Share Email Print

Proceedings Paper

Far-infrared spectra of galaxies
Author(s): Charles H. Townes; Norbert Geis
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The far infrared is an important region for the study ofgalaxies. These infrared wavelengths easily penetrate the clouds ofdust and molecules found in many galaxies which tend to hide their intaiors. In addition, the temperatures and densities ofgalacticmolecular clouds fall in a gieral range which easily excites far infrared spectra, but which do not produce much radiation at short wavelengths. Hence spectroscopic measuremaits in this region are of great value in providing measures of temperature, density and velocity. Howeva, it is only within the last ten or fifteen years that much galactic spectroscopy has been done in this region partly because technology for sensitive detectors and for some of the other opticai components has only recently been developed. In addition, the opacity of our atmosphere requires special platforms, generally above most ofthe atmosphere, for astronomical observations at these wavelengths. Fig. 1 shows the transparency ofour atmosphere at various wavelengths in the visible and infrared from the ground, from the height of the observatories on Mauna Kea, and from the height of NASA's aircraft observatory, the Kuiper Observatory, which generally operates at 40,00045,000 feet. Clearly, between about 40 and 300 microns wavelength, essentially no signals are transmitted to the top ofMauna Kea, while thae is substantial transmission to the high flying Kuiper Observatoiy, a C141 four engine aircraft carrying a 36 inch telescope. The telescope is gyrostabihzed and at altitude looks out ofthe side ofthe aircraft with no obstructing windows. Fig. 2 shows a somewhat more detailed picture ofthis transmission. Even at 14 kilometas transmission is by no means perfect the are many regions where spectral lines are well transmitted but also other regions whe radiation is absorbed by atmospheric spectral lines, particularly those ofwater vapor. From the point of view of transmission, an ideal base from which to observe the far infrared is a spacecraft or satellite, such as the Infrared Astronomical Satellite (IRAS), which flew a 60 inch cooled telescope for several years in orbit. However, its function was primarily to obtain systanatic information ofa broad-band nature in the infrared, which it did well, ratha than spectra. Up to the present, most ofthe spectral work done in the far infrared has been carried out in the Kuipa Observatory already mentioned. There is also the possibility ofdoing useful far infrared work from the South Pole. The land mass at the South Pole is actually below sea level, but a pile of about 10,000 feet of snow and ice on top of it gives the structures built atthe South Pole an altitude of about 9,300 feet. This altitude, combined with vy low temperatures which freeze out wat vapor, provide enough transparency in certain localized frequency regions that the South Pole can also be quite useful for particular spectral lines which fall in these atmospheric windows.1 Balloon borne telescopes, which fly considerably higher than the Kuiper Observatoiy, can also be useful. However, they do not carry human attendants to adjust or modil3r the equipment, whereas the airplane can take a sizable crew of passengers who work with the telescopes and spectrometers while observing. The plane also characteristically lands 'and takes offagain for further observations with a turnaround lime ofabout two days, and hence with it the experimental apparatus can be tried, modified and repaired, and flown again rather efficiently. There are three rather different types of spectrometers presently in use for astronomical measurements in the far infrared. The first is a grating systan constructed by Erickson and his associates at the NASA-Ames Research Center, which is shown in Fig. 3 . Ituses a linear array ofdctectors, with three different types ofdetectors for various wavelength ranges and as many as 13 detectors responding at any one frequency. It has excellent sensitivity and a resolving powerof about 5,000. A second type of spectrometer involves the use of Fabry-Perots, and is shown schematically in Fig. 4. Such a Fabiy-Perot system has the advantage that it can examine spectral lines in a number of geometric positions at once. The system shown in Fig. 4, which has been built jointly by groups at the University of California at Berkeley and at the Max Planck Institute in Garching, Gennany, has a 5 x 5 array of detectors and hence maps spectral lines over sizable regions rather rapidly. With three Fabiy-Perots in series, it has a resolving power as high as about 100,000. Its NEP approaches i0 waUs/sec'. The third type of spectrometer, shown in Fig. 5, uses heterodyne detection, with far infrared gas lasers as local oscillators. At infrared wavelengths as long as 300 run, harmonics ofGunn diodes have also been used for a heterodyne detection spectrometa. Hetaodyne detection can achieve exceedingly high spectral resolution and thus resolve any spectral line. For a narrow spectral region it has somewhat higher sensitivity than direct detection, but it is somewhat less saisitive for rather broad lines. The heterodyne system shown was built by Betz3, who is now at the University of Colorado.

Paper Details

Date Published: 19 August 1994
PDF: 25 pages
Proc. SPIE 2211, Millimeter and Submillimeter Waves, (19 August 1994); doi: 10.1117/12.182988
Show Author Affiliations
Charles H. Townes, Univ. of California/Berkeley (United States)
Norbert Geis, Univ. of California/Berkeley (United States)

Published in SPIE Proceedings Vol. 2211:
Millimeter and Submillimeter Waves
Mohammed N. Afsar, Editor(s)

© SPIE. Terms of Use
Back to Top