Interferometer

INTERFEROMETER. An instrument which makes use of the interference of light waves (see LIGHT) to measure very small differences in length or very small differences in wave length. On the one hand, if very homogeneous light (i.e., light which covers an extremely small range of wave-length, often re ferred to as light of one wave-length) is used, a difference of optical path can be measured by the help of the interference fringes formed: on the other hand, if the light contains two or more separate, but very near, wave-lengths, the component wave lengths can be detected, and their separation measured, by the separation of the fringe systems. (See SPECTROSCOPY.) The in terferometer has assumed great importance in modern physics, firstly from its use in attempts to measure a difference of optical path due to motion through a hypothetical ether, in experiments of the type initiated by Michelson and Morley (see ETHER, RELA TIVITY), secondly from its use in the determination of the ulti mate standard of length, that is, in measuring the length of the standard metre in terms of wave-lengths of a standard light (see SPECTROSCOPY), and thirdly from its use in resolving very close spectral lines, to which reference has just been made.

Separation of Interfering Beams.

The interferometer of Fabry and Perot, in which the interference is produced by re peated reflection of the light between two accurately plane glass plates, is described in the article LIGHT, where reference is also made to another multiple reflection interferometer, the Lummer plate. In both these instruments the reflections lead to what is virtually a large number of sources, with a large path difference, amounting to some tens of thousands of wave lengths, between successive beams. In contrast to such instruments, where the interference is due to the combined effect of a large number of beams produced by the reflections, are instruments in which two beams only, between which interference takes place, are pro duced from the original beam, the two beams following widely separated paths, so that, for instance, obstacles can be placed in one beam and their retarding effects measured by the conse quent displacement of the interference fringes. In the instrument of Jamin the separation is effected by the use of thick plates, the interference from which was first investigated by Brewster. Two similar thick glass plates, with accurately parallel surfaces, are arranged as in fig. I. The one of the two interfering beams is that which is reflected at the first surface of the first reflector and at the second surface of the second reflector : the other under goes reflection at the second surface of the first reflector and at the first surface of the second reflector. If the plates are exactly parallel and exactly equal in thickness the two paths are equal and the two beams should leave the second plate exactly in phase; if there is any lack of parallelism or in homogeneity of the glass interference fringes will be seen. If any substance with a refrac tive index greater than that of air be introduced in the path of one beam only the retardation produced will cause a shift of the fringes, from which the refractive index can be calculated. To enable the measurements to be made with greater ease a com pensator is provided, consisting of two thin glass plates, one in each path, which can be independently tilted to make any desired angle with beam: the greater the inclination the greater the thickness of glass traversed by the beam, so that any desired re tardation can be introduced. The instrument has been largely used to measure the refractive indices of gases, for which pur pose glass-ended tubes are introduced into the path of the beams, one being evacuated, while the gas is slowly introduced into the other, the passage of the fringes being counted. Since the instru ment was originally used for this purpose by Jamin it is often known as an interference refractometer. Extensive measurements of gaseous refractive indices have been made with it by C. Cuthbertson.

In the interferometer, also used for the accurate measurement of refractive indices of gases and weak solutions, devised by Lord Rayleigh (3rd Baron), light proceeding from a slit at A, fig. 2, perpendicular to the plane of the paper, falls upon a colli mating lens B, which is blocked out by a screen except for two parallel slits at C and D. The parallel beams CE, DF trans mitted by these slits are brought to a focus at G by the lens EF, and form interference bands in the focal plane. These bands, which are examined with a high power, eye-piece, undergo dis placement if any substance which causes relative retardation of the light is introduced into one of the paths. Consider, for ex ample, what happens at the point G itself, which is the image of A. If everything is symmetrical so that the paths ACEG, ABFG are exactly equal, there is brightness, but if, for example, CE be subjected to a relative retardation amounting to half a wave length, we have darkness at G, the band being shifted through half a band interval.

The Michelson Interferometer.

The best known form of interferometer, and the one to which the designation was origi nally applied, is that devised by Michelson for the Michelson Morley experiment (see ETHER, RELATIVITY, MICHELSON-MOR LEY EXPERIMENT), on the effect of the motion of the luminif erous medium on the velocity of light. In this the interfering beams are not only widely separated, but travel, as is required by the experiment, at right angles to one another for the greater part of their path. Light from the source S, which may be an ex tended luminous surface (a can dle, lamp or a lens with an arc light at the focus) falls at 45° on the surface of a plane-parallel glass plate (fig. 3), the surface of which is covered with a film of silver or platinum, of such thick ness that the reflected and trans mitted pencils are of approxi mately equal intensity. The transmitted beam falls normally on the mirror C, the reflected beam on B, both returning to the separating surface A, whence they both proceed in the direction AE. Since the beam reflected from C has had to pass three times through the plate A, whereas the beam from B only passes once through A, a compensating plate P' similar to the plate A, is introduced, which is traversed twice by the beam which goes to B. The resulting interference fringes may be projected on a screen or observed by eye, with or without an observing telescope.

The construction of the instrument is indicated in fig. 4. A heavy casting serves to support the optical parts, and the car riage holding the movable mirror C moves on very accurately ground ways. The motion is communicated by means of a screw provided with a worm wheel and a divided circle so that the motion of the carriage may be accurately measured. The sta tionary mirror D is provided with screws for adjustments about vertical and horizontal axes. The compensating plate B is held by a vertical steel rod, twisting which produces any required small alteration in the path. All of the optical surfaces are very accu rately plane, the errors being of the order of a twentieth of a light wave, or less.

The adjustment of the instru ment is effected as follows. The distances of the mirrors C and B from the half-silvered surface of A are made approximately equal (say, to within a millime ter), and an approximately homo geneous source of light (sodium flame, or better, a Cooper-Hewitt mercury arc) is placed in front of A, as indicated by S. The two images of a needle point placed near S are then brought into co incidence by the adjusting screws of the mirror D, when the interference fringes should ap pear. They are usually narrow, curved, and not very distinct ; but by slowly altering the adjustment of the mirror D they may be given any suitable width, and by diminishing the path difference by turning the screw S the fringes become more distinct. As the path difference approaches zero, the change of inclination of the fringes accompanying a change in position of the eye diminishes; and when this change vanishes, the (coloured) fringes in white light appear, or may be found in a few turns of the worm wheel which gives the slow motion to the screw S. The use of the instrument in the Michelson-Morley ex periment is described under RELATIVITY, in measuring the stand ard metre in terms of lightwaves under SPECTROSCOPY.