocks, oxygen is also observedas positive secondary ions, though with lower intensity.Therefore, oxygen is quali7ed as reference element besidessilicon for the comparison of negative and positive sec-ondary ion spectra. Because of the very low abundance of

17O (CI: 0.038%) and the high 16O1H− intensity, only 16Oand 18O can be measured satisfactorily in TOF-SIMS.

3.5.6. Fluorine, chlorine, bromine, and iodineAll halogens form preferentially negative secondary ions.

Chlorine, the element with the highest electron a4nity, hasthe maximum negative secondary ion yield. Nevertheless,especially samples with high .uorine concentrations oftenshow signals at 19 amu also in the positive secondary ionspectrum. Fluorine and iodine are both monoisotopic ele-ments, whereas chlorine and bromine have two stable iso-topes each. Convenient isotopic ratios of about three for35Cl=37Cl and about one for 79Br=81Br, respectively, allowmeasuring all four nuclides. Since 79Br is often disturbed bymolecular interferences, 81Br is the isotope of choice for aproper bromine analysis (Stephan et al., 1994b).Especially chlorine and .uorine are omnipresent in our

environment. Therefore, many samples show halogen con-tamination on the surface, often together with the alkalissodium and potassium. Consequently, cleanness of thesample surface is especially important for a properhalogen analysis.

3.5.7. Sodium, potassium, rubidium, and cesiumThe same high demands on surface purity apply to the

alkalis, especially sodium and potassium, which are oftenobserved as surface contamination. The alkalis have thehighest positive secondary ion yield, analogous to the highyield for negative ions for halogens. Hydrides and doublycharged secondary ions are not observed. The monoisotopicelement sodium can unequivocally be analyzed at 23 amu,


where no molecular interference appears. Potassium as 39K+

can easily be separated from C3H+3 at a low mass resolu-

tion (m=Vm = 652). 40K (0.01167% of K) is not distin-guishable from 40Ca (96.941% of Ca) due to a requiredmass resolution of m=Vm = 28400 and its low abundance.Although the abundance of 41K is higher (6.7302%), mass41 amu is typically dominated by 40C1H (required massresolution m=Vm = 4770). Rubidium has two primordialisotopes 85Rb (72.165%) and 87Rb (27.835%). The latteris interfering with 87Sr (7.00%, required mass resolutionm=Vm=296000). Cesium is monoisotopic and can be seenat 133 amu.

3.5.8. Magnesium, calcium, strontium, and bariumThe alkaline-earth metals are the group with the second

highest yield for positive secondary ions. They often formhydride-ions like, e.g., MgH+, complicating isotope analy-ses. 24Mg1H+ overlaps with 25Mg+; 25Mg1H+ with 26Mg+,and 7nally 26Mg1H+ coincides with 27Al+ (required massresolutions m=Vm = 3550; 2350, and 3040). The latter in-terference complicates the analysis of aluminum. However,assuming a 25Mg=24Mg-ratio of 0.127, which mostly is jus-ti7ed for the achievable accuracy, one can calculate a possi-ble 26Mg-enrichment and also the contribution of 26Mg1H+

to mass 27 amu:

27Al+ = I27 − I2426Mg24Mg

MgH+

Mg+(13)

with

MgH+=Mg+ =I25I24

− 0:127 (14)

and

26Mg=24Mg =I26I24

− 0:127MgH+

Mg+: (15)

Similar to this, 40Ca1H+ interferes with 41K like dis-cussed above. Besides hydrides, also oxides are observed.To separate 24Mg16O+ from 40Ca and 40Ca16O+ from56Fe, mass resolutions of m=Vm = 2300 and 2480, respec-tively, are required. Hydroxides like 40Ca16O1H+ at 57 amualso emerge in the mass spectra. 44Ca, the other isotopeavailable for quantitative calcium analysis interferes with28Si16O (m=Vm = 2690). Therefore, 40Ca is more reli-able for a quantitative analysis especially of silicates. 44Cacan be useful only in Mg-rich, Si-poor samples like somecarbonates.Doubly charged ions like 24Mg2+ or 40Ca2+ are also

observed for alkaline-earth metals. Since the mass scalein TOF-SIMS spectra is actually a mass-to-charge scale,they appear at 12 and 20 amu, respectively. Consequently,24Mg2+ interferes with 12C+ (m=Vm=1600), but this con-tribution can be calculated from the 25Mg2+ intensity at12:5 amu.Strontium is measured mainly as 88Sr+ (82:58%) and

86Sr+ (9:86%); 87Sr+ (7:00%) interferes with 87Rb+ (seeabove). Barium is detected primarily as 138Ba+ (71:70%)

and 137Ba+ (11:23%) but can also be seen as 136Ba+

(7:854%), 135Ba+ (6:592%), and 134Ba+ (2:417%). 130Ba+

(0:106%) and 132Ba+ (0:101%) are too rare in most spectrato be detected.

3.5.9. AluminumThe monoisotopic element aluminum can be measured

at nominal mass 27 amu considering corrections from26Mg1H+ according to Eq. (13). AlO+ at 43 amu; Al2+ at13:5 amu and even Al3+ at 9 amu are also observed. Thehydride might be present at 28 amu but is often not resolvedfrom 28Si+ (m=Vm= 2250).

3.5.10. SiliconSilicon has three stable isotopes, 28Si (92.23%), 29Si

(4.67%), and 30Si (3.10%). Although it forms predomi-nately positive secondary ions, also negative ions occur.For silicates it is suited as reference element besides oxygenfor the comparison of negative and positive secondary ionspectra (see above). Furthermore, because of its ubiquityin rocks, it is used as normalization element for elementabundances in meteorites or the solar system (Anders andGrevesse, 1989). Similar to magnesium, hydrides occur andinterfere with 29Si; 30Si, and 31P.

3.5.11. PhosphorousThe monoisotopic element phosphorous forms normally

31P−-ions, but also positive secondary ions are observed insamples with high phosphorous concentrations like phos-phates. Similar to 27Al+ and its interference with 26Mg1H+,31P+ is in.uenced by 30Si1H+, which can be corrected anal-ogous to Eq. (13).

3.5.12. SulfurSulfur, which forms negative secondary ions, has two

isotopes, 32S (95.02%) and 34S (4.21%), that are measurablein TOF-SIMS. Especially for 32S, the O2 molecule plays anannoying role in the mass spectrum, but it can be separatedsu4ciently at moderate mass resolution (m=Vm= 1800).

3.5.13. Scandium, titanium, vanadium, and chromiumThe monoisotopic element scandium has typically very

low abundances in most rocks (Sc=Si = 3:42× 10−5, chon-dritic atomic abundance according to Anders and Grevesse,1989). In addition, 45Sc+ interferes with 28Si16O1H+ andcan therefore only be seen in TOF-SIMS spectra with highmass resolution.In Ti-rich samples, all 7ve stable isotopes (46Ti: 8:0%,

47Ti: 7:3%, 48Ti: 73:8%, 49Ti: 5:5%; and 50Ti: 5:4%) canbe found as positive secondary ions. Since besides 48Ca,no major interference is observed, and the share from 48Cacan easily be calculated from the total calcium, 48Ti is usedto determine the element abundance. On the other hand,48Ti16O+ interferes with 64Zn+ as described below.


Because of the low abundance of 50V (0.250%) and itsinterference with 50Ti and 50Cr, vanadium can only be mea-sured as 51V. Chromium, 7nally, is measured as 52Cr, butit also contributes signi7cantly to the peaks at 50, 53, and54 amu. However, 54Cr+ can be neglected in most cases dueto much higher contributions from 54Fe+.

3.5.14. Manganese, iron, cobalt, and nickelIron usually dominates the positive mass spectrum at

54 and 56 amu due to its high cosmic abundance. Never-theless, in Ca-rich, Fe-poor samples, 40Ca16O+ contributessigni7cantly to the mass peak at 56 amu. Another possi-ble interference is the 28Si+2 dimer. Whenever these peaksare not separable (required mass resolution m=Vm = 2480for 40Ca16O+ and 2960 for 28Si+2 ), the less-abundant54Fe-isotope (5.8% instead of 91.72% for 56Fe) has to betaken for calculation of the iron abundance after correc-tion for contribution from 54Cr: 57Fe+ (2.2%) is in mostcases not discernable from 56Fe1H+ (m=Vm = 7730) andoften in.uenced by 40Ca16O1H+ (m=Vm = 1900). Espe-cially the FeH=Fe-ratio is important to determine, since54Fe1H+ is the major interference at mass 55 amu for55Mn+ (m=Vm = 5850). Similarly, the analysis of cobalt,which is also monoisotopic, is complicated by 58Ni1H+.Nickel can be measured as 58Ni+ and 60Ni+, the formerwith interferences from 58Fe+ and 57Fe1H+.

3.5.15. Copper and zincCopper and zinc have both low cosmic abundances

(Cu=Si = 5:22 × 10−4; Zn=Si = 1:260 × 10−3, chondriticatomic abundances according to Anders and Grevesse,1989) and are therefore di4cult to determine. From theeasy to con7rm isotopic ratio of approximately 7 : 3 for63Cu=65Cu, an unequivocal identi7cation of copper is some-times possible. Zinc, although having more stable isotopes,is harder to identify because of interferences from TiO+.

3.5.16. GalliumGallium was mainly used as primary ion species for the

TOF-SIMS analyses in this study. Although the ion sourceis equipped with “monoisotopic” 69Ga, some remaining71Ga (∼0:2%) was still present, since the industrial isotopeseparation process is not perfect. Therefore, both isotopescannot be used for the analysis of indigenous gallium.

3.5.17. SilverSilver has a comparatively high ionization yield although

it has a relatively high ionization energy (7:58 eV). Silveror silver oxide is often used as substrate for the analysis oforganic substances since it also has a positive eKect on theionization probability of other elements and molecules (ma-trix eKect). Both isotopes can be measured (107Ag: 51.839%and 109Ag: 49.161%) and show typically no interferences.

3.5.18. LanthanidesRare earth elements could be seen in the positive sec-

ondary ion spectrum. Nevertheless, an unambiguous deter-mination of these elements is not possible so far, due to awealth of isobaric interferences and the presence of hydridesand oxides—oxides of the low-mass lanthanides overlapwith those of the high-mass ones.

3.5.19. LeadAs a surface analytical technique, TOF-SIMS often “sees”

lead present in the positive secondary ion spectrum result-ing from surface contamination, because this highly volatileelement is omnipresent in our environment.

3.5.20. Other elementsNoble metals besides silver have typically low secondary

ion yields and are therefore not suited for TOF-SIMS anal-ysis. Noble gases normally are not ionized in the SIMSprocess. Other elements not mentioned speci7cally here aretypically too rare in cosmic materials. Nevertheless, someof them can certainly be analyzed with TOF-SIMS.

3.6. SIMS sensitivities

For a quantitative analysis with TOF-SIMS, to transformsecondary ion intensities into element ratios, relative sen-sitivities must be determined. Therefore, carefully selectedstandards have to be analyzed. Since the actually investi-gated volume is extremely small, only a few atomic mono-layers thin, sample homogeneity and purity are crucial. Asdescribed before, amorphous or polycrystalline standards aredesired. However, certain materials, glass as well as someminerals, used as standards in other techniques, do not meetthe requirements of TOF-SIMS concerning homogeneity ona micrometer scale. Therefore, the quality of a standard mustalways be monitored in the TOF-SIMS analysis.In Fig. 22, typical TOF-SIMS sensitivities are shown for

positive and negative secondary ions. These results werederived from TOF-SIMS analyses of eight glass standards(Jochum et al., 2000) with known composition. They meetthe above-mentioned strict requirements for standards con-cerning homogeneity as was shown during the TOF-SIMSanalysis. The results are comparable to DF-SIMS sensitivi-ties in the literature (e.g., Meyer, 1979; Lange et al., 1986).Unfortunately, .uorine and sulfur concentrations are not

known so far for these glass standards. Therefore, the data seton negative secondary ion sensitivities is quite incomplete.Previous investigations of halogens and sulfur showed thatthe selection of appropriate standards for these elements israther di4cult. Sensitivities here vary over more than oneorder of magnitude.All glass standards were measured twice, with simulta-

neous argon sputtering and directly after sputtering. Sen-sitivities before sputtering could not be determined sinceno stable ion signals were reached (cf. Fig. 15). Except forO−-ions, diKerences between sensitivities during and after


Fig. 22. Relative TOF-SIMS sensitivities for positive (solid symbols) and negative (open symbols) secondary ions relative to silicon derived from theanalysis of eight glass standards with the gallium primary ion source. The sensitivities were determined from measurements with simultaneous argonsputtering and directly after sputtering. Only for negative oxygen ions signi7cant diKerence between both measurements appears (the open square givesthe O−-sensitivity after sputtering, the open circle during sputtering). Carbon sensitivities are determined from silicon carbide samples.

sputtering seem not to exist for most elements, althoughsecondary ion signals typically decrease after sputtering(Fig. 15). For negative oxygen ions, the sensitivity relativeto silicon increases by a factor of ∼1:7 after sputteringcompared to the sensitivity during argon sputtering.Detailed results from the analyses of the individual glass

standards are listed in the appendix. For chlorine, only thesensitivity obtained during sputtering from the most Cl-richglass (ATHO-G, 400 ppm Cl) is shown in Fig. 22. All othersamples showed traces of surface adsorbed chlorine. Forother elements, geometric means for the sensitivity valuesgiven in the appendix are shown in Fig. 22. Error bars markthe range of values for diKerent measurements. Since no car-bon is present in the glass standards, C-sensitivities relativeto silicon were determined from the analysis of several sili-con carbide grains, although it is expected that matrix eKectsin silicone carbide hardly resemble those in glass.

3.7. Isotope analysis

Only a few elements in typical rocks allow isotopic anal-ysis at moderate mass resolution. For hydrogen (negativesecondary ions), lithium, and boron, practically no interfer-ences are observed in the spectrum. Starting with carbon,hydrides become a major problem, because they are notseparable with moderate mass resolution. Hydrides are lessimportant for elements with isotopes that are separatedby two mass units, e.g., copper or silver. Since these aretypically present as trace elements in natural samples, other

mass interferences like oxides often prevent properisotopic analyses.If isotopic analysis with high accuracy is required, short

primary ion pulses are indispensable to achieve high massresolution. Even with these, neighboring peaks are often notsu4ciently separated. Therefore, only mathematical peakseparation techniques provide further help. Since peaks inTOF-SIMS are typically not Gaussian and even not sym-metric, the expected peak shapes have to be known for aproper deconvolution. Peak asymmetry is in part caused bydead time eKects that must be corrected by using Poissonstatistics as described above. Other eKects cannot so easilybe corrected: Besides instrumental parameters like, e.g., ex-traction and re.ection 7elds the peak shape also depends onthe initial energy distribution of the respective ion species.However, if an undisturbed isotope of the same element ispresent, its peak may be used as an internal standard peak.This is the best available approximation under realistic con-ditions, since all isotopes of one element should have thesame energy distribution. Peaks of other elements are notsuitable, because their energy distribution might be diKerent.To separate, e.g., the hydride peak 24Mg1H+ from 25Mg+,

the 24Mg+-peak can be used to 7t the low-mass edge ofthe spectrum at 25 amu (Fig. 23). Here, the scaling fac-tor for 24Mg+ gives directly the 25Mg=24Mg isotopic ra-tio. The diKerence between the two spectra represents the24Mg1H+-peak.This peak deconvolution technique has been success-

fully applied to various isotope systems. For the eight glass


Fig. 23. A demonstration of the peak separation of 25Mg+ and 24Mg1H+.The shape of the 24Mg+-peak (dotted line, top x- and right y-axis) wasused as a model for the 25Mg+-peak. The hydride peak (solid line)emerges as the diKerence between the actual signal at mass 25 amu(shaded area) and the 24Mg+-peak, scaled to the same height. The spec-trum was measured with bunched gallium primary ions (∼600 ps pulsewidth) on glass BM90=21-G (cf. appendix).

standards, already used to determine elemental sensitiv-ities (see above), expected terrestrial values have beenreproduced within deviations of a few percent for severalisotopic ratios (Table 2). The accuracy achievable with thispeak separation technique strongly relies on the abundanceof the respective element in the sample and therefore variesfor diKerent samples.

3.8. Organic molecules

Since the number of possible interferences for molecu-lar ions can be very high, appropriate strategies had to bedeveloped for their proper identi7cation. Nevertheless, anaccurate mass determination is prerequisite to discriminatemost possible interferences. Fragmentation during primaryion bombardment of most organic compounds occurs, 7llingthe respective mass spectrum with numerous peaks. Seriesof fragments with well-de7ned mass diKerences are strikingfeatures of organic mass spectra. They often allow the iden-ti7cation of organic molecules or entire molecule classes.Quanti7cation of these results is even more complicated

than for element ratios, often limited to substances sitting onsimple, well-de7ned substrates, and requires internal stan-dards (HagenhoK et al., 1992; HagenhoK, 1997).

3.9. Data processing

The amount of information obtained from a TOF-SIMSanalysis can be tremendous. In principle, each primary ionshot delivers a complete mass spectrum, although only afew secondary ions might be distributed over the entire massrange. Nevertheless, each secondary ion has to be registeredindividually to avoid loss of information. Mass spectrafor a carefully selected series of primary ion shots can be

Table 2Reproducibility of isotopic ratio TOF-SIMS measurements from the anal-ysis of eight diKerent glass standardsa

Isotopic ratio �-values Standard Mean statistical(per mill) deviation error (per mill)

(per mill)

(Positive secondary ions)�(25Mg=24Mg) +5 9 2�(26Mg=24Mg) −13 7 2�(29Si=28Si) −23 14 4�(30Si=28Si) −67 14 5�(41K=39K) −42 56 7�(42Ca=40Ca) −9 13 9�(43Ca=40Ca) −45 41 21�(44Ca=40Ca) −76 49 7�(46Ti=48Ti) +97 121 27�(47Ti=48Ti) +75 57 25�(49Ti=48Ti) −2 91 26�(53Cr=52Cr) −10 60 39�(57Fe=56Fe) +1 30 14

(Negative secondary ions)�(13C=12C) ±0 94 168�(17O=16O) −72 81 44�(18O=16O) −88 36 13�(29Si=28Si) −46 41 28�(30Si=28Si) −71 21 33�(34S=32S) −83 117 108�(37Cl=35Cl) −21 37 30

aData measured by T. Henkel (pers. comm.). The second column givesthe average deviation from standard isotopic ratios in per mill as �-values

�(xA=yA) = 1000(

(xA=yA)(xA=yA)Std:

− 1):

The standard deviation of this average is shown in column three, whereascolumn four provides the mean statistical error of individual measure-ments. Since carbon was only present as surface contamination on theglass standards, the deviations from cosmic ratios are not representativefor samples containing carbon in the bulk.

obtained during subsequent data processing. Time pro7lesas those in Fig. 15 show the time dependence of the re-spective secondary ion signals. Secondary ion images foran unlimited number of mass intervals can be obtainedby scanning the primary ion beam over the sample. Theseimages show lateral element or molecule distributions(cf. Fig. 24).The amount of data, typically several hundred megabytes

in one measurement, can be handled and processed by thepresent generation of personal computers. A 7rst evaluationcan be performed during the measurement, providing totalmass spectra, selected ion images, or depth pro7les. Never-theless, for a detailed evaluation, all available informationon every secondary ion has to be monitored.Automated techniques are here desired for quali7ed data

reduction and evaluation in the future to get the immensedata stream under control, without losing any important in-formation about the target.


Fig. 24. Secondary ion distribution images from a CAI in Cold Bokkeveld. Field of view is 500× 500 �m2. Each image consists of 256× 256 pixels.The number of primary ion shots per pixel is 1600 for positive and 1120 for negative secondary ions. Secondary ion species are given below each image(e.g., 16O−). Individual secondary ion images are normalized to the number given below the image, usually the intensity in counts of the most intensepixel (489 for 16O−), shown in red. A color bar represents the linear scale from black (equals zero) to red. The second number below each image(9:57× 106 for 16O−) is the intensity integrated over the entire image. (D. Rost and S. Schirmeyer, pers. comm.)

4. Meteorites

Meteorites are the principal objects of research in cosmo-chemistry. They are the major source of information aboutour solar system, its origin and early history about 4:6 Gaago. The investigation of small, micrometer-sized entitieswithin meteorites has become increasingly important duringthe last decades. Here, TOF-SIMS was expected to be an ap-propriate technique due to its imaging capabilities. Elemen-tal mapping, isotope analysis, as well as the investigationof organic matter can be performed on phases with lineardimensions even smaller than one micrometer. Several dif-ferent applications of TOF-SIMS in meteorite research arepresented exemplarily in the following.

4.1. Ca,Al-rich inclusions in carbonaceous chondrites

Ca,Al-rich inclusions (CAIs) are among the most primi-tive objects found in meteorites (MacPherson et al., 1988).As high-temperature condensation products, they stemfrom the earliest stages of solar system formation and aretherefore of special interest in cosmochemistry. After theirformation, many CAIs experienced complex histories of al-teration. These objects where among the 7rst extraterrestrialsamples chosen for TOF-SIMS analysis (Stephan et al.,1991) to test the applicability of this technique in meteoriteresearch. CAIs from CM, CO, and CV meteorites wereinvestigated in a later, more extended study (Schirmeyeret al., 1996a, b, 1997). Here a striking feature is the dis-

tribution of lithium, which was found enriched in Fe-richphyllosilicates within CAIs from several diKerent carbona-ceous chondrites but not in phyllosilicates located in thesurrounding matrix. Examples of ion images for severalsecondary ion species including 7Li+ are shown in Fig. 24.They were obtained from a mineralogical thin section froma CAI within the CM chondrite Cold Bokkeveld (Fig. 25).The CAI itself can clearly be seen in the aluminum image.Although in the 7Li+-image only 2790 ions were detected,a concentration of the signal within several regions of theCAI is obvious. A closer examination of the results showthat lithium is especially enriched in Fe-rich phyllosilicateslocated between the Ca-rich pyroxene rim and the interiorspinel of the CAI (Schirmeyer et al., 1997). Although thishydrous phase is similar to the surrounding water-bearingminerals in the host rock, these minerals diKer by more thantwo orders of magnitude in their lithium contents.From TOF-SIMS studies on various carbonaceous chon-

drites, it can be ruled out that lithium was already presentin the proposed precursor material of these hydrous phases,most likely melilite (Schirmeyer et al., 1996b, 1997). It musthave been incorporated into the phyllosilicates during aque-ous alteration processes.Phyllosilicates in carbonaceous chondrites were formed

by aqueous alteration in the early solar nebula and=orwithin meteorite parent bodies (BischoK, 1998). Still somecontroversy exists about the time and environment of theirformation, and, more generally, about the possibility ofaqueous alteration prior to the accretion of the 7nal parent


Fig. 25. SEM-BSE (scanning electron microscope, backscattered elec-trons) image of the CAI shown in Fig. 24. (S. Schirmeyer, pers. comm.)

body. While Metzler et al. (1992) found clear evidencefor such preaccretionary processes for CM chondrites(BischoK, 1998), solely asteroidal alteration was also pro-posed (Browning et al., 1996; Browning and Keil, 1997;Buseck et al., 1997; Zolensky, 1997).The TOF-SIMS observation of greatly diKerent lithium

abundances in Fe-rich phyllosilicates within CAIs and thosein the adjacent matrix strongly suggest that both phases wereformed in diKerent environments. At least the phyllosilicatesof the CAI must have been formed prior to the incorporationof the CAI into the meteorite parent body probably undernebular conditions (Schirmeyer et al., 1997). This supportsthe conclusions of MacPherson and Davis (1994) who sug-gested alteration of refractory inclusions from the CM chon-drite Mighei in the solar nebula, and not in the present or aprevious generation of parent bodies.Lithium isotopic ratios are of great interest due to their

cosmological implications. In general, both lithium isotopescan be formed through three diKerent processes: (1) galacticcosmic-ray spallation of heavier nuclides, mainly 12C; 14Nand, 16O, (2) � + �-fusion reactions, and (3) during pri-mordial nucleosynthesis just after the big bang (Meneguzziet al., 1971; Reeves et al., 1973; Olive and Schramm,1992; Steigman and Walker, 1992; Thomas et al., 1994a).While big bang nucleosynthesis produces almost pure7Li (7Li=6Li-ratio ∼104), interactions of galactic cosmicray spallation and fusion reactions lead to 7Li=6Li-ratios ofthe order of one (Steigman and Walker, 1992). Therefore,the 7Li=6Li-ratio in the solar system (∼12) can only beexplained by a mixture of lithium from diKerent nucleosyn-thetic processes. Deviations from the CI-value in primitivesolar system material would be of great interest, but havenot been unambiguously found so far.Since lithium has two stable isotopes, 6Li and 7Li with

CI-abundances of 7.5 and 92.5% (Anders and Grevesse,1989), respectively, and no mass interferences occurs in thesecondary ion spectra, it seemed to be predestined for isotope

measurements with TOF-SIMS. Anyhow, within statisticalerror limits no evidence for deviations from the chondriticlithium isotope ratio other than caused by instrumental massfractionation was observed with TOF-SIMS for the Li-richphases in carbonaceous chondrites (Schirmeyer et al., 1997).These results were also con7rmed by DF-SIMS analysis(Schirmeyer et al., 1997).

4.2. Organic molecules in Allan Hills 84001, Murchison,and Orgueil

Since the early 19th century, it is known that carbona-ceous chondrites contain organic matter (Berzelius, 1834;“Gibt dies m�oglicherweise einen Wink �uber die Gegenwartorganischer Gebilde auf anderen Weltk�orpern?”). In thesecond half of the 20th century, due to new analytical tech-niques and a rising interest in space science, a lot of ef-forts were made to characterize these organic components(Hayes, 1967; Anders et al., 1973; Cronin et al., 1988). Inthe 1960s, amino acids were found in several carbonaceouschondrites, but contamination during handling and storage ofthe samples could not be unambiguously ruled out (Hayes,1967). Reports of non-terrestrial fossil bacteria (Claus andNagy, 1961) in those years were also soon compromisedas terrestrial organisms (Hayes, 1967). In the early 1970s,analyses of the Murchison meteorite fallen in 1969 showedclear evidence for extraterrestrial amino acids, aliphatic andaromatic hydrocarbons in this CM chondrite (Kvenvoldenet al., 1970; OrWo et al., 1971; Pering and Ponnamperuma,1971).Although no connection of these organic compounds to

extraterrestrial life can be drawn (Chyba and Sagan, 1987)—in contrary they have been linked to the interstellar medium(Allamandola et al., 1987)—prebiotic organic matter fromcomets and asteroids delivered to Earth by meteorites andinterplanetary dust might have been important for the originof life on our planet (Anders, 1989; Chyba et al., 1990;Chyba and McDonald, 1995).For our neighboring planet, organic compounds known

as polycyclic aromatic hydrocarbons (PAHs) in the Mar-tian meteorite ALH 84001 (Thomas et al., 1995) have beensuggested as being indicative for the existence of past lifeon Mars (McKay et al., 1996; Clemett et al., 1998). Sincethen, the signi7cance of PAHs as biomarkers in this casehas been questioned by many authors (Anders, 1996; Bell,1996; Becker et al., 1997, 1999; Stephan et al., 1998a–c,1999a; Zolotov and Shock, 1999).Chemically, PAHs represent the transition state between

aliphatic hydrocarbons and graphite. While the former arecharacterized by a H=C-ratio of ∼2, this ratio becomes zerofor the latter. For PAHs, H=C is below 0.8 and generally de-creases with increasing mass. PAHs are omnipresent in oursolar system and even beyond (Allamandola, 1996). In prin-ciple, they can derive from biogenic as well as non-biogenicsources. On Earth, PAHs derive mainly from anthropogenicemission caused by the combustion of fossil fuels and other


Fig. 26. Secondary ion (100 × 100 �m2, 200 shots=pixel, 256 × 256 pixels) and SEM-BSE images for an area of ALH 84001 dominated by acarbonate surrounded by orthopyroxene. The PAH image is obtained by summing up signals from characteristic PAH masses between 115 and 339 amu(cf. Figs. 28 and 29).

natural biogenic sources (La.amme and Hites, 1978; Wake-ham et al., 1980a, b). PAHs attributed to a non-biogenicorigin can be found in ordinary and carbonaceous chondrites(OrWo et al., 1971; Pering and Ponnamperuma, 1971; Hahnet al., 1988; Zenobi et al., 1989; Wing and Bada, 1991;Clemett et al., 1992; de Vries et al., 1993). They havebeen detected in interplanetary dust particles (Clemettet al., 1993), comet Halley (Moreels et al., 1994), and alsoin circumstellar graphite grains extracted from primitivemeteorites (Clemett et al., 1996; Messenger et al., 1998).Therefore, the presence of PAHs in ALH 84001 is not

indicative for past life on Mars. Nevertheless, is has beenargued that their spatial association with other proposedbiomarkers suggests a biogenic origin of PAHs in this Mar-tian meteorite (McKay et al., 1996). These biomarkers, mostprominently the “nanofossils”, are all related to chemicallyzoned Ca–Mg–Fe-carbonates (Mittlefehldt, 1994; Scottet al., 1998; Treiman and Romanek, 1998; Warren, 1998),which might have been deposited from hydrothermal .uidson Mars (McKay et al., 1996). These distinct observa-tions were made by use of entirely diKerent techniques,which, simply from their spatial resolution, hampers acorrelation of their results. Due to the small size of thesupposed microfossils, 100 nm in longest dimension and20–80 nm across, a scanning electron microscope with 7eldemission gun (FEG-SEM) with a lateral resolution of about2 nm was used (McKay et al., 1996). On the other hand,

PAHs were identi7ed by �L2MS analysis with a lateral res-olution of about 40 �m (Thomas et al., 1995). A meaningfulmass spectrum was obtained by mapping a surface regionof 750 × 750 �m2 at a spatial resolution of 50 × 50 �m2

(McKay et al., 1996). Therefore, a correlation of PAHs evenwith carbonates, between 1 and ∼250 �m in size (McKayet al., 1996), could not be scrutinized by this technique.Here, TOF-SIMS has its advantages. The high lat-

eral resolution in combination with parallel detection ofatomic as well as molecular secondary ions should bepredestined for an examination of the proposed spatialassociation. First TOF-SIMS studies of ALH 84001 wereperformed on polished thin sections of this meteorite(Stephan et al., 1998a, b). Although a homogenizationof the lateral PAH distribution by the polishing processwas expected, PAHs were found to be distributed inho-mogeneously, especially after removal of the uppermostmonolayers through sputtering. However, a spatial associa-tion with carbonates could not be con7rmed in this study.Although omnipresent in this Martian meteorite, in car-bonates, compared to orthopyroxene and feldspathic glass,PAHs even seem to be slightly depleted. Spatial distribu-tions of PAHs and major rock-forming elements are shownin Fig. 26 for one section of ALH 84001. The area is domi-nated by a carbonate, prevailing in the Ca+-image, which issurrounded by orthopyroxene (dominant in the Si+-image).In the meantime, the search for PAHs was extended to


Fig. 27. Secondary ion images from a fracture surface of ALH 84001 (150× 150 �m2, 3200 shots=pixel, 256× 256 pixels). Prevailing mineral phasesare orthopyroxene (dominant in Mg+-, Si+-, and Fe+-images), feldspathic glass (Na+-, Al+-, and K+-images), and carbonate (Ca+-image). Onecarbonate-rich area is enriched in barium. Small chromite grains are apparent in the Cr+-image. They are rich in titanium. Lithium is also enriched indistinct spots, but is not obviously correlated to any other element.

fractured surfaces. Preliminary results (Fig. 27) here alsoshow no indication of PAH enrichments associated withcarbonates. PAHs seem to be omnipresent on this surface.Nevertheless, no homogeneous distribution of PAHs wasobserved in the corresponding image. The variability can atleast in part be attributed to topographic eKects, since hereno .at surface was investigated.Amore detailed analysis of the TOF-SIMS spectra is help-

ful to elucidate the origin of the PAHs in ALH 84001. Thesemass spectra (Fig. 28) diKer signi7cantly from those ob-tained by �L2MS (McKay et al., 1996; Clemett et al., 1998;Becker et al., 1999). This can be explained by diKerencesin ionization processes. Fragmentation of the PAHs occursduring primary ion bombardment, increasing the low-massPAH signals.

Pure PAH substances, pentacene (C22H14) and coronene(C24H12), as well as two carbonaceous chondrites containingPAHs, Murchison and Orgueil, were analyzed to investigatethis fragmentation more thoroughly (Stephan et al., 1998c,1999a). Pentacene and coronene showed fractionation dur-ing primary ion bombardment, but to diKerent extents(Fig. 29). The linear line-up of benzene rings in pentaceneis more susceptible to fragmentation than their circular ar-rangement in coronene. However, for both substances thevery same PAH fragments were observed as in ALH 84001and the other two meteorites. From these results, one caninfer that PAHs similar to those observed by �L2MS hadproduced the observed TOF-SIMS spectra upon sputtering.Since unambiguous proof for a connection between PAHs

and indigenous features in ALH 84001 is missing, it has


Fig. 28. Typical TOF-SIMS spectrum of mass range 110–345 amu for positive secondary ions obtained from a section of ALH 84001. Relative maximaare separated by mass diKerences of 11 or 13 amu. Often several maxima appear separated from each other by a mass diKerence corresponding to H2.Since several structural isomers exist for all mass peaks, only examples for major peak identi7cations are given.

been suggested that PAHs as well as amino acids in ALH84001 originate from the icy environment like observedin other Antarctic meteorites (Becker et al., 1997; Badaet al., 1998). Based on stable carbon isotope measurements,it was proposed that besides terrestrial contamination, a sec-ond organic carbon component exists, which might be ex-traterrestrial, probably Martian (Jull et al., 1998; Beckeret al., 1999).To investigate the origin of PAHs in ALH 84001—

terrestrial or Martian—extraterrestrial PAHs from thecarbonaceous chondrites Murchison and Orgueil were an-alyzed with TOF-SIMS (Stephan et al., 1999a). The massspectra of PAHs from these two meteorites resemble re-markably well those obtained from ALH 84001 (Fig. 29).All three meteorites show essentially the same PAH frag-mentation and, therefore, a similar, non-biogenic origin canbe deduced. For biogenic PAHs we would expect higherdegrees of alkylation and the presence of aromatic hetero-cycles like observed in terrestrial samples (Wakeham et al.,

1980a, b). Heteroatoms would drastically increase the ion-ization yields of the PAHs. Because of this and since theywould appear at diKerent characteristic masses, they cannotbe overlooked in the mass spectra.So far, the comparison between the TOF-SIMS spec-

tra of the three meteorites is not quantitative. To calculatethe mean H=C-ratio for all observed PAH fragments, is a7rst attempt to overcome this de7ciency. Here, the threemeteorites yield similar values, 0.69 for ALH 84001 and0.65 for both carbonaceous chondrites. These values diKersigni7cantly from those obtained for pentacene (0.58) andcoronene (0.52). This diKerence and the similarities betweenall 7ve PAH spectra are also emphasized in Fig. 30, wherethe data points from Fig. 29 for ALH 84001, Orgueil, andthe two pure PAH substances are directly compared withthose of Murchison. It is expected that future TOF-SIMSanalyses of terrestrial biogenic PAHs will help to assessthese results and to discover the true nature of PAHs in ALH84001.


Fig. 29. Relative abundances of PAH ions in TOF-SIMS spectra obtained from ALH 84001, Murchison, and Orgueil compared with those obtained frompure PAHs, pentacene and coronene. Coronene and pentacene structures are shown as well as the structural formula of anthracene (C14H10) a majorfragment of pentacene during sputtering. Besides the number of carbon atoms above the upper x-axis, the numbers of hydrogen atoms are given for eachmajor peak next to the ALH 84001 data points.

However, the present results, great similarities betweenPAHs in ALH 84001 and carbonaceous chondrites, supportsthe suggestion that these organic molecules derived from theexogenous delivery of meteoritic or cometary debris to thesurface of Mars (Becker et al., 1999), similar to those com-pounds delivered to the early Earth (Anders, 1989; Chybaet al., 1990). Alternatively, an abiotic synthesis of PAHs onMars is also conceivable, provided that this chemical processis similar to the formation process of PAHs in Murchisonand Orgueil. This has also been proposed recently (Zolotovand Shock, 1999). Independent of the discussion about thesigni7cance of other suggested biomarkers in ALH 84001(e.g., Bradley et al., 1996, 1997, 1998; Leshin et al., 1998;Scott, 1999), TOF-SIMS studies revealed that no spatial

association between PAHs and carbonates exists inthis meteorite.

4.3. Further meteorite studies with TOF-SIMS

4.3.1. OrgueilCI chondrites represent chemically the most pristine solar

system material available for laboratory analysis. Except forthe so-called CHON elements (carbon, hydrogen, oxygen,and nitrogen), lithium, and the noble gases, their composi-tion is representative for the entire solar system (Anders andGrevesse, 1989).Only little is known about the actual distribution of

minor and trace elements in CI meteorites, despite their


Fig. 30. Comparison of relative intensities of PAHs in ALH 84001 andOrgueil meteorites, as well as of pentacene and coronene standards, withthe PAH intensities in Murchison.

importance for the understanding of early solar system evo-lution. The knowledge of the actual distribution of traceelements in chondritic matter is of great interest also forthe understanding of apparent enrichments of volatile traceelements in interplanetary dust (see below).Imaging TOF-SIMS at high lateral resolution was there-

fore used to analyze the element distribution within theOrgueil CI meteorite (Stephan et al., 1997a), which isgenerally used as the cosmic element abundance standard(Anders and Grevesse, 1989). The TOF-SIMS investigationrevealed a rather homogeneous distribution for the majorelements magnesium, aluminum, silicon, iron, and nickel.For sodium and potassium, enrichments throughout the en-tire section were observed in correlation with .uorine andchlorine. This might be explained by surface contamination,a general problem in TOF-SIMS studies (see above). Otherelements are often concentrated in small, up to 100 �msized regions. Calcium occurs in carbonates, as observedbefore in CI chondrites (EndreX and BischoK, 1996).Chromium, typically together with iron, can be found inchromites. Iron and nickel enrichments are often correlatedwith sulfur, indicative for the presence of sul7des. Lithium,titanium, manganese, copper, bromine, and iodine showedenrichments in small regions, only a few micrometers insize. Further mineralogical studies are needed to elucidatethe host phases of the enriched elements.It should be mentioned that although Orgueil is the ele-

ment standard in cosmochemistry, this meteorite is not suitedas an element standard for TOF-SIMS, due to lack of homo-geneity. Elements that are only present in distinctive mineral

phases diKer often drastically, up to orders of magnitude, intheir secondary ion yield from elements in the bulk matrix.

4.3.2. Chassigny and NakhlaA general interdisciplinary study on Martian meteorites

has just been started recently. A central role in this investi-gation plays TOF-SIMS (Stephan et al., 1999b). Besides theanalysis of ALH 84001, two further Martian meteorites wereinvestigated so far, namely Chassigny and Nakhla. In com-bination with SEM, electron microprobe, and TEM studies,so-called symplectic exsolutions in olivines were analyzedthat set constraints on the formation of these Martian rocks,the oxidizing conditions and the cooling history (Greshakeet al., 1998). Fig. 31 shows secondary ion images as wellas an SEM-BSE image from one of these objects found inNakhla. A calcium-rich pyroxene (augite) can be seen asdark areas in the SEM image. Here calcium showed the high-est secondary ion intensities. The bright phase in the SEMimage is dominated by chromium-rich magnetite, prevailingin the Cr+- and Fe+-images with a slight enrichment in alu-minum. Magnesium, silicon, and iron secondary ion imagesshow the highest intensities outside the exsolution lamella inthe olivine. Therefore, these images have been normalizedto the most intense pixels inside the interesting region andnot to the most intense pixels of the entire images. The factthat iron yields higher secondary ion intensities in olivinethan in magnetite, although the iron concentration is higherin the latter, can be attributed to matrix eKects.

5. Interplanetary dust

In 1683, when the Italian–French astronomer G.D. Cassini7rst described the zodiacal light, he also gave the rightexplanation for this phenomenon (in Cassini, 1730): It iscaused by sunlight scattered from dust particles concentratedin the ecliptic plane. Since 1981, interplanetary dust parti-cles (IDPs) are collected routinely in the stratosphere usinghigh-.ying aircraft (Warren and Zolensky, 1994). They rep-resent, besides meteorites and lunar rocks, the third class ofextraterrestrial material available for laboratory investiga-tion so far (Brownlee et al., 1977). Due to their small size—typical IDPs have a size of ∼10 �m—these samples requireanalytical techniques with high lateral resolution and highsensitivity.IDPs are of high interest in planetary research since

they potentially allow to investigate the most pristine solarsystem bodies, asteroids and comets (Brownlee, 1994; Der-mott et al., 1996). In particular for cometary material, theyrepresent the only available source so far for an earthboundinvestigation. But also asteroidal IDPs should signi7cantlydiKer from meteoritic material. While meteorites are gen-erally accepted to derive primarily from the inner asteroidbelt (Wetherill and Chapman, 1988), dust from all over thesolar system, including the outer belt, should spiral towardsthe sun due to Poynting-Robertson-drag (Robertson, 1937;


Fig. 31. Secondary ion images from a symplectic exsolution from the Nakhla Martian meteorite (20 × 20 �m2, 1600 shots=pixel, 256 × 256 pixels).Lower intensities towards the very right and bottom of the images are artifacts from ion beam shifts during analysis.

Klacka, 1994) and therefore contribute to the near-Earthdust population. Some of this dust eventually enters theEarth’s atmosphere and can be collected in the stratosphere.IDPs still preserve information on their precursors, althoughseveral selection and alteration processes occur during theirlifetime. For a general overview of IDPs refer to, e.g.,Zolensky et al. (1994), Brownlee (1996), Rietmeijer (1998),or Jessberger et al. (2001).Major goals in IDP research are still to unambiguously

diKerentiate cometary from asteroidal IDPs and to distin-guish between features caused by atmospheric entry andcollection in the stratosphere, and indigenous properties.Nevertheless, the “chondritic porous” subset of IDPs, an-hydrous particles with highly porous microstructures, highcarbon abundances, and often high atmospheric entry ve-locities are most promising candidates for cometary dust.They consist mainly of Mg-rich, Fe-poor silicates, remark-

ably similar to those observed in dust around stars and fromcomets Halley and Hale-Bopp (Jessberger and Kissel, 1991;Bradley et al., 1999a). Comets are probably the most prim-itive bodies in the solar system and therefore predestinedas deep-freeze storehouses for presolar material. The searchfor interstellar matter in cometary IDPs has been one inspir-ing motivation for the analysis of these tiny grains (Bradleyand Ireland, 1996). Recent observations link them—or atleast parts of them known as GEMS (glass with embed-ded metal and sul7des)—to interstellar amorphous silicategrains (Bradley, 1994b; Bradley et al., 1999b). Togetherwith presolar grains extracted from primitive meteorites,constituents of cometary IDPs are the most primitive mate-rials available for laboratory investigations and maybe thebest analogue to present interstellar dust.Due to its capabilities of analyzing small particles

(Zehnpfenning et al., 1994; Rost et al., 1998a), TOF-SIMS


was expected to be an appropriate technique for the inves-tigation of IDPs.

5.1. Volatile elements

Many stratospheric IDPs are more rich in volatile ele-ments than CI chondrites (Arndt et al., 1996a), while ma-jor elements usually show chondritic abundances (Schrammet al., 1989). Attempts to explain the enrichments rangefrom postulating a new type of volatile-rich chondrite-likematter (Flynn and Sutton, 1992; Flynn et al., 1996a) toinvoking atmospheric contamination processes (Jessbergeret al., 1992). Before far reaching conclusions on nebularprocesses can be drawn, stratospheric processes, contamina-tion during capture and handling, artifacts from various se-lection eKects or from analytical techniques and even fromdata treatment have to be excluded (Stephan et al., 1997b).Due to their complex history, IDPs in the laboratory have

already experienced several potential sources of contamina-tion. For small particles, contamination in general is a moresevere problem, because of the high surface-to-volume ra-tio, than for the comparably large “normal” meteorites. Highporosity of IDPs also enhances the surface receptive to con-tamination (Stephan et al., 1992a, 1994b). Furthermore, el-ements with very low abundances in chondritic matter aremost susceptible to contamination because then even a smallabsolute amount added can yield high enrichments.Bromine shows the highest enrichments in IDPs, with en-

richment factors of up to 104 × CI (Arndt et al., 1996a).Since bromine, on the other hand, has only a CI-abundanceof 3.56 ppm (Anders and Grevesse, 1989) and halogens arewidespread in the stratosphere (Cicerone, 1981), contamina-tion processes have to be considered as being responsible forhigh bromine concentrations in stratospheric particles. Firstdirect experimental evidence for such a contamination wereBr-salt nano-crystals attached to IDP W7029E5 (Rietmei-jer, 1993) and a halogen-rich exterior rim of IDP L2006G1found with TOF-SIMS (Stephan et al., 1994c). The distri-butions of secondary halogen ions from a section of thisIDP reveal an outer ring structure for .uorine, chlorine, andalso bromine, though the latter image is disturbed by ratherhigh background (Fig. 32). Assuming a spherical particle,originally 20 �m in diameter, and a continuous 1 �m thicksurface layer surrounding it, this layer represents ∼25 vol%of the entire particle. Although only a rough estimate, thesenumbers illustrate the in.uence that surface contaminationmight have on bulk chemistry of small particles.Other clear evidence for contamination as a major clue to

halogen enrichments of stratospheric IDPs was found withtest particles exposed to the stratosphere on a dust collectorby Arndt et al. (1996b). Besides this, weakly bound brominewas observed in large IDPs (Flynn et al., 1996b).To test more directly for surface contamination with

TOF-SIMS, the original surfaces of stratospheric particleswere analyzed and compared with sections from the verysame particles (Rost et al., 1998b, 1999a, b). Here halogens

were observed on several stratospheric particle surfacesalthough some analyses were obstructed by non-removablesilicone oil residues. —Stratospheric dust collectors arecoated with silicone oil, which is also used during particlehandling. —A surface correlation of halogens on some par-ticles was unequivocally proven by extensive sputtering,reducing and 7nally removing the halogens at least on cer-tain particles (Rost et al., 1999a). In addition, several dustparticles in sections showed again outer rings of halogencontaminants (Rost et al., 1998b, 1999b).Another source of contamination, not in the stratosphere

but from laboratory handling, was observed in TOF-SIMSstudies in form of surface correlated beryllium. This re-sults from the SEM-EDS analyses, which in case of theseparticles were performed on beryllium substrates (Rostet al., 1999a). After such analyses, some particles tend tostick to the substrate and carry beryllium when they wereremoved.For other elements, 7nally, no unambiguous evidence

for or against contamination is available so far. For mostvolatiles an unequivocal enrichment cannot be claimed sincethe data set available in the literature typically is too incom-plete for a clear statement (Arndt et al., 1996a) and limita-tions of the applied analytical and mathematical techniquescomplicate the interpretation (Stephan et al., 1997b). Traceelement distributions are largely variable and re.ect chemi-cal inhomogeneity on a micron scale (Bohsung et al., 1993;Stephan et al., 1993a, 1994b; Arndt et al., 1996a). Never-theless, since most IDPs chemically are remarkably similarto CI chondrites, they represent an adequate sample of prim-itive solar system material.

5.2. Mineral identi5cation

TOF-SIMS has enabled the discovery of several uniquemineral phases by secondary ion imaging, although thecapabilities of quanti7cation are limited for this technique.Calcium phosphate, probably apatite, was found for the7rst time to occur in an IDP (L2006E10, Fig. 33 top)by TOF-SIMS investigation (Stephan et al., 1994c) andhas meanwhile been reported in another IDP (Rost et al.,1999b). For U2071B6 (Fig. 33 bottom), a manganese sul-7de (most likely alabandite) was observed (Rost et al.,1998b). While apatite is widespread in many diKerent typesof meteorites, alabandite usually is only found in mete-orites that formed under highly reducing conditions, e.g.,enstatite chondrites or aubrites (Brearley and Jones, 1998;Mittlefehldt et al., 1998). However, no further connectionbetween these meteorite classes and this speci7c IDP can bedrawn from TEM studies so far (W. KlQock, pers. comm.).Although an unambiguous identi7cation of the actual min-

eral phase is not possible through TOF-SIMS analysis alone,indicative elemental correlations can help to recognize in-teresting regions within a sample, which can be analyzedlater by appropriate techniques like TEM for an unequivocalmineral assignment.


Fig. 32. Secondary ion images of positive as well as negative secondary ions from IDP L2006G1. The investigated area (30 × 30 �m2; 128 × 128pixels) was bombarded with Ga+ primary ions (1000 shots=pixel and 800 shots=pixel for positive and negative secondary ions, respectively). Althoughthe background for Br−-ions (here the added signals from 79Br− and 81Br−) is relatively high, a correlation with the halogens .uorine and chlorine isobvious.

5.3. The collector project

As long as no interplanetary dust particle has been cap-tured gently in outer space (cf. Brownlee et al., 1996, 1997),retrieved to Earth, and recovered successfully from the cap-ture medium (probably aerogel; Tsou, 1996), IDPs collectedin the stratosphere are the 7rst-choice samples of interplan-etary dust. Therefore, a comprehensive consortium studywas initiated to investigate all particles from one collectionsurface (U2071) with as many diKerent techniques as pos-sible (Stephan et al., 1994d). This suite of analytical meth-ods (cf. Table 3) includes TOF-SIMS for surface studies(Stephan et al., 1995a; Rost et al., 1996, 1999a, b) as well asfor the investigation of interior element distributions (Rostet al., 1998b, 1999b) in sections within the same particles.Not only the investigation of contamination processes asmentioned above is the aim of this ongoing study, but toinvestigate the mineralogical and chemical properties of anunbiased sample of stratospheric dust particles as thoroughlyas possible. By studying all particles from one collector,

one avoids human selection eKects prevailing in most stud-ies. Only “chondritic looking” (black and .uKy) particlesare picked from the collector .ags in most cases (J.L. War-ren, pers. comm.) and only particles pre-classi7ed as “cos-mic” due to their chondritic major element composition arefurther investigated. Therefore, one major objective of thisstudy is to track down non-chondritic interplanetary dustparticles.Table 3 gives an overview of the present status of this con-

sortium study. All 326 particles with diameters¿ 10 �m oncollection surface U2071 were investigated under an opticalmicroscope, and 90 of them were selected for the 7rst roundof investigations. After SEM imaging and EDS analysis ofmajor element concentrations including carbon (Thomaset al., 1993, 1994b), these particles were pre-classi7edas cosmic particles (C), terrestrial contamination, natu-ral (TCN) and arti7cial (TCA), aluminum oxide spheres(AOS; solid fuel rocket exhaust), and non-classi7ed parti-cles. The classi7cation scheme was used according to theCosmic Dust Catalogs (Warren and Zolensky, 1994).


Fig. 33. Secondary ion images from IDPs L2006E10 (7rst row; Stephan et al., 1994c) and U2071B6 (second row; Rost et al., 1998b) show the 7rstreported observation of calcium phosphate (presumably apatite) and manganese sul7de (most likely alabandite) in IDPs. Fields of view are 25× 25 �m2

(L2006E10) and 30×30 �m2 (U2071B6). Both samples were scanned with 128×128 pixels, and 1000 shots=pixel (L2006E10) and 2560=2400 shots=pixel(U2071B6, positive=negative secondary ions).

Table 3Analytical sequence and present status of the consortium study on particles from stratospheric collector U2071a

Analytical technique Particles from collector U2071

Optical microscope 326 particles ¿10 �mSEM (images) 90 particlesSEM-EDS 90 particlesTOF-SIMS (surface) B3, B6, B7a, C3, D1, D4, D8, E6, E8, F3, G8, H1a, H9, I9, J9a, L1, L2PIXE (bulk) B3, B6, C3, D1, D8, E6, F3, G8, H1a, H9, I9, L2STIM (bulk) B3, B6, C3, D1, D8, E6, F3, G8, H1a, H9, I9, L2TEM (section) B6, C3, D1, D4, E6, E8, F3, G8, H9, I9, L1TFO-SIMS (section) B6, C3, D1, D4, E6, E8, H9, I9, L1PIXE (section) in preparation

aUnderlined names represent particles pre-classi7ed as cosmic (type C) IDPs. Other types are terrestrial contamination, natural (TCN: E6) and arti7cial(TCA: I9), aluminum oxide spheres (AOS: B3, D8, G8, L2), as well as non-classi7ed particles (type ?: D1). The TEM studies are presently in progress.

Surface analysis with TOF-SIMS followed to study con-tamination eKects (see above). Subsequently, the particleswere investigated with PIXE (proton induced X-ray emis-sion) analysis including STIM (scanning transmission ionmicroprobe). PIXE allows trace element studies with highsensitivity down to ppm levels for elements with atomicnumbers ¿ 11 (Traxel et al., 1995; Bohsung et al., 1995).STIM yields particle densities and masses (Maetz et al.,1994; Arndt et al., 1996c, 1997). This 7rst PIXE investiga-tion was done at limited proton current and dose to avoiddestruction of hydrous minerals (Maetz et al., 1996). Theinformation depth in PIXE is of the order of 30 �m, thetypical dimension of the dust particles. The information inSEM-EDS comes from the upper 1–2 �m, and TOF-SIMS

only investigates the surface. Consequently, even for ele-ments that are detectable with all three analytical techniques,the results are not straightforward to compare (Stephanet al., 1995b).After these analyses, the particles are imbedded in epoxy

and cut with an ultramicrotome. The microtome sectionswere prepared for TEM studies and the particle residuesin the epoxy were analyzed again with TOF-SIMS. Thisallows a direct comparison of mineralogical results (TEM)and element distribution images (TOF-SIMS) of adjacentsections.It is further planned to produce sections several microm-

eters thick from the remaining epoxy stubs for further PIXEstudies. As the last analytical step for these particles, PIXE


can now be performed at high proton current, dose, andtherefore increased sensitivity. Since the 7nal stage in the se-quence of analytical steps is now reached, precautions con-cerning mineral destruction have not to be considered anylonger.The general principle for this sequence of analytical tech-

niques is to go from non-destructive via less destructive tothe most destructive techniques. The analysis of contamina-tion eKects in the stratosphere has to be made as early aspossible to minimize the risk of additional contamination inthe laboratory. It is expected that the great variety of ana-lytical techniques applied to individual IDPs in this ongoingstudy will 7nally be instrumental in a comprehensive under-standing of the origin and history of interplanetary dust.

5.4. Particle impact residues from space probes

Each surface exposed in space is subject to dust par-ticle bombardment. In low Earth orbit, this dust consistsof interplanetary dust as well as man-made space debris.To discern between these populations and to reveal theirrespective frequencies were major objectives of the capturecell experiment A0187-2 on the LDEF (long duration expo-sure facility) satellite (Amari et al., 1991, 1993; Bernhardet al., 1993a, b; HQorz et al., 1995).At typical relative velocities of several kilometers per sec-

ond, an impinging dust particle evaporates almost entirely.Nevertheless, some material may be deposited in or aroundthe impact crater. These impact residues provide informa-tion on the chemical composition of the precursor particles,although tremendous elemental fractionation processes oc-curring during particle impact have to be considered (Langeet al., 1986). The typical amount of sample material de-posited by high-velocity impact is con7ned to 1–5 atomiclayers. TOF-SIMS with its high surface sensitivity hereseems to be ideally suited for the analysis of these impactresidues. Due to parallel detection of all secondary ions withone polarity, TOF-SIMS allows to investigate numerous ele-ments even within these extreme thin layers. This might helpto elucidate the origin of the respective impacting particle.TOF-SIMS studies were performed on two impact

residues from the above-mentioned LDEF capture cell ex-periment A0187-2 (Stephan et al., 1992b) as well as onseveral impact craters from the Hubble space telescope so-lar cell array (unpublished data) to reveal the origin of theimpacting particles.Prior to the TOF-SIMS analysis of the LDEF impacts,

both samples were investigated with DF-SIMS (Amariet al., 1991). From these analyses, the two projectiles thathad produced the impact features were classi7ed as proba-bly chondritic, based on relative abundances of magnesium,aluminum, calcium, titanium, and iron. However, an unam-biguous identi7cation of the true nature of the impactingparticle is hampered by the huge elemental fractionation.With TOF-SIMS, high secondary ion intensities for cop-

per were found in both LDEF residues (Stephan et al.,

1992b). Although a proper quanti7cation is not possible,copper was identi7ed as the dominating element. These7ndings seem to exclude an extraterrestrial origin of the re-spective particles, since high abundances of elements withextreme low cosmic abundances like copper are indicativefor space debris. Copper has a mean chondritic abundanceof only 126 ppm (Anders and Grevesse, 1989) and shouldnot be present as major element in interplanetary dust. Theseresults clearly demonstrate the advantages of TOF-SIMS es-pecially when sample material is limited. DF-SIMS only al-lowed to analyze a limited number of elements within thethin layers. This is often not su4cient for a proper classi7ca-tion of impact residues. The parallel detection in TOF-SIMSled to the discovery of the copper enrichment. Beforehand,such enrichments were not expected and therefore copperwas not selected for DF-SIMS analysis.

5.5. Further TOF-SIMS studies

As mentioned before, to decipher the origin of an inter-planetary dust particle, its connection to asteroids or cometsis a major aim of IDP research. The TOF-SIMS study of ananhydrous particle supported the idea of a relation betweenchondritic porous IDPs and comets (Stephan et al., 1993b;Jessberger et al., 2001). The element abundances of thisparticle, L2006B21, are similar to the element abundancesobserved in comet Halley’s dust (Jessberger et al., 1988).Besides interplanetary dust collected in the stratosphere

and impact residues from surfaces exposed in low Earth or-bit, also micrometeorites extracted from Antarctic ice havebeen investigated by TOF-SIMS (Stephan et al., 1995b).So-called COPS phases, iron oxide rich in carbon, phos-phorous, and sulfur (Engrand et al., 1993), were found onthe surface and in voids of the micrometeorites. The lateraldistribution supported the idea that COPS phases grew onthe surface of micrometeorites, possibly by modi7cation ofindigenous material, and by incorporation of material dur-ing atmospheric passage and Antarctic residence (Mauretteet al., 1994; Kurat et al., 1994; Stephan et al., 1995b; Jess-berger et al., 2001).First attempts to study organic molecules or fullerenes in

IDPs using TOF-SIMS have been made in the past (Radicatidi Brozolo and Fleming, 1992; Stephan et al., 1996), butno systematic investigation has been performed so far. Herethe general contamination problem complicates the analy-ses. Residual silicone oil as well as carbon-containing em-bedding materials obstruct a proper examination.Oxygen isotopic imaging with TOF-SIMS has been per-

formed for four IDPs in order to search for sub-micrometer-sized stardust (Messenger, 1999). The comparison withDF-SIMS results of the same particles (Messenger, 1998)showed no signi7cant diKerences in 18O=16O isotopic ra-tios. Nevertheless, within the large statistical error limits(30–90 per mill in �18O), the values obtained by TOF-SIMSare also not discernable from solar ratios.


6. Presolar grains

The year 1987 became a major turning point in cosmo-chemistry through the 7rst discovery and isolation of indi-vidual pristine presolar dust grains (for reviews on presolardust cf. Anders and Zinner, 1993; Ott, 1993; Zinner, 1998).This discovery was the reward for more than 20 years of ef-forts to locate the carriers of several highly anomalous noblegas components in meteorites (Tang et al., 1988; Tang andAnders, 1988a, b). Major types of presolar dust found so farare diamond (Lewis et al., 1987), silicon carbide (Berna-towicz et al., 1987; Zinner et al., 1987), graphite (Amariet al., 1990), corundum (Huss et al., 1992; Hutcheon et al.,1994; Nittler et al., 1994), and silicon nitride (Nittler et al.,1995). Titanium carbide was found inside graphite and sil-icon carbide grains (Bernatowicz et al., 1991, 1992). Alu-minum within silicon carbide and graphite grains showed acorrelation with nitrogen, an indication for the presence ofaluminum nitride (Zinner et al., 1991a; Virag et al., 1992;Stephan et al., 1997c).

6.1. Isotope measurements

Conventional DF-SIMS has played the major role in thestudy of presolar grains, because this technique allows tomeasure isotopic compositions of individual grains downto ∼1 �m in size (Zinner, 1998). These isotopic composi-tions, often diKerent from mean solar system values by sev-eral orders of magnitude, clearly prove the extrasolar originof these grains. Due to the tremendous magnitude of theanomalies, the required precision for this kind of analysis israther low.TOF-SIMS seemed to be a promising technique for the

analysis of individual presolar grains. Especially the highlateral resolution together with the simultaneous detectionof the entire mass range were expected to allow new insightsinto the properties of individual presolar grains. Isotopicstudies should be possible but were not the major purposeof these investigations.However, in the 7rst TOF-SIMS study of presolar grains

(Stephan and Jessberger, 1996; Stephan et al., 1997c) it wastried to reproduce already known carbon isotopic ratios forsilicon carbide (SiC) grains from the Murchison meteorite,previously investigated by DF-SIMS (Hoppe et al., 1994).As discussed before, the major problem for carbon iso-tope analysis at high lateral resolution (necessary for smallpresolar grains) is the separation of 13C from 12C1H. Sincethis is typically not possible satisfactorily with TOF-SIMS,mathematical peak deconvolution techniques have to be ap-plied. In principle, similar to the technique described above(Fig. 23), two peak 7tting routines are possible (Fig. 34):The 12C-peak can be used (1) as a prototype for the 13C-peakor (2) to 7t the 12C1H-peak. Although the 7rst approach(Fig. 34 top) seems to be more promising—both carbon iso-topes should indeed have the same peak shapes—the secondmethod (Fig. 34 bottom) is often more appropriate. The lat-

Fig. 34. Negative secondary ion mass spectrum at 13 amu obtained froma presolar SiC grain (KJG2-1412). The peaks can be deconvoluted withtwo diKerent methods. In the upper spectrum, the 12C-peak (dotted line,top x- and right y-axis) was 7t to the left edge of the spectrum at 13 amu(shaded area, bottom x- and left y-axis). Here, the scaling factor 9:9±0:9directly gives the 12C=13C isotopic ratio. The diKerence between the twospectra (solid line) represents the 12C1H-peak. In the lower spectrum, the12C-peak is compared with the right edge of the shaded area resultingfrom the 12C1H-peak. Here the 13C-peak is the diKerence of both spectra(solid line). The 12C=13C-ratio is 10:9 ± 2:1. Both results within theiruncertainties agree with the DF-SIMS result, 12C=13C = 9:795.

ter allows using more actual channels of the spectrum forthe 7tting routine and is therefore less susceptible to statis-tical variations.TOF-SIMS allowed a proper analysis of the carbon iso-

topic ratio for 13 diKerent SiC grains so far (Fig. 35). Inmost cases, these results agree well with those obtained byDF-SIMS (Hoppe et al., 1994). Nevertheless, the limitedmass resolution only allows the analysis of grains signi7-cantly enriched in 13C compared to chondritic isotope ratios.A signi7cant increase in mass resolution is impeded by therequired high lateral resolution.The analysis of nitrogen isotopic ratios was unsatisfactory

due to the limitation in mass resolution. Since 12C15N− can-not be resolved from 13C14N−, the relatively high statisticalerror in the 13C analysis prevents an appropriate determina-tion of 15N.Silicon and magnesium isotopic measurements were re-

cently performed successfully with TOF-SIMS on presolarSiC grains from Murchison and Semarkona meteorites by


Fig. 35. Comparison of 12C=13C isotopic ratios from TOF-SIMS withthose from DF-SIMS for 13 diKerent SiC grains from the Murchison KJGfraction (Amari et al., 1994). All analyses were performed on grains with12C=13C-ratios below the chondritic (CI) value.

Fahey and Messenger (1999). The burst mode (see above)was used in the investigation to increase the secondary iondata rates. From 26Mg secondary ion imaging, it was pos-sible to identify several so-called X-grains (Zinner et al.,1991b; Amari et al., 1992; Hoppe et al., 1996) among siliconcarbides from Murchison (Fahey and Messenger, 1999).

6.2. Element mapping with TOF-SIMS

However, the imaging capabilities of TOF-SIMS aremuch more impressive than its abilities in isotope analysis.Interstellar grains within interstellar grains (Bernatowiczet al., 1991) have been identi7ed with the TEM as titaniumcarbide (TiC) grains inside graphite and SiC grains (Berna-towicz et al., 1991, 1992). Although a general correlationof aluminum and nitrogen was found for SiC grains (Zinneret al., 1991a; Virag et al., 1992), no separate Al-bearingphase like aluminum nitride (AlN) has been detected withthe TEM (Bernatowicz et al., 1992). Therefore, imaging

Fig. 36. TOF-SIMS images of SiC grain KJG2-0411 show a rather heterogeneous distribution of aluminum, calcium, and titanium. Field of view is12× 12 �m2. The sample was scanned with 128× 128 pixels and 800 shots=pixel.

TOF-SIMS has been used to search for possible subunitswithin SiC grains (Stephan et al., 1997c).TOF-SIMS studies of presolar SiC grain KJG2-0411 re-

vealed an inhomogeneous distribution for several elements(Stephan et al., 1997c). Secondary ion images of aluminum,silicon, calcium, and titanium are shown in Fig. 36. Sili-con is distributed homogeneously, whereas aluminum is en-riched in a distinct area, which appeared after sputtering.Consecutive analyses of the grain showed that Al-rich ma-terial is concentrated in the center of this presolar grain.A correlation between Al=Si- and CN=Si-ratios (Fig. 37)indicates again the presence of aluminum nitride (Stephanet al., 1997c). Whether AlN is present as distinct phase or insolid solution with the SiC as suggested before (Bernatow-icz et al., 1992), cannot be clari7ed with TOF-SIMS. Theincrease of the Al=Si-ratio towards the center of the grainmight be due to destruction of AlN on or near the surfaceduring chemical processing when the grains were extractedfrom the meteorite (Stephan et al., 1997c). Although sig-nal intensities are rather low, calcium showed also a corre-lation with aluminum and nitrogen (Fig. 36). Titanium, onthe other hand, is enriched in a diKerent region of the samegrain, maybe indicative for the presence of TiC.In a type X presolar SiC grain, isotopic heterogeneities

were found recently with TOF-SIMS (Henkel et al., 2000).These analyses clearly demonstrate the potential of

TOF-SIMS for the investigation of presolar grains. Highlateral resolution in combination with parallel detection ofall secondary ion species is ideally suited to localize andidentify interesting grains or subgrains, even if the ability tomeasure isotopes is limited. Although no presolar grain hasbeen found in situ with TOF-SIMS so far, this techniqueshould be in principle suited for this kind of search. In situlocalization would also allow investigating the real linkbetween SiC and AlN, because chemical processing duringisolation of the grains may selectively destroy AlN. Themajor advantage of in situ analysis is, however, that infor-mation about the petrographic environment of the presolargrains could be obtained. This would help to clarify possible


Fig. 37. Al=Si and CN−=Si− show a linear correlation that indicates AlNas the dominating aluminum-bearing phase in KJG2-0411. Data point (A)corresponds to the Si-rich, Al-poor area in the lower left of the particlein Fig. 36. (B) is the Al-rich area to the right. (C) is the titanium hotspot. (D) represents an earlier measurement of the entire particle beforesputtering.

connections of presolar grains with other constituents of therespective meteorites, e.g., CAIs, chondrules, or matrix.

7. TOF-SIMS in comparison

To put TOF-SIMS into an analytical context, a com-parison with other micro-analytical techniques used incosmochemistry—DF-SIMS, TOF-SNMS, laser desorp-tion TOF-SNMS, EMPA (electron microprobe analysis),PIXE, SXRF (synchrotron X-ray .uorescence analysis),and RBS (Rutherford backscattering)—is given in Table 4.A more comprehensive description of conventional SIMStechniques, various instruments, and their applications incosmochemistry can be found in Ireland (1995). Micro-probe techniques in Earth sciences are explained in detailby Potts et al. (1995). The numbers for spatial resolution,information depth, and detection limits provided in Table 4are typical values for instruments used in this 7eld, whichmay not always re.ect the most recent developments inthe respective techniques. However, these numbers clearlydemonstrate their major diKerences, which often complicatea proper comparison of the various results. The diKerentmethods also vary in the type of information they pro-vide. For the analysis of large organic compounds, only“soft” ionization methods are suitable and the simultane-ous detection of all ion species like in time-of-.ight massspectrometry is required.

8. TOF-SIMS in space

For analysis in space, rugged and simpli7ed techniquesare required. Limitations mainly arise from size, mass, en-

ergy, and data transmission constraints. Compared to othertypes of mass spectrometers, the time-of-.ight principleshould be advantageous. Since no magnet or other heavycomponents are needed, the mass of the instrument can bekept comparatively low. Power consumption can also below for this kind of mass spectrometer. Time-of-.ight massspectrometers were therefore used and are still used aboardseveral space probes. Three missions to comet 1p=Halleyapplied TOF mass spectrometry after impact ionizationof dust particles with high relative velocities (Kissel andKrueger, 1987a; Hornung and Kissel, 1994). The PIA (par-ticle impact analyzer) instrument on Giotto and the PUMA(pyle-udarnyi mass analizator, pyle-udarnyi massanalizator, Russian for dust impact mass analyzer) instru-ments on both Vega missions to comet Halley (Kissel et al.,1986a, b) delivered a wealth of information on the elemen-tal and isotopic composition of Halley’s dust (Jessberger etal., 1988, 1989; Jessberger and Kissel, 1991). A modernvariant of these instruments, CIDA (cometary and interstel-lar dust analyzer), was launched in February 1999 aboardthe Stardust space vehicle to comet 81p=Wild 2 (Brownleeet al., 1996, 1997).The 7rst SIMS instrument envisioned for space .ight was

an astronaut operated DF-SIMS instrument for on-site anal-ysis of the lunar surface in the framework of the Apollomissions (RQudenauer, 1994), which however never .ew.The DION system on both Soviet space probes to the Mar-tian satellite Phobos was the “largest” SIMS instrument everbuilt. Secondary ions sputtered from the surface were tobe analyzed in a quadrupole mass spectrometer at a cruis-ing altitude of 50 m above Phobos (Balebanov et al., 1986;Hamelin et al., 1990; RQudenauer, 1994). A time-of-.ightinstrument was also on board both Phobos missions, theLIMA-D experiment (Sagdeev et al., 1986; Pellinen et al.,1990). The plan was to use the cruising altitude of 50 m as.ight path for ions generated by high-energy laser pulses.Unfortunately, both spacecraft were lost prior to arrival atPhobos.A new era in TOF-SIMS will hopefully commence in

2011, when the 7rst instrument of this type is planned toreach comet 46p=Wirtanen on ESA’s Rosetta spacecraft. Af-ter reaching its target, it is planned to collect cometary dustintact at low relative speed. The dust particles then willbe analyzed in a TOF-SIMS instrument named CoSIMA(cometary secondary ion mass analyzer; Kissel et al., 1988).CoMA (cometary matter analyzer), an earlier version ofthis instrument, was also designed by J. Kissel to .y onNASA’s cometary rendezvous and asteroid .yby mission(CRAF) that was cancelled in 1992 (Zscheeg et al., 1992;Beck, 1994; Beck and Kissel, 1994). With the CoSIMA in-strument, a lateral resolution of 5 �m with 100 ns primaryion pulse width can be obtained. In a second mode, with3 ns pulse width, 20 �m spatial resolution is possible. Themass resolution, up to m=Vm = 2000, is su4cient for theseparation of most hydrocarbon interferences from elemen-tal peaks. The CoSIMA results will certainly not reach the

894T.Stephan

/Planetary

andSpace

Science

49(2001)

859–906

Table 4Comparison of micro-analytical techniques used in cosmochemistry

Technique Primary particles Secondary particles Information obtained on Spatial resolution Information depth Destructiveness Detection limits

Elements Isotopes Molecules Organics

TOF-SIMS Ions Ions Z ¿ 1 Yes Yes Yes 0:2–10 �m ¡ 1 nma Yesa ¿ 10 ppbDF-SIMS Ions Ions Z ¿ 1 Yes Yes No 0:05b–1 �m ¿ 100 nm Yes ¿ 1 ppbTOF-SNMSc Ions Neutrals Z ¿ 1d Yesd Yesd Yesd 0:2–10 �m ¡ 1 nma Yesa ¿ 10 ppbLaser desorption Photons Neutrals Z ¿ 1d Yesd Yesd Yesd 1–40 �mf ¿ 1 nm Yes ¿ 1 ppbTOF-SNMSe

EMPA Electrons X-rays Z ¿ 4 (Be) No No No 2 �m 1 �m Nog 200–1000 ppmPIXE Protons X-rays Z ¿ 12 (Mg) No No No 1 �m 30 �m Nog 0:1–100 ppmSXRF X-raysh X-rays Z ¿ 16 (S) No No No 1 �m 50 �m No 0:1–5 ppmRBS Protonsi Protonsi Z ¿ 1 No No No 1 �m 15 �m Nog 0:1–1%aOnly few atomic monolayers.bWith the Cameca Nanosims 50.cResonant or non-resonant laser ionization of secondary neutrals sputtered by primary ions.dDepending on photon energy and .ux density.eResonant or non-resonant laser ionization of neutrals after laser desorption.f 1 �m for the CHARISMA instrument (Nicolussi et al., 1998a), 40 �m for �L2MS (Kovalenko et al., 1992).gStructural modi7cations possible.hSynchrotron radiation.iEnergy measurement of backscattered protons or heavier particles.


same quality as laboratory TOF-SIMS analyses, but thiscannot be expected from an instrument with mass less than20 kg and primary power between 8.2 and 19:5 W (techni-cal data J. Kissel, pers. comm.). However, it is expected,that profound information on the composition of cometarymatter will be obtained with CoSIMA. The experience withTOF-SIMS analysis of extraterrestrial matter gathered hereon Earth and reported in the present study will de7nitelyhelp to interpret the results that will be acquired in space.Especially the surface analysis of IDPs deals with problemsas expected for the space experiment, e.g., rough surfaces.Since complex mixtures of organic material and silicates areexpected for cometary dust (Kissel and Krueger, 1987b),laboratory TOF-SIMS investigation of organic compoundsin analogous material, e.g., chondrites and IDPs is essential(Heiss et al., 1998).

9. Summary

Time-of-.ight secondary ion mass spectrometry has beensuccessfully introduced into cosmochemistry during the pastdecade and has demonstrated its potential. Its major advan-tage results from the simultaneous detection of all secondaryions with one polarity. This in combination with high lat-eral resolution quali7es TOF-SIMS for the analysis of smallsamples, becoming more and more important in the inves-tigation of extraterrestrial matter.Although mass resolution and quanti7cation capabili-

ties in TOF-SIMS are often inferior to conventional ionmicroprobes, the advantages of TOF-SIMS open speci7c7elds of application, not or less accessible with DF-SIMS.Consequently, these two SIMS techniques are rather com-plementary than in competition with each other. Parallel de-tection of all ions expressly does not require a pre-selectionof masses to be analyzed. Therefore, TOF-SIMS is predes-tined to make unexpected discoveries. But then, for quan-ti7cation, subsequent investigations with other techniquesincluding DF-SIMS may be required.Further technical developments in TOF-SIMS will help

to overcome some limitations: A new generation of liquidmetal ion guns allows to increase the primary ion current—and the secondary ion intensity—even at high lateral resolu-tion. Decreasing the primary beam size below 100 nm willextend the application to very small entities. Especially laserpost-ionization is expected to improve quanti7cation. Nev-ertheless, even today TOF-SIMS is a powerful tool for theinvestigation of natural terrestrial and extraterrestrial sam-ples. Applications near the borderline between life sciencesand cosmochemistry—studying the origin of life in our so-lar system—might be one promising 7eld for TOF-SIMS.First results, although negative for ALH 84001, were high-lighting the potentials of TOF-SIMS.Particle analysis is already one of the strong points of

TOF-SIMS. A maximum of information on single grains

without their complete destruction is required. Here, a wealthof new discoveries from TOF-SIMS studies can be expectedin the future.

Acknowledgements

The author is indebted to many people who made thiswork possible. A special thank goes to E.K. Jessberger forhis extensive support. Superb working conditions were alsoprovided by A. Benninghoven, D. StQo\er, and J. Kissel.Many colleagues and friends inMQunster and Heidelberg sup-ported me throughout the past decade and contributed withtheir work to this study, especially D. Rost, the compan-ion for more than 6 years of TOF-SIMS research, and C.H.Heiss, T. Henkel, and S. Schirmeyer. Expert sample prepa-ration by U. Heitmann is gratefully acknowledged. I alsothank S. Amari, P. Hoppe, F.J. Stadermann, and E.K. Zin-ner for not only providing samples, but also for making theirDF-SIMS data available, and for oKering inside informa-tion on the ion microprobe technique. I thank K.P. Jochumfor providing the glass standards used in this study. Themanuscript bene7ted from critical reviews of two anony-mous reviewers. This work was supported by the DeutscheForschungsgemeinschaft through grants Je96=6 and Je96=10and a special grant from the Ministerium f�ur Schule undWeiterbildung, Wissenschaft und Forschung des LandesNordrhein-Westfalen.

Appendix

A suite of eight diKerent glass standards was used to cali-brate our TOF-SIMS. These glasses were prepared by fusingsamples of basalt, andesite, komatiite, peridotite, tonalite,and rhyolite. They have been analyzed by diKerent bulk andmicroanalytical techniques in various laboratories to deter-mine their composition of major, minor, and trace element.A description of the samples, their preparation, and analyt-ical results are given by Jochum et al. (2000).Secondary ion sensitivities relative to silicon are com-

prised in the following tables for all eight glass standards.Data are given for positive as well as negative secondaryions, with simultaneous argon sputtering and after sputter-ing, respectively. All glass standards were analyzed withthe gallium source at ∼600 ps primary pulse width, exceptStHs6=80-G, which was measured at ∼5 ns pulse width.Errors from the uncertainties in compositional data orTOF-SIMS counting statistics are not given, since they aretypically much smaller than the diKerence between valuesfrom individual standards.Element abundances in Tables A1–A8 are given accord-

ing to Jochum et al. (2000) in wt% for major elements andppm (�g=g) for trace elements. Uncertain values are givenin parentheses. For the chlorine sensitivity in Fig. 22, theminimum value obtained from standard ATHO-G duringsputtering (Table A1) was used.


Table A1Secondary ion sensitivities for glass KL2-G prepared from a basalt from the Kilauea volcano of Hawaii

Element Abundance Positive with Positive after Negative with Negative afterAr-sputtering Ar-sputtering Ar-sputtering Ar-sputtering

Li 5.4 6.3 6.5Be 0.88 1.4 1.1B 2.6 1.3 0.77O (%) 45.2 0.00086 0.00079 30 45Na (%) 1.68 21 20Mg (%) 4.37 5.8 6.5Al (%) 6.93 7.6 7.4Si (%) 23.4 1 1 1 1P 1080 0.030 0.026 1.9 1.6Cl 30 (340) (4900)K 3900 34 33Ca (%) 7.79 12 12Sc 32.3 7.9 4.1Ti (%) 1.53 4.7 4.3V 370 3.6 3.6Cr 306 2.7 2.6Mn 1270 3.6 3.9Fe (%) 8.24 1.9 1.9Co 41.6 1.3 1.3Ni 117 0.85Cu 95 0.90 1.2Rb 8.9 59 60Sr 364 12 11Ba 123 9.0 8.7

Table A2Secondary ion sensitivities for glass ML3B-G prepared from a basalt from the Mauna Loa volcano of Hawaii


Li 4.2 13 23Be 0.75 1.1 2.4B 2.2 0.84 1.0O (%) 45.8 0.0010 0.00089 32 53Na (%) 1.74 18 29Mg (%) 3.94 8.1 10Al (%) 7.04 7.3 7.5Si (%) 23.7 1 1 1 1P 960 0.025 3.4 2.6K 3200 37 56Ca (%) 7.43 15 15Sc 31.4 7.7 5.1Ti (%) 1.26 7.5 6.1V 260 5.7 6.3Cr 170 7.6 10.6Mn 1370 3.8 4.8Fe (%) 8.39 1.8 2.6Co 41.4 0.79 2.0Cu 115 0.33 0.47Rb 5.78 49 93Sr 315 14 14Cs 0.142 65 59Ba 80.1 10 18


Table A3Secondary ion sensitivities for glass StHs6=80-G prepared from an andesite from the St. Helens (USA) eruption


Li 19 11 14Be 1.4 1.8 1.5B 13 0.52 0.71O (%) 47.4 0.00091 0.00063 23 57Na (%) 3.39 22 22Mg (%) 1.19 7.0 8.1Al (%) 9.33 6.6 7.1Si (%) 29.8 1 1 1 1P 700 1.4Cl 260 (720)K 10700 35.5 26Ca (%) 3.76 13.6 15Ti (%) 0.42 6.4 6.1V 96 4.0Mn 590 5.8 3.4Fe (%) 3.39 3.1 3.4Rb 29.5 58 37Sr 486 16 15Cs 1.89 47Ba 302 14 14

Table A4Secondary ion sensitivities for glass GOR128-G prepared from a komatiite from the Gorgona Island, Colombia


Li 7.3 22 25Be 0.04 3.8B 22 0.58 0.63O (%) 44.6 0.0012 0.0010 24 40Na (%) 0.417 31 32Mg (%) 15.6 9.0 9.6Al (%) 5.2 8.7 8.3Si (%) 21.5 1 1 1 1P 120 0.023 2.1K 300 50 53Ca (%) 4.39 19 16Sc 31.1 11 5.5Ti (%) 0.168 7.5 6.0V 170 5.9 5.1Cr 2150 5.8 5.6Mn 1390 4.2 4.3Fe (%) 7.58 2.1 2.1Co 79.5 1.4 1.8Ni 1070 1.2 0.74Cu 63 0.90 2.0


Table A5Secondary ion sensitivities for glass GOR132-G prepared from a komatiite from the Gorgona Island, Colombia


Li 7.9 15 23B 18 0.44 0.62O (%) 44.1 0.0011 0.0010 26 40Na (%) 0.592 25 29Mg (%) 13.5 8.5 10Al (%) 5.77 8.0 8.4Si (%) 21.3 1 1 1 1P 170 1.7K 270 37 45Ca (%) 6.07 17 15Sc 34.6 7.2Ti (%) 0.18 6.4 5.8V 189 5.4 5.7Cr 2410 5.5 6.2Mn 1180 4.4 5.1Fe (%) 7.93 1.8 2.6Co 88.2 1.1 2.1Ni 1170 0.67 1.3Cu 180 1.5Rb 2.14 78Sr 15.6 13Cs 8.08 47 56

Table A6Secondary ion sensitivities for glass BM90=21-G prepared from a peridotite from the Ivrea Zone, Italy


Li 1.4 20 15B 2.8 0.93 1.4O (%) 45.9 0.00097 0.00092 23 35Na (%) 0.0838 24 19Mg (%) 20.6 8.1 8.3Al (%) 1.23 7.6 6.9Si (%) 24.9 1 1 1 1K 31 65 59Ca (%) 1.5 16 11Sc 11.3 7.5 7.1Ti (%) 0.037 6.5 4.9V 40 8.1 6.2Cr 2100 5.6 5.1Mn 829 4.5 4.4Fe (%) 5.24 2.3 2.6Co 88.5 1.5 1.9Ni 1880 0.83 1.2Cs 1.2 45


Table A7Secondary ion sensitivities for glass T1-G prepared from a tonalite from the Italian Alps


Li 19.8 8.5 9.8Be 2.4 1.3 1.7B 4.6 0.88 0.79O (%) 46.6 0.00083 0.00078 33 56Na (%) 2.32 18 16Mg (%) 2.25 7.1 7.8Al (%) 9.05 7.6 6.6Si (%) 27.3 1 1 1 1P 772 0.043 0.039 3.0 1.6Cl 86 (860) (3000)K 15900 34 27Ca (%) 5.05 14 11Sc 26.7 8.6 4.2Ti (%) 0.445 6.7 5.2V 190 5.6 4.8Cr 20 6.5 6.2Mn 1020 4.1 4.3Fe (%) 5.02 2.2 2.4Co 18.9 1.3 1.8Cu 21 0.90Rb 79.3 40 30Sr 282 15 11Cs 2.85 44 37Ba 393 17 10

Table A8Secondary ion sensitivities for glass ATHO-G prepared from a rhyolite from Iceland


Li 27.6 5.9 7.7Be 3.7 0.78 1.5B 5.8 0.46 0.75O (%) 49.0 0.00073 0.00071 30 53Na (%) 3 11 24Mg (%) 0.068 4.5 6.9Al (%) 6.16 5.4 5.5Si (%) 35.5 1 1 1 1P 110 0.066 0.040 2.9Cl 400 140 (450)K 22100 14 24Ca (%) 1.18 9.5 12Sc 5.03 6.5 4.1Ti (%) 0.145 5.6 4.6V 4.4 3.2 5.1Cr 5.3 3.6 4.2Mn 790 3.6 5.3Fe (%) 2.42 2.5 3.6Co 2.2 2.0Rb 63.8 14 25Sr 95.8 9.7 9.8Cs 1.31 57Ba 549 10 8.8


References

Allamandola, L.J., 1996. PAHs, they’re everywhere! In: Greenberg, J.M.(Ed.), The Cosmic Dust Connection. Kluwer Academic Publishers,Dordrecht, pp. 81–102.

Allamandola, L.J., Sandford, S.A., Wopenka, B., 1987. Interstellarpolycyclic aromatic hydrocarbons and carbons in interplanetary dustparticles and meteorites. Science 237, 56–59.

Amari, S., Anders, E., Virag, A., Zinner, E., 1990. Interstellar graphitein meteorites. Nature 345, 238–240.

Amari, S., Foote, J., Simon, C., Swan, P., Walker, R.M., Zinner, E.K.,Jessberger, E.K., Lange, G., Stadermann, F., 1991. SIMS chemicalanalysis of extended impact features from the trailing edge portionof experiment A0187-2. Proceedings of the First LDEF Post-RetrievalSymposium, pp. 503–516.

Amari, S., Foote, J., Swan, P., Walker, R.M., Zinner, E.K., Lange, G.,1993. SIMS chemical analysis of extended impacts on the leadingand trailing edges of LDEF experiment A0187-2. Proceedings of theSecond LDEF Post-Retrieval Symposium, pp. 513–528.

Amari, S., Hoppe, P., Zinner, E., Lewis, R.S., 1992. Interstellar SiC withunusual isotopic compositions: Grains from a supernova? Astrophys.J. 394, L43–L46.

Amari, S., Lewis, R.S., Anders, E., 1994. Interstellar grains in Meteorites:I. Isolation of SiC, graphite, and diamond; size distributions of SiCand graphite. Geochim. Cosmochim. Acta 58, 459–470.

Anders, E., 1989. Pre-biotic organic matter from comets and asteroids.Nature 342, 255–257.

Anders, E., 1996. Evaluating the evidence for past life on Mars. Science274, 2119–2121.

Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoriticand solar. Geochim. Cosmochim. Acta 53, 197–214.

Anders, E., Hayatsu, R., Studier, M.H., 1973. Organic compounds inmeteorites. Science 182, 781–790.

Anders, E., Zinner, E., 1993. Interstellar grains in primitive meteorites:Diamond, silicon carbide, and graphite. Meteoritics 28, 490–514.

Arlinghaus, H.F., Guo, X.Q., 1998. Depth pro7ling and imaging with laserpostionization SNMS. In: Gillen, G., Lareau, R., Bennett, J., Stevie,F. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMS XI.Wiley, Chichester, pp. 677–680.

Arlinghaus, H.F., Whitaker, T.J., Joyner, C.F., Kwoka, P., Jacobson, B.,Tower, J., 1997. Present status of laser postionization analyses. In:Benninghoven, A., HagenhoK, B., Werner, H.W. (Eds.), SecondaryIon Mass Spectrometry. Proceedings SIMS X. Wiley, Chichester, pp.123–130.

Arndt, P., Bohsung, J., Maetz, M., Jessberger, E.K., 1996a. The elementalabundances in interplanetary dust particles. Meteorit. Planet. Sci. 31,817–833.

Arndt, P., Jessberger, E.K., Maetz, M., Reimold, D., Traxel, K., 1997. Onthe accuracy of element concentrations and masses of micron sizedsamples determined with the Heidelberg proton microprobe. Nucl.Instr. and Meth. B 130, 192–198.

Arndt, P., Jessberger, E.K., Warren, J., Zolensky, M., 1996b. Brominecontamination of IDPs during collection (abstract). Meteorit. Planet.Sci. 31, A8.

Arndt, P., Maetz, M., Reimold, D., Wallianos, A., Jessberger, E.K.,1996c. Mass and multielement analyses of interplanetary dust particleswith PIXE and STIM at the Heidelberg proton microprobe (abstract).Meteorit. Planet. Sci. 31, A8–A9.

Bada, J.L., Glavin, D.P., McDonald, G.D., Becker, L., 1998. A searchfor endogenous amino acids in Martian meteorite ALH84001. Science279, 362–365.

Balebanov, B.M., Beghin, C., Belousov, K.G., Bujor, M., Evlanov, E.N.,Hamelin, M., Inal-Ipa, A., Khromov, V.I., Kohnev, V.A., Langevin,Y., Lespagnol, J., Managadze, G.G., Martinson, A.A., Michau, J.L.,Minkala, J.L., Pelinen, R., Podkolzin, S.N., Pomathiod, L., Riedler,W., Sagdeev, R.Z., Schwingenschuh, K., Steller, M., Thomas, R.,Zubkov, B.V., 1986. Presentation of the DION experiment: Preliminary

results of SIMS method on mineralogical samples. Scienti7c andMethodological Aspects of the PHOBOS Study. Proceedings of theInternational Workshop, Space Research Institute, USSR Academy ofScience, Moscow, USSR, pp. 245–260.

Beck, P., 1994. CoMA: A high resolution time of .ight secondary ion massspectrometer for in situ analysis of cometary matter. In: Benninghoven,A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS IX. Wiley, Chichester, pp. 923–926.

Beck, P., Kissel, J., 1994. CoMA:A cometarymatter analyzer for in situ ana-lysis with high mass resolution (abstract). Lunar Planet. Sci. 25, 75–76.

Becker, L., Glavin, D.P., Bada, J.L., 1997. Polycyclic aromatichydrocarbons (PAHs) in Antarctic Martian meteorites, carbonaceouschondrites, and polar ice. Geochim. Cosmochim. Acta 61, 475–481.

Becker, L., Popp, B., Rust, T., Bada, J.L., 1999. The origin of organicmatter in the Martian meteorite ALH84001. Earth Planet. Sci. Lett.167, 71–79.

Bell, J.F., 1996. Evaluating the evidence for past life on Mars. Science274, 2121–2122.

Benninghoven, A., 1973. Surface investigation of solids by the staticalmethod of secondary ion mass spectroscopy (SIMS). Surface Sci. 35,427–457.

Benninghoven, A., 1975. Developments in secondary ion mass spectro-scopy and applications to surface studies. Surface Sci. 53, 596–625.

Benninghoven, A., 1985. Static SIMS applications—From silicon singlecrystal oxidation to DNA sequencing. J. Vac. Sci. Technol. A 3,451–460.

Benninghoven, A., 1994. Surface analysis by secondary ion massspectroscopy (SIMS). Surface Sci. 299/300, 246–260.

Bernatowicz, T.J., Amari, S., Lewis, R.S., 1992. TEM studies of acircumstellar rock (abstract). Lunar Planet. Sci. 23, 91–92.

Bernatowicz, T.J., Amari, S., Zinner, E.K., Lewis, R.S., 1991. Interstellargrains within interstellar grains. Astrophys. J. 373, L73–L76.

Bernatowicz, T., Fraundorf, G., Tang, M., Anders, E., Wopenka, B.,Zinner, E., Fraundorf, P., 1987. Evidence for interstellar SiC in theMurray carbonaceous meteorite. Nature 330, 728–730.

Bernhard, R.P., HQorz, F., Zolensky, M.E., See, T.H., Barrett, R.A., 1993a.Composition and frequency of impact residues detected on LDEFsurfaces. Proceedings of the First European Conference on SpaceDebris, pp. 189–194.

Bernhard, R.P., See, T.H., HQorz, F., 1993b. Projectile compositionsand modal frequencies on the “chemistry of micrometeoroids”LDEF experiment. Proceedings of the Second LDEF Post-RetrievalSymposium, pp. 551–573.

Berzelius, J.J., 1834. Ueber Meteorsteine. Ann. Phys. Chem. 33, 113–148.BischoK, A., 1998. Aqueous alteration of carbonaceous chondrites:

Evidence for preaccretionary alteration—A review. Meteorit. Planet.Sci. 33, 1113–1122.

Blaise, G., 1994. New description of the charge mechanism of insulators.In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS IX. Wiley,Chichester, pp. 45–48.

Bohsung, J., Arndt, P., Jessberger, E.K., Maetz, M., Traxel, K., Wallianos,A., 1995. High resolution PIXE analyses of interplanetary dust particleswith the new Heidelberg proton microprobe. Planet. Space Sci. 43,411–428.

Bohsung, J., Jessberger, E.K., Traxel, K., 1993. PIXE analysesof interplanetary dust particles using the new Heidelberg protonmicroprobe (abstract). Meteoritics 28, 326–327.

Bottazzi, P., Ottolini, L., Vannucci, R., Zanetti, A., 1994. An accurateprocedure for the quanti7cation of rare earth elements in silicates.In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS IX. Wiley,Chichester, pp. 927–930.

Bradley, J.P., 1994a. Nanometer-scale mineralogy and petrography of7ne-grained aggregates in anhydrous interplanetary dust particles.Geochim. Cosmochim. Acta 58, 2123–2134.

Bradley, J.P., 1994b. Chemically anomalous, preaccretionally irradiatedgrains in interplanetary dust from comets. Science 265, 925–929.


Bradley, J.P., Harvey, R.P., McSween Jr., H.Y., 1996. Magnetite whiskersand platelets in the ALH84001 Martian meteorite: Evidence ofvapor phase growth. Geochim. Cosmochim. Acta 60, 5149–5155.

Bradley, J.P., Harvey, R.P., McSween Jr., H.Y., 1997. No ‘nanofossils’in martian meteorites. Science 390, 454–456.

Bradley, J., Ireland, T., 1996. The search for interstellar componentsin interplanetary dust particles. In: Gustafson, B. _A.S., Hanner, M.S.(Eds.), Physics, Chemistry, and Dynamics of Interplanetary Dust, ASPConference Series, Vol. 104. ASP, San Francisco, pp. 275–282.

Bradley, J.P., Keller, L.P., Snow, T.P., Hanner, M.S., Flynn, G.J., Gezo,J.C., Clemett, S.J., Brownlee, D.E., Bowey, J.E., 1999b. An infraredspectral match between GEMS and interstellar grains. Science 285,1716–1718.

Bradley, J.P., McSween Jr., H.Y., Harvey, R.P., 1998. Epitaxial growthof nanophase magnetite in Martian meteorite Allan Hills 84001: Impli-cations for biogenic mineralization. Meteorit. Planet. Sci. 33, 765–773.

Bradley, J.P., Snow, T.P., Brownlee, D.E., Hanner, M.S., 1999a.Mg-rich olivine and pyroxene grains in primitive meteoritic materials:comparison with crystalline silicate data from ISO. In: d’Hendecourt,L., Joblin, C., Jones, A. (Eds.), Solid Interstellar Matter: The ISORevolution. Springer, Berlin, Heidelberg, pp. 297–315.

Brearley, A.J., Jones, R.H., 1998. Chondritic meteorites. In: Papike,J.J. (Ed.), Planetary Materials, Reviews in Mineralogy, MineralogicalSociety of America, Washington, D.C., pp. (3)1–(3)398.

Browning, L.B., Keil, K., 1997. Were CM chondrites aqueously alteredin the quiescent interior of their parent body? (abstract). Lunar Planet.Sci. 28, 163–164.

Browning, L.B., McSween Jr., H.Y., Zolensky, M.E., 1996. Correlatedalteration eKects in CM carbonaceous chondrites. Geochim.Cosmochim. Acta 60, 2621–2633.

Brownlee, D., 1994. The origin and role of dust in the early solar system.In: Zolensky, M.E., Wilson, T.L., Rietmeijer, F.J.M., Flynn, G.J.(Eds.), Analysis of Interplanetary Dust, AIP Conference Proceedings,Vol. 310. AIP, New York, pp. 5–8.

Brownlee, D.E., 1996. The elemental composition of interplanetary dust.In: Gustafson, B. _A.S., Hanner, M.S. (Eds.), Physics, Chemistry, andDynamics of Interplanetary Dust, ASP Conference Series 104. ASP,San Francisco, pp. 261–273.

Brownlee, D.E., Burnett, D., Clark, B., Hanner, M.S., HQorz, F., Kissel,J., Newburn, R., Sandford, S., Sekanina, Z., Tsou, P., Zolensky, M.,1996. Stardust: comet and interstellar dust sample return mission.In: Gustafson, B. _A.S., Hanner, M.S. (Eds.), Physics, Chemistry, andDynamics of Interplanetary Dust, ASP Conference Series 104. ASP,San Francisco, pp. 223–226.

Brownlee, D.E., Tomandl, D.A., Olszewski, E., 1977. Interplanetarydust; A new source of extraterrestrial material for laboratory studies.Proceedings of the Eighth Lunar Science Conference, pp. 149–160.

Brownlee, D.E., Tsou, P., Burnett, D., Clark, B., Hanner, M.S., HQorz, F.,Kissel, J., McDonnell, J.A.M., Newburn, R.L., Sandford, S., Sekanina,Z., Tuzzolino, A.J., Zolensky,M., 1997. The Stardust mission: Returningcomet samples to Earth (abstract). Meteorit. Planet. Sci. 32, A22.

Buseck, P.R., Devouard, B., Hua, X., 1997. Comparison of the mineralogyof 7ne-grained rims and adjacent matrix in CM2 chondrites (abstract).Meteorit. Planet. Sci. 32, A24–A25.

Cassini, G.D., 1730. DWecouverte de la lumiere celeste qui paroist dans lezodiaque. Mem. Acad. Roy. Sci. Paris VIII 1666–1699, 119–209.

Chaussidon, M., Libourel, G., 1993. Boron partitioning in the uppermantle: an experimental and ion probe study. Geochim. Cosmochim.Acta 57, 5053–5062.

Chaussidon, M., Robert, F., Mangin, D., Hanon, P., Rose, E., 1998.Microscale boron isotope analysis in meteoritic matter. In: Gillen,G., Lareau, R., Bennett, J., Stevie, F. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS XI. Wiley, Chichester, pp. 43–46.

Chyba, C.F., McDonald, G.D., 1995. The origin of life in the solar system:Current issues. Annu. Rev. Earth Planet. Sci. 23, 215–249.

Chyba, C.F., Sagan, C., 1987. Cometary organics but no evidence forbacteria. Nature 329, 208.

Chyba, C.F., Thomas, P.J., Brookshaw, L., Sagan, C., 1990. Cometarydelivery of organic molecules to the early Earth. Science 249, 366–373.

Cicerone, R.J., 1981. Halogens in the atmosphere. Rev. Geophys. SpacePhys. 19, 123–139.

Claus, G., Nagy, B., 1961. A microbiological examination of somecarbonaceous chondrites. Nature 192, 594–596.

Clemett, S.J., Dulay, M.T., Gillette, J.S., Chillier, X.D.F., Mahajan, T.B.,Zare, R.N., 1998. Evidence for the extraterrestrial origin of polycyclicaromatic hydrocarbons (PAHs) in the Martian meteorite ALH84001.Faraday Disc. Chem. Soc. 109, 417–436.

Clemett, S.J., Maechling, C.R., Zare, R.N., Alexander, C.M.O., 1992.Analysis of polycyclic aromatic hydrocarbons in seventeen ordinaryand carbonaceous chondrites (abstract). Lunar Planet. Sci. 23,233–234.

Clemett, S.J., Maechling, C.R., Zare, R.N., Swan, P.D., Walker, R.M.,1993. Identi7cation of complex aromatic molecules in individualinterplanetary dust particles. Science 262, 721–725.

Clemett, S.J., Messenger, S., Chillier, X.D.F., Gao, X., Walker,R.M., Zare, R.N., 1996. Indigenous polycyclic aromatic hydrocarbonmolecules in circumstellar graphite grains (abstract). Lunar Planet. Sci.27, 229–230.

Compston, W., Williams, I.S., Meyer, C., 1984. U-Pb geochronology ofzircons from lunar breccia 73217 using a sensitive high mass-resolutionion microprobe. Proceedings of the 14th Lunar Planetary ScienceConference. J. Geophys. Res. 89, B525–B534.

Cronin, J.R., Pizzarello, S., Cruikshank, D.P., 1988. Organic matter incarbonaceous chondrites, planetary satellites, asteroids and comets. In:Kerridge, J.F., Matthews, M.S. (Eds.), Meteorites and the Early SolarSystem. Univ. Arizona Press, Tucson, pp. 819–857.

Crozaz, G., Zinner, E., 1985. Ion probe determinations of the rare earthconcentrations of individual meteoritic phosphate grains. Earth Planet.Sci. Lett. 73, 41–52.

de Chambost, E., Hillion, F., Rasser, B., Migeon, H.N., 1992. TheCAMECA IMS 1270: A description of the secondary ion opticalsystem. In: Benninghoven, A., Janssen, T.F., TQumpner, J., Werner,H.W. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMSVIII. Wiley, Chichester, pp. 207–210.

de Chambost, E., Schuhmacher, M., LQovestam, G., Claesson, S.,1997. Achieving high transmission with the Cameca IMS1270. In:Benninghoven, A., HagenhoK, B., Werner, H.W. (Eds.), SecondaryIon Mass Spectrometry. Proceedings SIMS X. Wiley, Chichester,pp. 1003–1006.

Dermott, S.F., Grogan, K., Gustafson, B. _A.S., Jayaraman, S., Kortenkamp,S.J., Xu, Y.L., 1996. Sources of interplanetary dust. In: Gustafson,B. _A.S., Hanner, M.S. (Eds.), Physics, Chemistry, and Dynamics ofInterplanetary Dust, ASP Conference Series 104. ASP, San Francisco,pp. 143–153.

de Vries, M.S., Reihs, K., Wendt, H.R., Golden, W.G., Hunziker, H.E.,Fleming, R., Peterson, E., Chang, S., 1993. A search for C60 incarbonaceous chondrites. Geochim. Cosmochim. Acta 57, 933–938.

Didenko, P.I., Koval, V.B., 1994. SIMS measurement of metalisotope-ratios in geological samples. In: Benninghoven, A., Nihei, Y.,Shimizu, R., Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry.Proceedings SIMS IX. Wiley, Chichester, pp. 931–933.

Eiler, J.M., Valley, J.W., Graham, C.M., 1998. Oxygen and carbon isotopeanalysis by SIMS: A case study of the Martian meteorite ALH84001.In: Gillen, G., Lareau, R., Bennett, J., Stevie, F. (Eds.), SecondaryIon Mass Spectrometry. Proceedings SIMS XI. Wiley, Chichester,pp. 47–50.

EndreX, M., BischoK, A., 1996. Carbonates in CI chondrites: clues toparent body evolution. Geochim. Cosmochim. Acta 60, 489–507.

Engrand, C., Maurette, M., Kurat, G., BrandstQatter, F., Perreau, M.,1993. A new carbon-rich phase (“COPS”) in Antarctic micrometeorites(abstract). Lunar Planet. Sci. 24, 441–442.

Esposito, F., Spinelli, N., Velotta, R., Berardi, V., 1991. Dead timecorrection of time distribution measurements. Rev. Sci. Instrum. 62,2822–2827.


Eugster, O., Armstrong, J.T., Wasserburg, G.J., 1985. Identi7cation ofisobaric interferences. A computer program for ion microprobe massspectral data. Int. J. Mass Spectrom. Ion Process. 66, 291–312.

Fahey, A.J., 1998. Measurements of REE abundances in NIST SRM-600series glasses and direct measurements of REE oxide to REE ionsignals. In: Gillen, G., Lareau, R., Bennett, J., Stevie, F. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS XI. Wiley,Chichester, pp. 789–792.

Fahey, A.J., Messenger, S., 1999. Isotopic measurements of Murchisonand Semarkona residues by TOF-SIMS (abstract). Lunar Planet. Sci.30, #1101.

Floss, C., JolliK, B., 1998. Rare earth element sensitivity factors in calcicplagioclase (anorthite). In: Gillen, G., Lareau, R., Bennett, J., Stevie,F. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMS XI.Wiley, Chichester, pp. 785–788.

Flynn, G.J., Bajt, S., Sutton, S.R., 1996b. Evidence for weakly boundbromine in large interplanetary dust particles collected from thestratosphere (abstract). Lunar Planet. Sci. 27, 367–368.

Flynn, G.J., Bajt, S., Sutton, S.R., Zolensky, M.E., Thomas, K.L.,Keller, L.P., 1996a. The abundance pattern of elements having lownebular condensation temperatures in interplanetary dust particles:evidence for a new chemical type of chondritic material. In: Gustafson,B. _A.S., Hanner, M.S. (Eds.), Physics, Chemistry, and Dynamics ofInterplanetary Dust, ASP Conference Series 104. ASP, San Francisco,pp. 291–294.

Flynn, G.J., Sutton, S.R., 1992. Element abundances in stratosphericcosmic dust: indications for a new chemical type of chondritic material(abstract). Lunar Planet. Sci. 23, 373–374.

Froude, D.O., Ireland, T.R., Kinny, P.D., Williams, I.S., Compston, W.,Williams, I.R., Myers, J.S., 1983. Ion microprobe identi7cation of4,100–4,200 Myr-old terrestrial zircons. Nature 304, 616–618.

Greshake, A., Stephan, T., Rost, D., 1998. Symplectic exsolutions inolivine from the Martian meteorite Chassigny: evidence for slowcooling under highly oxidizing conditions (abstract). Lunar Planet. Sci.29, #1069.

HagenhoK, B., 1997. Quanti7cation in molecular SIMS. In: Benninghoven,A., HagenhoK, B., Werner, H.W. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS X. Wiley, Chichester, pp. 81–88.

HagenhoK, B., Kock, R., Deimel, M., Benninghoven, A., Bauch, H.-J.,1992. Quanti7cation of molecular SIMS by internal standards. In:Benninghoven, A., Janssen, T.F., TQumpner, J., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS VIII. Wiley,Chichester, pp. 831–834.

HagenhoK, B., van Leyen, D., Niehuis, E., Benninghoven, A., 1989.Time-of-.ight secondary ion mass spectrometry of insulators withpulsed charge compensation by low-energy electrons. J. Vac. Sci.Technol. A 7, 3056–3064.

Hahn, J.H., Zenobi, R., Bada, J.L., Zare, R.N., 1988. Applicationof two-step laser mass spectrometry to cosmogeochemistry: Directanalysis of meteorites. Science 239, 1523–1525.

Hamelin, M., Balebanov, V.M., BWeghin, C., Evlanov, E.N., Grard, R.,Inal-Ipa, A., Khromov, V.N., Kotchnev, V.A., Langevin, Y., Liede, I.,Managadze, G.G., Michau, J.L., Pellinen, R., Piironen, J., Pomathoid,L., Raitala, J., Riedler, W., Roux, A., Sagdeev, R.Z., Schwingenschuh,K., Thomas, R., Trotignon, J.G., Zubkov, B.V., 1990. SIMS remoteanalysis of the Phobos surface: The DION experiment. Adv. SpaceRes., 10, (3)49–(3)52.

Hayes, J.M., 1967. Organic constituents of meteorites—a review.Geochim. Cosmochim. Acta 31, 1395–1440.

Heiss, C.H., Stephan, T., Jessberger, E.K., Wanczek, K.P., Kissel, J., 1998.Analysis of organic cometary compounds with TOF-SIMS (abstract).1st European Workshop on Secondary Ion Mass Spectrometry, p. 32.

Henkel, T., Stephan, T., Jessberger, E.K., Hoppe, P., Strebel, R., 2000.Structured presolar silicon carbide grains? (abstract). Meteorit. Planet.Sci. 35, A69–A70.

Hillion, F., Daigne, B., Girard, F., Slodzian, G., 1997. The Cameca“Nanosims 50” experimental results. In: Benninghoven, A., HagenhoK,

B., Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry.Proceedings SIMS X. Wiley, Chichester, pp. 979–982.

Hillion, F., Daigne, B., Girard, F., Slodzian, G., Schuhmacher, M., 1994.A new high performance instrument the Cameca “Nanosims 50”.In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS IX. Wiley,Chichester, pp. 254–257.

Hoppe, P., Amari, S., Zinner, E., Ireland, T., Lewis, R.S., 1994. Carbon,nitrogen, magnesium, silicon, and titanium isotopic compositionsof single interstellar silicon carbide grains from the Murchisoncarbonaceous chondrite. Astrophys. J. 430, 870–890.

Hoppe, P., Strebel, R., Eberhardt, P., Amari, S., Lewis, R.S., 1996. TypeII supernova matter in a silicon carbide grain from the Murchisonmeteorite. Science 272, 1314–1316.

Hornung, K., Kissel, J., 1994. On shock wave impact ionization of dustparticles. Astron. Astrophys. 291, 324–336.

HQorz, F., See, T.H., Bernhard, R.P., Brownlee, D.E., 1995. Naturaland orbital debris particles on LDEF’s trailing and forward-facingsurfaces. Proceedings of the Third LDEF Post-Retrieval Symposium,pp. 415–430.

Hurst, G.S., Payne, M.G., Kramer, S.D., Young, J.P., 1979. Resonanceionization spectroscopy and one-atom detection. Rev. Mod. Phys. 51,767–819.

Huss, G.R., Hutcheon, I.D., Wasserburg, G.J., Stone, J., 1992. Presolar(?) corundum in the Orgueil meteorite (abstract). Lunar Planet. Sci.23, 563–564.

Hutcheon, I.D., Huss, G.R., Fahey, A.J., Wasserburg, G.J., 1994. Extreme26Mg and 17O enrichments in an Orgueil corundum: identi7cation ofa presolar oxide grain. Astrophys. J. 425, L97–L100.

Ireland, T.R., 1995. Ion microprobe mass spectrometry: Techniques andapplications in cosmochemistry, geochemistry and geochronology. In:Hyman, M., Rowe, M. (Eds.), Advances in Analytical Geochemistry,Vol. 2. JAI Press, Greenwich, Connecticut, pp. 1–118.

Jessberger, E.K., Bohsung, J., Chakaveh, S., Traxel, K., 1992. The volatileelement enrichment of chondritic interplanetary dust particles. EarthPlanet. Sci. Lett. 112, 91–99.

Jessberger, E.K., Christoforidis, A., Kissel, J., 1988. Aspects of the majorelement composition of Halley’s dust. Nature 332, 691–695.

Jessberger, E.K., Kissel, J., 1991. Chemical properties of cometary dustand a note on carbon isotopes. In: Newburn, R.L., Neugebauer, M.,Rahe, J. (Eds.), Comets in the Post-Halley Era. Kluwer AcademicPublishers, Dordrecht, pp. 1075–1092.

Jessberger, E.K., Kissel, J., Rahe, J., 1989. The composition of comets.In: Atreya, S.K., Pollack, J.B., Matthews, M.S. (Eds.), Origin andEvolution of Planetary and Satellite Atmospheres. Univ. Arizona Press,Tucson, pp. 167–191.

Jessberger, E.K., Stephan, T., Rost, D., Arndt, P., Maetz, M., Stadermann,F.J., Brownlee, D.E., Bradley, J.P., Kurat, G., 2001. Properties ofinterplanetary dust: information from collected samples In: Dermott,S.F., Fechtig, H., GrQun, E., Gustafson, B. _A.S. (Eds.), InterplanetaryDust. Univ. of Arizona Press, Tucson, pp. 253–294.

Jochum, K.P., Dingwell, D.B., Rocholl, A., Stoll, B., Hofmann, A.W.,Becker, S., Besmehn, A., Bessette, D., Dietze, H.-J., Dulski, P.,Erzinger, J., Hellebrand, E., Hoppe, P., Horn, I., Janssens, K., Jenner,G.A., Klein, M., McDonough, W.F., Maetz, M., Mezger, K., MQunker,C., Nikogosian, I.K., Pickhardt, C., Raczek, I., Rhede, D., Seufert,H.M., Simakin, S.G., Sobolev, A.V., Spettel, B., Straub, S., Vincze, L.,Wallianos, A., Weckwerth, G., Weyer, S., Wolf, D., Zimmer, M., 2000.The preparation and preliminary characterisation of eight geologicalMPI-DING reference glasses for in-situ microanalysis. GeostandardsNewsletter 24, 87–133.

Jull, A.J.T., Courtney, C., JeKrey, D.A., Beck, J.W., 1998. Isotopicevidence for a terrestrial source of organic compounds found in Martianmeteorites Allan Hills 84001 and Elephant Moraine 79001. Science279, 366–369.


JQurgens, U., 1993. Ober.Qachenanlyse von Si-Wafern mit derFlugzeit-SIMS. Dissertation, WestfQalische Wilhelms-UniversitQatMQunster. 120p.

JQurgens, U., Cramer, H.-G., Heller, T., Niehuis, E., Zhang, Z.,Benninghoven, A., 1992. Ultra trace detection of metal contaminationon wafer surfaces. In: Benninghoven, A., Janssen, T.F., TQumpner, J.,Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS VIII. Wiley, Chichester, pp. 277–280.

Kampwerth, G., Terhorst, M., Heller, T., Niehuis, E., Benninghoven,A., 1990. TOF-SNMS by nonresonant laser-postionization. In:Benninghoven, A., Evans, C.A.Jr., McKeegan, K.D., Storms, H.A.,Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS VII. Wiley, Chichester, pp. 203–206.

Kissel, J., Krueger, F.R., 1987a. Ion formation by impact of fast dustparticles and comparison with related techniques. Appl. Phys. A 42,69–85.

Kissel, J., Krueger, F.R., 1987b. The organic component in dust fromcomet Halley as measured by the PUMA mass spectrometer on boardVega 1. Nature 326, 755–760.

Kissel, J., Brownlee, D.E., BQuchler, K., Clark, B.C., Fechtig, H., GrQun,E., Hornung, K., Igenbergs, E.B., Jessberger, E.K., Krueger, F.R.,Kuczera, H., McDonnell, J.A.M., Mor7ll, G.M., Rahe, J., Schwehm,G.H., Sekanina, Z., Utterback, N.G., VQolk, H.J., Zook, H.A., 1986a.Composition of comet Halley dust particles from Giotto observations.Nature 321, 336–337.

Kissel, J., Sagdeev, R.Z., Bertaux, J.L., Angarov, V.N., Audouze, J.,Blamont, J.E., BQuchler, K., Evlanov, E.N., Fechtig, H., Fomenkova,M.N., von Hoerner, H., Inogamov, N.A., Khromov, V.N., Knabe, W.,Krueger, F.R., Langevin, Y., Leonas, V.B., Lavasseur-Regourd, A.C.,Managadze, G.G., Podkolzin, S.N., Shapiro, V.D., Tabaldyev, S.R.,Zubkov, B.V., 1986b. Composition of comet Halley dust particles fromVega observations. Nature 321, 280–282.

Kissel, J., Zscheeg, H., RQudenauer, F.G., 1988. Pulsed operation of aliquid metal ion source. Appl. Phys. A 47, 167–169.

Klacka, J., 1994. Dust particle in the solar gravitational andelectromagnetic 7elds. Earth, Moon, and Planets 64, 55–58.

KQotter, F., Benninghoven, A., 1997. Time-of-.ight mass spectrometry ofsputtered neutrals after electron beam postionization. In: Benninghoven,A., HagenhoK, B., Werner, H.W. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS X. Wiley, Chichester, pp. 1011–1014.

Kovalenko, L.J., Maechling, C.R., Clemett, S.J., Philippoz, J.-M., Zare,R.N., Alexander, C.M.O., 1992. Microscopic organic analysis usingtwo-step laser mass spectrometry: Application to meteoritic acidresidues. Anal. Chem. 64, 682–690.

Kovalenko, L.J., Philippoz, J.-M., Bucenell, J.R., Zenobi, R., Zare, R.N.,1991. Organic chemical analysis on a microscopic scale using two-steplaser desorption=laser ionization mass spectrometry. Space Sci. Rev.56, 191–195.

Kurat, G., Koeberl, C., Presper, T., BrandstQatter, F., Maurette, M., 1994.Petrology and geochemistry of Antarctic micrometeorites. Geochim.Cosmochim. Acta 58, 3879–3904.

Kvenvolden, K., Lawless, J., Pering, K., Peterson, E., Flores, J.,Ponnamperuma, C., Kaplan, I.R., Moore, C., 1970. Evidence forextraterrestrial amino-acids and hydrocarbons in the Murchisonmeteorite. Nature 228, 923–926.

La.amme, R.E., Hites, R.A., 1978. The global distribution of polycyclicaromatic hydrocarbons in recent sediments. Geochim. Cosmochim.Acta 42, 289–303.

Lange, G., Eigner, S., Igenbergs, E., Jessberger, E.K., Kuczera, H.,Maas, D., Sutton, S., Weishaupt, U., Zinner, E., 1986. Ion microprobesensitivities and their application to multielement analysis of LDEFimpact residues (abstract). Lunar Planet. Sci. 17, 456–457.

Leshin, L.A., McKeegan, K.D., Carpenter, P.K., Harvey, R.P., 1998.Oxygen isotopic constraints on the genesis of carbonates from Martianmeteorite ALH84001. Geochim. Cosmochim. Acta 62, 3–13.

Lewis, R.S., Tang, M., Wacker, J.F., Anders, E., Steel, E., 1987.Interstellar diamonds in meteorites. Nature 326, 160–162.

Ma, Z., Thompson, R.N., Lykke, K.R., Pellin, M.J., Davis, A.M., 1995.New instrument for microbeam analysis incorporating submicronimaging and resonance ionization mass spectrometry. Rev. Sci. Instrum.66, 3168–3176.

MacPherson, G.J., Davis, A.M., 1994. Refractory inclusions in theprototypical CM chondrite. Mighei. Geochim. Cosmochim. Acta 58,5599–5625.

MacPherson, G.J., Wark, D.A., Armstrong, J.T., 1988. Primitive materialsurviving in chondrites: Refractory inclusions. In: Kerridge, J.F.,Matthews, M.S. (Eds.), Meteorites and the Early Solar System. Univ.Arizona Press, Tucson, pp. 746–807.

Maetz, M., Arndt, P., Bohsung, J., Jessberger, E.K., Traxel, K., 1994.Comprehensive analysis of six IDPs with the Heidelberg protonmicroprobe (abstract). Meteoritics 29, 494–495.

Maetz, M., Arndt, P., Greshake, A., Jessberger, E.K., KlQock, W., Traxel,K., 1996. Structural and chemical modi7cations of microsamplesinduced during PIXE analysis. Nucl. Instr. and Meth. B 109=110,192–196.

Mamyrin, B.A., Krataev, V.I., Shmikk, D.V., Zagulin, V.A., 1973. Themass-re.ectron, a new nonmagnetic time-of-.ight mass spectrometerwith high resolution. Sov. Phys. JETP 37, 45–48.

Maurette, M., Immel, G., Hammer, C., Harvey, R., Kurat, G., Taylor, S.,1994. Collection and curation of IDPs from the Greenland and Antarcticice sheets. In: Zolensky, M.E., Wilson, T.L., Rietmeijer, F.J.M.,Flynn, G.J. (Eds.), Analysis of Interplanetary Dust, AIP ConferenceProceedings, Vol. 310. AIP, New York, pp. 277–289.

McKay, D.S., Gibson Jr., E.K., Thomas-Keprta, K.L., Vali, H., Romanek,C.S., Clemett, S.J., Chillier, X.D.F., Maechling, C.R., Zare, R.N., 1996.Search for past life on Mars: Possible relic biogenic activity in Martianmeteorite ALH84001. Science 273, 491–504.

McKeegan, K.D., Leshin, L.A., Russel, S.S., MacPherson, G.J., 1998.Oxygen isotopic abundances in calcium-aluminum-rich inclusions fromordinary chondrites: Implications for nebular heterogeneity. Science280, 414–418.

McKeegan, K.D., Walker, R.M., Zinner, E., 1985. Ion microprobe isotopicmeasurements of individual interplanetary dust particles. Geochim.Cosmochim. Acta 49, 1971–1987.

Meneguzzi, M., Audouze, J., Reeves, H., 1971. The production of theelements Li, Be, B by galactic cosmic rays in space and its relationwith stellar observations. Astron. Astrophys. 15, 337–359.

Messenger, S., 1998. Oxygen isotopic imaging of interplanetary dust(abstract). Meteorit. Planet. Sci. 33, A106–A107.

Messenger, S., 1999. Oxygen isotopic imaging of interplanetary dust byTOF-SIMS (abstract). Lunar Planet. Sci. 30, #1600.

Messenger, S., Amari, S., Gao, X., Walker, R.M., Clemett, S.J.,Chillier, X.D.F., Zare, R.N., Lewis, R.S., 1998. Indigenous polycyclicaromatic hydrocarbons in circumstellar graphite grains from primitivemeteorites. Astrophys. J. 502, 284–295.

Metzler, K., BischoK, A., StQo\er, D., 1992. Accretionary dust mantlesin CM chondrites: evidence for solar nebular processes. Geochim.Cosmochim. Acta 56, 2873–2897.

Meyer Jr., C., 1979. Trace element analysis of silicates by ion microprobe.In: Benninghoven, A., Evans Jr., C.A., Powell, R.A., Shimizu, R.,Storms, H.A. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS II. Springer, New York, pp. 67–69.

Mittlefehldt, D.W., 1994. ALH84001, a cumulate orthopyroxenite memberof the martian meteorite clan. Meteoritics 29, 214–221.

Mittlefehldt, D.W., McCoy, T.J., Goodrich, C.A., Kracher, A., 1998.Non-chondritic meteorites from asteroidal bodies. In: Papike, J.J. (Ed.),Planetary Materials. Reviews in Mineralogy. Mineralogical Society ofAmerica, Washington, DC, pp. (4)1–(4)195.

Moreels, G., Clairemidi, J., Hermine, P., Brechignac, P., Rousselot, P.,1994. Detection of a polycyclic aromatic molecule in comet P=Halley.Astron. Astrophys. 282, 643–656.

Nicolussi, G.K., Davis, A.M., Pellin, M.J., Lewis, R.S., Clayton, R.N.,Amari, S., 1997a. The isotopic composition of molybdenum in single


presolar silicon carbide grains (abstract). Lunar Planet. Sci. 28,1021–1022.

Nicolussi, G.K., Davis, A.M., Pellin, M.J., Lewis, R.S., Clayton, R.N.,Amari, S., 1997b. S-process zirconium in individual presolar siliconcarbide grains (abstract). Lunar Planet. Sci. 28, 1023–1024.

Nicolussi, G.K., Pellin, M.J., Davis, A.M., Lewis, R.S., Clayton, R.N.,1998b. Zr and Mo isotopes in single presolar graphite grains: a recordof stellar nucleosynthesis (abstract). Lunar Planet. Sci. 29, #1415.

Nicolussi, G.K., Pellin, M.J., Lewis, R.S., Davis, A.M., Amari, S.,Clayton, R.N., 1998a. Molybdenum isotopic composition of individualpresolar silicon carbide grains from the Murchison meteorite. Geochim.Cosmochim. Acta 62, 1093–1104.

Nicolussi, G.K., Pellin, M.J., Lykke, K.R., Trevor, J.L., Mencer, D.E.,Davis, A.M., 1996. Surface analysis by SNMS: Femtosecond laserpostionization of sputtered and laser desorbed atoms. Surf. InterfaceAnal. 24, 363–370.

Niehuis, E., 1990. Exact mass determination using TOF-SIMS. In:Benninghoven, A., Evans Jr., C.A., McKeegan, K.D., Storms, H.A.,Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS VII. Wiley, Chichester, pp. 299–304.

Niehuis, E., 1994. Surface analysis by nonresonant multiphoton ionizationof sputtered neutrals. In: Benninghoven, A., Nihei, Y., Shimizu, R.,Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS IX. Wiley, Chichester, pp. 429–433.

Niehuis, E., 1997. Microanalysis with gallium liquid metal ion guns. In:Benninghoven, A., HagenhoK, B., Werner, H.W. (Eds.), SecondaryIon Mass Spectrometry. Proceedings SIMS X. Wiley, Chichester, pp.47–54.

Niehuis, E., Heller, T., Feld, H., Benninghoven, A., 1987. Design andperformance of a re.ectron based time-of-.ight secondary ion massspectrometer with electrodynamic primary ion mass separation. J. Vac.Sci. Technol. A 5, 1243–1246.

Nittler, L.R., Alexander, C.M.O., Gao, X., Walker, R.M., Zinner, E.K.,1994. Interstellar oxide grains from the Tieschitz meteorite. Nature370, 443–446.

Nittler, L.R., Hoppe, P., Alexander, C.M.O., Amari, S., Eberhardt, P.,Gao, X., Lewis, R.S., Strebel, R., Walker, R.M., Zinner, E., 1995.Silicon nitride from supernovae. Astrophys. J. 453, L25–L28.

Oechsner, H., 1994. Secondary neutral microprobes. In: Benninghoven,A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS IX. Wiley, Chichester, pp. 316–322.

Olive, K.A., Schramm, D.N., 1992. Astrophysical 7Li as a product ofBig Bang nucleosynthesis and galactic cosmic-ray spallation. Nature360, 439–442.

OrWo, J., Gibert, J., Lichtenstein, H., Wirkstrom, S., Flory, D.A., 1971.Amino-acids, aliphatic and aromatic hydrocarbons in the Murchisonmeteorite. Nature 230, 105–106.

Ott, U., 1993. Interstellar grains in meteorites. Nature 364, 25–33.Pellin, M.J., Young, C.E., Calway, W.F., Burnett, J.W., JHrgensen, B.,

Schweitzer, E.L., Gruen, D.M., 1987. Sensitive, low damage surfaceanalysis using resonance ionization of sputtered atoms. Nucl. Instr.and Meth. B 18, 446–451.

Pellinen, R., Piironen, J., SilWen, J., Sagdeev, R.Z., Managadze, G.G.,Shutyaev, I., Timofeev, P., Bondarenko, A., Ter-Mikaeliyan, V.,1990. Laser-ionization studies with the technical models of theLIMA-D=PHOBOS experiment. Adv. Space Res. 10, (3)57–(3)61.

Pering, K.L., Ponnamperuma, C., 1971. Aromatic hydrocarbons in theMurchison meteorite. Science 173, 237–239.

Potts, P.J., Bowles, J.F.W., Reed, S.J.B., Cave, M.R. (Eds.), 1995.Microprobe Techniques in the Earth Sciences. Chapman & Hall,London. 419 p.

Radicati di Brozolo, F., Fleming, R.H., 1992. Mass spectrometricobservation of organic species in a single IDP thin section (abstract).Lunar Planet. Sci. 23, 1123–1124.

Radicati di Brozolo, F., Harris, D.W., Chakel, J.A., Fleming, R.H., Bunch,T.E., 1991. Imaging analysis of LDEF craters (abstract). Meteoritics26, 387.

Reeves, H., Audouze, J., Fowler, W.A., Schramm, D.N., 1973. On theorigin of light elements. Astrophys. J. 179, 909–930.

Rietmeijer, F.J.M., 1993. The bromine content of micrometeorites:Arguments for stratospheric contamination. J. Geophys. Res. 98,E7409–E7414.

Rietmeijer, F.J.M., 1998. Interplanetary dust particles. In: Papike, J.J.(Ed.), Planetary Materials, Reviews in Mineralogy. MineralogicalSociety of America, Washington, DC, pp. (2)1–(2)95.

Robertson, H.P., 1937. Dynamical eKects of radiation in the solar system.Mon. Not. Roy. Astron. Soc. 97, 423–438.

Rost, D., Stephan, T., Jessberger, E.K., 1996. Surface analysis ofstratospheric dust particles with TOF-SIMS: new results (abstract).Meteorit. Planet. Sci. 31, A118–A119.

Rost, D., Stephan, T., Jessberger, E.K., 1998a. Imaging of small particleswith TOF-SIMS (abstract). 1st European Workshop on Secondary IonMass Spectrometry, p. 31.

Rost, D., Stephan, T., Jessberger, E.K., 1999a. Surface analysis ofstratospheric dust particles. Meteorit. Planet. Sci. 34, 637–646.

Rost, D., Stephan, T., Jessberger, E.K., Nakamura, K., KlQock, W., 1998b.New TOF-SIMS analyses of sections from stratospheric dust particles(abstract). Lunar Planet. Sci. 29, #1637.

Rost, D., Stephan, T., Wies, C., Jessberger, E.K., 1999b. Analysisof sections and surfaces of interplanetary dust particles (abstract).Meteorit. Planet. Sci. 34, A99.

RQudenauer, F.G., 1994. SIMS in space. In: Benninghoven, A., Nihei, Y.,Shimizu, R., Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry.Proceedings SIMS IX. Wiley, Chichester, pp. 901–905.

Sagdeev, R.Z., Prokhorov, A.M., Managadze, G.G., Kissel, J., Pellinen,R., Balebanov, V.M., Tabaldyev, S.R., Shutyaev, I.Yu., Bondarenko,A.L., Timofeev, P.P., Ter-Mikaelyan, V.M., Terent’ev, V.I., Natenzon,M.Ja., von Hoerner, H., Riedler, B., Berg, R., Zlatev, S., Kynchev,A., Belyakov, P., Simeonov, L., Silen, J., Phironen, J., Lehto,A., Salmu, Ch., Peltonen, P., Dul’nev, G.N., Pashenin, P.P.,Malyutin, A.A., Il’ichev, N.N., Arumov, G.P., Pershin, S.M.,1986. LIMA-D experiment: methodology laboratory tests. Scienti7cand Methodological Aspects of the PHOBOS Study. ProceedingsInternational Workshop, Space Research Institute, USSR Academy ofScience, Moscow, USSR, pp. 220–230.

SchiK, L.I., 1936. Statistical analysis of counter data. Phys. Rev. 50,88–96.

Schirmeyer, S., BischoK, A., Stephan, T., Jessberger, E.K., 1996a.Lithium-bearing phases in Ca,Al-rich inclusions from CM-chondrites:indication of nebular alteration processes (abstract). Lunar Planet. Sci.27, 1141–1142.

Schirmeyer, S., BischoK, A., Stephan, T., Jessberger, E.K., 1996b. Lithiumdistribution within the carbonaceous chondrites LancWe (CO3) andAllende (CV3): Preliminary results (abstract). Meteorit. Planet. Sci.31, A123–A124.

Schirmeyer, S., Hoppe, P., Stephan, T., BischoK, A., Jessberger, E.K.,1997. A lithium-bearing Ca,Al-rich inclusion from the CM-chondriteCold Bokkeveld studied by TOF-SIMS and conventional SIMS(abstract). Lunar Planet. Sci. 28, 1253–1254.

Schnieders, A., 1999. Quantitative surface analysis by laser postionizationof sputtered neutrals. Dissertation, WestfQalische Wilhelms-UniversitQatMQunster. 141p.

Schnieders, A., Kollmer, F., Benninghoven, A., 1998. Fundamentalinvestigations with imaging laser-SNMS. In: Gillen, G., Lareau, R.,Bennett, J., Stevie, F. (Eds.), Secondary Ion Mass Spectrometry.Proceedings SIMS XI. Wiley, Chichester, pp. 661–664.

Schnieders, A., MQollers, R., Terhorst, M., Cramer, H.-G., Niehuis, E.,1996. Quanti7cation of metal trace contaminants on Si wafer surfacesby laser SNMS and TOF-SIMS using sputter deposited submonolayerstandards. J. Vac. Sci. Technol. B 14, 2712–2724.

Schramm, L.S., Brownlee, D.E., Wheelock, M.M., 1989. Major elementalcomposition of stratospheric micrometeorites. Meteoritics 24, 99–112.

Schueler, B.W., 1992. Microscope imaging by time-of-.ight secondaryion mass spectrometry. Microsc. Microanal. Microstruct. 3, 119–139.


Schueler, B.W., Sander, P., Reed, D.A., 1990. A time-of-.ight secondaryion microscope. Vacuum 41, 1661–1664.

Schuhmacher, M., de Chambost, E., McKeegan, K.D., Harrison, T.M.,Migeon, H.N., 1994. In situ U=Pb dating of zircon with the CAMECAims 1270. In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner,H.W. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMSIX. Wiley, Chichester, pp. 919–922.

Schwieters, J., Cramer, H.-G., Heller, T., JQurgens, U., Niehuis, E.,Zehnpfenning, J., Benninghoven, A., 1991. High mass resolutionsurface imaging with a time-of-.ight secondary ion mass spectroscopyscanning microprobe. J. Vac. Sci. Technol. A 9, 2864–2871.

Schwieters, J., Cramer, H.-G., JQurgens, U., Niehuis, E., Rulle, H.,Heller, T., Zehnpfenning, J., Benninghoven, A., 1992. TOF-SIMSscanning microprobe for high mass resolution surface imaging. In:Benninghoven, A., Janssen, T.F., TQumpner, J., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS VIII. Wiley,Chichester, pp. 497–500.

Scott, E.R.D., 1999. Origin of carbonate-magnetite-sul7de assemblages inMartian meteorite ALH84001. J. Geophys. Res. 104, E3803–E3813.

Scott, E.R.D., Krot, A.N., Yamaguchi, A., 1998. Carbonates in fracturesof Martian meteorite Allan Hills 84001: Petrologic evidence for impactorigin. Meteorit. Planet. Sci. 33, 709–719.

Sigmund, P., 1969. Theory of sputtering. I. sputtering yield of amorphousand polycrystalline targets. Phys. Rev. 184, 383–416.

Slodzian, G., Chaintreau, M., Dennebouy, R., Rasser, B., 1998. Isotoperatio measurements over micro-areas. The challenge of high precisionrepeatability. In: Gillen, G., Lareau, R., Bennett, J., Stevie, F. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS XI. Wiley,Chichester, pp. 29–34.

Slodzian, G., Lorin, J.C., Havette, A., 1980. Isotopic eKect on theionization probabilities in secondary ion emission. J. Physique Lett.41, L555–L558.

Stadermann, F.J., Walker, R.M., Zinner, E., 1999a. Sub-micron isotopicmeasurements with the CAMECA NanoSIMS (abstract). Lunar Planet.Sci. 30, #1407.

Stadermann, F.J., Walker, R.M., Zinner, E., 1999b. NanoSIMS: The nextgeneration ion probe for the microanalysis of extraterrestrial material(abstract). Meteorit. Planet. Sci. 34, A111–A112.

Steigman, G., Walker, T.P., 1992. Production of Li, Be, and B in theearly galaxy. Astrophys. J. 385, L13–L16.

Stephan, T., 1992. Statistical analysis of single-ion-counting dataand isotopic measurements with TOF-SIMS. In: Benninghoven, A.,Janssen, T.F., TQumpner, J., Werner, H.W. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS VIII. Wiley, Chichester, pp. 115–118.

Stephan, T., Arndt, P., Jessberger, E.K., Maetz, M., Reimold, D., Walter,J., 1995b. Multielement analysis of Antarctic micrometeorites usingSEM, EDXA, EMPA, TOF-SIMS, and PIXE (abstract). Lunar Planet.Sci. 26, 1353–1354.

Stephan, T., BischoK, A., Cramer, H.-G., Zehnpfenning, J., 1991.TOF-SIMS, applications in meteorite research, 7rst results (abstract).Meteoritics 26, 397.

Stephan, T., Heiss, C.H., Rost, D., Jessberger, E.K., 1999a. Polycyclicaromatic hydrocarbons in meteorites: Allan Hills 84001, Murchison,and Orgueil (abstract). Lunar Planet. Sci. 30, #1569.

Stephan, T., Jessberger, E.K., 1996. TOF-SIMS analysis of interstellarSiC grains (abstract). Lunar Planet. Sci. 27, 1267–1268.

Stephan, T., Jessberger, E.K., Keller, L.P., Flynn, G.J., Bajt, S., Chapman,H.N., 1996. Fullerenes in interplanetary dust? (abstract). Meteorit.Planet. Sci. 31, A134.

Stephan, T., Jessberger, E.K., KlQock, W., Rulle, H., Zehnpfenning, J.,1994b. TOF-SIMS analysis of interplanetary dust. Earth Planet. Sci.Lett. 128, 453–467.

Stephan, T., Jessberger, E.K., Rost, D., Heiss, C.H., 1999b. TOF-SIMSanalysis of Martian rocks. International Symposium: Mars ExplorationProgram & Sample Return Missions, No. 09=S4(32), CNES, Paris.

Stephan, T., Jessberger, E.K., Rulle, H., Thomas, K.L., KlQock, W.,1994c. New TOF-SIMS results on hydrated interplanetary dust particles(abstract). Lunar Planet. Sci. 25, 1341–1342.

Stephan, T., KlQock, W., Jessberger, E.K., Rulle, H., Zehnpfenning, J.,1993a. Multielement analysis of interplanetary dust particles usingTOF-SIMS (abstract). Lunar Planet. Sci. 24, 1349–1350.

Stephan, T., KlQock, W., Jessberger, E.K., Thomas, K.L., Keller, L.P.,Behla, F., 1993b. Multielement analysis of carbon-rich interplanetarydust particles with TOF-SIMS (abstract). Meteoritics 28, 443–444.

Stephan, T., KlQock, W., Jessberger, E.K., Zehnpfenning, J., 1992a.Analysis of stratospheric interplanetary dust particles with TOF-SIMS,SEM, and TEM (abstract). Meteoritics 27, 292.

Stephan, T., Rost, D., Heiss, C.H., Jessberger, E.K., Greshake, A., 1998c.The lateral distribution of polycyclic aromatic hydrocarbons in AllanHills 84001—implications for their origin. Workshop on the IssueMartian Meteorites: Where Do We Stand and Where Are We Going?LPI Contribution No. 956, Lunar and Planetary Institute, Houston, pp.50–51.

Stephan, T., Rost, D., Jessberger, E.K., 1997b. Volatile-elementenrichments in interplanetary dust due to nebular processes? In:Zolensky, M.E., Krot, A.N., Scott, E.R.D. (Eds.), Workshop onParent-Body and Nebular Modi7cation of Chondritic Materials. LPITechnical Report 97-02, Part 1, Lunar and Planetary Institute, Houston,pp. 59–60.

Stephan, T., Rost, D., Jessberger, E.K., Budell, R., Greshake, A., Zinner,E.K., Amari, S., Hoppe, P., Lewis, R.S., 1997c. TOF-SIMS analysisof SiC grains with high lateral resolution (abstract). Lunar Planet. Sci.28, 1371–1372.

Stephan, T., Rost, D., Jessberger, E.K., Greshake, A., 1998a. Polycyclicaromatic hydrocarbons in ALH84001 analyzed with time-of-.ightsecondary ion mass spectrometry (abstract). Lunar Planet. Sci. 29,#1263.

Stephan, T., Rost, D., Jessberger, E.K., Greshake, A., 1998b. Polycyclicaromatic hydrocarbons are everywhere in Allan Hills 84001 (abstract).Meteorit. Planet. Sci. 33, A149–A150.

Stephan, T., Rost, D., Jessberger, E.K., KlQock, W., 1997a. Time-of-.ightsecondary ion mass spectrometry TOF-SIMS analysis of the OrgueilCI meteorite at high lateral resolution (abstract). Meteorit. Planet. Sci.32, A124–A125.

Stephan, T., Thomas, K.L., Warren, J.L., 1994d. Comprehensiveconsortium study of stratospheric particles from one collector(abstract). Meteoritics 29, 536–537.

Stephan, T., Zehnpfenning, J., Benninghoven, A., 1994a. Correction ofdead time eKects in time-of-.ight mass spectrometry. J. Vac. Sci.Technol. A 12, 405–410.

Stephan, T., Stadermann, F., Cramer, H.-G., Zehnpfenning, J., 1992b.TOF-SIMS analysis of LDEF impact residues (abstract). Lunar Planet.Sci. 23, 1357–1358.

Stephan, T., Rost, D., Jessberger, E.K., 1995a. Surface analysis ofstratospheric particles with TOF-SIMS—bromine enrichments due tocontamination (abstract). Meteoritics 30, 583.

Storms, H.A., Peterson, C.L., 1994. Contributions of deadtime and massbias uncertainties to isotopic composition measurement error. In:Benninghoven, A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS IX. Wiley,Chichester, pp. 39–44.

Tang, M., Anders, E., 1988a. Isotopic anomalies of Ne, Xe, and C inmeteorites. II. Interstellar diamond and SiC: Carriers of exotic noblegases. Geochim. Cosmochim. Acta 52, 1235–1244.

Tang, M., Anders, E., 1988b. Isotopic anomalies of Ne, Xe, and Cin meteorites. III. Local and exotic noble gas components and theirinterrelations. Geochim. Cosmochim. Acta 52, 1245–1254.

Tang, M., Lewis, R.S., Anders, E., Grady, M.M., Wright, I.P., Pillinger,C.T., 1988. Isotopic anomalies of Ne, Xe, and C in meteorites. I.Separation of carriers by density and chemical resistance. Geochim.Cosmochim. Acta 52, 1221–1234.

Terhorst, M., Cramer, H.-G., Niehuis, E., 1997. Dual source ion columnfor depth pro7ling by TOF-SIMS. In: Benninghoven, A., HagenhoK, B.,Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry. ProceedingsSIMS X. Wiley, Chichester, pp. 427–430.


Terhorst, M., Kampwerth, G., Niehuis, E., Benninghoven, A., 1992.Sputtered neutrals mass spectrometry of organic molecules usingmultiphoton postionization. J. Vac. Sci. Technol. A 10, 3210–3215.

Terhorst, M., MQollers, R., Niehuis, E., Benninghoven, A., 1994.Time-of-.ight mass spectrometry of laser postionized sputteredmolecules. In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner,H.W. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMSIX. Wiley, Chichester, pp. 355–358.

Thomas, K.L., Blanford, G.E., Keller, L.P., KlQock, W., McKay, D.S.,1993. Carbon abundance and silicate mineralogy of anhydrousinterplanetary dust particles. Geochim. Cosmochim. Acta 57, 1551–1566.

Thomas, K.L., Keller, L.P., Blanford, G.E., McKay, D.S., 1994b.Quantitative analyses of carbon in anhydrous and hydratedinterplanetary dust particles. In: Zolensky, M.E., Wilson, T.L.,Rietmeijer, F.J.M., Flynn, G.J. (Eds.), Analysis of Interplanetary Dust,AIP Conference Proceedings, Vol. 310. AIP, New York, pp. 165–172.

Thomas, K.L., Romanek, C.S., Clemett, S.J., Gibson, E.K., McKay, D.S.,1995. Preliminary analysis of polycyclic aromatic hydrocarbons in theMartian (SNC) meteorite ALH 84001 (abstract). Lunar Planet. Sci.26, 1409–1410.

Thomas, D., Schramm, D.N., Olive, K.A., Mathews, G.J., Meyer, B.S.,Fields, B.D., 1994a. Production of lithium, beryllium, and boron frombaryon inhomogeneous primordial nucleosynthesis. Astrophys. J. 430,291–299.

Tomeoka, K., Buseck, P.R., 1984. Transmission electron microscopy ofthe “LOW-CA” hydrated interplanetary dust particle. Earth Planet. Sci.Lett. 69, 243–254.

Tomeoka, K., Buseck, P.R., 1985. Indicators of aqueous alteration in CMcarbonaceous chondrites: Microtextures of a layered mineral containingFe, S, O and Ni. Geochim. Cosmochim. Acta 49, 2149–2163.

Tomeoka, K., Buseck, P.R., 1988. Matrix mineralogy of the Orgueil CIcarbonaceous chondrite. Geochim. Cosmochim. Acta 52, 1627–1640.

Traxel, K., Arndt, P., Bohsung, J., Braun-Dullaeus, K.U., Maetz, M.,Reimold, D., Schiebler, H., Wallianos, A., 1995. The new Heidelbergproton microprobe: the success of a minimal concept. Nucl. Instr. andMeth. B 104, 19–25.

Treiman, A.H., Romanek, C.S., 1998. Bulk and stable isotopiccompositions of carbonate minerals in Martian meteorite Allan Hills84001: no proof of high formation temperature. Meteorit. Planet. Sci.33, 737–742.

Tsou, P., 1996. Hypervelocity capture of meteoroids in aerogel. In:Gustafson, B. _A.S., Hanner, M.S. (Eds.), Physics, Chemistry, andDynamics of Interplanetary Dust, ASP Conference Series 104 ASP,San Francisco, pp. 237–242.

Virag, A., Wopenka, B., Amari, S., Zinner, E., Anders, E., Lewis, R.S.,1992. Isotopic, optical, and trace element properties of large singleSiC grains from the Murchison meteorite. Geochim. Cosmochim. Acta56, 1715–1733.

Wakeham, S.G., SchaKner, C., Giger, W., 1980a. Polycyclic aromatichydrocarbons in recent lake sediments—I. Compounds havinganthropogenic origins. Geochim. Cosmochim. Acta 44, 403–413.

Wakeham, S.G., SchaKner, C., Giger, W., 1980b. Polycyclic aromatichydrocarbons in recent lake sediments—II. Compounds derived frombiogenic precursors during early diagenesis. Geochim. Cosmochim.Acta 44, 415–429.

Warren, P.H., 1998. Petrologic evidence for low-temperature, possibly.ood evaporitic origin of carbonates in the ALH84001 meteorite. J.Geophys. Res. 103, E16759–E16773.

Warren, J.L., Zolensky, M.E., 1994. Collection and curation ofinterplanetary dust particles recovered from the stratosphere by NASA.In: Zolensky, M.E., Wilson, T.L., Rietmeijer, F.J.M., Flynn, G.J.(Eds.), Analysis of Interplanetary Dust, AIP Conference Proceedings,Vol. 310. AIP, New York, pp. 245–254.

Werner, H.W., Morgan, A.E., 1976. Charging of insulators byion bombardment and its minimization for secondary ion massspectrometry (SIMS) measurements. J. Appl. Phys. 47, 1232–1242.

Werner, H.W., Warmoltz, N., 1984. Beam techniques for the analysis ofpoorly conducting materials. J. Vac. Sci. Technol. A 2, 726–731.

Wetherill, G.W., Chapman, C.R., 1988. Asteroids and Meteorites. In:Kerridge, J.F., Matthews, M.S. (Eds.), Meteorites and the Early SolarSystem. Univ. Arizona Press, Tucson, pp. 35–67.

Williams, P., 1998. Sputtering artifacts in SIMS depth pro7les. In: Gillen,G., Lareau, R., Bennett, J., Stevie, F. (Eds.), Secondary Ion MassSpectrometry. Proceedings SIMS XI. Wiley, Chichester, pp. 3–10.

Wing, M.R., Bada, J.L., 1991. Geochromatography on the parent bodyof the carbonaceous chondrite Ivuna. Geochim. Cosmochim. Acta 55,2937–2942.

Wittmaack, K., 1994. Electron impact postionization of sputtered neutrals:Basic aspects and recent progress. In: Benninghoven, A., Nihei, Y.,Shimizu, R., Werner, H.W. (Eds.), Secondary Ion Mass Spectrometry.Proceedings SIMS IX. Wiley, Chichester, pp. 309–315.

Wittmaack, K., 1998. Physical and chemical parameters determining ionyields in SIMS analyses: A closer look at the oxygen-induced yieldenhancement eKect. In: Gillen, G., Lareau, R., Bennett, J., Stevie,F. (Eds.), Secondary Ion Mass Spectrometry. Proceedings SIMS XI.Wiley, Chichester, pp. 11–18.

Zehnpfenning, J., Cramer, H.G., Heller, T., Niehuis, E., Rulle, H., Stephan,T., Benninghoven, A., 1994. Particle analysis by imaging TOF-SIMS.In: Benninghoven, A., Nihei, Y., Shimizu, R., Werner, H.W. (Eds.),Secondary Ion Mass Spectrometry. Proceedings SIMS IX. Wiley,Chichester, pp. 453–456.

Zenobi, R., Philippoz, J.-M., Buseck, P.R., Zare, R.N., 1989. Spatiallyresolved organic analysis of the Allende meteorite. Science 246,1026–1029.

Zinner, E., 1989. Isotopic measurements with the ion microprobe.Proceedings Workshop on New Frontiers in Stable Isotope Research,USGS Bulletin, Vol. 1890, pp. 145–162.

Zinner, E., 1991. Cometary and interstellar dust grains: Analysis by ionmicroprobe mass spectrometry and other techniques. Space Sci. Rev.56, 147–156.

Zinner, E., 1998. Trends in the study of presolar dust grains from primitivemeteorites. Meteorit. Planet. Sci. 33, 549–564.

Zinner, E., Crozaz, G., 1986. A method for the quantitative measurementof rare earth elements in the ion microprobe. Int. J. Mass Spectrom.Ion Process. 69, 17–38.

Zinner, E., Tang, M., Anders, E., 1987. Large isotopic anomalies of Si,C, N and noble gases in interstellar silicon carbide from the Murraymeteorite. Nature 330, 730–732.

Zinner, E., Tang, M., Anders, E., 1989. Interstellar SiC in the Murchisonand Murray meteorites: isotopic composition of Ne, Xe, Si, C, and N.Geochim. Cosmochim. Acta 53, 3273–3290.

Zinner, E., Amari, S., Anders, E., Lewis, R., 1991a. Large amounts ofextinct 26Al in interstellar grains from the Murchison meteorite. Nature349, 51–54.

Zinner, E., Amari, S., Lewis, R.S., 1991b. Silicon carbide from asupernova? (abstract). Meteoritics 26, 413.

Zolensky, M.E., 1997. Aqueous alteration of carbonaceous chondrites:evidence for asteroidal alteration. In: Zolensky, M.E., Krot, A.N., Scott,E.R.D. (Eds.), Workshop on Parent-Body and Nebular Modi7cation ofChondritic Materials LPI Technical Report 97-02, Part 1, Lunar andPlanetary Institute, Houston, pp. 70–71.

Zolensky, M.E., Wilson, T.L., Rietmeijer, F.J.M., Flynn, G.J. (Eds.),1994. Analysis of interplanetary dust. AIP Conference Proceedings,Vol. 310. AIP, New York. 357 p.

Zolotov, M., Shock, E., 1999. Abiotic synthesis of polycyclic aromatichydrocarbons on Mars. J. Geophys. Res. 104, E14033–E14049.

Zscheeg, H., Kissel, J., Natour, Gh. 1992. CoMA—a high resolutiontime-of-.ight secondary ion mass spectrometer (TOF-SIMS) for insitu analysis of cometary matter. Asteroids, Comets, Meteors 1991,Lunar and Planetary Institute, Houston, pp. 683–686.

tof-simsincosmochemistrygeosci.uchicago.edu/~tstephan/publications/tof... ·...

Documents