Abstract
The class of grazing incidence spectrometers proposed by Cash [ Appl. Opt. 21, 710 ( 1982)] and by McClintock and Cash [ Proc. Soc. Photo-Opt. Instrum. Eng. 331, 12 ( 1982)] for use in the extreme and far ultraviolet has been analyzed. We calculate efficiencies which are lower by a factor of 4 than those estimated by the above authors. This discrepancy is due dominantly to their use of incorrect reflectances. We also calculate a factor of 10 smaller bandpass (3%) and a factor of 2 lower spectral resolution (λ/Δλ ≃ 4000) at the spectrum edges. These discrepancies are due to underestimated camera mirror aberrations and neglect of echelle ripple considerations. We consider modifications to overcome these limitations. A WolterSchwarzschild type-I camera mirror of large radius increases the bandpass by a factor of 10. However, the proposed spectral resolution over this bandpass requires a very long instrument and/or a significant further decrease in efficiency. A sample design consists of a normal incidence camera and a grazing incidence 0.5-sec of arc collecting mirror and delivers λ/Δλ = 20,000 over a 40% range in wavelength. However, the normal incidence reflection decreases the efficiency by an additional factor of 3, only wavelengths longward of ~400 Å are directly accessible, and the total instrument length is 6.7 m. Such limitations preclude the usefulness of this design class in several applications currently under study in astronomical space instrumentation. Other classes of grazing incidence spectrometers which are not subject to these limitations are being investigated.
© 1984 Optical Society of America
Full Article | PDF ArticleMore Like This
Webster Cash
Appl. Opt. 21(4) 710-717 (1982)
W. Werner
Appl. Opt. 23(14) 2408-2411 (1984)
Michael C. Hettrick
Appl. Opt. 23(18) 3221-3235 (1984)