Sources for Spectroscopic Observation

SOURCES FOR SPECTROSCOPIC OBSERVATION While any luminous body is a suitable subject for spectroscopic investigation, there are several methods of exciting substances to luminosity which are especially convenient, or are of a character to give valuable information about the emitting atoms or mole cules. Apart from the spectra of gases, and such phenomena as fluorescence, the three chief sources employed in a laboratory are the bunsen or oxyhydrogen flame, the electric arc and the electric spark. In general, the same substance will show differences in its spectrum when examined in these different ways, but some lines are usually common to the three sources. Lines which occur in the flame are distinguished as flame lines, but only those which have their greatest intensity in the arc are properly called arc lines; lines which are more strongly developed, or only appear, in the spark are called spark lines. The spark lines thus defined are the "enhanced lines" of Sir Norman Lockyer, and the distinction is now of great importance. In the language of modern spectroscopy, flame lines are the more easily excited lines of the neutral atom in question, while the arc usually exhibits the complete spectrum of the neutral atom together with the more easily excited lines of the ionized atom ; the spark gives a special development of the lines of the ionized atom, besides retaining the more persistent lines of the neutral atom. Portions of the spark and arc spectra of strontium are shown in Plate II., 3a and 3b; it will be seen, for example, that the line at A4,607, which is a flame line, is reduced in intensity in the spark, while the enhanced lines at X4,o78 and X4,216 are relatively much intensified in the spark.

Flames.

The simplest procedure for the observation of flame spectra is to make use of a bunsen burner, and to introduce the various substances to be studied by means of a platinum wire, or on the charred end of a match. This method suffices to give spectra of the alkali and alkaline earth metals, and of thallium salts, as well as of numerous compounds such as copper chloride.

At the relatively low temperature of flames a large number of substances give banded spectra, as the molecules of the elements or compounds introduced are only partially dissociated into their component atoms. Some of these bands are due to oxides which

are produced in the flame.

The Electric Arc.

A very valuable source for the observa tion of spectra is the electric arc, which, besides the advantage of great luminosity, will fuse and volatilize any substance which may be introduced between the poles. For the investigation of metallic spectra it is best to use poles made of the metals them selves, if they do not melt too readily, in order to avoid con tamination. In general, however, some of the substance is placed on the lower (positive) pole of the carbon arc. The bands of the Swan spectrum of carbon and those of cyanogen which are given by the carbon itself would often be troublesome but for the fact that they tend to disappear if a sufficient amount of material be introduced. Lines due to impurities in the carbon rods can be distinguished by observing or photographing the spectrum before the introduction of the substance under examination. Acheson graphite rods can often be used with advantage, as the impurities are usually very slight. Rods of pure silver have also been used as a means of supporting the substance.

The appearance of so-called "reversed lines" in arc spectra is quite common. This is accounted for by a broadening of a line from the denser vapour in the core of the arc, and the super position upon it of a narrower absorption line produced by the cooler and less dense vapour in the surrounding region.

It should be further noted that different parts of an arc show considerable variations in spectrum. The fully developed arc spectrum is restricted to the central core, while the surrounding "arc-flame" yields a spectrum corresponding to a lower tempera ture. The arc-flame may, in fact, often be conveniently used in place of the oxy-hydrogen flame. These differences are readily observed when an image of the arc is focussed on the slit of the spectrograph. By the same method it may be observed, more especially when the poles of the arc are metallic, that certain lines only appear, or are better developed in the immediate neighbourhood of the poles; these are the so-called "polar lines," and are generally found to be lines which are more strongly developed in the spark.