N
Micrometeorite Background

By Section:
     1. Damage Areas
     2. Mass Flux
     3. Micrometeorite Material Retention
     4. Perforating Impacts--Testing
     5. Beta Meteoroids

The micrometeorite flux is a potential source of background for a solar wind measurement. There are three principal issues: (1) can the micrometeorite flux be ignored? (2) even if the micrometeorite flux is important, will a significant fraction of the impacting mass be retained? As discussed below, the total micrometeorite mass flux cannot be regarded as negligible relative to the solar wind; thus sticking of micrometeorite material to the collector is an important issue. (3) Additionally, could micrometeorites possibly jeopardize the mission as a whole if collector materials burst into small fragments which impeded crucial capsule or canister closing mechanisms? To answer this question we have performed several impact laboratory tests.

1. Damage areas.
Issue (1) has two parts: one is how much collector area will be removed or damaged directly by cratering or spalling of collector material by meteoroid impacts. The second issue is the mass flux of meteoroids hitting (and an extreme upper limit for the amount retained on) the collectors. The spall damage has only been estimated presently for Si, which should constitute the bulk of the collector material. Results for Ge should be similar. Upon micrometeoroid impact, the silicon wafer is compressed by a shock wave emanating from the impact point. Peak pressures are high enough to liquefy and vaporize the silicon near the impact point. A tensile force is produced by the shock wave action which can yield the surface thereby creating surface spalling. Based on hypervelocity testing of fused silica windows, Cour-Palais [1987] developed an expression for the penetration depth as a function of projectile density, diameter, and velocity [p = 0.53r^0.5 d^1.06 v^0.67]. Testing in the JSC Hypervelocity Impact Test Facility indicates a spall zone diameter of d_spall = 13.664 d V_n^2/3, where V_n is the normal impact velocity. Secondary mass ejected by these impacts is calculated from m_spall = d²_spall p p_wafer / 12, where p_wafer is the wafer density and p is given by the equation above. For 20 km/sec impacts, allowing for impactor masses ranging from 10^-20 to 10 g, we determined the cumulative number of impacts as a function of impactor mass. Estimates are based on the Helios and Heos data [Grün, 1985], as well as LDEF [Love and Brownlee, 1993] with corrections to simulate interplanetary space and our spacecraft geometry. The fraction of collector surface damaged is 0.37% (Grün model) to 4.1% (Love-Brownlee). Loss of a minor fraction of the collection surface is not a problem. The solar wind concentrator will be designed to withstand the expected number of micrometeorites to its electrically active surfaces. The concentrator target--the most precious part of the payload--will be facing backwards, protected by a backing plate, so it will be protected from micrometeoroids.

2. Mass Flux.
We now discuss the magnitude of the micrometeorite mass flux in comparison to the mass flux of solar wind (second part of item 1 above). To compare the micrometeorite and solar wind fluxes we choose Si as the comparison element. To the extent that micrometeorites are similar to chondritic material the relative elemental abundances of non-volatile elements should be similar for micrometeorites and the solar wind; thus the choice of comparison element is not important. We adopt a solar wind proton flux of 3 x l0⁸ /cm²-sec and assume (based on photospheric abundances) Si/H = 3 x l0^-5. We assume that the micrometeorites are 20% Si by mass; all silicate materials will be similar.

Too few events are available from earlier spacecraft data to accurately determine the flux-mass relationship, so we adopt the mass dependence indicated by the slope of the microcrater diameter distribution from the lunar surface, primarily by rock 12054 (Morrison and Zinner, 1977). However, to be conservative we have increased the absolute value of the flux by a factor of 6 to match the Helios datum. The resulting flux-mass curve has been integrated to give the total mass or Si atom micrometeorite input to a given collector. More recent results from LDEF (Sullivan and McDonnell, 1992) indicate a higher flux of particles in the > 10^-7 g mass range was estimated previously (e.g., Grün et al., 1985). At 10^-5 g, corresponding to around 100 mm particles, the new estimate of 0.6 m^-2yr^-1 is a factor of four above the previous estimate, and means that over the whole collector surface we could expect that one such event would be captured during a two-year exposure. Even an impact of this size would directly damage only the immediate surrounding mm unless a significant vapor cloud developed; a vapor cloud is not expected for this size meteorite impact.

The slope of the mass distribution is sufficiently shallow that the area-averaged Si flux is determined by the largest impact events, i.e. the total number of Si atoms is determined by the upper limit of the integral over the mass distribution. The choice of upper limit was made as the micrometeorite mass for which there was a 10% probability/year for the detector area of interest. Since this upper mass limit will increase with area, the micrometeorite background flux will increase with area analyzed for solar wind. The smaller the area, the less important will be the micrometeorite background. We have calculated the relative micrometeorite to solar wind Si flux for areas of 0.1, 1, and 100cm² (the latter is probably the maximum practical area which could be analyzed for solar wind) as 0.04, 0.2, and 5, respectively.

Given the large uncertainties, it is clear that the micrometeorite Si flux cannot be ignored for 100cm2 or larger areas. For an uncollimated sun-oriented solar wind collector, the anisotropy in the micrometeorite flux will decrease these ratios by about a factor of 2. To cause a 5% correction to the solar wind flux the required micrometeorite retention percentages are 100%, 25%, and 1% for 0.1, 1, and 100 cm², respectively.

3. Micrometeorite material retention.
Intuitively, and by analogy to lunar microcraters, (where projectile residues are not found), retention of projectile material during hypervelocity impact in zero gravity is expected to be small, at least for insulating collector materials. Hörz et al. (1983) report 20-50% retention of material from mg-sized silicate projectiles for laboratory impacts into Au and Cu at velocities up to 6.5 km/sec. In addition, for impacts of metal projectiles onto metal targets, large retentions are found at least up to 13 km/sec, even for small projectiles (see for example, Dietzel et al., 1972), though the flux of metallic micrometeorites appears to be low. The generally high retention of meteoritic residue on metal targets is a possibly serious complication in that Au and other noble metals are a potentially important class of high purity collector materials.

Si is a brittle metal, closer to rock than Au, so retention is expected to be low. We have examined LDEF impact craters into Ge (obtained courtesy of R Walker) and could detect no evidence of residues using photoelectron spectroscopy which is sensitive to surface deposits. Also, relative to the laboratory studies, interplanetary impacts will tend to be at higher velocities with lower projectile retention factors. Moreover, other experiments with micron-sized silicate projectiles in Au show much less retention (F. Hörz, private communication).

For an exposure time of two years at most a few mg-sized particles will hit a meter-sized, collector array. For contemplated analysis areas of 1-100 cm² the probability of a mg-sized impact is small. Thus experiments with 1-100 micron-sized projectiles are more relevant. The Hörz et al experiments suggest that micrometeorite retention fractions for silicate projectiles should be low. We assume that if retention of silicate projectile material in ductile metal targets is negligible (following Hörz), then retention of silicate projectile material in brittle, insulating targets, or brittle metals such as Si, will also be negligible.

In summary it appears that micrometeorite background would be important only for the analysis of large collector areas, and possibly not even then depending on the details of the collector material, the elements analyzed, and the exact retention probabilities.

We propose to pursue the issue of projectile retention primarily by discussion with experts in these areas but experiments will be carried out if there are gaps which can be feasibly filled. We will also review the new developments on the micrometeorite flux issue. In the final analysis microscopic examination of the collector areas analyzed will be carried out and any large micrometeorite impact pits can be avoided.

4. Perforating Impacts--Testing.
Issue (3) above is the question of what happens with relatively large micrometeorite impacts. Impact experiments have shown that perforation usually occurs when the target thickness is less than twice the penetration depth. The calculations above indicate that by this criterion, between 7 and 29 perforating impacts should occur on the Genesis collection surfaces, based on the Grün [1985] and Love and Brownlee [1993] models, respectively. These numbers would drop rapidly if we decided to use thicker Si wafers than the semiconductor industry standard. At the very worst, perforated collector pieces could shatter, with fragments going everywhere, possibly risking failure of crucial mechanisms for stowing the collectors or closing the return capsule, jeopardizing the mission. While this is unlikely, another scenario would be fracture into a number of pieces, so that most of the perforated collector (~100 cm²) would be lost. To determine whether either of these scenarios is likely, several hypervelocity tests were performed on Si wafers at the Johnson Space Center Impact Laboratory.

Three tests were performed in October, 1995 using nylon spheres launched by two-stage light-gas gun to impact with a velocity near 7 km/s. The projectile sizes were 164 microns, 238 microns, and 344 microns, all within the size needed for perforation. The wafer was held by rubber washers which should have mechanical properties roughly similar to the wafer fasteners for the spacecraft. The 164 micron projectile just perforated the wafer. A clean, single crack was observed in all tests. The observed crack in each test wafer seemed to be cleavage along the direction of crystal orientation. The washer support held the wafer intact so that no collector was lost. It therefore appears that minimal damage will be incurred by perforating impacts.

5. Beta Meteoroids.
There is a set of more­or-less solar directed micrometeorite population (beta meteoroids), but these are small particles and the mass flux is probably much less than we have estimated above. Nevertheless, these have been given special attention because of their directionality.

In this section the flux of ß-meteoroids is calculated separately to determine whether their contribution will threaten the purity of a solar wind sample. Beta-meteoroids are very small particles for which radiation pressure rather than gravity is the dominant force. Consequently, they come from the sun-ward direction, into which the solar wind collector will be directly facing. The mass range for such particles is below approximately 1.4 x 10^-12 g for a particle density of 2 g/cm³ originally in a circular orbit (Zook and Berg, 1975); elliptical orbits allow slightly more massive particles. The lower mass limit derives from the fact that radiation pressure drops off rapidly at around 10^-15 g as particle sizes fall below the dominant wavelength (Zook and Berg, 1975); Mie scattering off smaller particles is very inefficient.

Beta-meteoroids have been difficult to detect because of their very small mass. Data from Pioneers 8 and 9, and Helios 1 led Grün et al. (1985) to estimate their cumulative flux F(m) (flux of particles greater than the given mass) as km^-0.89 m^-2s^-1, where k = 1.41 x 10^-17 and m is in grams. Geometrical differences in collector orientations will add a factor of pi for a continuous sun-facing collector relative to that of Grün et al. (1985). Integration of flux over the mass range given above yields a mass flux of 1.1 x 10^-17 g/m²s, resulting in a 2-year total flux of 3.5 x 10^-14 g/cm². Adjusting the upper and lower masses, for example, to include highly eccentric as well as hyperbolic orbits, would not make a difference of over 50% from this result. If the ß-meteoroid composition is close to chondritic, resulting from the collisional breakup of Poynting-Robertson particles, it should be approximately 30% silicon. By comparison, the expected solar wind Si flux over 2 years is 2.8 x 10^-11 g/cm², orders of magnitude higher. The upper mass cut-off in the ß-meteoroid size distribution would have to be 1-2 orders of magnitude higher for them to be an important background source.

The only way ß-meteoroids might be a significant problem is if their composition is anomalous. One possible origin is Poynting-Robertson particles that are brought sufficiently close to the sun to vaporize all but the refractory elements, which are then blown away as ß-meteoroids. This could lead to a contamination of the refractory elements in our collector. However, it is unlikely that the majority of the ß-meteoroids are produced in this way. The collisional cross sections in the inner solar system are sufficiently high to ensure that Poynting- Robertson particles are reduced to ß-meteoroid size before their orbit decays to below 0.03 AU, the solar proximity needed to vaporize most elements (Grün et al., 1985). Additionally, refractory particles of higher density have a much smaller mass window in which radiation pressure dominates.

Because each ß-particle is of much larger size relative to solar wind particles, it may be instructive to consider the flux F(m) of such particles impinging on the collector, and the relative analysis area needed to make the contamination effects of such impacts negligible. The power law flux equation gives (with the added factor of pi) F(m) = 6.2 cm^-2 for m > 10^-15 g (Grün et al., 1985). Even the largest ß-meteoroids, at 1.4 x 10^-12 g. of roughly chondritic composition would give no more than 2% contamination to a 1 cm² solar wind analysis assuming complete retention. A microanalysis technique such as SIMS would be contaminated if the beam spot would happen to impinge directly on the ß-meteoroid impact. However, this event would be very unlikely, and could be guarded against by checking for microcraters or requiring two or more microanalysis spots to agree in composition.