The link:
patents.ibm.com
Fundamentally, it has been found that adjacent cavity modes are plane polarized with their planes of polarization at right angles to each other. In order to isolate and capture the two cavity modes, the output signal of the laser 12 is split into two beams in beam splitters 24,26.
Where did the second mode come from?
Why should the two modes be "captured"?
A polarizer 28, 30.inserted into the path of each beam. Each polarizer 28,30 is oriented orthogonally with respect to the other, to isolate the two adjacent cavity modes. Two matched InGaAs detectors are used to detect the two modes. The difference in the output of the detectors 28, 30 is proportional to The difference in power of the two modes.
Does this answer the above question?
The output of the two detectors 28, 30 is, provided as an input to a differential input voltage controlled oscillator (VCO) 32. The differential input from the two detectors 28, 30 is used by the differential input VCO 32 to adjust the power to the heating coil 20 to, in turn, change the length of the optical cavity of the laser 12. This, in turn, alters the mode frequencies of the. laser 12. The operation of the first feedback loop 13 has been found to stabilize the frequency of the laser 12 to within 10 Angstroms.
Electrons don't move at the speed of light. Will the integrated feedback from the VCO have a steep enough hysteresis slope with respect to the cavity wrapped heater to induce sufficiently fast heater adaption inverse to heat variance?
To further stabilize the laser 12, a second feedback loop 14 is provided. Within the second feedback loop 14, an output beam of the laser 12 is compared with an output beam of a reference paradigm laser 40 and the difference provided as a second feedback signal as a further means of achieving control over the stability of the laser 12.
Using a beam splitter 26, a small portion of the output beam of the laser 12 is provided as a first input to an optoelectronic clock 34 through a first monochrometer 34, a second input to the optoelectronic clock 34 is provided from the paradigm laser 40, through a second monochrometer 38. Under the embodiment, the first. and second monochrometers 34, 38 may be constructed substantially the same.
Within the monochrometer, the beam goes through a slit 50 (FIG. 2), passes through a beam expander-collimator 52 and then on to a grating 54. From the grating 54, the beam goes to a parabolic mirror 56 with a 24 cm focal length.
The slit 50 is fabricated with characteristics which limit the bandwidth of incident radiation. Under the embodiment, the dimensions of the slit 50 may be determined using an appropriate bandwidth limiting technique (.e.g., see Richardson, D., Diffraction Gratings, Applied Optics and Optical Engineering, Vol. V, Part II, Optical Instruments, R. Kingslake, ed., Academic Press, New York, N.Y, (1969)).
The beam expander and collimator 52 are of conventional design for a 13,000 Angstrom signal, using quartz optics. The grating 54 may be fabricated with a blaze angle of 24 degrees and a line spacing of 13.4 x 10^4 lines/inch or 527.559 lines/mm.
Where light is incident at an angle, a, relative to the surface, normal to a reflecting grating, ruled with a spacing, d, the path difference for the light incident on any two adjacent grooves is d*sin a. When this light is diffracted from the rulings at some other angle, b, the path difference for the light is further increased by the amount of d*sin b. The reflected light of wavelength, L, will be in phase over the entire wave front when the path difference for rays incident on adjacent grooves is an integral multiple of the wavelength. For light of a given wavelength incident at a particular angle, the light reflected from all of the grooves will be in phase only at certain angles. The number of wavelengths of a path difference from adjacent grooves is called the order of interference, m. Using the described variables, a grating equation can be written as follows
m*L = d*(sin a = sin b).
When b is equal to the blaze angle c, then Lb, the first order blaze wavelength can be described as:
DEF: Blaze Wavelength
The light wavelength for which the direction of reflectance from the groove face is identical to the angle of diffraction for a specified angle of incidence.
Lb = 2*d*sin c.
For a grating 54 where c is 24 degrees and the grating spacing is 527.559 lines/mm, the spacing, d, is equal to 18,955.3 Angstroms.
From the equation, Lb can be determined as follows
Lb = 2*d*sin c = 2*1.89553 x 10^-(10)meters*(sin 24 degrees) = l5,419.56 Angstroms.
If the blazed wavelength, Lb, for the condition a = b is known, and the blazed wavelength Lb' is for other combinations of a and b is desired, then the desired values may be determined as follows
L' = Lb*cos((a' + b)/2)
For the particular case of Lb = 1.3 x 10^(-6) meters, where we want to find the angle normal to the grating that is suitable for our wavelength, the angle may be determined by
Lb'/Lb = cos((a'+ or - b)/2)
which may be simplified to produce,
1.31/1.541956 = .843085 = cos(a' + or - b)/2) and
arccos(0.843085) = 32.53266 degrees.
From this value, a' may be determined as follows
1
a' = 2*(32.53266) + 24 degrees of blaze angle = 89.0653 degrees.
Another issue regarding the grating 54 is that of resolution. Using a 6 inch wide, grating, the grating resolution can be determined using Fraunhofer diffraction theory from the following expression
L/dL = N*d*(sin a + sin b)/L
where
N = number of grooves = 527.5510 lines/inch = 80,400 lines and
d = spacing between grooves = 1.8955 x 10^(-6) Meter = 18,955.2 Angstroms
Substituting N and d into the equation results in the expression as follows
80,400 lines = L/dL = 1.89552 x 10^(-6) meters*(0.9998669 - 0.4067366)/1.30 x 10^(-6) meters
L^2/0.09039294 meters = dL
L/dL = 0.09039294 meters/1.3 x 10^(-6) meters and
Angstroms = 1.8696 x 10^(-11)meters/10^-(10)meters/Angstrom = .18696.
The resolving power of a grating is a measure of its ability to separate adjacent spectrum lines. It is expressed as L/dL, where L + dL is the wavelength of a spectrum line that is just barely distinguishable from a line at wavelength L.
Using the techniques developed herein, it should be evident that large gratings used at high angles are needed to achieve high resolving power. The actual attainment of high resolving power with a grating depends upon the optical quality of the grating surface, the uniformity of the spacing of the grooves, and the associated optical components. The equation suggests that one should be ale to have an infinite resolving power simply by increasing the total number of grooves in a given width. However, there is a fundamental rule that has to be applied, (i.e., that it is absolutely necessary that the incident wavelength be less than the groove spacing (d)).
Otherwise what would happen?
If not for the fundamental rule, a grating having 4500 lines/mm would have been selected.
Why?
Unfortunately, this would have provided a groove spacing of 2220 Angstroms which is 5.85 times too small to provide an effective solution to the problem.
Why 2220 Angstroms?
Why 5.85 too small?
Using the techniques described herein, the resolution of the laser 12 has been improved to provide a linewidth that is approximately 0.18696 Angstroms or 1.8696 x 10^(-11) meter. That is, the linewidth of the grating 54 spans the range of between 13,000.18 Angstroms and l2999.81 Angstroms. Effectively the line width is stable to 5 parts in 10^11. The approximate bandwidth can be shown as follows
d(phi)*L = 3 x 10^8/(4*(dL/L)*Ym) = 2.6075 x 10^15
where Ym = .002, which is a dimensionless material dispersion coefficient.
DEF: Monochromator
Also known as monochromatic illuminator. An instrument for isolating narrow portions of the spectrum. The spectrum of any light source is formed by a prism or grating, and an exit slit placed in the spectrum selects a narrow band of wavelengths for emission. By moving the spectrum of a source internally past the slit, the color of the emitted light can be varied at will. As most monochromators emit several percentages of unwanted light, either white or of the wrong wavelength, along with the desired wavelength, two monochromators are often used in tandem, both being set to transmit the same wavelength. In this way the percentage of unwanted light can be drastically reduced.
Last question: What is the focusing parabolic mirror doing aside from focusing? Does it collect the entire spread of the expanded diffracted beam for focusing?
|
