Imágenes de páginas
PDF
EPUB

SPECTRUM ANALYSIS.

285

foundations of the modern sciences of Spectrum Analysis and of Solar and Stellar Chemistry, and has led to the discovery, that numerous elements which exist in the earth, also exist in the sun and fixed stars.

283. Spectrum analysis may be considered to have commenced by the discovery of Bunsen and Kirchhoff' in 1859, that every chemical substance, when incandescent, produces its own particular spectrum. Thus if any salt of sodium is burnt in a spirit lamp, the spectrum, instead of being continuous, consists of two bright yellow lines, which occupy the same position as the D lines in the solar spectrum. Potassium gives two red lines, which coincide with the lines A and B, and a violet line between G and H. Lithium gives a strong red line between B and C, and three fainter lines in the orange, green and greenish blue. Sodium, potassium and lithium accordingly colour a flame yellow, violet and red respectively. Strontium also colours a flame red, but its spectrum can never be mistaken for that of lithium, since the spectrum of the former consists of a number of lines in the red and orange, and a single line in the blue. The spectrum of barium consists of a number of lines in the yellow and green, and this substance colours a flame green. Gases also, when incandescent, exhibit spectra. That of hydrogen consists of a red, blue-green and an indigo line, each of which respectively coincides with the lines C, F and G; whilst the spectra of oxygen and of nitrogen are more complicated.

284. Spectrum analysis furnishes an exceedingly delicate test of the presence or absence of any element in any chemical compound; for if we place a small portion of the compound in the flame of a spirit lamp, the presence of the element will be detected by its spectrum. It has by this means been discovered, that sodium is one of the most widely distributed substances; also lithium, which was formerly supposed to be rare, is found to exist in numerous minerals, and also in plants and in the bodies of animals. Spectrum analysis has also led to the discovery of new elements; for if the spectrum of a substance shows lines, which do not correspond to the lines of any known element, the obvious. inference is, that these lines are due to the presence of some previously unknown element. It was in this way, that Bunsen

Chemical Analysis by Spectrum Observations; translated, Phil. Mag. Aug. 1860, p. 89; Nov. 1861, p. 329; Dec. 1861, p. 498.

discovered two new alkaline metals in the mineral springs of Baden-Baden and of Dürkheim, which he named cæsium and rubidium. The spectrum of the former consists of two lines in the blue between F and G, and a number of lines between B and E; whilst the spectrum of the latter consists of two lines in the red below A, two lines in the violet a little above G, and a number of lines between C and F. Thallium, which was discovered by Crookes' in 1861, is distinguished by a green line a little below E. The spectrum of indium, which was discovered by Reich and Richter in 1864, consists of a blue line and an indigo line; whilst that of gallium, which was discovered by Lecoq de Boisbaudan in 1875, consists of two bright violet lines.

The methods of spectrum analysis afford so delicate a test of the presence of an element, that 3 x 10-6 of a millegramme of sodium, 10-5 of a millegramme of lithium, and 6 x 10-5 of a millegramme of strontium can be detected.

285. Up to the present time, we have confined our attention to the visible portion of the spectrum; we must now show that ethereal waves exist, whose periods are greater than those of the extreme red waves, and less than those of the extreme violet. These portions of the spectrum are called the infra-red and the ultra-violet respectively.

286. The infra-red waves are waves of dark heat, and were first discovered by Sir W. Herschel in 1800 by means of their thermal effects. He found, that when a thermometer was held for a short time in that portion of the spectrum, which lies below the visible red, an increase of temperature was observed; but perhaps the most striking way of showing the existence of the infra-red waves, is by means of the following experiment due to Tyndall. A solution of iodine in disulphide of carbon is opaque to the luminous waves, but is transparent to the infra-red waves; accordingly if sunlight, or the light from an electric lamp, is passed through such a solution, the infra-red waves are alone transmitted. If therefore the rays are now brought to a focus by means of a convex lens, their thermal effects can be exhibited by placing at the focus a little gunpowder, which immediately explodes, or by lighting a cigar. Tyndall also found that the amount of radiation, which passed through the solution, was ths of the whole.

1 Phil. Mag. vol. xxI. (1861), p. 301.

2 Tyndall, Phil. Trans. 1870, p. 333.

ULTRA-VIOLET AND INFRA-RED WAVES.

287

287. The ultra-violet or actinic waves are noted for the chemical effects which they produce. This can be shown by passing the light from an electric lamp, or from burning magnesium, which contains a large amount of ultra-violet light, through any mixture which absorbs these waves, and then allowing the light to fall upon a photographic plate, when it will be found that very little effect has been produced. Hence it is necessary to develop photographs in a room with yellow blinds, which are opaque to these waves.

288. It was formerly supposed, that the infra-red portion of the spectrum produced the maximum heating effect; that the yellow was the most luminous portion; whilst the maximum chemical effect was contained in the violet and ultra-violet portions. More recent investigations have shown, that this view is not correct, and that every portion of the spectrum exhibits all three effects. If two spectra of equal lengths, formed by a prism and by a diffraction grating, be examined, it will be found that in the spectrum formed by the prism, the maximum heating effect is in the red, whilst in the spectrum formed by the grating, it is in the yellow. The reason of this is, that the spectrum formed by the prism is more spread out towards the violet end and more compressed towards the red end, than the spectrum formed by the grating. This was first pointed out by Draper. The distribution of heat in the spectrum has been investigated by Prof. Langley, by means of an instrument invented by himself and called a bolometer, which is of so delicate a character, that a difference of temperature amounting to 10 of a degree centigrade can easily be noted. By means of this instrument, Langley detected the heating effects of the ultra-violet waves, although the radiation was so weak, that if it fell uninterruptedly for over 1000 years upon a kilogramme of ice, the latter would not be entirely melted.

289. The actinic effects of the infra-red waves have been investigated by Captain Abney, who has succeeded in preparing sensitive photographic films capable of being affected by them.

290. The researches of Langley, Abney and others have greatly extended our knowledge of the solar spectrum, and have shown that instead of being confined to rather less than an octave, the portion hitherto examined extends from about wave-lengths 1400 to 18000 tenth-metres, that is to about four octaves.

291. Having discussed the solar spectrum, and explained how spectrum analysis enables us to detect the presence of chemical elements which exist in the earth, and can therefore be experimented upon; we must now enquire, how the presence or absence of these elements in the sun and fixed stars, can also be ascertained.

292. Frauenhofer observed, that the bright yellow line of the spectrum of sodium coincided with the line D of the solar spectrum, and he suggested that the dark lines of the latter were due to some agency, which lay outside the earth's atmosphere. Brewster, who was acquainted with Frauenhofer's researches, observed that the two red lines in the spectrum of potassium coincided with the lines A and B, and was struck with the coincidence; but neither Frauenhofer nor Brewster appear to have had any idea of the true cause. This seems to have been first pointed out by Sir G. Stokes' about 1852, and depends upon the following considerations.

It is a well-known dynamical theorem, due to the late Sir J. Herschel, that when a material system is acted upon by a periodic force, whose period is equal or nearly equal to one of the free periods of the system, the amplitude of the corresponding forced vibration will be large. Stokes accordingly suggested, that sodium by virtue of its molecular structure was capable of vibrating in the periods of the light, which corresponds to the two D lines of the solar spectrum. If therefore light from a sodium flame be passed through a stratum of vapour of sodium, the molecules of the latter will be thrown into a state of vibration, and owing to the coincidence of the free and forced periods, the amplitudes of the vibrations of the molecules of sodium vapour will be large. The sodium vapour will thus become the seat of energy, and since this energy is necessarily supplied by the energy of the waves of sodium light, which are incident upon the stratum, the amount of energy in the form of waves of light which emerges from the stratum, will be very much less than that which entered; and accordingly if the stratum be sufficiently thick, no energy will emerge, and light will be absorbed. It appears that in 1849 Foucault' experimentally proved, that light from a sodium flame was absorbed by sodium vapour, but the experiment attracted no

1 Phil. Mag. vol. xx. (1860), p. 20.

2 Journal de l'Institut, Feb. 7th, 1849.

KIRCHHOFF'S LAWS.

289

attention, and the subject was not properly investigated until it was taken up by Kirchhoff in 1859.

293. Kirchhoff first examined the spectrum of incandescent sodium vapour, and found that the two bright lines of its spectrum coincided with the two D lines of the solar spectrum. He also examined the spectra of calcium, magnesium, iron and a variety of other substances, and found that the bright lines of their spectra coincided with certain definite dark lines of the solar spectrum. From these observations he deduced the following two very important laws:

I.

Whenever a group of dark lines in the solar spectrum coincides with the bright lines in the spectrum of any incandescent substance, that substance is present in the sun.

II. A substance which emits light of definite periods when incandescent, absorbs light of the same periods as it would emit when incandescent.

294. Ångström and Thalen shortly afterwards showed, that hydrogen was present in the sun, and Norman Lockyer has added lead, potassium and a variety of other substances to the list.

The spectra of many of the fixed stars have also been examined; and it has been found that hydrogen exists in Sirius, and also in the Nebula.

295. The physical theory suggested by Stokes, coupled with the experiments of Kirchhoff', furnish a satisfactory explanation of these phenomena. The sun possesses an incandescent gaseous atmosphere, which surrounds a solid nucleus having a still higher temperature. If we could examine the spectrum of the solar atmosphere, we should observe bright bands, which correspond to the substances contained in it; but the more intense luminosity of the solid nucleus does not permit the spectrum of the sun's atmosphere to appear. The solar atmosphere accordingly absorbs the light, which the gases it contains would emit, and consequently dark bands appear in the solar spectrum, which correspond to the light absorbed. If any particular substance, such as sodium, is present in the sun's nucleus, the sun's atmosphere will also contain

1 Phil. Mag. vol. xIx. (1860), p. 193; vol. xx. (1860), p. 1; vol. xxi. (1861), p. 185. See also, Kirchhoff and Bunsen, Phil. Mag. xx. (1860), p. 89; Kirchhoff, Gesammelte Abhandlungen, p. 598.

B. O.

19

« AnteriorContinuar »