American Institute of Physics
SEARCH AIP
home contact us sitemap
Physics News Update
Number 450, September 30, 1999 by Phillip F. Schewe and Ben Stein

CHILLING MIRRORS WITH LIGHT. In astronomy the effect of atmospheric turbulence on the quality of images acquired by ground-based telescopes can be greatly reduced by "adaptive optics", a corrective process in which parts of the telescope mirror are flexed mechanically by piezoelectric motors in an amount typically equal to a fraction of the wavelength of the incoming light. In interferometric measurements adjustments in mirrors are also desirable, not because of turbulence in the intervening medium but because of thermal noise in the mirror itself. The LIGO and VIRGO interferometers (Update 442), searching for gravity waves, need very still mirrors, the better to observe the flexing of space-time on a scale far smaller than the size of an atom. A new technique might help in this regard. Physicists at the Ecole Normale Superieure and Université P. et M. Curie in Paris (Antoine Heidmann and Michel Pinard, heidmann@spectro.jussieu.fr, 011-33.1-4427-4405), can measure the thermal agitation of mirrors and reduce this unwanted noise by a factor of 20, with pressure from laser light. This corresponds to a spatial sensitivity of the mirror at a level of a billionth of an angstrom. (P.F. Cohadon et al., Physical Review Letters October 18; see figure at Physics News Graphics.)

COUNTING UP TO 100 MILLION. The science of measurement, metrology, has been moving away from standards based on artifacts such as a meter stick and toward the use of quantum phenomena to provide reliable, accurate and, if possible, portable calibrations that can be used by researchers in the field. Examples are resistance defined in terms of the quantum Hall effect (Update 205) and voltage in terms of the Josephson effect (Update 406). Consider capacitance, the measure of how well a tiny reservoir can store electrical charge. NIST already has the best capacitance standard, accurate to 0.02 parts per million (ppm). But this device is cumbersome and, more importantly, its accuracy is frequency dependent. For rendering the value of capacitance in circuits operating outside a certain frequency range, the standard is no better than 2 ppm. A promising new approach to capacitance (pioneered at NIST; contact Mark Keller, 303-497-5430) uses a single-electron transistor (SET), which contains at its heart a tiny refuge for electrons where the arriving charges can be counted one at a time, all the way up to 100 million or more. When combined with an accurate voltage measurement this becomes an accurate capacitance standard (C=Q/V). The SET approach has now achieved a measurement accuracy of about 2 ppm, and the NIST researchers hope soon to reach 0.1 ppm. The setup is relatively portable and its output is largely independent of frequency. (Keller et al., Science 10 Sept.)

QUANTUM COOL Physicists at Simon Fraser University in Vancouver are trying to get electrical circuits to cool themselves electrostatically. To do this they employ both quantum and classical physics. First, the classical: a gas can cool down by pushing against a piston; some of the gas's thermal energy is converted into mechanical energy. Second, the quantum: electrons flowing from one GaAs layer into another via another a thin layer of AlGaAs will move with optimum efficiency if the electron energy matches a preferred "resonant" energy in the AlGaAs layer. This three-layer setup, called a quantum well, is at the heart of grocery-store laser scanners and CD players. As circuitry shrinks, disposing of waste heat from even tiny electric currents becomes an ever greater problem. The Simon Fraser researchers are proposing that the electrons in a quantum well cool themselves by moving against not a piston but against an opposing electric field, a field in addition to the one moving the electrons through their circuit. This way of combining the quantum (the electrons as waves tunneling through a thin layer) and the classical (the electrons as a working fluid in a sort of Carnot heat engine) might lead to a completely new category of microelectronic quantum device. (Luis Rego and George Kirczenow, kirczenow@sfu.edu, Applied Physics Letters, tent. 11 Oct.)