Ocean Optics MMS Raman - Manual (Page 22)

B: Introduction to Multimodal Sampling
MMS and Conventional Multiplex Spectroscopy
Multiplex design has been applied to spectroscopy for over 50 years, including both Fourier and
Hadamard transform designs. In general, these designs have been applied to infrared spectroscopy and
have emphasized minimizing the number of electronic detectors used. For example, dynamic mask
Hadamard transform spectrometers employ spatial light modulators (SLM) such as MEMS based devices
or electro-optic elements to temporally multiplex spectral channels. SLM is a device that controls the
transmission or reflection of light electronically and is placed between the grating and the detector. The
SLM spectrally encodes the dispersed light from the grating and combines them in coded fashion (like a
Hadamard transform) onto the detector, one row at a time. Since the coding is done between the grating
and the detector, there is no real etendue advantage in such spectrometers. Furthermore, data acquisition
is performed serially.
In contrast, MMS is a true 2-dimensional parallel acquisition and processing system that captures both
spatial and spectral information simultaneously throughout the entire aperture. This leads to a number of
performance advantages (not the least of which is higher etendue and SNR) as well as unique features
such as multi-input and hyperspectral imaging. These important differences enable MMS designs to
outperform conventional multiplex as well as fiber and slit entrance spectrometer designs. In addition to
its performance advantages, MMS can be implemented using commonly available low cost components.
Performance Comparison
In this section, the performance of MMS spectrometer is compared with that of a conventional slit and
fiber input spectrometers. System components for each system (f number, grating, lenses, filters, detector
etc.) are identical. For these experiments, a Raman spectrometer with an excitation wavelength of 663 nm
is used as the test-bed.
This system implements an f/2 optical design and a Kodak CCD detector, cooled to -18oC. The system
was set up to accommodate each of three entrance designs: 1) pin hole, 2) vertically binned slit, and 3)
coded aperture. To provide a fair performance comparison of the three sampling methods, the width of the
pin hole, slit width and feature size of coded aperture are equivalent. Thus the optical resolution of each
configuration is equal. Additionally, the height of the slit and aperture are equal.
Etendue
Etendue is a well accepted measure for optical throughput as it specifies the geometric capability of an
optical system to transmit radiation. The numeric value of the etendue is a constant of the system and is
calculated as the product of the entrance aperture (or slit area) and the solid angle through which light is
accepted.
Assume that the input aperture implements an order N matrix. Such an aperture typically has 2N x N
aperture elements, with each element proportional to the pixel size. The etendue of such a spectrometer is
given by:
Etendue (MMS) = 0.5 x 2N x N x Ω
where Ω is the input solid angle and the factor of 0.5 takes into account the fact that only 50% of the
aperture elements are transparent.
20 000-40000-000-02-0906