Advanced integrated circuits when exposed to a space environment are prone to single event radiation effects. Accelerators can provide particle beams which test electrical components in a simulation of the space environments. Methods are available in most cases to relate the accelerator results to predicted performance in space.

Three advanced 16-bit NMOS microprocessors have been observed to suffer single event upset at a rate varying between one upset for every 8 × 1010 to one for every 2 × 1012 n/cm2-upset for cyclotron-produced neutrons with an average energy of 14 MeV. These rates are expected to vary, probably upward, with different types of programs. The errors are inferred to occur in memory-like components of the...

Several different types of random-access-memories (RAMs) have been tested for soft upset susceptibility under a variety of different particle bombardments including thermal neutrons, GeV protons, and protons and neutrons below 100 MeV and with few exceptions found to suffer single event upsets. Devices tested included 4K, 16K and 64K dynamic RAMs and 4K NMOS and 256×4 CMOS static RAMs. Mean upset ...

In an effort to better understand radiation damage to electronic materials, we use the binary-collision simulation code MARLOWE to model displacement cascades in silicon. We examine the average number of displacements produced by knock-on atoms as a function of their energy. The resulting vacancy-interstitial pairs are classified according to separation radius. We also examine a few particular cases of 100 keV cascades in silicon in order to highlight the importance that channeling has on the shape of displacement cascades.

Dynamic 16K random access memories (RAMs) have been irradiated with neutrons having mean energies of 6.5, 9 and 14 MeV and with 32 MeV protons and have been found to undergo single event upset. For both particles, one upset is expected for approximately 108 particles/cm2. The upsets are statistical and the affected cells can be reset and continue normal operation. Both HIGH and LOW storage elements are upset though at different rates. The cause of the upsets is most probably a multi-MeV alpha particle created by an (n, alpha) or (p, alpha) or similar nuclear reaction. The alpha particle discharges either the storage capacitor, the floating bit line, or the reference capacitor used by the sense amplifier.

In the above paper,’ Higa attempted to explain nonlinear conduction in metal contacts by appealing to electron-tunneling theory. He interpreted the sublinear slope (d31/dV3 < 0) of his current-voltage plot for an aluminum-toaluminum contact on the basis of tunneling through a 4 to 7.5 A aluminumaxide film. Such an interpretation is unsatisfactory. His conclusions are based upon equations for electron tunneling that are incorrect when used to calculate nonlinear effects across ultrathin gaps. For some parameters, these equations predict sublinear tunneling currents only because they have been inconsistently expanded. As a result, these -equations predict nonphysical results when used for extreme values. Higa’s (17), except for a factor of 4, was originally given by Holm [ 11 who described it as approximate. Forlani and Minnaja [2] have shown this equation to be equivalent to the lowest order expandon in (A lp:I2)-’, although higher order terms are implicitly included in the exponentials. In Higa’s notation, A is proportional to oxide thickness and L~O is the interface work function. Simmons (31 expanded the exponentials of the Holm equation to third order in the applied voltage and retained coefficients to third order in (A ppbI2)-l. He thereby saved terms of the same order as terms previously discarded. This expansion constitutes Higa’s (18). To obtain a consistent result, we will expand Forlani and Minnaja’s [2, eq. (18)]. This equation neglects the effect of a f~te Fermi energy and also contains three typographical errors. The result for a symmetric barrier is very simply given in Higa’s notation as This equation is correct to fourth order in V and all orders of (A lph12)-’ resulting from the usual expansion of the WKB exponential. Each coefficient of this equation is necessarily positive, thus guaranteeing superlinear behavior. Hence tunneling mot be used to explain Higa’s experiment. It is difficult to comment on Higa’s experiment because of a lack of details. Such experiments do need to be satisfactorily explained to better understand the contact problem. We suggest that possible explanations of his observed sublinear behavior may be a) temperature dependent conductivity in narrow contact filaments, [4] b) space charge effects (either in the oxide or included voids) or c) charge buildup in the metal filaments due to current variations on the scale of the electron mean free path. A possible cause of the wideband noise may be the already observed [SI dielectric breakdown or electroforming in relatively thick oxide films.

The relative neutron irradiation hardness of photoconductive and photovoltaic InSb infrared detectors is examined for various operating modes and conditions. It is found that photoconductors are much less vulnerable to fast neutrons at 78K than are photovoltaic detectors. The damage mechanisms for photovoltaic detectors are increased leakage currents coupled with a decrease of minority carrier lif...

Displacement damage in InSb is calculated for 1MeV and 14MeV incident neutrons. Ranges and energy deposition profiles are calculated with the E-DEP-1 computer code and combined with published neutron cross sections to obtain atomic displacement rates and range distributions. The damage ratio of 14MeV to 1MeV neutrons is estimated as close to unity. Comparison results are presented for Si. Experimental verification for unity damage ratio is given.

Optical quenching of photoluminescence by radiation with photon energy less than the band-gap energy has been observed in glassy ${\mathrm{As}}_{2}$${\mathrm{Se}}_{3}$ and ${\mathrm{As}}_{2}$${\mathrm{Se}}_{3}$\ifmmode\cdot\else\textperiodcentered\fi{}${\mathrm{As}}_{2}$${\mathrm{Te}}_{3}$. The quenching spectra provide evidence of optical absorption by a narrow band of localized states in the gap which are involved in the radiative recombination process and which are attributable to defects or impurities rather than to tails in the density of states extending into the gap.