A discussion of the performance characteristics of an infra-red microspectrometer is given in terms of the cross-sectional area and minimum volume, V, which can be observed with satisfactory signal-to-noise ratio. Methods of increasing the ratio of measured absorption to spectrometer noise are discussed and five ways are enumerated. It is pointed out that for a microscope objective of numerical aperture NAm associated with a spectrometer of numerical aperture NAs, the useful magnification is (NAm)/(NAs). The design of reflecting-type infra-red microscope objectives having numerical apertures up to 1.5 is described.
A description of an experimental infra-red microspectrometer is given. Use is made of a commercial spectrometer and conversion from a micro to a macro instrument can be made in a few minutes. In the experimental arrangement no changes were introduced that affected the operation of the spectrometer as a macro instrument. Calculations indicate that by making use of other components now available, such as a hotter source and a smaller, more sensitive thermal detector, it will be possible to obtain infrared spectral measurements of specimens whose linear dimensions approximate those set by diffraction.
Examples of infra-red spectra of crystals, fibers, and tissues of microscopic area are shown. Comparisons of spectral data with those obtained for macro samples are made and an indication of the experimental limitations of the technique is given.
ELKAN R. BLOUT, GEORGE R. BIRD, and DAVID S. GREY, "Infra-Red Microspectroscopy," J. Opt. Soc. Am. 40, 304-312 (1950)