Optics: measuring and testing – By dispersed light spectroscopy – Utilizing a spectrometer
Reexamination Certificate
2000-02-23
2002-02-26
Turner, Samuel A. (Department: 2877)
Optics: measuring and testing
By dispersed light spectroscopy
Utilizing a spectrometer
C356S329000, C356S073000
Reexamination Certificate
active
06351307
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to interferometric and dispersive spectroscopy of broadband waves such as light, and more specifically the interferometric measurement of effects which can be made to produce phase shifts such as Doppler velocities, distances and angles, and furthermore the mapping of spectra.
2. Description of Related Art
Spectroscopy is the art of measuring the wavelength or frequency characteristics. There are two complementary forms of spectroscopy method currently used today. In the oldest form, a prism or grating disperses input illumination (let us call it light) into independent channels organized by wavelength or frequency. A spectrum is created which is the intensity versus wavelength channel. This is a scalar versus wavelength channel. In the other method, called Fourier transform spectroscopy, an interferometer having a variable path length difference (called the delay) interferes the illumination with a delayed copy of itself, creating an interferogram. The Fourier transform of this yields the spectrum. Previously, the two methods have not been used together where the interferometry and dispersiveness have had equal emphasis.
An important practical use of spectroscopy is the measurement of Doppler shifts. In addition to many industrial applications of Doppler velocimetry, astronomers measure the Doppler velocity of stars in order to deduce the presence of planets orbiting around the star. The stellar spectrum contains numerous dark absorption lines against a bright continuum background. These spectral lines are randomly distributed about 1 Angstrom apart from each other. A slight change in the average position of these lines is the Doppler effect to be measured. The average width of these stellar lines is about 0.12 Angstroms in the visible, which corresponds to an equivalent Doppler velocity width of about 6000 m/s. Hence, measuring Doppler velocities below 6000 m/s is extremely challenging and requires carefully dividing out the intrinsic behavior of the instrument from the raw data.
A 1 m/s velocity resolution is desired in order to reliably detect the presence of Jupiter and Saturn-like planets, which produce 12 m/s and 3 m/s changes respectively in the stars intrinsic velocity. Current astronomical spectrometers are based on the diffraction grating. These have a best velocity resolution of 3 m/s, but is often 10 m/s in practice. This resolution is insufficient to reliably detect Saturn-like extrasolar planets. This limit is related to the difficulty in controlling or calibrating the point spread function (PSF).
The PSF is the shape of the spectrum for a perfectly monochromatic input. Ideally this is a narrow peak of well-determined shape. Unfortunately, the PSF of actual gratings varies significantly and in a complicated way against many parameters such as temperature, time, and average position in the spectrum. It is a complicated function that requires many mathematical terms to adequately approximate it. This is fundamentally due to the hundreds or thousands of degrees of freedom of the diffraction grating—at least one degree of freedom per groove of the grating. These degrees of freedom must be carefully calibrated, otherwise drifts can cause apparent Doppler velocities much larger than the effect being sought. The calibration process is time consuming.
Another disadvantage of conventional astronomical spectrometers is their large size, which can be several meters in length. Large distances between optical components, which need to be held to optical tolerances, require very heavy and expensive mounts and platforms to prevent flexure. This dramatically increases expense and prevents portability. Practical use aboard spacecraft or aircraft is prevented. The high expense limits the number of spectrometers which can be built to a few, only by well-endowed institutions.
Other disadvantages include a very limited field of view, which is called etendue and is the area of the input beam times its solid angle. This is due to the narrowness of the slit at the instrument entrance that defines the range of entry angles. In a grating or prism based instrument the entry angle and the wavelength, and hence deduced Doppler velocity, are directly linked. The slit needs to be narrow to provide better than 0.05 Angstrom resolution to resolve the stellar spectral lines. Atmospheric turbulence causes the star image to dance around, sometimes off the slit opening. This reduces the effective instrument throughput. Furthermore, changes in intensity profile across the slit have to be carefully deconvolved from the data, since the Doppler velocity gradient across the slit is approximately 3000 m/s. Thus achieving 3 m/s velocity accuracy is extremely difficult with a dispersive spectrometer, and 1 m/s has never been achieved.
An interferometer is attractive for spectroscopy because its angular dependence can be made very small or zero. This allows wider slits, and hence accommodating blurrier star images at high throughput, for the same equivalent spectral resolution. Secondly, its PSF is a sinusoid, which is a simple mathematical function having only 3 degrees of freedom (phase, amplitude and intensity offset). This makes calibration of instrument and processing of data fast, since standard vector mathematics can be used. Secondly, this makes it easy to reject noise not having the expected sinusoidal shape. Furthermore, the spectral resolution can be made almost arbitrarily large simply by increasing the delay (difference in path length between the two interferometer arms). The interferometer is compact and inexpensive, because the optical components need only be a few millimeters or centimeters from each other.
The important difficulty of an interferometer measuring broadband illumination is poor fringe visibility. Fringe phases naturally changes with wavelength. When component fringes of many wavelengths combine on the same detector, they reduce the visibility of the net fringe. For this reason conventional interferometer based instruments such as Fourier transform spectrometers without any wavelength restricting filters are rarely used in low light applications.
A solution to this problem is to combine a wavelength disperser with the interferometer so that fringes of different wavelengths do not fall on the same place on the detector. The combination of disperser and Fabry-Perot interferometer is described in the book “Principles of Optics” by Max Born and Emil Wolf, Pergamon Press, 6th edition, on page 336, section 7.6.4 and their FIGS. 7.63, 7.66 and references therein. Distinctions exist between apparatus described in “Principles of Optics” and the present invention. The Born & Wolf device produces fringes that are narrow and peak-like, not sinusoidal. Consequently, the fringe shape is not described by a 2-element vector. This reduces accuracy when trying to measure small phase shifts. Furthermore, phase stepping is not involved. Thirdly, a heterodyning action is not employed to shift high-resolution spectral details to low spectral resolution.
In some kinds of metrology a secondary effect, such as temperature, pressure or acceleration, is measured by the change it induces in the delay of an interferometer, such as through changing the position of a reflective surface or altering a refractive index. The delay is then sensed by the phase of a fringe. In current devices monochromatic illumination, such as laser illumination, is needed to produce visible fringes from non-zero delays. (The delays are often non-zero for practical reasons, or to have a significant range of travel.) However, the use of monochromatic illumination creates fringe skip ambiguities which make the absolute size of the effect being measured ambiguous. Only small changes can be reliably measured. Broadband light solves the fringe skip problem, but produces insufficient fringe visibility because its coherence length (about 1 micron) is usually very much shorter than the delay.
In a related metrology, fringe shifts can be used to measure an
The Regents of the University of California
Thompson Alan H.
Turner Samuel A.
Wooldridge John P.
LandOfFree
Combined dispersive/interference spectroscopy for producing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Combined dispersive/interference spectroscopy for producing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Combined dispersive/interference spectroscopy for producing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2982474