APPARATUS FOR MEASURING THE REFRACTIVE INDEX OF GASEOUS MEDIA
This invention relates to optical measuring instruments and, in particular, to apparatus for measuring the refractive index of gases.
In British Patent No. 2120383B, there is described a portable optical instrument for precisely determining the refractive index of air using a laser source which emitted one wavelength together with a double path optical interferometer and a gas cell of known length in one of the paths. The measurement procedure required the cell to be initially evacuated following which the sample gas was leaked into the cell and the corresponding change in optical path length was determined by continuously counting interference fringes. From the knowledge of the cell length and the change induced in the optical path length, a value for the refractive index of the gas could be easily obtained. However, any further refractive index measurements normally required this complete measurement procedure to be repeated. We have now devised an interferometer which, while having similar refractive index measurement accuracy, has a number of significant improvements:
1) it uses multiwavelength laser technology which reduces the overall cost of the instrumentation. 2) it employs a dual chamber gas cell which contains a medium with a constant refractive index in the inner chamber. This, together with 1), eliminates the requirement for the regular evacuation of the cell which was necessary for the operation of the original refractometer.
3) it can be operated in two modes to produce refractive index measurements. According to the present invention there is provided optical apparatus for measuring the refractive index of gaseous media including radiation source means to produce a beam of
polychromatic radiation, beam splitter means to produce a pair of component beams from said beam of polychromatic radiation, dual chamber cell means have first and second chambers, deflector means to direct a first of said pair of component beams through said first chamber and said second of said pair of component beams through said second chamber, recombining means to recombine said first and said second component beams after passage through said chamber, dispersion means differentially to deflect radiation of different wavelengths in said first and second component beams after recombination and sensing means disposed to received radiation deflected by said dispersion means.
The invention will be specifically described with reference to the accompanying drawings, in which:- Figure la and b depicts an optical measuring instrument in accordance with a specific embodiment of the invention; Figure 2 is an explanatory diagram showing various signals associated with the instrument of Figure 1; and Figure 3 depicts another embodiment of the invention.
Referring now to the drawings, Figure 1 depicts an optical measuring instrument which uses several laser wavelengths.
Laser light from a source L is reflected by a mirror Mχ and is then incident on a fused silica beam splitter BS which has a semi-reflecting front surface and a fully reflecting back surface. The reflected beam 1 from the front surface travels down the outer arm 3 of a dual compartment cell 5 and is returned by a corner cube reflector 7 along a parallel path to the beam splitter. The transmitted beam 9 through the beam splitter is reflected by the fully aluminised back surface and travels down the inner arm 11 of the instrument before being returned by the cube corner reflector 7 to the beam splitter BS where it is reco bined with the beam that has travelled along the outer arm. These superposed beams, which produce interference, are reflected by mirror M and separated into
discrete wavelength components by a dispersive prism P. The intensity of the light at each wavelength, which is directly related to the phase of the interference, is incident upon separate photodetectors (labelled Pχ to Pa in the drawing) that produce electrical signals for processing from which interference fringe fractions are determined. The instrument also uses a single compartment gas cell 13 through which only the transmitted beam 9 passes.
This configuration of refractometer is used to determine the change in optical path length between the two arms by using the measurement of the fringe fractions at each of the wavelengths used. For example, when a dual wavelength laser is used emitting wavelengths Xx and λ2 any change in optical path length (dl) in either of the arms of the instrument is given by (if air dispersion is ignored):
dl = (m1 + Φi)}^ = (m2 + Φ2)λ2 (1)
where x and 2 are the integer numbers of the interference orders and φx and φ2 are the fringe fractions for each wavelength.
This sequence of fringe fractions will be repeated every time the optical path length changes by λ1λ2/(λ1-λ2) . To remove this ambiguity either a third fringe fraction from an additional wavelength can be used or an approximate value of dl is employed which is determined by other means. Hence any change in optical path length can be obtained in this way without using continuous fringe counting.
For the specific case of the refractometer, a dual compartment gas cell is used which has an approximate length of 1=31.64 cm. This length is chosen so that one fringe of optical path length change is approximately equal to a change of 1 x 10~6 in refractive index. The inner chamber of the cell has a known constant optical path length whilst the outer chamber contains the air requiring refractive index measurement. From
both the measurement of the fringe fractions under these conditions and also when the refractometer has an identical optical path length in both arms, a value for the refractive index of air [n1(λ>] at wavelength λ can be precisely determined from the simple equation:
nx(X) = n2(λ) + dl(λ) + σ (2)
1
where n2(λ) is the refractive index of the medium contained in the inner cell chamber at wavelength λ, dl(λ) is the induced path length change and σ is a correction for any additional optical path length changes induced during the measurement procedure. The multiwavelength requirements discussed in the previous section ideally require a single laser source that continuously emits three separate wavelengths for the unambiguous determination of optical path length changes. However, a simple source of this type does not currently exist but can be simulated by using:
1) different laser sources to illuminate separate optical fibres which have a common output for illuminating the refractometer.
2) both a single laser source which emits two wavelengths and, to eliminate ambiguities, a sufficiently accurate value of dl(λ) derived from the known length 1 of the cell together with a value of air refractive index which may be calculated from Edlen's equation CB. Edlen "The refractive index of air" Metrologia 2, 71 (1966)3 using the measurement of atmospheric pressure and air temperature; or
3) both a single laser source which emits one wavelength together with an improved value of dl(λ), derived in the same manner as 2) above, but using higher accuracy atmospheric pressure and air temperature measurements.
This is only possible if it is known that any variations in the air constituents do not cause the measured and calculated optical path differences to vary by >0.5 fringe. The second of these alternatives is preferable since dual wavelength lasers are readily available with the same common wavelength (633nm) as that used in length measuring interferometers which, for the highest accuracy length measurements in air, would require correspondingly accurate air refractive index values. In addition a suitable value of dl(λ) can be easily obtained using inexpensive and rugged pressure and temperature sensors with respective accuracies of ±500 Pa and ±1°C.
The above embodiment of the invention uses a gas cell that has an inner chamber with a known constant optical path length and this must be maintained to within about 3nm over the ambient ranges of atmospheric pressure and air temperature in order to achieve a measurement accuracy of 1 x 10~8 in air refractive index. This can be realised by using a cell which has either a permanently evacuated central chamber which incorporates a getter pump for maintaining and monitoring the vacuum or a central chamber that is filled with dry gas at atmospheric pressure. Both of these approaches require an initial measurement of the length (1) of each cell to an accuracy of about ±lμm (equivalent to an uncertainty of 1 x 10~9 in refractive index measurement using a cell of length 31cm) followed by the insertion of the cell into the refractometer and the evacuation of both cell chambers to determine the dual wavelength "zero" fringe fractions which should be identical since both optical paths are exactly equal. This ensures that there are no anomalous dispersion effects which could lead to measurement errors.
For the first type of cell, the inner chamber is sealed with the vacuum being maintained and monitored by a getter pump and air is leaked into the outer chamber following which the dual
wavelength fringe fractions are determined. Atmospheric pressure and air temperature are also measured and these values are used to calculate an approximate air refractive index value from Edlen's equation which, with the measurement of the cell length, allows the change in optical path length Cdl(λ)] to be calculated to within a few fringes. Finally the dual wavelength fringe fractions, with this approximate value of air refractive index, enables the exact optical path difference to be determined and hence the refractive index of the ambient air is now readily given by equation (2) since n2(λ) = 1.000. The refractometer is now ready for the continuous measurement of refractive index with the requirement for weekly or monthly repeat determinations of the "zero" fringe fractions.
In the second case, after the evacuation of both cell chambers and the determination of the dual wavelength "zero" fringe fractions, the inner chamber is filled with dry air from a gas cylinder following which this chamber is sealed at atmospheric pressure. The dual wavelength fringe fractions together with the pressure and temperature of the air in the cell are similarly determined. Using a similar technique as that described in the first case the refractive index [n2(λ)] of the air contained in the cell can now be determined precisely from the following equation since n3(λ) = 1.000:
n2(λ) = n3(λ) + dl(λ) + σ (3)
1
Ambient air is admitted into the outer chamber and again the fringe fractions, atmospheric pressure and air temperature are determined from which the refractive index of the air is obtained from equation (2) in the manner previously described. The refractometer is now ready for the continuous measurement of air refractive indices with similar measurement requirements for the "zero" fringe fractions as those discussed previously. In order to ensure that the refractive index in the inner chamber
remains constant fused silica is used to fabricate the cell. This results is an insignificant refractive index correction due to variations in the cell volume induced by ambient temperature changes. In summary the first type of cell has the advantage that the refractive index of the medium is exactly 1.000 and. therefore, does not require the additional measurement stage of using equation (3) whilst the second cell is both easier to fabricate and maintain in a lead-free condition. The dispersion of air must also be considered which for the two wavelengths proposed (612 and 633nm) introduces a difference of about 3 x 10~7 in refractive index at NPT for which a correction must be applied. However if the cell is used which has an air filled central chamber this correction will be reduced by the factor pχ/p2 where pχ is the original pressure of the air in the central chamber and p2 is the ambient atmospheric pressure. This factor can be easily determined since pχ and p2 will already be routinely measured with a sufficient accuracy.
A refractometer can be operated in two modes to produce refractive index data. The first version, shown in Figure 1, allows the optical path in the inner arm to be varied by a few interferometer fringes in order to determine the fringe fractions. This path length variation, which is illustrated in Figure 2, is produced by uniformly varying the pressure in the single compartment cell and monitoring the induced change in optical path as a function of time.
Figure 2 shows two amplified interferometer signals produced by the photodetectors located in the output beams of the refractometer with each photodetector monitoring the interfering output from each wavelength as, in the illustration, the optical path length is varied. Prior to modulation, at time t=0, a clock which generates equally spaced pulses is reset. The number of pulses is counted for the time each fringe signal takes to cross the zero volt line as the optical path length is changed in a given direction - these are at times t=a, b, c and
d in the drawing. In addition the number of pulses per fringe is counted and these are determined in the illustration between times t=a and c and t=b and d for λj and λ2 respectively. The fringe fractions at each wavelength may now be easily determined as the simple fractions ΔNj/Nj and ΔN2/N2 for and λ2 respectively. In this way fringe fraction data is obtained from which the air refractive index can be determined using those techniques described earlier.
The second embodiment of the invention is shown in Figure 3 which illustrates the required change in the arrangement for dual wavelength operation. The plane of polarisation of the light emitted by the laser is required to have an angle of 45° to the plane of the diagram. A quarter-wave plate 15 designed for use at the two wavelengths is inserted into one of the arms of the refractometer and this introduces a 90° phase delay between the 's' and 'p' polarised components of the laser light. The normal interfering 17 and non-interfering 19 output beams are used which are separated into their discrete wavelength components by a dispersive prism. The non-interfering output beams are incident on photodetectors Pχ and P2 whilst the interfering output is separated into 's' and 'p' polarised components by a polarising beam splitter. The intensity of these polarised components is monitored by photodetectors P3 to P6. This arrangement provides phase quadrature fringe detection and using the techniques described by PLM Heydemann in Applied Optics 20, 3382 (1981). The technique relies on calibrating the fringe detection electronics by modulating the optical path length in the refractometer in the same manner as that described previously. If a third wavelength is available an additional three photodetectors are required to measure the intensity of the non-interfering and interfering components.
In the first embodiment, path length modulation is carried out every time fringe fraction data is required and in this way drift in the detection electronics is eliminated, whilst in the second embodiment an initial calibration of the interferometer
signals allows fringe fractions to be determined for several hours without recalibration provided the laser intensity and the photodetector electronics are sufficiently stable.