CompuServe Messages

Yes — if a photon happens to interact/dump its energy in the vicinity of the teeny tiny gate capacitor in a CMOS memory cell, the associated bit can easily change state. It is quite possible for a beta particle to do the same thing, as a beta is nothing more than an anarchistic electron, and ionizes like the devil. However, beta (and alpha) particles would need to originate within the chip rather than penetrating from the external environment, since its mean free path is so short in plastic or silicon. Even though impurities in the chip may be below the parts per billion range, given a meg of bytes times n atoms in the vicinity of a gate, natural decay can certainly occur often enough to cause a countable number of bits to toggle to the opposite state. Actually, the vulnerability of memory arrays to ionizing radiation is not altogether a bad thing. This effect is exploited by the CCDs (charge coupled devices) in some video cameras and astronomical grade photon catchers. Somewhere in my stack (pile? heap? compost?) of Byte back issues, Ciarcia published the design of a video camera which used an ordinary memory chip (sans cover) as the detector. Mebbe we can set up an Amiga graphic bit plane as a photon detector, and watch for little pieces of Chernobyl #4 to drift by overhead. 😉