Our MS257 1/4 m Monochromator is also carefully designed for use as a spectrograph. In a monochromator, the design concentrates on the path of light from the input slit, off the grating and to the output slit. Light of other wavelengths is absorbed. At any grating setting, only a very small range of angles around the diffraction angle D, figuratively one wavelength, passes through the monochromator.
We do not use a slit at the output of a spectrograph. Instead we look simultaneously at a wide range of angles D, and therefore at the range of wavelengths that satisfy the grating equation for this range of angles D. The optics of our spectrographs are designed for this. The output is not a slit, but a long strip over which the various wavelengths are spread in a known fashion.
With a spectrograph, the term bandpass refers to the entire wavelength range in the long strip output to the detector. Thus the bandpass in a spectrograph is very wide, limited only by the output aperture. However, it is not enough to simply remove the slit from a monochromator and use this larger aperture. Generally, wavelengths away from the center of a monochromator slit are focused slightly inside the instrument. Spectrographs are designed to correct this curvature; they have a flat output field matched to flat CCDs and PDAs.
Very often, the length of the detector limits the bandpass. For example, the MS257 has a 28 mm flat field, however our standard InstaSpec CCDs are only 25.4 mm in length. Therefore, bandpass must be defined with a particular detector in mind, i.e. 82 nm with a 1 inch array.
The entrance slit width generally determines the resolution of a spectrograph. However, the limiting resolution is reached when the entrance slit is reduced to the width of a single pixel in the detector array. Nyquist Sampling Theory requires us to calculate the resolution over two pixels. Thus the limiting resolution for a 77781/77782 Spectrograph with a diode array and 25 µm pixels would be about 0.2 nm, twice the 0.1 nm limit possible with the same instrument used as a scanning monochromator with 25 µm slits.
Although some improvement can be achieved using arrays with very small pixels, such as our LineSpec CCD, most instruments reach their aberration limited resolution with 10 - 25 µm input slits. Beyond this point, narrower slits and/or pixels only reduce system throughput.
In a spectrograph/detector array system, stray light can be a more significant problem than it is with scanning monochromators. The array is a much larger target for stray radiation; there is no exit slit to limit the field of view for the detector. Some signal is reflected back off the array where it is scattered as stray light, and the array itself typically has less dynamic range than single element detectors.
It is difficult to characterize the stray light behavior of spectrographs. As with monochromators, variation in spectral content of your source, grating choice, and type of signal, contribute to stray light. The structure and choice of array, wavelength region, and internal specular reflections add to this changing picture for spectrographs.
All Oriel Spectrographs are designed for low stray light. The MS260i, MS127i, and Imaging MS257 conform to design rules that eliminate re-entrant spectra. Re-entrant spectra is due to diffracted radiation being reflected off the detector or one of the mirrors, back to the grating, where it is diffracted again, and then focused onto the array. Without proper elimination these ghost spectral lines appear, disappear and move, with wavelength changes.
The term imaging spectrograph has different meanings depending upon the field of application. Here we use the term to describe the point-to-point replication of the input slit at the output plane of the spectrograph. This concept is best understood by comparison; for simplicity our source will be monochromatic radiation, focused on the entrance slit at the instruments F/#. In a conventional spectrograph this source radiation is focused in a narrow, vertical line on the exit plane of the instrument, at the appropriate position for that wavelength. If you mask half of the input slit on a conventional monochromator youll see a decrease in intensity at the output plane with very little spatial difference in the focused line. By contrast, the imaging spectrograph focuses the source radiation on the exit plane so the focused spot parallels the shape and distribution of the input slit. In fact, with an imaging system, if you mask the top half of your input slit, half of your output image disappears. Because of this spatial relationship between input and output, imaging systems are valuable tools for measuring multiple samples simultaneously. Fig. 1 illustrates the results of two fibers carrying monochromatic radiation, placed at the input plane of a conventional spectrograph and the same two fibers placed at the input plane of an imaging spectrograph, such as our MS257.Fig. 1 The monochromatic light from 2 fibers appears in the output planes.
Spatial resolution is the ability of an imaging spectrograph to distinguish between two features perpendicular to the spectral axis. There is no standard measurement. Some manufacturers refer to the number of independent fiber sources that can be resolved, but this is only meaningful for specific fiber diameters and doesn't describe the signal leakage from one channel to its neighbor.
The most significant measurement is the aberration limited spatial resolution. This value is defined as the full width at half height of the smallest feature that can be resolved. If an effective point source (usually a 10-25 µm pinhole) is placed at the entrance slit and illuminated at the instruments F/#, what is the FWHM measured perpendicular to the spectral axis? For the 77782 Imaging Spectrograph this value is 40 µm.