Identification of Mineral Impactors in Hypervelocity Impact Craters in Aluminum by Raman Spectroscopy of Residues

Identification of mineral impactors in hypervelocity impact craters in aluminum by Raman spectroscopy of residues

Item Type Article; text

Authors Burchell, M. J.; Foster, N. J.; Kearsley, A. T.; Creighton, J. A.

Citation Burchell, M. J., Foster, N. J., Kearsley, A. T., & Creighton, J. A. (2008). Identification of mineral impactors in hypervelocity impact craters in aluminum by Raman spectroscopy of residues. Meteoritics & Planetary Science, 43(12), 135-142.

DOI 10.1111/j.1945-5100.2008.tb00614.x

Publisher The Meteoritical Society

Journal Meteoritics & Planetary Science

Download date 03/10/2021 07:24:48

Item License http://rightsstatements.org/vocab/InC/1.0/

Version Final published version

Link to Item http://hdl.handle.net/10150/656386 Meteoritics & Planetary Science 43, Nr 1/2, 135–142 (2008) Abstract available online at http://meteoritics.org

Identification of mineral impactors in hypervelocity impact craters in aluminum by Raman spectroscopy of residues

M. J. BURCHELL1*, N. J. FOSTER1, A. T. KEARSLEY2, and J. A. CREIGHTON1

1Centre for Astrophysics and Planetary Sciences, School of Physical Sciences, Ingram Building, University of Kent, Canterbury, Kent CT2 7NH, UK 2Impact and Astromaterials Research Centre, Department of Mineralogy, The Natural History Museum, London SW7 5BD, UK *Corresponding author. E-mail: [email protected] (Received 20 February 2007; revision accepted 13 July 2007)

Abstract–Here we demonstrate the use of Raman spectroscopy techniques to identify mineral particle fragments after their impact into aluminum foil at ~6 km s−1. Samples of six minerals (olivine, rhodonite, enstatite, diopside, wollastonite, and lizardite) were fired into aluminum foil and the resulting impact craters were studied with a HeNe laser connected to a Raman spectrometer. Raman spectra similar to those of the raw mineral grains were obtained from the craters for impacts by olivine, rhodonite, enstatite, wollastonite, and diopside, but no Raman signals were found from lizardite after impact. In general, the impactors do not survive completely intact, but are fragmented into smaller fractions that retain the structure of the original body. Combined with evidence for SEM and FIB studies, this suggests that in most cases the fragments are relatively unaltered during impact. The survival of identifiable projectile fragments after impact at ~6 km s−1 is thus established in general, but may not apply to all minerals. Where survival has occurred, the use of Raman spectroscopic techniques for identifying minerals after hypervelocity impacts into a metallic target is also demonstrated.

INTRODUCTION of Kearsley et al. 2007a) a substantial residue fraction (>6%) can be found from even high-velocity impact (5–6 km s−1) The capture of cosmic dust in space can be achieved by a and can yield the compositions of stoichiometric minerals. A variety of means. Among the most simple is placing a smooth number of essentially nondestructive in situ analysis metallic (ductile) material in space and retrieving it at a later techniques can be used on crater residues, the most obvious date. Analysis of the surface will then show small impact being scanning electron microscopy and energy dispersive craters formed by the high-speed impacts. The speed of the X-ray microanalysis (SEM-EDX; e.g., Kearsley et al. 2007b) impact will depend on the relative speeds of the impactor and and time-of-flight secondary ion mass spectrometry (ToF- the target. In low Earth orbit (LEO), these speeds will SIMS; e.g., Hoppe et al. 2006), both of which can yield typically be 7–11 km s−1 for man-made debris and higher detailed elemental information on the composition of the (mean of 20–25 km s−1) for interplanetary dust particles residue, and in the latter case, isotopic analysis, albeit with (IDPs). Such impacts are in the speed range that is often some surface ablation. termed “hypervelocity,” meaning that the speed is above the 1 The recent return to Earth of cometary dust by the NASA or 2 km s−1 that is the normal maximum traditionally Stardust spacecraft (Brownlee et al. 2006) has generated achievable on Earth. The results of an impact on a solid target increased efforts in analysis of cosmic dust. Although most of in this speed regime are that a crater forms in the target, the the cometary particles returned by Stardust were captured in projectile is crushed, heated, melted, even vaporized, and aerogel (a low-density, porous solid), there were also many much of the projectile material may be removed from the craters in aluminum foils carried by the spacecraft (Hörz et al. impact site as ejecta. This makes identifying the composition 2006). One unusual feature of dust collection by Stardust was of the projectile difficult. However, in hypervelocity impacts, that the impact speed was fixed by the encounter speed of the it has long been known that some remnant of the impactor can spacecraft during its fly-by of the comet and is thus well line the resultant crater in the form of an impact melt, and on constrained as 6.1 km s−1, and is easily achievable in the some substrates (e.g., the borosilicate glass solar cell surfaces laboratory. Detailed analysis techniques have been developed

135 © The Meteoritical Society, 2008. Printed in USA. 136 M. J. Burchell et al.

Fig. 1. a) Soda lime glass residues in an impact crater. b) Wollastonite residues in impact craters. c) Olivine craters. d) Enstatite craters. All craters are in Al 1100 foil. (a) and (b) are secondary electron images taken at 2kV accelerating voltage, impact residues appear as dark areas. (c) and (d) are optical images from a Leica MZ stereo microscope. and tested in laboratory simulations of impacts at this speed to reflects the vibrational energy levels of the bond structure in study the impact residues in craters (e.g., Kearsley et al. the illuminated specimen. In many cases, the Raman 2007b), but these techniques rely on identifying the impactor by spectrum can be sufficiently distinct to serve as a means of its elemental composition. Here we assess whether an alternate identifying the sample, and there is a very extensive literature analysis method, Raman spectroscopy, can be successfully on this subject (e.g., Lewis and Edwards 2001; Smith and employed to recognize characteristic molecular bonding attributes Dent 2005), as well as several online spectral databases (e.g., within the projectile residue in hypervelocity impact craters on http://rruff.info/index.php). Use of laser illumination aluminum targets. combined with a microscope allows study of materials at the Accordingly, in this paper we briefly discuss previous micron scale and has been a standard tool of mineralogical applications of Raman spectroscopy to the study of solar analysis for many decades. It has been applied to minerals in system materials, present evidence that intact fragments of general (see, e.g., Frost et al. 1999 for a brief review) as well projectiles may survive impacts on metal surfaces, and as interplanetary dust particles (IDPs) (e.g., Wopenka 1988) include examples of impacts in the laboratory on Stardust- and meteoritic materials (e.g., Zinner et al. 1995). It is also grade aluminum foils at speeds of 6 km s−1. The impact particularly useful for the study of carbonaceous materials craters in these samples were then subjected to examination (see, e.g. Beyssac et al. 2003 and references therein). Use of under a microRaman system with a view to determining if a Raman techniques on particles captured in hypervelocity material structure based, rather than elemental based, impacts on aerogel was shown to be possible by Burchell identification of the impactor can be made from the residues. et al. (2001) for mineral grains and also for organic materials (Burchell et al. 2004, 2006). In the case of organics, this was RAMA SPECTROSCOPY OF MIERALS successfully applied to particles captured in the Stardust aerogel (Sandford et al. 2006). However, there has previously Raman spectroscopy involves the illumination of a been no demonstration as to whether impact crater residues in sample by a monochromatic light source. Some of the light is aluminum foil are even in principle amenable to Raman inelastically scattered and the spectrum of this component analysis. Identification of mineral impactors in hypervelocity impact craters in aluminium 137

Fig. 2. Ion-induced secondary electron image of residue (dark, shown arrowed) in a vertical FIB section cut into metal (bright) of the floor of an impact crater (basalt glass projectile at ~6.1 km s−1).

IMPACTS ON METALS Fig. 3. Backscattered electron image of vertical section of impact crater in aluminum foil from impact of diopside at 6 km s−1. The widely held belief that impact on metallic substrates Residues are shown arrowed. involves very high peak shock pressures may have contributed to this oversight. It has been implicitly assumed The recent use of ion-stimulated (e.g., Graham et al. that not only will there will be very little projectile material 2006) or low–accelerating voltage electron-stimulated retained in the crater, but that what is present will have secondary electron imagery has allowed direct visualization undergone shock processing at pressures of many GPa and of the distribution of impact residues from silicate minerals that any residue will be impact melt or will have condensed and glasses across the surface of aluminum foil impact from a vapor state (produced by the shock of the impact). For craters. From basaltic and artificial soda-lime glasses, there is the majority of impacts in LEO this may well be true, but in evidence of an extensive but patchy, smooth, quenched melt the case of the Stardust encounter, the impact velocity sheet stretching from the crater floor, across the walls, and (6.1 km−1) is likely to generate peak pressures less than onto the uplifted rim. This is confirmed here by study of 80 GPa, which is insufficient to completely melt, let alone impacts at 6 km s−1 onto aluminum foil with soda-lime glass vaporize, typical silicate impactors. The role of complex and wollastonite projectiles (Figs. 1a and 1b show soda-lime internal grain boundaries may also result in inhomogeneous glass and wollastonite residues in craters). Focused ion beam partitioning of strain within the impactor, with “sacrificial” (FIB) trenches through basalt glass residue (Fig. 2) show that melting and vaporization in parts, and survival of solid (albeit the impactor melt may be interlayered with aluminum target deformed) fragments from others. material. Angular glass impactor fragments were also Experimentally, a study of cratering by glass projectiles abundant, adhering to the internal crater walls. As the original in aluminum by Hörz et al. (1995) showed that at 2 km s−1, projectile was amorphous (i.e., had no discernible crystalline projectile fragments could still be found in the crater, but that structure), it is very difficult to assign diagnostic features above this speed, melted glass dominated the residue. Indeed, other than gross morphology, but the suggestion is that some at speeds of 6 km s−1 this impact melt was itself significantly of the original fragment survived unmelted. Crystalline reduced in quantity, with none at the bottom of the craters and silicate impactors also create a sheet of quenched melt, but most of it having been ejected. It may thus seem highly their craters also often contain remnant crystalline fragments unlikely that an analysis technique (such as Raman of the impactor, either embedded within melt or as coarse spectroscopy) requiring projectile fragments at greater than clumps (Fig. 3). The retention of oriented, characteristic the micron scale would be applicable. However, Hernandez crystal cleavage morphology in residues of wollastonite et al. (2006) reported that they observed discrete projectile (Fig. 4) strongly suggests that the stoichiometric remnants fragments of metal (stainless steel) projectiles in craters in the have not recrystallized from melt, but are genuine solid laboratory for impact speeds of up to 5 km s−1, and that survivors from the original particle. hydrocode simulations suggested these should still be present In separate work, transmission electron microscopy of at 6 km s−1. In their work, fragment size and number density Stardust small crater sections prepared by FIB microscopy decreased as impact speed increased, but fragments were still also shows the survival of crystalline silicate and sulfide found at the bottom of the crater as well as on the walls. The mineral grains within quenched melt (Zolensky et al. 2006; targets were nickel, copper, and stainless steel (compared to Borg et al. 2007). In the case of mineral projectiles (rather aluminum for Stardust), but the results do suggest that than metals or glasses), such fragments may be amenable to projectile fragments (rather than melt) may still line craters Raman analysis. Indeed, Raman spectroscopy can be a even in impacts at 6 km s−1. sensitive indicator of the extent of impact shock 138 M. J. Burchell et al.

Fig. 4. Fine structure on walls of crater in aluminum foil made by impact of wollastonite at 6 km s−1. The crater is shown full size (left) and the region highlighted is expanded (right). In the expanded view fine elongate fragments are shown arrowed (apparently embedded in impact melt) with similar orientation to the main axis of the original (elongated) crater.

Table 1. found in meteorites and which gives a moderate strength Approximate Raman signals), wollastonite (CaSiO3, which we use as a Impact speed grain size Crater size projectile with high aspect ratio as it cleaves readily into Projectile (km s−1) (μm) (μm) elongated fragments and is fairly refractory), and lizardite Olivine 5.89 75–175 220–300 (Mg3Si2O5[OH]4, a phyllosilicate with about 13% water Olivine and 6.06 75–90 and 53–125 250–300 content, and a polymorph of serpentine). Although most shots rhodonite had only one mineral type per shot, one shot included a Enstatite 5.85 >38 60–300 mixture of rhodonite and olivine as a blind test of the analysis. Diopside 6.01 <125 200 Also given in Table 1 are the particle sizes used along with Wollastonite 5.74 20–350 100–150 the resultant crater sizes. These sizes are large compared to Lizardite 5.85 53–125 ~300 most of the craters that impacted the Stardust spacecraft; however, 63 craters larger than 20 μm in diameter have been processing, for example, in feldspars, where the distinctive identified so far, seven of which (sized 44–142 μm in fine peak structure becomes suppressed as maskelynite is diameter) have been studied in detail (Kearsley et al. 2008). formed. Targets were sheets of flight spare Stardust aluminum foil. This was type Al l100, with a thickness of 0.1 mm and a METHOD surface area of 1 × 1 cm (see Kearsley et al. 2007b for further details). The targets were mounted on a larger aluminum plate The impacts were generated using a two-stage light-gas and placed in the gun so that impacts were at normal gun at the University of Kent (Burchell et al. 1999). As well incidence. During the shots, the target chamber was as firing single projectiles in a shot, this gun can also fire a evacuated to typically 0.2 mbar. cloud of projectiles in a “buckshot” (mono-dispersive) or The Raman system used a Jobin-Yvon microRaman “powder” (poly-dispersive) type of shot. This was done for module coupled with an Olympus BX40 microscope. It also the experiments reported herein, yielding several craters to comprised a HR640 spectrometer with 1200 g/mm grating analyze from each shot. The shot speed was found in each and a liquid nitrogen cooled CCD (1024 × 256 pixels). shot using passage of the projectiles through laser light Illumination was from a HeNe (632.8 nm), laser which curtains (see Burchell et al. 1999 for details) and the delivered 3 mW of power at the sample in a spot size of 4– dispersion of speeds across the cloud of projectiles was less 5 microns, yielding a power intensity at the sample of than 2%. Details of the shot program are given in Table 1. approximately 189 MW m−2. Collection times for raw grains Projectile materials were selected to include a variety of were typically 4 × 30 s and 8 × 30 s for craters. minerals, i.e., olivine ([Mg,Fe]2SiO4, a common mineral, found in meteorites and with a high melting point), diopside RESULTS (MgCaSi2O6, a rock-forming mineral found in meteorites and the Ca-Mg end member of the pyroxene family), rhodonite After each shot, the target foils were examined and ([Mn,Fe,Mg, Ca]SiO3, a mineral used as an internal impact craters found. Typical craters are shown in Figs. 1 standard in our laboratory and which gives strong Raman and 4. Craters were usually 60 to 300 microns across. A signals), enstatite (MgSiO3, the Mg endmember pyroxene, crater-projectile size calibration for impacts at this speed in Identification of mineral impactors in hypervelocity impact craters in aluminium 139

Fig. 5. Raman spectra from craters in Al 1100 foils from made from impacts of (a) olivine, (b) rhodonite, (c) diopside, (d) enstatite, and (e) wollastonite projectiles. In each case the spectrum from the crater is the upper trace, accompanied by a lower trace of a spectrum from a raw grain of the same material pre-shot.

Al 1100 was given in Kearsley et al. (2006). It suggests that (pre-shot) with the signals obtained from the craters. crater rim-rim diameters should be 4.6 times projectile Several craters on each target were studied in this fashion; diameter for glass impactors with density 2.4 g cm−3, not all gave Raman spectra. For example, on the first extrapolation from density-dependent crater scaling olivine shot one out of the three craters studied gave a (Kearsley et al. 2007b) gives approximately 5 times good signal (shown here), one a weak signal (just above projectile diameter for the five minerals in our experiments. noise levels) and one no signal. It may be possible that Using this relation suggests that the impactors here were of longer integration times on the spectrometer would improve order 12 to 60 μm in diameter, which is roughly in line with these statistics. In the shot which fired both rhodonite and expectations (Table 1). olivine, all craters studied gave spectra, several for olivine Raw Raman spectra were obtained from each mineral and several for rhodonite. Given the size of the craters sample before firing. After a shot the craters were then compared to the Raman beam size, it was possible to take examined with the Raman system. For the olivine, spectra from several distinct sites inside individual craters. rhodonite, and diopside shots, identifiable Raman spectra of Raman spectra showing rhodonite, for example, were the projectile material were readily obtained from the obtained from the floor (bottom) of a crater and from the craters. Examples are given in Fig. 5. There is a clear rising walls (edges), indicating a wide dispersion of the identification of the Raman peaks from the raw grains residue on the crater walls. 140 M. J. Burchell et al.

For the enstatite shot, initially no Raman signals for particularly to extend the range of appropriate mineral species enstatite were obtained from the craters. Re-examination of the (such as feldspars and sulfides) or to various members inside craters with longer accumulation times (15 × 60 s) did then a mineral family (for example, Kuebler et al. 2005 have result in successful acquisition of an enstatite spectrum from shown that it is possible to use Raman spectroscopy on raw a 68 μm diameter crater. This spectrum (Fig. 5d) was grains to sensitively determine the composition of olivines in obtained from the internal wall of the crater, where a feature terms of their Mg and Fe content; in future work we will whose morphology was suggestive of an accumulation of compare the shifts and any broadening from residual strain residue had been seen under the microscope. It may be that for with those due to varying family member composition), to some materials, longer integration times combined with increase the sensitivity of the Raman analysis (for example, selectivity in the site to be illuminated are thus required to by longer accumulation times to see if this reveals any weak obtain signals above the noise level. By contrast, the spectral features), and using impactors of varying sizes to wollastonite impact craters that were studied only gave good determine the size dependence of structural preservation. Raman spectra for wollastonite from one out of six craters Preliminary further work using smaller projectiles of olivine, (using the same integration times as for olivine, rhodonite, and has shown that craters just 25 μm in diameter (and hence diopside). projectile diameters of approximately 5 μm) still yield In the lizardite shots, no Raman spectra showed distinct recognizable Raman spectra for olivine. peaks indicative of the projectiles. Several craters were tested The lack of distinctive Raman peaks of any kind from and multiple sites were tested in the craters and on their rims, lizardite impact residue is a particularly significant result as it with longer accumulation times tried, but still no spectra for is a member of the serpentine group of hydrous phyllosilicate lizardite were obtained, nor indeed were any distinctive minerals and decomposes with loss of water at a relatively Raman peaks found in the observed spectra. Raw grains of low temperature, usually with generation of olivine. lizardite did, however, readily give Raman spectra. Explaining the absence of a signal from the lizardite is not straightforward. Material strength and melting point for DISCUSSION example are a function of pressure, and, during impacts, extreme pressures occur for short durations. Therefore a full Following the work of Hernandez et al. (2006) the simulation of the material response to extreme shock suggestion arose that discrete grains of projectile material pressures is required. It may be that lizardite as a may survive impacts at speeds above 5 km s−1. As indicated in phyllosilicate is less prone to survival under these impact the introduction here, at speeds of 6 km s−1, laboratory conditions. It appears that this mineral group, common in impacts into aluminum foil now provide strong evidence for Type 1 and 2 carbonaceous chondrite meteorites and usually discrete regions in the impact melt lining craters, which is attributed to hydration of anhydrous mafic silicate suggestive of fragments of the incident projectile. precursors during parent-body (asteroidal) processing, Furthermore, FIB cross sections from Stardust craters clearly may not survive impact under Stardust encounter demonstrate survival of crystalline material (see above). The conditions in a form that can be recognized by Raman result found here is that for olivine, diopside, enstatite, and spectroscopy of craters. It thus may not be possible to use rhodonite impacting at 6 km s−1, Raman spectra from residue Raman spectroscopy to distinguish residue from after the impact can be obtained with the same characteristics phyllosilicates (indicating substantial aqueous activity) from as the impactor. This does not a priori confirm that amorphous anhydrous mafic silicate glass of pre-nebular macroscopic fragments of projectile survive intact at these origin. Further quantitative SEM-EDS work is needed on the speeds. It could be argued that re-crystallized melt has lizardite craters to see if their residue contents give a different formed. However, this would have to retain the same analytic totals to the original projectiles, and whether this can composition and structure as the original material and we be attributed to loss of water or even generation of anhydrous consider it unlikely. Furthermore, the projectile would have silicate mineral species to be heated to a sufficiently high enough temperature In summary, however, it is clear that several of the most (compared to its melting point) for a prolonged period for full common minerals to be found in solar system bodies, melting of the entire projectile to have occurred. It is more especially olivine and pyroxenes, can survive as identifiable likely that the projectile has been partly melted and partly crystalline residues after hypervelocity impacts at 6 km s−1. broken into fragments, with local regions of higher/lower pressure and temperatures occurring non-uniformly across CONCLUSIONS the body. In addition, rapid cooling will have occurred due to contact with the metal target with a lower melting point. The results show that for 5 minerals (olivine, diopside, The success of Raman spectroscopy on aluminum craters rhodonite, wollastonite, and enstatite), Raman spectra may currently applies to five minerals: olivine, diopside, readily be obtained after a projectile strikes an aluminum rhodonite, wollastonite, and enstatite. More work is required, target at speeds of some 6 km s−1. For some minerals, all of Identification of mineral impactors in hypervelocity impact craters in aluminium 141 the craters studied readily gave recognizable Raman spectra REFERENCES for the impactor. In others, however, longer integration times were needed to produce spectra and some yielded no Beyssac O., Goffe B., Petitet J.-P., Froigneux E., Moreau M., and recognizable spectra, even though neighboring craters in the Rouzard J.-N. 2003. On the characterization of disordered and same shot did so. This suggests that either the survival of heterogeneous carbonaceous materials by Raman spectroscopy. crystalline fragments is not uniform or that they may be Spectrochimica Acta Part A 59:2267–2276. Borg J., Hörz F., Bridges J. C., Burchell M. J., Djouadi Z., Floss C., buried in the impact melt below a layer of aluminum (the Graham G. A., Green S. F., Heck P. R., Hoppe P., Huth J., target material). The smallest craters so far studied that gave Kearsley A., Leroux H., Marhas K., Staderman F. J., and Raman signals for the impactors were 25 μm in diameter, Teslich N. 2007. SEM-EDS analyses of small craters in Stardust corresponding to grains of olivine approximately 5 μm aluminum foils: Implications for the Wild 2 dust distribution in diameter. In the case of one mineral, lizardite, no (abstract #1592). 38th Lunar and Planetary Science Conference. CD-ROM. Raman spectra were obtained from the impact craters. Brownlee D., Tsou P., Aléon J., Alexander C. M. O’D., Araki T., With just one counter-example, it is hard to be definitive Bajt S., Baratta G. A., Bastien R., Bland P., Bleuet P., Borg J., as to the cause, but it may lie in the delicate Bradley J. P., Brearley A., Brenker F., Brennan S., Bridges J. C. crystallographic structure of this phyllosilicate projectile Browning N., Brucato J. R., Brucato H., Bullock E., Burchell (with consequent weak bonding between planes) or the M. J., Busemann H., Butterworth A., Chaussidon M., Cheuvront A., Chi M., Cintala M. J, Clark B. C., Clemett S. J., presence of a significant water content altering the behavior Cody G., Colangeli L., Cooper G., Cordier P. G., Daghlian C., on impact. Dai Z., D’Hendecourt L., Djouadi Z., Dominguez G., These results suggest that projectile fragments may Duxbury T., Dworkin J. P., Ebel D., Economou T. E., Fairey survive impact, but that this may not be true for all materials. S. A. J., Fallon S., Ferrini G., Ferroir T. Fleckenstein H., Floss C., This novel result is of interest in its own right. It both changes Flynn G., Franchi I. A., Fries M., Gainsforth Z., Gallien J.-P., Genge M., Gilles M. K., Gillet P., Gilmour J., Glavin D. P., the view as to the degree of melt and alteration in impacts at Gounelle M., Grady M. M., Graham G. A., Grant P. G., Green these speeds and also introduces a new analysis technique for S. F., Grossemy F., Grossman L., Grossman J., Guan Y., studying impact crater residues. It suggests that the NASA Hagiya K., Harvey R., Heck P., Herzog G. F., Hoppe P., Stardust space mission analysis program should use Raman Hörz F., Huth J., Hutcheon I. D., Ishii H., Ito M., Jacob D., techniques to examine the Al 1100 foils exposed during the Jacobsen C., Jacobsen S., Joswiak D., Kearsley A. T., Keller L., Khodja H., Kilcoyne A. L. D., Kissel J., Krot A., Langenhorst F., flyby of comet 81P/Wild 2 in 2004 (at an encounter speed of Lanzirotti A., Le L., Leshin L., Leitner J., Lemelle L., Leroux H., 6.1 km s−1). There are many large craters reported on those Liu M.-C., Luening K., Lyon I., MacPherson G., Marcus M. A., foils, with at least seven greater than 50 microns in size (i.e., Marhas K., Matrajt G., Meibom A., Mennella V., Messenger K., in the size range used here) already studied in great detail with Mikouchi T., Mostefaoui S., Nakamura T., Nakano T., SEM-EDS. A major portion of the grains impacting Stardust Newville M., Nittler L. R., Ohnishi I., Ohsumi K., Okudaira K., Papanastassiou D. A., Palma R., Palumbo M. O., Pepin R. E., were silicate materials such as olivine and pyroxene Perkins D., Perronnet M., Pianetta P., Rao W., Rietmeijer F., (Brownlee et al. 2006; Hörz et al. 2006) and olivine and Robert F., Rost D., Rotundi A., Ryan R., Sandford S. A., diopside (the Ca-Mg end member of the pyroxene family) Schwandt C. S., See T. H., Schlutter D., Sheffield-Parker J. A., provided positive results in the analysis here. Such an analysis Simionovici S., Sitnitsky S. I., Snead C. J., Spencer M. K., applied to Stardust craters would offer the chance of a Stadermann F. J., Steele A., Stephan T., Stroud R., Susini J., Sutton S. R., Taheri M., Taylor S., Teslich N., Tomeoka K., spectroscopic identification of the impactors based on their Tomioka N., Toppani A., Trigo-Rodríguez J. M., Troadec D., structure, rather than just on elemental composition, as is Tsuchiyama A., Tuzolino A. J., Tyliszczak T., Uesugi K., common at present. Although the bulk of the mass (over 90%) Velbel M., Vellenga J., Vicenzi E., Vincze L., Warren J., that hit Stardust occurred in the larger particles (and hence Weber I., Weisberg M., Westphal A. J., Wirick S., Wooden D., larger craters), the great majority of particles (and craters) are Wopenka B., Wozniakiewicz P. A., Wright I., Yabuta H., Yano H., Young E. D., Zare R. N., Zega T., Ziegler K., much smaller. However, for particles of 1 μm and below it is Zimmerman L., Zinner E., and Zolensky M. 2006. Comet Wild not yet clear if this technique can be applied as an analysis 2 under a microscope. Science 314:1711–1716. tool. Burchell M. J., Cole M. J., McDonnell J. A. M., and Zarnecki J. C. 1999. Hypervelocity impact studies using the 2 MV Van de Acknowledgments—We thank the Particle Physics and Graaff accelerator and two-stage light-gas gun of the University of Kent at Canterbury. Measurement Science Astronomy Research Council for grants to operate the light- Technology 10:41–50. gas gun. We thank Mr. M. Cole for the firing of the gun. Burchell M. J., Creighton J. A., Cole M. J., Mann J., and Kearsley F. Hörz is thanked for supplying the Stardust-grade aluminum A. T. 2001. Capture of particles in hypervelocity impacts in foil. Richard Chater of the Department of Materials at Imperial aerogel. Meteoritics & Planetary Science 36:209–221. College London kindly prepared and imaged the FIB trench Burchell M. J., Creighton J. A., and Kearsley A. T. 2004. Identification of organic particles via Raman techniques after shown in Fig. 2. N. J. F. thanks the University of Kent alumni capture in hypervelocity impacts on aerogel. Journal of Raman for funding. Spectroscopy 35:249–253. Burchell M. J., Mann J., Creighton J. A., Kearsley A. T., Graham G., Editorial Handling—Dr. Scott Sandford and Franchi I. A. 2006. Identification of minerals and 142 M. J. Burchell et al.

meteoritic materials via Raman techniques after capture in P. R., Huth J., Stadermann F. J., Floss C., Marhas K., Stephan T., hypervelocity impacts on aerogel. Meteoritics & Planetary and Leitner J. 2008. Dust from comet Wild 2: Interpreting Science 41:217–232. particle size, shape, structure, and composition from impact Frost R., Kloprogge T., and Schmidt J. 1999. Non-destructive features on the Stardust aluminum foils. Meteoritics & identification of minerals by Raman microscopy. Internet Planetary Science 43. This issue. Journal of Vibrational Spectroscopy 3, edition 4. http:// Kuebler B. L., Jolliff B. L., Wang A., and Haskin L. A. 2005. www.ijvs.com/volume3/edition4/section1.html#Feature. Extracting olivine (Fo-Fa) compositions from Raman spectral Graham G. A., Teslich N., Dai Z. R., Bradley J. P., Kearsley A. T., peak positions (abstract #2086). 36th Lunar and Planetary and Hörz F. 2006. Focused ion beam recovery of Science Conference. CD-ROM. hypervelocity impact residue in experimental craters on Lewis I. R. and Edwards H. G. M. 2001. Handbook of Raman metallic foils. Meteoritics & Planetary Science 41:159–165. spectroscopy. New York: CRC Press. 1072 p. Hernandez V. S., Murr L. E., and Anchondo I. A. 2006. Experimental Sandford S. A., Aléon J., Alexander C. M. O’D., Araki T., Bajt S., observations and computer simulations for metallic projectile Baratta G. A., Borg J., Brucato J. R., Burchell M. J., fragmentation and impact crater development in thick metal targets. Busemann H., Butterworth A., Clemett S. J., Cody G., International Journal of Impact Engineering 32:1981–1999. Colangeli L., Cooper G., D’Hendecourt L., Djouadi Z., Dworkin Hoppe P., Stadermann F. J., Stephan T., Floss C., Leitner J., Marhas J. P., Ferrini G., Fleckenstein H, Flynn G. J., Franchi I. A., K. K., and Hörz F. 2006. SIMS studies of Allende projectiles Fries M., Gilles M. K., Glavin D. P., Gounelle M., Grossemy F., fired into Stardust-type aluminum foils at 6 km/s. Meteoritics & Jacobsen C., Keller L. P., Kilcoyne A. L. D., Leitner J., Planetary Science 41:197–209. Matrajt G., Meibom A., Mennella V., Mostefaoui S., Nittler Hörz F., Cintala M. J., Bernhard R. P., Cardenas F., Davidson W. E., L. R., Palumbo M. E., Robert F., Rotundi A., Snead C. J., Haynes G., See T. H., and Winkler J. L. 1995. Penetration Spencer M. K., Steele A., Stephan T., Tyliszczak T., Westphal experiments in aluminum 1100 targets using soda-lime glass A. J., Wirick S., Wopenka B., Yabuta H., Zare R. N., and projectiles. NASA TM-104813. 328 p. Zolensky M. 2006. Organics captured from comet Wild 2 by the Hörz F., Bastien R., Borg J., Bradley J. P., Bridges J. C., Brownlee Stardust spacecraft. Science 314:1720–1724. D. E., Burchell M. J., Cintala M. J., Dai Z. R., Djouadi Z., Smith E. and Dent G. 2005. Modern Raman spectroscopy: A Dominguez G., Economou T. E., Fairey S. A. J., Floss C., practical approach. Hoboken, New Jersey: Wiley. 210 p. Franchi I. A., Graham G. A., Green S. F., Heck H., Hoppe P., Wopenka B. 1988. Raman observations on individual Huth J., Ishii H., Kearsley A. T., Kissel J., Leitner J., Leroux H., interplanetary dust particles. Earth and Planetary Science Marhas M., Messenger K., Schwandt C. S., See T. H., Snead S., Letters 88:221–231. Stadermann F. J., Stephan T., Stroud R., Teslich N., Trigo- Zinner E., Amari S., Wopenka B., and Lewis R. S. 1995. Interstellar Rodríguez J. M., Tuzzolino A. J., Troadec D., Tsou P., Warren J., graphite in meteorites: Isotopic compositions and structural Westphal A., Wozniakiewicz P. J., Wright I., and Zinner E. 2006. properties of single graphite grains from Murchison. Meteoritics Impact features on Stardust: Implications for Wild 2 dust. 30:209–226. Science 314:1716–1719. Zolensky M. E., Zega T. J., Yano H., Wirick S., Westphal A. J., Kearsley A. T., Burchell M. J., Hörz F., Cole M. J., and Schwandt Weisberg M. K., Weber I., Warren J. L., Velbel M. A., C. S. 2006. Laboratory simulations of impacts on aluminum foils of Tsuchiyama A., Tsou P., Toppani A., Tomioka N., Tomeoka K., the Stardust spacecraft: Calibration of dust particle size from comet Teslich N., Taheri M., Susini J., Stroud R., Stephan T., Wild 2. Meteoritics & Planetary Science 41:167–180. Stadermann F. J., Snead C. J., Simon S. B., Simionovici A., See Kearsley A. T., Graham G. A., McDonnell J. A. M., Taylor E. A., T. H., Robert F., Rietmeijer F. J. M., Rao W., Perronnet M. C., Drolshagen G., Chater R. J., McPhail D., and Burchell M. J. Papanastassiou D. A., Okudaira K., Ohsumi K., Ohnishi I., 2007a. The chemical composition of micrometeoroids impacting Nakamura-Messenger K., Nakamura T., Mostefaoui S., upon the solar arrays of the Hubble Space Telescope. Advances Mikouchi T., Meibom A., Matrajt G., Marcus M. A., Leroux H., in Space Research 39:590–604. Lemelle L., Le L., Lanzirotti A., Langenhorst F., Krot A. N., Kearsley A. T., Graham G. A., Burchell M. J., Cole M. J., Dai Z., Keller L. P., Kearsley A. T., Joswiak D., Jacob D., Ishii H., Teslich N., Chater R. J., Wozniakiewicz, Spratt J., and Jones G. Harvey R., Hagiya K., Grossman L., Grossman J. N., Graham. 2007b. Analytical scanning and transmission electron microscopy G. A., Gounelle M., Gillet Ph., Genge M. J., Flynn G., Ferroir T., of laboratory impacts on Stardust aluminum foils: Interpreting Fallon S., Ebel D. S., Dai Z. R., Cordier P., Clark B., Chi M., impact crater morphology and the composition of impact Butterworth A. L., Brownlee D. E., Bridges J. C., Brennan S., residues. Meteoritics & Planetary Science 42:191–210. Brearley A., Bradley J. P., Bleuet P., Bland P. A., and Bastien R. Kearsley A. T., Borg J., Graham G. A., Burchell M. J., Cole M. J., 2006. Mineralogy and petrology of comet 81P/Wild 2 nucleus Leroux H., Bridges J. C., Hörz F., Wozniakiewicz P. J., Bland samples. Science 314:1735–1739. P. A., Bradley J. P., Dai Z. R., Teslich N., See T., Hoppe P., Heck