Chapter XXV: Introduction (23)

By making use of one of the Bunsen burners, the lines which are characteristic of some ten or twelve metals are readily seen when one of their more volatile salts is converted into vapour. For this purpose their chlorides are usually employed, but the reactions are common to all their salts. It is necessary that the metal should exist in the flame in the state of highly heated vapour or gas, in order that its characteristic rays should be given off. We usually introduce compounds of these metals into the flame; but there is reason to believe that these are decomposed in the flame, and the disassociated metal takes the form of glowing gas, a small quantity of which suffices for the production of the bright lines. No doubt the other constituent of the compound, the chlorine for example, is also set free in the gaseous form; but since the spectrum of the metal only is visible, we may infer that at the temperature of the flame, the non-metallic elements are not sufficiently luminous to produce a spectrum. When we repeat the experiments with salts of the less volatile metals, we obtain no spectra—the temperature of the flame not being sufficiently high to convert these into vapour. Other methods have, therefore, to be resorted to, and advantage is taken of the fact discovered by Faraday, that an electric spark is nothing but highly heated matter. The spectroscope gives us reason to believe that this matter, which is formed of the substances between which the spark passes, is in the gaseous state; for it is found, on examining sparks passing between two pieces of each metal, that characteristic bright lines are produced. If one of the metals already named is submitted to this examination, the same lines are found which are seen in the spectra produced by the salts of the metal volatized in the flame, but in some cases additional bright lines appear in the spark spectrum. With the heavier metals the spark, or the electric arc, is, however, the only means of igniting their vapours. The usual mode of doing this is to make the discharges of a large induction coil pass between the two fine wires of the metals, placed about a quarter of an inch apart. A Leyden jar is commonly employed to condense the discharge, and thus produce a still higher temperature. Mr. Browning has contrived the neat little apparatus shown in Fig. 223, in which the jar is superseded by a more compact and convenient condenser inside of the box, so that it is only necessary to attach one terminal of the coil to the binding-screw, seen outside of the end of the box, and place the other wire from the coil in the binding-screw of one or the other of the pieces of apparatus supported by the upright rod. Of these it is the one on the right which at present engages our attention. Within a small glass cylinder are two sliding rods, terminated by screw-clips, which hold finely-pointed pieces of the metal under examination. The slit of the spectroscope is placed close to the glass cylinder, and when a very rapid succession of sparks is passing, the bright lines are seen continuously. The spectra of metals examined in this way are found to yield a very large number of lines. Thus the spectrum of calcium has 75 lines, and that of iron no fewer than 450 lines. Our limits will not permit of an account of many interesting particulars relating to these spectra, which include those of all the 50 metallic elements. It should, perhaps, be stated that a modified mode of producing spectra by sparks is sometimes found useful. This consists in causing sparks to pass between a solution of some salt of the metal and a piece of platinum wire. The apparatus for this purpose is that shown on the left side of the upright in Fig. 223.

FIG. 223.—_Apparatus for Spark Spectra._
]

It remains to describe the method of producing spectra of the gaseous non-metallic elements, such as oxygen, nitrogen, hydrogen, &c. For this purpose electricity is again made use of. It has been found that while an electric discharge cannot take place across a perfect vacuum, and air or gas, at ordinary densities, offers much resistance to the passage of electricity, on the other hand, a highly rarefied gas permits the discharge to take place through it with great facility. This is seen in Geissler’s tubes, where a succession of discharges from a Ruhmkorff’s coil causes the tubes to appear filled with light—due to the heating to incandescence of a very minute quantity of the gas. The eye readily recognizes difference of colour in the light given off by the different gases, and when this light is examined by the spectroscope, bright lines, characteristic of each gas, are observed. Nos. 12 and 13, in Plate XVII., are the spectra of hydrogen and of nitrogen respectively, which appear when the gases are examined in the manner just described. In this manner the spectra of chlorine, bromine, iodine, oxygen, sulphur, phosphorus, &c., may be studied. Silicon and some other solid non-metallic elements present great difficulties to the spectroscopist, for these elements cannot be volatized at any temperature we can command, and the spectra of their elements can only be inferred from those of their compounds. But unfortunately the spectra are found to vary with the nature of the compound, and thus it happens that in the case of carbon, for example, no definite spectrum can be assigned to the element. The flame of coal-gas, burning in the air, as in the Bunsen burner, gives the spectrum No. 14; but if this is compared with the spectrum of the flame of burning _cyanogen_ (a compound of carbon and nitrogen), the two are found to differ greatly. The cyanogen spectrum has the two pale broad bands of violet-blue, the four blue lines, the two green lines, and the brightest of the greenish yellow which are seen in the coal-gas spectrum. But it has in addition a characteristic series of violet lines, a series of bright blue, two or three crimson and red lines, and bands in the orange, and several green lines, none of which occur in the coal-gas spectrum. These additional lines are not due to nitrogen, for, with perhaps the exception of some red lines, they do not coincide in position with any of the nitrogen lines. The spectrum of hydrogen, No. 12, should be noticed, as its three lines are very distinct, and it will be observed that they exactly coincide in their position with the three Fraunhofer lines, C, F, and G, in No. 1.

There is another branch of this extensive subject to which we have now to invite the reader’s attention. The power of certain gases to absorb or stop certain rays of an otherwise continuous spectrum has already been mentioned; but this property is by no means confined to gases, for certain liquids and solids do this in a high degree. There is a remarkable metallic element, named _didymium_. It is a rare substance, and its presence cannot with certainty be detected by any ordinary tests. Its salts, however, form solutions _without colour_, or nearly so, which have the power of strongly absorbing certain rays. If we hold before the slit of the spectrum a small tube containing a solution of any one of the salts, and allow the rays from the sun, or from a luminous gas or candle-flame, to pass through it, we see the spectrum crossed by certain well-defined very dark bands. A spectrum of this kind is called an _absorption spectrum_, and the position, number, width, &c., of dark bands are found to be as peculiar to each substance as are the bright lines in the spectra of the elements. The method of observing them when produced by solutions is very simple. The liquid is contained in a small test-tube, which is placed in front of the slit; or, more conveniently, the liquid is put into a _wedge-shaped_ vessel, and thus the thickness of the stratum of liquid through which the rays pass can easily be varied, so that the best results may be obtained. The absorption spectra are produced by many compound substances. A striking absorption spectrum is seen when a solution in alcohol of the green colouring matter of leaves (_chlorophyll_) is examined; for several distinct bands are seen, one in the red being especially well marked. Many other coloured bodies exhibit characteristic absorption bands, as, for example, permanganate of potash, uranic salts, madder, port wine, and magenta. The bands are so peculiar for each substance, that if so-called port wine, for example, owe its colour to colouring matter other than that of the grape, such as logwood, &c., the adulteration can be instantly detected by a glance at the absorption spectrum. As, however, the absorption bands are not, like the bright lines of metals, definite images of the slit, but rather broad portions of the spectra, it is very desirable in examining such spectra to compare them directly with those of known substances, by throwing two spectra into one field, by means of a side reflecting prism, as already described.

Perhaps one of the most interesting examples of absorption spectra is that of blood. A single drop of blood in a tea-cupful of water will show its characteristic spectrum when it is properly examined. If the blood is arterial or oxidized blood, two well-marked dark bands are visible; but if venous or deoxidized blood be used, we see, instead of the two dark bands, a single one in an intermediate position. These differences have been proved to be due to oxidization and deoxidization of a constituent of the blood, called _hæmoglobin_, and by using appropriate chemical reagents, the same specimen of blood may be made to exhibit any number of alternations of the two spectra, according as oxidants or reducing reagents are employed. It would be possible by an examination of the absorption spectrum of a drop of arterial blood to pronounce that a person had died of suffocation from the fumes of burning charcoal. In such case, the supply of oxygen being cut off, the hæmoglobin of the whole of the blood in the system becomes deoxidized.

The beautiful delicacy of these spectrum reactions has permitted the spectroscope to be applied to the microscope with signal success by Mr. Browning, working in conjunction with Mr. Sorby, who has devoted great attention to this subject. The Sorby-Browning instrument is a direct-vision spectroscope, with a slit, lens, &c., placed above the eye-piece of the microscope. By receiving the light through a single drop of an absorptive liquid placed under the object-glass of the microscope, the characteristic bands are made visible. The micro-spectroscope is also a valuable instrument for examining the absorption bands which are found in the light reflected from solid bodies, for the smallest fragment suffices to fill the field of the microscope. Mr. Sorby is able to obtain most unmistakably the dark bands peculiar to blood from a particle of the matter of a blood-stain weighing less than 1/1000th part of a grain. It is plain from this that the spectroscope must sometimes prove of great service in giving evidence of crime from traces which would escape all ordinary observation.

The micro-spectroscope, in its most complete form, is represented in Fig. 224. As may be seen from the figure, the apparatus consists of several parts. The prism is contained in a small tube, which can be removed at pleasure; below the prism is an achromatic eye-piece, having an adjustable slit between the two lenses; the upper lens being furnished with a screw motion to focus the slit. A side slit, capable of adjustment, admits, when required, a second beam of light from any object whose spectrum it is desired to compare with that of the object placed on the stage of the microscope. This second beam of light strikes against a very small prism suitably placed inside the apparatus, and is reflected up through the compound prism, forming a spectrum in the same field with that obtained from the object on the stage. A is a brass tube carrying the compound direct-vision prism, and has a sliding arrangement for roughly focussing.

FIG. 224.—_The Sorby-Browning Micro-Spectroscope._
]

B, a milled head, with screw motion to finely adjust the focus of the achromatic eye-lens.

C, milled head, with screw motion to open or shut the slit _vertically_. Another screw, H, at right angles to C, regulates the slit horizontally. This screw has a larger head, and when once recognized cannot be mistaken for the other.