FLUORESCENCE. In a paper read before the Royal Society of Edinburgh in 1833, Sir David Brewster described a remarkable phenomenon he had discovered to which he gave the name of "internal dispersion." On admitting a beam of sunlight, condensed by a lens, into a solution of chlorophyll, the green colouring matter of leaves (see fig. I), he was surprised to find that the path of the rays within the fluid was marked by a bright light of a blood-red colour, strangely contrasting with the beautiful green of the fluid when seen in moderate thickness. Brewster afterwards observed the same phenomenon in various vegetable solutions and essential oils, and in some solids, amongst which was fluor-spar. He believed this effect to be due to coloured particles held in suspension. A few years later, Sir John Herschel independently discovered that if a solution of quinine sulphate, which, viewed by transmitted light, appears colourless and transparent like water, were illuminated by a beam of ordinary daylight, a peculiar blue colour was seen in a thin stratum of the fluid adjacent to the surface by which the light entered. The blue light was unpolarized and passed freely through many inches of the fluid. The incident beam, after having passed through the stratum from which the blue light came, was not sensibly enfeebled or coloured, but yet it had lost the power of FIG. I.
producing the characteristic blue colour when admitted into a second solution of quinine sulphate. A beam of light modified in this mysterious manner was called by Herschel "epipolized." Brewster showed that epipolic was merely a particular case of internal dispersion, peculiar only in this respect, that the rays capable of dispersion were dispersed with unusual rapidity.
The investigation of this phenomenon was afterwards taken up by Sir G. G. Stokes, to whom the greater part of our present knowledge of the subject is due. Stokes's first paper "On the Change of the Refrangibility of Light" appeared in 1852. He repeated the experiments of Brewster and Herschel, and considerably extended them. These experiments soon led him to the conclusion that the effect could not be due, as Brewster had imagined, to the scattering of light by suspended particles, but that the dispersed beam actually differed in refrangibility from the light which excited it. He therefore termed it "true internal dispersion" to distinguish it from the scattering of light, which he called "false internal dispersion." As this name, however, is apt to suggest Brewster's view of the phenomenon, he afterwards abandoned it as unsatisfactory, and substituted the word "fluorescence." This term, derived from fluor-spar after the analogy of opalescence from opal, does not presuppose any theory. To examine the 'nature of the fluorescence produced by quinine, Stokes formed a pure spectrum of the sun's rays in the usual manner. A test-tube, filled with a dilute solution of quinine sulphate, was placed just outside the red end of the spectrum and then gradually moved along the spectrum to the other extremity. No fluorescence was observed as long as the tube remained in the more luminous portion, but as soon as the violet was reached, a ghost-like gleam of blue light shot right across the tube. On continuing to move the tube, the blue light at first increased in intensity and afterwards died away, but not until the tube had been moved a considerable distance into the ultra-violet part of the spectrum. When the blue gleam first appeared it extended right across the tube, but just before disappearing it was confined to a very thin stratum on the side at which the exciting rays entered. Stokes varied this experiment by placing a vessel filled with the dilute solution in a spectrum formed by a train of prisms. The appearance is illustrated diagrammatically in fig. 2. The greater part of the light passed freely as if through water, but from about half-way between the Fraunhofer lines G and H to far beyond the extreme violet, the incident rays gave rise to light of a sky-blue colour, which emanated in all directions from the portion of the fluid (represented white in fig. 2) which was under the influence of the incident rays. The anterior surface of the blue space coin FIG. 2. cided, of course, with the inner surface of the glass vessel. The posterior surface marked the distance to which the incident rays were able to penetrate before they were absorbed. This distance was at first considerable, greater than the diameter of the vessel, but decreased with great rapidity as the refrangibility of the incident light increased, so that from a little beyond the extreme violet to the end, the blue space was reduced to an excessively thin stratum. This shows that the fluid is very opaque to the ultraviolet rays. The fixed lines in the violet and invisible part of the solar spectrum were represented by dark lines, or rather planes, intersecting the blue region. Stokes found that the fluorescent light is not homogeneous, for on reducing the incident rays to a narrow band of homogeneous light, and examining the dispersed beam through a prism, he found that the blue light consisted of rays extending over a wide range of refrangibility, but not into the ultra-violet.
Another method, which Stokes found especially useful in examining different substances for fluorescence, was as follows. Two coloured media were prepared, one of which transmitted the upper portion of the spectrum and was opaque to the lower portion, while the second was opaque to the upper and trans parent to the lower part of the spectrum. These were called by Stokes "complementary absorbents." No pair could be found which were exactly complementary, of course, but the condition was approximately fulfilled by several sets of coloured glasses or solutions. One such combination consisted of a deep-blue solution of ammoniacal copper sulphate and a yellow glass coloured with silver. The two media together were almost opaque. The light of the sun being admitted through a hole in the window-shutter, a white porcelain tablet was laid on a shelf fastened in front of the hole. If the vessel containing the blue solution was placed so as to cover the hole, and the tablet was viewed through the yellow glass, scarcely any light entered the eye, but if a paper washed with some fluorescent liquid were laid on the tablet it appeared brilliantly luminous. Different pairs of complementary absorbents were required according to the colour of the fluorescent light. This experiment shows clearly that the light which passed through the first absorbent and which would have been stopped by the second gave rise in the fluorescent substance to rays of a different wave-length which were transmitted by the second absorbent. Scattered light, with which the true fluorescent light was often associated, was eliminated by this method, being stopped by the second absorbent.
Stokes also used a method, analogous to Newton's method of crossed prisms, for the purpose of analysing the fluorescent light. A spectrum was produced by means of a slit and a prism, the slit being horizontal instead of vertical. The resulting very narrow spectrum was projected on a white paper moistened with a fluorescent solution, and viewed through a second prism with its refracting edge per pendicular to that of the first prism. In addition to the sloping spectrum seen under ordinary circumstances, another spectrum due to the fluorescent light alone, made its appearance, as seen in figs. 3 and 4. In this spectrum the colours do not run from left to right, but in horizontal lines. Thus the dark lines of the solar spectrum lie across the colours. The spectra in figs. 3 and 4 were obtained by V. Pierre with an improved arrange ment of Stokes's method. It will be seen that, in the case of chlorophyll, the whole spectrum, far into the ultra-violet, gives rise to a short range of red fluorescent light, while the effective part of the exciting light in the case of aesculin (a glucoside occurring in horse-chestnut bark) begins a little above the fixed line G and the fluorescent light covers a wide range extending from orange to blue.
Besides the substances already mentioned, a large number of vegetable extracts and some inorganic bodies are strongly fluorescent. Stokes found that most organic substances show signs of fluorescence. Green fluor-spar from Alston Moor exhibits a violet, uranium glass a yellowish-green fluorescence. Tincture of turmeric gives rise to a greenish light, and the extract of seeds of Datura stramonium a pale green light. Ordinary paraffin oil fluoresces blue. Barium platinocyanide, which is much used in the fluorescent screens employed in work with the Röntgen rays, shows a brilliant green fluorescence with ordinary light. Crystals of magnesium platinocyanide possess the remarkable property of emitting a polarized fluorescent light, G H FIG. 3. - Spectrum of Chlorophyll.
H FIG. 4. - Spectrum of Aesculin.
C the colour and plane of polarization depending on the position of the crystal with respect to the incident beam, and, if polarized light is used, on the plane of polarization of the latter.
In all the substances examined by Stokes, the fluorescent light appeared to be of lower refrangibility than the light which excited it. Stokes considered it probable that this lowering of the refrangibility of the light was a general law which held for all substances. This is known as Stokes's law. It has been shown, however, by E. Lommel and others, that this law does not hold generally. Lommel distinguishes two kinds of fluorescence. The bodies which exhibit the first kind are those which possess strong absorption bands, of which only one remains appreciable after great dilution. These bodies are always strongly coloured and show anomalous dispersion and (in solids) surface colour. In such cases, the maximum of intensity in the fluorescent spectrum corresponds to the maximum of absorption. Stokes's law is not obeyed, for a fluorescent spectrum can be produced by means of homogeneous light of lower refrangibility than a great part of the fluorescent light. The second kind of fluorescence is the most common, and is exhibited by bodies which show absorption only in the upper part of the spectrum, i.e. they are usually yellow or brown or (if the absorption is in the ultra-violet) colourless. The absorption bands also are different from those of substances of the first kind, for they readily disappear on dilution. A third class of bodies is formed by those substances which exhibit both kinds of fluorescence.
No complete theory of fluorescence has yet been given, though various attempts have been made to explain the phenomenon. Fluorescence is closely allied to phosphorescence, the difference consisting in the duration of the effect after the exciting cause is removed. Liquids which fluoresce only do so while the exciting light is falling on them, ceasing immediately the exciting light is cut off. In the case of solids, on the other hand, such as fluor-spar or uranium glass, the effect, though very brief, does not die away quite instantaneously, so that it is really a very brief phosphorescence. The property of phosphorescence has been generally attributed to some molecular change taking place in the bodies possessing it. That some such change takes place during fluorescence is rendered probable by the fact that the property depends upon the state of sensitive substance; some bodies, such as barium platinocyanide, fluorescing in the solid state but not in solution, while others, such as fluorescein, only fluoresce in solution. Fluorescence is always associated with absorption, but many bodies are absorbent without showing fluorescence. A satisfactory theory would have to account for these facts as well as for the production of waves of one period by those of another, and the non-homogeneous character of the fluorescent light. Quite recently W. Voigt has sought to give a theory of fluorescence depending on the theory of electrons. Briefly, this theory assumes that the electrons which constitute the molecule of the sensitive body can exist in two or more different configurations simultaneously, and that these are in dynamical equilibrium, like the molecule in a partially dissociated gas. If the electrons have different periods of vibration in the different configurations, then it would happen that the electrons whose period nearly corresponded with that of the incident light would absorb the energy of the latter, and if they then underwent a transformation into a different configuration with a different period, this absorbed energy would be given out in waves of a period corresponding to that of the new configuration.
The phenomenon of fluorescence can be utilized for the purpose of illustrating the laws of reflection and refraction in lecture experiments since the path of a ray of light through a very dilute solution of a sensitive substance is rendered visible. The existence of the dark lines in the ultra-violet portion of the solar spectrum can also be demonstrated in a simple manner. In addition to the foregoing applications, Stokes made use of this property for studying the character of the ultra-violet spectrum of different sources of illumination and flames. He suggested also that the property would in some cases furnish a simple test for the presence of a small quantity of a sensitive substance in an organic mixture. Fluorescent screens are largely used in work with Röntgen rays. There appears to be some prospect of light being thrown on the question of molecular structure by experiments on the fluorescence of vapours. Some very interesting experiments in this direction have been performed by R. W. Wood on the fluorescence of sodium vapour.
REFERENCES.-Sir G. G. Stokes, Mathematical and Physical Papers, vols. iii. and iv.; Muller-Pouillet, Lehrbuch der Physik, Bd. ii. (1897); A. Wullner, Lehrbuch der Experimentalphysik, Bd. iv. (1899); A. A. Winkelmann, Handbuch der Physik, Bd. vi. (1906); R. W. Wood, Physical Optics (1905). (J. R. C.)
or