Search

We measure the response of silicon photodiodes to irradiation by H/sup +/, He/sup +/, C/sup +/, N/sup +/, O/sup +/, Ne/sup +/, and Ar/sup +/ ions with energies up to 60 keV. The unique properties of these photodiodes, including an ultrathin SiO/sub 2/ dead layer and 100% internal carrier collection efficiency, allow direct measurement of the total energy lost to nuclear (nonionizing) and electronic (ionizing) energy loss processes, which are important for quantifying effects such as damage and charge deposition. When plotted as a function of E/mZ/sup 1/2/, where E, m, and Z are the incident ion energy, mass, and atomic number, respectively, the responsivity is found to follow a single curve that represents all ion species and energies used in this study. This enables rapid, accurate estimation of damage and charge deposition by an ion as a function of penetration depth in silicon. A comparison of the measurements with the stopping and range of ions in matter (SRIM) Monte Carlo simulation code shows that SRIM significantly overestimates the fraction of the incident energy lost to electronic stopping processes for E/mZ/sup 1/2/

We report energy distributions of electrons emitted from Al and Mg surfaces bombarded by 3--100-keV/amu hydrogen and deuterium ions. The energy spectra contain structure consistent with the decay of bulk plasmons, even below the velocity threshold expected from current theories. To explain the results we compare the importance of additional mechanisms involving electron capture, lattice-assisted excitation, and excitation by fast secondary electrons.

We measure the change in the response of 100% internal carrier collection efficiency silicon photodiodes having 60 /spl Aring/ SiO/sub 2/ passivation layers due to the damage induced by bombardment with 10-60 keV ions of H, He, N, Ne, and Ar. We find an initially exponential decrease in responsivity with increasing ion fluence /spl Phi/ and use this to define a damage constant /spl beta/. The correlation of /spl beta/ with the nuclear stopping power of the incident ion instead of with the total energy lost to nuclear stopping indicates that damage in a channel lying within the n-type silicon near the Si-SiO/sub 2/ interface dominates the radiation-induced change in the photodiode response. We use a fluid model of electron transport in the channel to derive a universal curve to describe the damage as a function of ion fluence and to show that the damage constant /spl beta/ is proportional to the damage cross section. Over the energy range of this study, damage cross sections of N/sup +/, Ne/sup +/, and Ar/sup +/ are 10-100 times that of He/sup +/, and /spl sim/1000 times that of H.

Recent advances in solid-state detector (SSD) technology have demonstrated the detection of ions and electrons down to 1 keV. However, ions at keV energies lose a substantial amount of energy ΔN in a SSD through Coulombic interactions with target nuclei rather than through interactions that contribute to the SSD output pulse, whose magnitude is a measure of the ion’s incident energy. Because ΔN depends on the ion species, detector material, and interaction physics, it represents a fundamental limitation of the output pulse magnitude of the detector. Using 100% quantum collection efficiency silicon photodiodes with a thin (40–60 A) SiO2 passivation layer, we accurately quantify ΔN for incident 1–120 keV ions and, therefore, evaluate the detection limits of keV ions using silicon detectors.

Small-angle neutron scattering was used to investigate the microstructure of the high-temperature Lα phase in the AOT/water/decane system [AOT = sodium bis(2-ethylhexyl) sulfosuccinate]. The system was modeled as randomly oriented lamellar stacks with a one-dimensional paracrystalline distortion. For samples injected into cells at high temperature, the two-dimensional Q maps show anisotropic scattering consistent with partial alignment of the phase, while samples injected near room temperature do not exhibit this feature. From the latter samples, it was found that a stack typically contains about 15 lamellae, about 80 A apart. Each layer is coated by a diffuse surfactant interface with a characteristic thickness of approximately 9 A. The fact that the scattering spectra do not exhibit maxima beyond the first order is explained by the observation that the Hosemann g factor gives about a 20% variation in the interlamellar spacing. The mean spacing is inversely proportional to the surfactant concentration, as expected for a lamellar phase. Over-estimation of the zero-order scattering predicted by the paracrystal model is attributed primarily to improper treatment of the contrast between each layer and the space-filling lamellar matrix.

Using silicon photodiodes with an ultrathin passivation layer, the average total energy lost to silicon target electrons (electronic stopping) by incident low energy ions and the recoil target atoms they generate is directly measured. We find that the total electronic energy deposition and the ratio of the total nuclear to electronic stopping powers for the incident ions and their recoils each follow a simple, universal representation, thus enabling systematic prediction of ion-induced effects in silicon. We also observe a velocity threshold at 0.05 a.u. for the onset of electronic stopping.

Ultraviolet (UV) blocking filters are needed for atomic flux imaging in environments where high levels of ultraviolet radiation are present. Freestanding gratings are a promising candidate for UV filtering. They have a high aspect ratio (∼13), narrow (∼40 nm) slots, and effectively block UV radiation. The grating fabrication process makes use of several etching, electroplating, and lithographic steps and includes an optional step to plug pinholes induced by particles during processing. Gratings were successfully manufactured and tested. Measured UV transmissions of ∼10−5 and particle transmissions of ∼10% are in agreement with theoretical predictions.