Search

We report the use fo delta-doped charge-coupled devices (CCDs) for direct detection of electrons in the 50-1500 eV energy range. These are the first measurements with a solid state device to detect electrons in this energy range.

In this paper, we will briefly discuss delta-doped CCDs and their application of low-energy electron detection. We show that modification of the surface this way can greatly improve sensitivity to low energy electrons.

A pulsed DT neutron generator system, similar to that used in commercial well logging, offers the possibility of performing accurate elemental analyses to depths of tens of centimeters in a few seconds with the probe on the body's surface.

The use of neutron and gamma ray measurements for the analysis of material composition has become well established in the last 40 years. Schlumberger has pioneered the use of this technology for logging wells drilled to produce oil and gas, and for this purpose has developed neutron generators that allow measurements to be made in deep (5000 m) boreholes under adverse conditions. We also make ruggedized neutron and gamma ray detector packages that can be used to make reliable measurements on the drill collar of a rotating drill string while the well is being drilled, where the conditions are severe. Modern nuclear methods used in logging measure rock formation parameters like bulk density and porosity, fluid composition, and element abundances by weight including hydrogen concentration. The measurements are made with high precision and accuracy. These devices (well logging sondes) share many of the design criteria required for remote sensing in space; they must be small, light, rugged, and able to perform reliably under adverse conditions. We see a role for the adaptation of this technology to lunar or planetary resource assessment missions.

The use of thin, self-supporting carbon catcher foils allows one to measure sputtering yields in a broad range of materials with high sensitivity. Analyzing the foils with Rutherford forward scattering, sputtered Al, Si and P surface densities down to 5 x 10 to the 13th per sq cm with uncertainties of about 20 percent have been measured.

Track damage and the associated ejection of atoms and molecules from insulators, which occur as a result of ion-induced electronic excitation, are of interest both in their own right and because of the mechanisms through which the energy in the excited electrons is transformed into atomic motion. In this paper an overview is given of the phenomena that are observed. We show that there is a remarkable similarity between the damage profile along the ion's track in the solid and the yield of ejected atoms at the energy that corresponds to each point on the track. It is also seen that the density of extended defects (or, correspondingly, the ejected particle yield) appears to have a 'universal' form that is weakly dependent on the type of material. In the model presented this is a consequence of the inner-shell ionization of light elements in the solid by the incident ion; the resulting Auger decay produces an intense ionization spike that locally triggers the track formation/erosion process. This model allows the estimation of erosion yields/damage profiles for different ions and materials.

A new model is proposed to explain both localized damage regions and preferential etching of damage tracks caused by the passage of energetic ions in insulators. The formation of each region of extended defects is initiated by the Auger decay of a vacancy produced in an inner electronic shell of an atom of the insulator by an incident ion. This decay produces an intense source of ionization within a small volume around the decaying atom, which causes decomposition of the material in a manner similar to that observed in pulsed laser irradiation. The resulting chemical or crystalline modification of the material is the latent track, which can be preferentially etched due to its changed structure.

The diffusion of Xe in olivine, a major mineral in both meteorites and linear samples, was studied. Xe ions were implanted at 200 keV into single-crystal synthetic-forsterite targets and the depth profiles were measured by alpha particle backscattering before and after annealing for 1 hour at temperatures up to 1500 C. The fraction of implanted XE retained following annealing was strongly dependent on the implantation dose. Maximum retention of 100 percent occurred for an implanting dose of 3 x 10 to the 15th power Xe ions/sq cm. Retention was less at lower doses, with approximately more than or = 50 percent loss at one hundred trillion Xe ions/sq cm. Taking the diffusion coefficient at this dose as a lower limit, the minimum activation energy necessary for Xe retention in a 10 micrometer layer for ten million years was calculated as a function of metamorphic temperature. Previously announced in STAR as N83-18528

Phenomena that involve accelerated ions in stellar processes that can be simulated with laboratory accelerators are described. Stellar evolutionary phases, such as the CNO cycle, have been partially explored with accelerators, up to the consumption of He by alpha particle radiative capture reactions. Further experimentation is indicated on reactions featuring N-13(p,gamma)O-14, O-15(alpha, gamma)Ne-19, and O-14(alpha,p)F-17. Accelerated beams interacting with thin foils produce reaction products that permit a determination of possible elemental abundances in stellar objects. Additionally, isotopic ratios observed in chondrites can be duplicated with accelerator beam interactions and thus constraints can be set on the conditions producing the meteorites. Data from isotopic fractionation from sputtering, i.e., blasting surface atoms from a material using a low energy ion beam, leads to possible models for processes occurring in supernova explosions. Finally, molecules can be synthesized with accelerators and compared with spectroscopic observations of stellar winds.

Electronic excitation induced by MeV/amu ion beams in a variety of materials has been employed successfully for a number of applications. The examples that will be presented are: modification of the surface reflectivities of optical materials; sputter-erosion of dielectrics; and enhancement of the adhesion of thin film coatings. All of these effects arise from the loss of energy by the ion beam to electrons in the target material; the mechanisms involved are at best qualitatively understood. This paper will stress not only the exploitation of such high energy bombardment techniques but will also briefly review attempts to expose the underlying causes.

Sputtering yields of solid SO2 by high energy ions were measured in order to study the mechanism for sputtering dielectrics with ions in the electronic stopping power region. The incident ions were helium and fluorine with energies ranging from 1.5 MeV to 25 MeV. Yields as high as 7000 SO2 molecules/incident F ion were measured; the 1.5 MeV He-4 beam had a sputtering yield of 50. The data are compared to yield measurements made on UF4 and H2O targets. There is a striking similarity in the yield as a function of the incident F energy for all three targets. The data compare favorably with theoretical yield curves based on a new model for the sputtering which considers the electronic excitations induced in the target by the incident beam. Measurements and calculations of this sort are also useful in understanding processes which occur on the surface of Jupiter's satellite Io, which is covered with SO2 frost and bombarded by energetic ions trapped in the Jovian magnetosphere.

It has been found that the adhesion of thin metal films on insulators, semiconductors, and metals could be improved by subjecting the material to a high-energy ion bombardment. Griffith et al. (1982) have first suggested a use of this technique with insulators. The present investigation has the objective to determine the mechanism for the adhesion enhancement. A description is presented of a preliminary transmission electron microscopy (TEM) study of thinned bonded samples of silver on silicon using electron diffraction. It is found that irradiation of a variety of thin film-substrate combinations by heavy ion beams will provide a remarkable improvement in the adherence of the film. The evidence for the mechanism involved in the enhancement of adhesion is discussed.

The sputtering of UF4 targets by energetic beams of O-16, F-19, and Cl-35 ions has been investigated for beam energies in the range 0.12 to 1.5 MeV/amu. The sputtering yields, which follow the same trend as the electronic part of the projectile energy loss in the material, are observed to have a strong dependence on the charge state of the incident ions. Data have been taken both in transmission and reflection (0 and 180 deg to the incident beam direction, respectively). Energy spectra of the neutral sputtered particles have been obtained for 5 MeV F-19 ions and for 13 MeV Cl-35 ions; in both cases the spectrum has a Maxwellian form. The data obtained are compared with several models of the high energy sputtering process.

A common aspect of the models of such processes as track formation and induced desorption of molecular ions is the requirement that the electronic excitation have a sufficiently long lifetime (of the order of picoseconds) so that energy can be transferred to atomic motion. Attention is called to several phenomena recently observed indicating that atomic motion/chemical rearrangement can occur even when the electronic excitation is shorter lived (of the order of femtoseconds). Two such processes, which are also induced by MeV ion bombardment, are track damage in heavily doped compound semiconductors and greatly enhanced adhesion of metal films to metallic and semiconducting substrates. Since neither of these effects can easily be accommodated within the existing theoretical models, it is thought possible that an even richer variety of ion-induced effects will be discovered.

Watson and Haff (1980) have discussed a theory which is designed to explain quantitatively isotopic fractionation effects observed during sputtering of simple or complex targets. This theory is based on the assumption that most of the atoms sputtered from a surface originate in the top monolayer. The present investigation is mainly concerned with a direct experimental test of that assumption. The sputtering of both solid and liquid phases of gallium, indium, and the gallium-indium eutectic alloy is studied. Results obtained with the aid of ion scattering and Auger spectroscopy show that, in agreement with rough theoretical expectations, the surface monolayer of a gallium-indium alloy with 16.5 percent indium in bulk contains more than 94 percent indium, while the next layer can be only slightly enriched.

Silicon, silicon dioxide, and silicon nitride have been sputtered with chlorine ions at 5 MeV and 20 MeV. While the yield from the silicon target was unmeasurably low, the insulating compounds exhibited the enhanced yields observed in other insulating targets. The yield follows the electronic stopping power and seems to be independent of the target's thermal properties. Some of the data suggest that the enhanced sputtering mechanism may be active in extremely thin films (not less than 3 monolayers).

The erosion of frozen SO2 due to bombardment by both light and heavy ions (He and F) was measured for bombarding energies of 0.08 to 1.3 MeV/amu. The number of SO2 molecules ejected from the target per incident ion (i.e., the sputtering yield) was 50 for 1.5 MeV He ions and 7300 for 6 MeV F ions. Ion bombardment followed by heating produced an oxygen/sulfur residue which was much more stable against subsequent ion bombardment than the initial frozen SO2. The erosion rate of SO2 frost on Jupiter's moon Io depends strongly on the elemental composition and energy spectra of the magnetospheric ion flux which bombards the surface. The combined effects of ion bombardment and heating which produced residues on the target substrates may also occur on Io from magnetospheric ion bombardment and heating by volcanism. The experimental results compare favorably with a new model of the sputtering process which considers the energy loss of the incident ion to electronic excitation in the target.

Sputtering yields of solid SO2 by high energy ions were measured in order to study the mechanism for sputtering dielectrics with ions in the electronic stopping power region. The incident ions were helium and fluorine with energies ranging from 1.5 MeV to 25 MeV. Yields as high as 7000 SO2 molecules/incident F ion were measured; the 1.5 MeV He4 beam had a sputtering yield of 50. The data are compared to yield measurements made on UF4 and H2O targets. There is a striking similarity in the yield as a function of the Energy for all three targets. The data compare favorably with theoretical curves based on a model for the sputtering which considers the electronic excitations induced the target by the incident beam. Measurements and calculations of the sort are also useful in understanding processes which occur on the surface of Jupiter's satellite Io, which is covered with SO2 frost and bombarded by energetic ions trapped in the Jovian magnetosphere.

In this paper recent experimental results on sputtering of UF4 and H2O (ice) by fast heavy ions are reviewed. Measurements have been made of the dependence of the sputtering yield on the incident ion type, charge state, and energy. In the case of UF4, the energy spectra of neutral sputtered particles have been obtained as well. There is a clear dependence of the sputtering yield on the electronic part of the stopping power, and the yield is strongly affected by the charge state of the incident ion, which shows that in the near-surface region from which sputtered particles arise, the charge state of the incident ion has not reached equilibrium. The shape of the energy spectra observed for UF4 targets is that expected from a thermal distribution rather than that of the collision cascade form typical of ordinary sputtering. A model of the sputtering-track registration process that has arisen from these data is shown to provide a framework of understanding the close relationship of these observations to the 'plasma desorption' of marcromolecules.

The diffusion properties of xenon implanted into feldspar, a major mineral in meteorites and lunar samples, are investigated in light of the importance of xenon diffusion in the interpretation of early solar system chronologies and the retention time of solar-wind-implanted Xe. Known doses of Xe ions were implanted at an energy of 200 keV into single-crystal plagioclase targets, and depth profiles were measured by alpha particle backscattering before and after annealing for one hour at 900 or 1000 C. The fraction of Xe retained following annealing is found to be strongly dependent on implantation dose, being greatest at a dose of 3 x 10 to the 15th ions/sq cm and decreasing at higher and lower doses. Xe retention is also observed to be unaffected by two-step anneals, or by implantation with He or Ar. Three models of the dose-dependent diffusion properties are considered, including epitaxial crystal regrowth during annealing controlled by the extent of radiation damage, the creation of trapping sites by radiation damage, and the inhibition of recrystallization by Xe during annealing

The sites and materials involved in solar system sputtering of planetary surfaces are reviewed, together with existing models for the processes of sputtering. Attention is given to the interaction of the solar wind with planetary atmospheres in terms of the role played by the solar wind in affecting the He-4 budget in the Venus atmosphere, and the erosion and differentiation of the Mars atmosphere by solar wind sputtering. The study is extended to the production of isotopic fractionation and anomalies in interplanetary grains by irradiation, and to erosion effects on planetary satellites with frozen volatile surfaces, such as with Io, Europa, and Ganymede. Further measurements are recommended of the molecular form of the ejected material, the yields and energy spectra of the sputtered products, the iosotopic fractionation sputtering causes, and the possibility of electronic sputtering enhancement with materials such as silicates.

Because of recent interest in the role played by the thermal properties of materials that exhibit high energy sputtering, Al2O3 and LiNbO3 were sputtered with chlorine ions at energies between 3 MeV and 25 MeV. To detect the sputtered Al and Nb we employ thin carbon catcher foils, which are analyzed with Rutherford scattering in the forward direction. Al surface densities of 10 to the 14th/sq cm and Nb surface densities of 10 to the 13th/sq cm are easily measured. The sputtering yields for both Al2O3 and LiNbO3 increase rapidly with increasing chlorine energy, and the Al and Nb yields are both approximately 0.2 at 20 MeV. Tests for dose, beam current, and contamination effects will be discussed.

The diffusion of Xe in olivine, a major mineral in both meteorites and lunar samples, was studied. Xe ions were implanted at 200 keV into single-crystal synthetic-forsterite targets and the depth profiles were measured by alpha particle backscattering before and after annealing for 1 hour at temperatures up to 1500 C. The fraction of implanted Xe retained following annealing was strongly dependent on the implantation dose. Maximum retention of 100% occurred for an implantion dose of 3 x 10 to the 15th power Xe ions/sq cm. Retention was less at lower doses, with (approximately more than or = 50% loss at one hundred trillion Xe ions/sq cm. Taking the diffusion coefficient at this dose as a lower limit, the minimum activation energy necessary for Xe retention in a 10 micrometer layer for ten million years was calculated as a function of metamorphic temperature.

The use of ion beams in the electronic stopping region to improve the adhesion of insulators to other materials is described. In particular, the bonding of Au films to Teflon, ferrite, and SiO2 was improved by bombarding them with He and Cl, respectively. Improvements in bonding were also observed for Au on glass, Au and Cu on sapphire, and Si3N4 on Si. The mechanism is apparently associated with sputtering and track forming processes occurring in the electronic stopping region. Some applications are discussed.

A model is presented which predicts isotope fractionation in material sputtered from mineral surfaces. According to the model, the fractionation is a function of the fractional abundances of each isotope, the low-energy collision cross sections and the atomic masses, and the fractionation pattern can be nonlinear. Calculations are illustrated for all sets of isotopes in the minerals perovskite, anorthite, akermanite, enstatite and troilite, and it is found that while O is always positively fractionated, with heavier isotopes sputtered preferentially, heavier elements are generally negatively fractionated. It is noted that the model may be tested by experiments on a system such as CaF2, CaI2, in which the Ca sputtered from CaF2 is predicted to be strongly fractionated in a negative sense and that sputtered from CaI2 is predicted to be strongly fractionated in a positive sense.

Fluorine surface deposits on Apollo 15 green glass, Apollo 17 orange glass, and on vesicle linings from Apollo 15 basalts were studied by a resonant nuclear reaction technique. Surface layers of about 10 to the 15th F atoms/sq cm were observed on unbroken spheroidal glass surfaces, whereas thinner (about 10 to the 14th atoms/sq cm) deposits were found on the vesicle linings.

The sputtering of the lunar surface by the solar wind is examined as a possible mechanism of mass fractionation. Simple arguments based on current theories of sputtering and the ballistics of the sputtered atoms suggest that most ejected atoms will have sufficiently high energy to escape lunar gravity. However, the fraction of atoms which falls back to the surface is enriched in the heavier atomic components relative to the lighter ones. This material is incorporated into the heavily radiation-damaged outer surfaces of grains where it is subject to resputtering. Over the course of several hundred years an equilibrium surface layer, enriched in heavier atoms, is found to form. The dependence of the calculated results upon the sputtering rate and on the details of the energy spectrum of sputtered particles is investigated. It is concluded that mass fractionation by solar wind sputtering is likely to be an important phenomenon on the lunar surface.

The resonant nuclear reaction F-19 (p, alpha gamma)0-16 has been used to perform depth sensitive analyses for both fluorine and hydrogen in lunar samples. The resonance at 0.83 MeV (center-of-mass) in this reaction has been applied to the measurement of the distribution of trapped solar protons in lunar samples to depths of about 1/2 micrometer. These results are interpreted in terms of terrestrial H2O surface contamination and a redistribution of the implanted solar H which has been influenced by heavy radiation damage in the surface region. Results are also presented for an experiment to test the penetration of H2O into laboratory glass samples which have been irradiated with 0-16 to simulate the radiation damaged surfaces of lunar glasses. Fluorine determinations have been performed in a 1 pm surface layer on lunar samples using the same F-19 alpha gamma)0-16 resonance. The data are discussed from the standpoint of lunar fluorine and Teflon contamination.

A velocity spectrum of neutral sputtered particles as well as a low resolution mass spectrum of sputtered molecular ions was measured for 4.74 MeV F-19(+2) incident of UF4. The velocity spectrum is dramatically different from spectra taken with low energy (keV) bombarding ions, and is shown to be consistent with a hot plasma of atoms in thermal equilibrium inside the target. A thermalized ion explosion model is proposed for high energy sputtering which is expected to describe track formation in dielectric materials. The model is shown to be consistent with the observed total sputtering yield and the dependence of the yield on the primary ionization rate of the incident ion.

An apparatus for determining helium to hydrogen ratios in a gaseous sample is presented. The sample is bombarded with alpha particles created by a self contained radioactive source and scattering products falling within a predetermined forward scattering angular range impact a detector assembly. Two detectors are mounted in tandem, the first completely blocking the second with respect to incident scattering products. Alpha particle/hydrogen or alpha particle/helium collisions are identified by whether scattering product impacts occur simultaneously in both detectors or only in the first detector. Relative magnitudes of the two pulses can be used to further discriminate against other effects such as noise and cosmic ray events.

A model which makes definite predictions for the fractionation of isotopes in sputtered material is discussed. The fractionation patterns are nonlinear, and the pattern for a particular set of isotopes depends on the chemical matrix within which those isotopes are contained. Calculations are presented for all nonmonoisotopic elements contained in the minerals perovskite, anorthite, ackermanite, enstatite, and troilite. All isotopes are fractionated at the level of approximately 4-6 deg/o per atomic mass unit. Oxygen is always positively fractionated (heavier isotopes sputtered preferentially), and heavier elements are generally negatively fractioned (light isotopes sputtered preferentially). The value of Delta (O-18:O-16) is always less by about 1.8 deg/o than a linear extrapolation based upon the calculated delta (O-17:O-16) value would suggest. The phenomenon of both negative and positive fractionation patterns from a single target mineral are used to make an experimental test of the proposed model.

It is found through an investigation, combining Monte Carlo simulations and analytical techniques, of the direct collisional interaction of an energetic particle flux with the neutral components of a planetary atmosphere, that solar wind sputtering could provide an important exospheric mass sink on both Mars and Venus under appropriate conditions. The computed rates of helium loss in the Venusian atmosphere and of carbon, nitrogen and oxygen in the Martian atmosphere imply that sputtering would have a significant effect on the noble gas budget of Venus and dominate the chemical and photochemical loss processes of Mars. Because of diffusive separation of lighter elements and isotopes, and because the gravitational binding energy is proportional to the mass, the erosion process preferentially removes the lighter components of the atmosphere. Solar wind sputtering could therefore compete with other erosion mechanisms in generating substantial fractionation effects.

Charged particle-induced nuclear reactions were used in the following problems: the determination of elemental abundances of boron and fluorine in carbonaceous chondritic meteorites; the identification of products of lunar vulcanism; and the study of solar wind-implanted atoms in lunar materials. The technique was seen as an important supplement to other methods of elemental and isotopic analysis. This was especially true for cases involving light elements at very low concentrations or where high resolution depth distribution information was needed in non-destructive analysis.

Several different models of ion-induced erosion of both the rocky and icy Galilean satellites are considered. The conclusion is that erasure of surface relief via this mechanism is likely to occur at a rate not exceeding about 100 m per billion years, and therefore is not of global topographic significance. Another effect of ion bombardment is the possible growth of a thin mineral armor on the surfaces of the icy satellites. A volume concentration of 0.1% of 1 micron sized mineral grains can lead to 50% armoring of the surface on time scales ranging from thousands to millions of years.

The capture coefficient probabilities for sputtered atoms of Nb-93 and Rh-103 incident on Al2O3 surfaces were measured using the backscattering of MeV heavy ions. In the circumstance where the collecting surface is thickly covered, the sticking probabilities integrated over the energy distribution of sputtered atoms are 0.97 plus or minus 0.01 for Nb-93 and 0.95 plus or minus 0.01 for Rh-103 respectively. In the limit of negligible areal coverage of the collector, the accuracy is less; in this case the sticking probabilities are 0.97 + 0.03 or -0.08 and 0.95 + 0.05 or -0.08.

The role of solar-wind sputtering in causing atmospheric mass loss from Mars is discussed. The atmospheric mass loss is studied by analogy to models of sputtering of solid surfaces and by use of empirical data for the lunar surface. Results show that Martian atmospheric mass loss, when integrated over the course of 4 times 10 to the ninth power years, amounts to 3.1 times 10 to the 41st power CO2 molecules. Preferential loss of light elements and isotopes by sputtering is also considered.

Results are presented for an experimental study of the sputtering of U-235 atoms from foil targets by hydrogen, helium, and argon ions, which was performed by observing tracks produced in mica by fission fragments following thermal-neutron-induced fission. The technique used allowed measurements of uranium sputtering yields of less than 0.0001 atom/ion as well as yields involving the removal of less than 0.01 monolayer of the uranium target surface. The results reported include measurements of the sputtering yields for 40-120-keV protons, 40-120-keV He-4(+) ions, and 40- and 80-keV Ar-40(+) ions, the mass distribution of chunks emitted during sputtering by the protons and 80-keV Ar-40(+) ions, the total chunk yield during He-4(+) sputtering, and some limited data on molecular sputtering by H2(+) and H3(+). The angular distribution of the sputtered uranium is discussed, and the yields obtained are compared with the predictions of collision cascade theory.

The fission track technique for detecting U-235 has been used in conjunction with a mechanical time-of-flight spectrometer in order to measure the energy spectrum in the region 1 eV to 1 keV of material sputtered from a 93% enriched U-235 foil by 80 keV Ar-40(+) ions. The spectrum was found to exhibit a peak in the region 2-4 eV and to decrease approximately as E exp -1.77 for E not less than 100 eV. The design, construction and resolution of the mechanical spectrometer are discussed and comparisons are made between the data and the predictions of the random collision cascade model of sputtering.

A nuclear reaction depth profiling technique previously described by Filleux et al. (1977) has been used to measure the depth distribution of C on grain surfaces for Apollo 11, 15, 16 and 17 soil breccias. The surface C concentration of all samples studied lies between 2 and 8 times 10 to the 15th atoms per sq cm, showing no correlation with the volume C, which varies over an order of magnitude. If the observed variation represents the presence of unexposed grains on the surfaces studied, these results indicate a steady state surface C concentration of 5 to 10 times 10 to the 15th atoms per sq cm, accumulated over a time scale short compared with that required for the formation of volume-related C and with the mean lifetime of grains at the lunar surface. About one-third to one-half of the total C in lunar soil seems to be surface-correlated.

An investigation is conducted concerning the mass-fractionation effects produced in connection with the bombardment of the moon by the solar wind. Most of the material ejected by sputtering escapes the moon's gravity, but some returning matter settles back onto the lunar surface. This material, which is somewhat richer in heavier atoms than the starting surface, is incorporated into the heavily radiation-damaged outer surfaces of grains. The investigation indicates that sputtering of the lunar surface by the solar wind will give rise to significant surface heavy atom enrichments if the grain surfaces are allowed to come into sputtering equilibrium.

The interaction between the solar wind and a planetary atmosphere is evaluated as a cause of atmospheric mass loss. For the case of Mars, calculations suggest that an amount of material has been sputtered which is of the same order as the mass of the present atmosphere.

Ca has been measured in a lunar soil in order to establish the presence of isotopically mass-fractionated components. Ca was extracted by a series of water leaches after the soils were 'activated' by brief exposures to fluorine gas. The O2 obtained by this fluorination is found to have delta (O-18) of +21 per mil and to be, therefore, significantly mass-fractionated. Ca obtained in the leaches was analyzed using the double-spike technique. Very small Ca isotope fractionation is found in the leaches of this soil of up to 1 per mil per mass unit difference. The small Ca effects are in marked contrast to the measured delta (O-18) for the same sample and to large effects observed in many soils for oxygen, silicon, sulfur, and potassium. The data on Ca provide stringent constraints on models which attempt to explain the isotope mass-fractionation effects in lunar soils.

The sputtering of the lunar surface by the solar wind is examined as a possible mechanism of mass fractionation. Two distinct contributions to this mechanism are considered: (1) the tendency for a greater fraction of the heavier sputtered atoms to fall back to the surface under lunar gravity; and (2) mass-dependent kinetic effects that arise in the sputtering process itself. Calculations predict that delta (O-18) is approximately 5.6% while delta (Si-30) is about 3.6% and that oxygen is depleted in the outer regions of grains relative to the bulk composition. The results are in reasonable agreement with experiment, and one must conclude that mass fractionation by solar wind is an important phenomenon on the lunar surface.

The fission track technique for detecting uranium 235 was used in conjunction with a mechanical time-of-flight spectrometer to measure the energy spectrum in the region 1 eV to 1 keV of material sputtered from a 93% enriched U-235 foil by 80 keV Ar-40(+) ions. The spectrum was found to exhibit a peak in the region 2-4 eV and to decrease approximately as E to the -1.77 power for E is approximately greater than 100 eV. The design, construction and resolution of the mechanical spectrometer are discussed and comparisons are made between the data and the predictions of the ramdom collision cascade model of sputtering.

Measurements of surface concentrations of carbon in lunar soils and soil breccias provide information on the origin of carbon in the regolith. The reaction C-12 (d, p sub zero) is used to measure 'surface' and 'volume' concentrations in lunar samples. This method has a depth resolution of 1 micron, which permits only a 'surface' and a 'volume' component to be measured. Three of four Apollo 16 double drive tube samples show a surface carbon concentration of about 8 by 10 to the 14th power/sq cm, whereas the fourth sample gave 4 by 10 to the 14th power/sq cm. It can be convincingly shown that the measured concentration does not originate from fluorocarbon or hydrocarbon contaminants. Surface adsorbed layers of CO or CO2 are removed by a sputter cleaning procedure using a 2-MeV F beam. It is shown that the residual C concentration of 8 by 10 to the 14th power/sq cm cannot be further reduced by increased F fluence, and it is therefore concluded that it is truly lunar. If one assumes that the measured surface C concentration is a steady-state concentration determined only by a balance between solar-wind implantation and sputtering, a sputter erosion rate of 0.1 A/yr is obtained. However, it would be more profitable to use an independently derived sputter erosion rate to test the hypothesis of a solar-wind origin of the surface carbon.

Matrix element of nuclear beta decay of N-12 and B-12