Introduction

Many scientists have waited a long time to be able to study cosmic dust in the their own laboratories. In the nineteenth century, the existence of interstellar matter was discovered. Evidence of this matter in space was looked for in meteorites, but scientific equipment was not advanced enough to differentiate between matter from space and matter from Earth. In the 1910s, stable isotopes were discovered (Thomson, 1912), which sparked a search for cosmic dust. Scientists started getting results in the early sixties (Reynolds, 1960). Presolar grains can be identified by isotopic ratios different from our Sun’s ratios (Bernatowicz & Walker, 1997). The first evidence for interstellar material in meteorites was found in 1973 (Clayton et al, 1973). In the mid-eighties, presolar grains were found inside meteorites (Wasserburg and Papanastassiou, 1982). This, along with the laboratory study of interplanetary dust, has opened up the field of materials astronomy, in which astrophysical objects are studied with microscopes as well as telescopes.

Interstellar graphite onions are one of the few presolar relics that have survived the trip to earth. They are called onions because they are round and their outside portions are layered. These graphite onions make up about one third of all presolar graphite (Zinner, 1998).

More than 80% of interstellar graphite onions have a strange core formed of carbon material that does not contain much order. Images of these cores suggest randomly oriented, fine crystals. Diffraction indicates graphene sheets, or graphenes, which are atom-thick sheets of carbon in hexagonal rings. Basically, graphene is graphite that does not have layering in the third dimension. (Bernatowicz et al, 1996).

Diffraction patterns suggest that the mass-weighted average diameter for graphene sheets is between three and four nanometers (Bernatowicz, Gibbons, Amari, & Lewis, 1995). Graphenes of this size contain several hundred carbon atoms and comprise the bulk of the mass in the cores of the interstellar graphite onions. Only one quarter of the mass comes from graphene sheets with diameters less than or equal to a nanometer. However, these tiny graphenes outnumber the large ones by far (Bernatowicz et al, 1996).

Materials astronomers have no simple explanation why the onion cores do not show this layering. Although curvature of the sheets (Bernatowicz et al, 1995), for example due to the occasional replacement of hexagonal rings in the graphene sheet with pentagonal or heptagonal ones (Bernatowicz et al, 1996) has been suggested as an explanation, this does not explain the abrupt transition to layered growth between core and rim. Moreover, the observed 4 nm coherence widths are much larger than the coherence widths of layered carbon nanotubes, and work by Wackenhut here shows that the curvature due to even a single pentagonal insert would destroy this coherence.

The cores of interstellar graphite onions have often been compared to polycyclic aromatic hydrocarbons (PAHs), which are a type of organic molecule that are sheets of carbon atoms with hydrogen at the edges (Messenger et al, 1995). PAHs have been discovered to contribute less than 10% of interstellar graphite onion cores’ volume (Bernatowicz et al, 1995) but 5 to 25% of the core mass (Bernatowicz et al, 1996).

The rims of these spherical graphite onions consist of three-dimensional van der Waals bonded graphite layers with 0.34 nm spacing (as shown in Fig. 1) that curve like the layers of an onion. The three-dimensional graphite of the rim always surrounds the two-dimensional graphite of the core, never vice versa. Three fourths of interstellar graphite onions have been found to have rim thicknesses between 0.1 and 0.4 microns. Small graphite onions do not have direct correlation between their size and rim thickness, while onions with a diameter of about one micron or more have thicker rims than smaller onions (Bernatowicz et al, 1996).

Astronomical observation has estimated the mean interstellar grain size to be 0.025 to 0.25 microns (Amari, Lewis, & Anders, 1994); however, the mean size for a graphite onion is 1.5 microns. The smallest interstellar graphite onion has been measured to be 0.3 microns (Bernatowicz et al, 1996), while the largest measured is 20 microns (Zinner, 1998).

The density of onion cores is found to be greater than that of amorphous carbon and a little less than the density of regular graphite (Bernatowicz et al, 1996). The density of the whole onion ranges from 2 to 2.2 grams per cubic centimeter (Anders & Zinner. 1993).

Experimental Methods

Because presolar grains are such a small part of meteorites (5 parts per million), they were previously overlooked. Scientists dissolved meteorites with acid to study the isotopic composition of selected components. The presolar grains identified so far are highly insoluble and hence did not dissolve with the rest of the meteorite. Scientists soon discovered that the mound of dust left over was mostly material that outdated our solar system (Bernatowicz & Walker, 1997).

Other types of presolar dust in meteorites go through many purification processes. These diminish the number of particles but decrease their contamination. Because interstellar graphite onions are so rare in meteorites, their quantity is preserved, independent of purity. Still, many small graphite onions are probably lost in their retrieval (Amari et al, 1994). The interstellar graphite onion samples studied today came from only two meteorites, Murchison and Tieschitz (Zinner, Amari, Wopenka, & Lewis, 1995). Whole micron-sized grains are kept for study in scanning electron microscopes and for ion microprobe analysis. Most samples used in transmission electron microscopes (TEMs) are separated by size, cut with a diamond ultramicrotome knife to about 100 nm thick, and deposited on a TEM grid (Bernatowicz & Walker, 1997).

Observations

I took a survey of the specimen of sliced onions KFC1AE from the Murchison meteorite in a 300 kV Philips TEM. Example of a cleanly sliced onion is shown in Fig. 2. Fifty-six onions were noted, described, measured, and mapped. Measurements were approximated using the 5-mm diameter ring on the screen of the microscope for reference. Mapping was done by marking the position of the specific onion on a low magnitude image of the grid square.

Data from this survey is available on the web. The size distribution of sliced onions ranged from 0.4 to 2.2 microns in diameter with a mean of about one micron. More than 90% of the onions seen showed signs of a core. Regions of thin core material suitable for high-resolution electron microscopy were also identified. High-resolution images taken in our lab in some cases (cf. Fig. 3) show the smooth graphite (hk0) fringes characteristic of core material.

Since slicing of the onions alters the size distribution and the core-rim ratio of the sample, we requested a sample of small, whole graphite onions deposited on a TEM grid from Roy Lewis at the University of Chicago. A survey has revealed thirteen whole onions, which have been noted, measured, and mapped. Measurements have been carried out the same way, but mapping was done by drawing the sample. For the whole onions, the size distribution has ranged from 0.3 to 3.5 microns with a mean of about 1.4 microns. We took a series of dark field images while rotating the incident beam to help us search for cores in the whole onions. Some of these evidenced inclusions of the sort expected for non core-rim onions, as shown in Fig. 4, while others evidenced a decrease in (002) diffraction toward the particle center even though no sharp core-rim boundary could be observed (cf. Fig. 5).

One of the smaller whole onions exhibited an obvious core, as shown in Fig. 6. A tilt of the specimen by 15 degrees has shown this onion’s core to be spherically symmetric, while its rim is not. Its diameter was an average of 0.5 microns. At the rim’s thickest point, this onion’s core-rim ratio is 0.44, but at the rim’s thinnest point, the onion’s core-rim ratio is 0.63.

DISCUSSION

The sliced onions without a core are likely the slices from the very top and bottom of the spherules. The many whole onions without a core are much more mysterious. At least two possible conclusions can be drawn from the lack of cores in the whole onion sample. First, onions from the unsliced set we were provided do not usually have cores. Alternatively, cores are difficult to detect in larger onions. In either case, the onion with an obvious core remains the exception in our collection of unsliced onions, even though the majority of onions in the set microtomed by Tom Bernatowicz are of the core-rim type.

David Dawkins, during his research here, deduced that the thickness of the rim of particles, subject to the expulsion by radiation during rim formation near the surface of the star, might depend on core-radius r, according to the simple relationship:

where D=deposition rate of graphite, L=luminosity of the star, r=radius of the particle, R=distance from the star to the particle, and r*=maximum radius of a particle (Dawkins, 1996). For red giants, r* is much greater than 1 micron, so the thickness of the rim is proportional to r^-1/2. For our single onion with a well-defined core, the average constant of proportionality is 0.12. The resulting deposition rate, so inferred, exceeds the rate one might expect in regions of “average density” in 2000 degree Celsius regions of a red giant atmosphere. If these particles do form in stellar atmospheres, regions of significantly increased carbon gas density are likely required.

Overall, we have inventoried 56 microtomed and 13 unsliced interstellar graphite onions. A subset of the former have regions of torn core material that ramp down to thicknesses less than the 10 nm suitable for high resolution study. Some images taken from these regions already show the power spectrum signature of the core material, and work to both analyze the contrast and compare with simulations is presently underway. For example, is the relationship between graphene sheets dendritic, is it random, or does it show evidence of bending due to pentagonal or heptagonal loop defects in the sheets? This information is crucial to understanding the processes by which they are formed, as well as the nature of the abrupt transition to layered growth in the onion rims. The unsliced onions show promise for providing core-rim ratios not broadened by the onion slicing process, which helps for comparison to models of transport out of the star’s gravitational well by radiation pressure after they are formed. More statistics are needed for firm conclusions at this point in time.

References

Amari, S., Lewis, R.S., Anders, E. (1994). Interstellar Grains in Meteorites: I. Isolation of SiC, Graphite, and Diamond; Size Distributions of SiC and Graphite. Geochimica et Cosmochimica Acta, 58, 459-70.

Anders, E., Zinner, E. (1993). Interstellar Grains in Primitive Meteorites: Diamond, Silicon Carbide, and Graphite. Meteoritics, 28, 490-514.

Bernatowicz, T.J., Gibbons, P.C., Amari, S., Lewis, R.S. (1995). On the Nature of Carbon Cores in Interstellar Graphite. Lunar Planetary Science Conference XXVI.

Bernatowicz, Thomas J., Cowsik, Ramanath, Gibbons, Patrick C., Lodders, Katharina, Fegley, Bruce, Jr., Amari, Sachiko, Lewis, Roy S. (1996). Constraints on Stellar Grain Formation from Presolar Graphite in the Murchison Meteorite. The Astrophysical Journal, 472, 760-782.

Bernatowicz, Thomas J., Walker, Robert M. (1997). Ancient Stardust in the Laboratory. Physics Today, Dec. 1997, 26-32.

Clayton, D., Grossman, L., Mayeda, T. (1973). Isotopic Variations in the Rock-forming Elements in Meteorites. Phil. Trans. Roy. Soc. Lond. A, 325, 483-401.

Dawkins, D. (1996). Determination of the core to diameter ratio of interstellar dust particles from a carbonaceous meteorite. http://newton.umsl.edu/~dawkins/grant.html.

Fraundorf, P., Brewer, K., Dawkins, D., Truong, M, Witt, D. (2000) The Core-rim Structure of Pre-solar Graphite Onions. Meteoritics and Planetary Science, 35, 5, A56-57.

Messenger, S., Amari, S., Gao, X., Walker, R.M., Clemett, S.J., Maechling, C.R., Chen, Y.H., Zare, R.N., Lewis, R. (1995). Organic Molecules in Interstellar Graphite Grains. Lunar Planetary Science Conference XXVI.

Reynolds, J.H. (1960). Isotopic Composition of Primordial Xenon. Physical Review Letters, 4, 351-354.

Thomson, J.J. (1912). Further Experiments on Positive Rays. Phil. Mag., 24, 209-253.

Wasserburg, G., Papanastassiou, D. (1982). Some Short-lived Nuclides in the Early Solar System. Essays in Nuclear Astrophysics (eds. C.A. Barnes et al), 77-140, Cambridge University Press.

Zinner, E., Amari, S., Wopenka, B., Lewis, R.S. (1995). Interstellar Graphite in Meteorites: Isotopic Compositions And Structural Properties of Single Graphite Grains from Murchison. Meteoritics, 30, 209-26.