DIFFRACTION GRATING (Fr., Reseau de diffraction ; Ger., Diffraktions-Gitter) An opaque screen containing a large number of fine skits, or a transparent screen having opaque lines engraved upon it very close together. A pencil of light is formed of a wave front or a series of overlapping waves which may be repre sented by the diagram A, in which r, is the light source and A B the main wave front in which every particle excites fresh secondary waves, as shown by the curve C D. Nearly all these secondary circles mutually interfere one with the other, except in the main wave front c D. This interference cannot be seen, but its exist ence can be proved by limiting the size of the main wave front by an opaque screen s s, which stops out some of the secondary waves but not all, and the latter are seen as delicate fringes P on each side of the main wave front. This can be experimentally proved in a very simple way. Take a black opaque card about 6 in. or 8 in. square, and cut in the centre a slit about in. long and about *in. wide. Take also a piece of glass about 3 in. square and either smoke it or cover it with black varnish, and with a fine needle-point scratch a thin, dear line about in. long. On holding the card at arm's length dose to a brilliant light, and examining it through the scratch on the glass, held dose to the eye, there will be seen a bright central image of the slit, and on each side of it faint black lines, which are the diffraction fringes. If in front of the light source a violet glass is placed and then a deep red one, there will be obtained images, as shown in the diagram B, in which o is the central image, v the violet bands or fringes, and R the red ones, the violet being nearer together than the red. If a green glass is used the green fringes would fall midway between the red and violet.
The explanation of this phenomenon is as follows : In diagram C let A B represent an opaque screen with an aperture C D, and r, the beam of light, which, proceeding in a straight line, forms a bright central image at it P. Now diffraction, or the bending of the light waves round the edge of the opaque screen, will cause secondary waves to proceed in all directions from every particle of ether lying between C D. For the sake of clearness, let us consider only the waves in one direction, and represent these as straight lines C G D H. Let us further assume that between c and D there are eight ether particles acting as sources of secondary waves. If we now draw c w at right angles to the path of the rays, it will be at once seen that the waves from D have further to travel than those from c by the distance D w. Let D w be a wave length, then obviously D w is exactly half a wave length ; drawing a perpendicular from w to the ether particles we at once see that the wave from 4 is exactly half a wave length from c, and the same distance in front of that from D. By the same reasoning it will be found that r is half a wave-length in front of 5, 2 half a wave-length in front of 6, and so on ; so that every ray is in opposite phase with another ray in the slit. Now two rays in oppo site phase (see " Interference of Light ") produce interference or darkness, so that on the screen GBH r there would be a dark band. By similar
them, the rulings acting as opaque screens ; those on metal are known as reflection gratings. Diagram B represents practically the spectra which are obtained with any diffraction grating. They are arranged on each side of the central white image o. The spectra nearest 0 are called the spectra of the first order, then there is a reasoning we could find beyond this particular angle a bright band where the secondary waves would be in the same phase.
It is obvious that the greater the number of apertures in an opaque screen the greater the number of secondary waves formed, and there fore the greater the chance of interference. Further than that, the narrower the slit the greater must be the obliquity or the angular distance from the central image, for the greater obliquity will be required to produce the neces sary difference between the paths of the rays from a narrower slit ; that is to say, the more slits there are in a unit length the greater the obliquity. Now we have already seen that the violet bands are closer together than the red, and that the green would lie in between, so that if we illuminate a series of slits by hetereogeneous white light the waves will be sifted out into their respective positions, and we obtain a spectrum in which the rays are arranged accord ing to their wave length.
Fraunhofer was the first to utilise the pheno menon of diffraction, and he made his gratings of silver wire wound round two fine threaded placed some distance apart. The next forward step was the ruling of a series of fine lines with a diamond on glass, and later still the glass was silvered. Then Rutherford, of New York, ruled the lines on plane speculum metal, and later on spherical mirrors of speculum metal. Diffraction gratings on glass are called transmission gratings, as the light passes through dark space filled by the invisible infra red and the invisible ultra violet ; then we have the spectra of the second order overlapped even in the visible spectrum by the violet of the third order. Then follow the fourth and other order spectra, the number being dependent on the brilliancy of the light and the character of the grating.
A moment's consideration will prove that the spectra formed by a diffraction grating cannot be as brilliant as d. prismatic spectrum, for although there is some loss of light with the latter, through the reflection from the front surface of the prism and by absorption in the glass itself, yet in the case of the diffraction grating not only does the bulk of the light pro ceed to the central image, which is useless, but the rest is split up into the various spectra on each side. Sometimes, too, in consequence of some peculiarity in the ruling, one or other of the spectra may be much more brilliant than the others.
A grating spectroscope or spectrograph is, however, much to be preferred to a prismatic, as the rays are arranged exactly according to their wave-length, whilst with the prismatic spectrum the violet and blue are spread out at the expense of the red and orange, which are cramped together. (See also " Spectrum," " Dif fraction Grating Replicas," etc.)