Page images

the variations for certain cases. The majority of the variable bands can be attributed to different substances; for example, the three bands of praseodidymium, 4819, 4690, 4450, and the group at 4755, and the bands of neodidymium, 5710, 5239, 5214, 5205, 5115, 4270. The spectra of the solutions of double nitrates are all identical, but the spectra of the crystals show distinct differences, a fact which would indicate that the double compound is decomposed when dissolved. The bands which have been enumerated are those which have a particular direction of absorption in the crystals, and several of them have been separated by chemical processes.

The bands show both a chemical and crystallographic individuality, but it is not at present possible to decide whether they are due to distinct simple substances. The substances which produce the bands have certain of the properties which characterise simple substances, and it may be that some of the substances to which the bands are due are either very stable combinations of two substances or polymeric modifications of the same substance. C. H. B.

Red Fluorescence of Alumina. By L. DE BOISBAUDRAN (Compt. rend., 104, 478-482, 554-556).-Alumina was prepared from aluminium chloride by decomposing it with water, evaporating the solution to dryness, and heating to a temperature sufficiently high to produce the red fluorescence with alumina from alum. The fluorescence of the alumina thus obtained was not red, but blue to greenish-blue, and its spectrum gave neither a line nor a band. The addition of 0.1 per cent. of chromic oxide, however, produced a bril liant red fluorescence far brighter than that obtained from pure alumina under any conditions, and the bands in the spectrum are well defined. With 001 per cent. of chromic oxide, the fluorescence is rose-red, and the band and line in the spectrum are very distinct. 0.005 per cent. of chromic oxide produces a mixture of the greenishblue and rose-red fluorescence, which is generally of a rosy-white tint.

Moderately calcined alumina mixed with about 1 per cent. of bismuth oxide shows at most a very feeble greenish fluorescence.

A strongly calcined mixture of bismuth oxide and alumina from the sulphate shows a reddish-lilac fluorescence. If the alumina has been prepared from alum, the fluorescence is a mixture of the palegreen and dull red fluorescences; if from aluminium chloride, the fluorescence is almost white with a greenish or violet tinge. When the exciting current is weak, the violet tinge only is seen. With a much smaller proportion of bismuth oxide (0.065 per cent.), the fluorescence is brighter, and has a greenish-blue tinge. It is distinctly brighter than with 1 per cent. of the oxide, and much brighter than with pure alumina.

The curious fluorescence, lilac in the cold but blue on heating, which is obtained with alumina prepared from the sulphate mixed with bismuth oxide, seems to be due to the simultaneous presence of bismuth and some other substance, but the latter is not potassium.

The red line seen in the spectrum of the fluorescence of alumina from alum, &c., disappears when the tube is moderately heated in a


vacuum, and at the same time the fluorescence becomes orange. broad red band in the spectrum at first becomes more intense, but if the tube is more strongly heated the band disappears.

Alumina mixed with a small proportion of ferric oxide and very strongly heated gives no red fluorescence. Alumina from alum containing 1 per cent. of cupric oxide and moderately calcined, shows a somewhat bright bluish-green fluorescence. After strongly heating, the red fluorescence is obtained.

Alumina from alum gives the most intense fluorescence when it is precipitated by ammonia, and at first gradually and then very strongly heated. Alumina from alum which has been very strongly heated, and containing 0.001 per cent. of manganese oxide, but free from potassium oxide, gives a pale-rose fluorescence with the red band in the spectrum well marked. Here and there the fluorescence is greenish.

The red obtained from pure alumina in the phosphoroscope is much feebler than that given by less pure varieties of the oxide, especially if the latter contain a small quantity of chromium. There is no spectrum evidence to show whether the two fluorescences are or are not identical.

Pure alumina prepared from redistilled aluminium chloride by precipitation with ammonia and calcination at a high temperature, gives a feeble red phosphorescence in the phosphoroscope. If the aluminium chloride is decomposed by a small quantity of water, the solution simply evaporated to dryness, and the residue strongly heated, the alumina thus obtained gives a very feeble greenish-white phosphorescence in the phosphoroscope. It would seem that the red phosphorescence observed in the first case is due to impurities introduced either in the form of dust or in the distilled water used for washing.

These results tend to support the author's view that small quantities of chromium are the determining cause of the red fluorescence.

C. H. B.

Causes which Determine the Phosphorescence of Calcium Sulphide. By A. VERNEUIL (Compt. rend., 104, 501-504).—The shell of Hypopus vulgaris (this vol., p. 2) has the composition CaCO, 98-21, Na2CO3 0·99, NaCl 0·06, insoluble matter 0·04, SiO2 0·02, MgO 001, P2O, traces, organic matter and loss 0·67 = 100.

Pure calcium carbonate mixed with double the proportion of sodium carbonate present in the shell and 0.02 per cent. of sodium chloride, and then heated with 30 per cent. of sulphur and 0.02 per cent. of bismuth nitrate, yields a product which phosphoresces in a manner very similar to that of the product obtained when the shell of Hypopus is treated in a similar manner.

Pure calcium sulphide gives no phosphorescence; if mixed with a small proportion of the sulphate it shows a feeble white phosphorescence; with sodium carbonate, a greenish-white phosphorescence; with bismuth added in the form of bismuth nitrate, a feeble white phosphorescence; with sodium carbonate and bismuth simultaneously, a blue phosphorescence. Sodium chloride increases the intensity of the phosphorescences produced by sodium carbonate and by bismuth. From these and former observations (loc. cit.), it appears that the

violet phosphorescence of calcium sulphide prepared from the shell of Hypopus vulgaris is due to the simultaneous presence of bismuth oxide, sodium carbonate, sodium chloride, and calcium sulphate. An increase in the proportion of sodium carbonate and chloride above that contained in the shell does not increase the brilliancy of the phosphorescence. Traces of silica, &c., probably exert a similar effect, but their amount is so small that their action is negligeable.

The action of these substances is probably due to the fact that they act as fluxes, and it seems not improbable that all substances which are capable of vitrifying the surface of the calcium sulphide without colouring it render it phosphorescent.

Calcium sulphide becomes phosphorescent to a greater or less extent if heated on platinum foil with a small quantity of borax, potassium carbonate, sodium chloride, sodium carbonate, sodium sulphate, sodium silicate, barium chloride, strontium chloride, calcium fluoride, barium fluoride, barium silicofluoride, cryolite, &c. C. H. B.

Phosphorescence of Calcium Sulphide. By E. BECQUERel (Compt. rend., 104, 551-554).-This paper is mainly a summary of the author's previous observations respecting the phosphorescence of calcium sulphide.

Verneuil's pure calcium sulphide is feebly luminous in the phosphoroscope after being exposed to the sun's rays.

The colour of the phosphorescence of sulphides varies with the temperature. This is well seen in the case of strontium sulphide prepared from strontium oxide and sulphur. At -20° the phosphorescence is violet-blue, at +40° pale-blue, at 90° greenish-yellow, at 150° orange, and the reverse changes are observed as the temperature falls. Strontium sulphide at different temperatures reproduces temporarily and successively under the influence of light almost all the colours of the spectrum. Similar effects are obtained permanently by admixture with foreign substances. C. H. B.

Rotatory Power of Compounds formed in Solutions of Tartaric Acid. By D. GERNEZ (Compt. rend., 104, 783-785).—Many substances which, like boric acid, have no action on polarised light, have the power of removing the anomalies of the law of dispersion of solutions of malic and tartaric acids, and of increasing the rotatory power of these substances to a considerable extent. Amongst these are amido-compounds, such as formamide, acetamide, urea; acids, such as arsenic, arsenious, molybdic, and antimonic acids; salts, such as alkaline arsenates, molybdates, and tungstates.

The effect is particularly well seen in the case of sodium molybdate, a very soluble salt. When tartaric acid and sodium molybdate exist together in solution in the proportion of equal equivalents, they form a substance which has a rotatory power 37.57 times as great as that of tartaric acid. The compound formed under these conditions will be Na2MOO,,2C.HO. In solutions which contain a lower proportion of sodium molybdate, the rotatory power is proportional to the amount of the inactive substance up to and even beyond half an equivalent.

The addition of sodium molybdate in quantity above that required by the formula given has no appreciable effect. C. H. B.

Galvanic Element. By W. BORCHERS (Dingl. polyt. J., 263, 32-34). The cell containing the exciting agent consists of an ordinary wrought-iron tube, closed at one end and in which a zinc or tin rod is suspended. The iron tube forms the positive pole. The exciting agent is a solution of sodium hydroxide and sodium nitrate, sodium chloride being added to increase the active power of the solution. For technical purposes the ratio of Na2O: NaNO, : NaCl = 90: 80: 300 may be used. D. B.

Standard Galvanic Cell. By GoUY (Compt. rend., 104, 781-783). -A convenient standard of electromotive force is furnished by a cell composed of zinc, zinc sulphate, mercuric oxide, and mercury. The bottom of a flask is covered to a depth of 2 to 3 cm. with carefully purified mercury, a platinum wire sealed into the glass and in contact with the mercury forming the positive pole. On the mercury is placed a layer of yellow mercuric oxide, and the flask is filled with a 10 per cent. solution of crystallised zinc sulphate (sp. gr. 106), and in this is immersed a rod of pure zinc, which should be amalgamated. In order to render the cell portable, the zinc may be enclosed in a tube with a narrow aperture closed with some porous material. This arrangement will give a cell with a very high resistance. It matters little whether the cells are exposed to air or are hermetically sealed. Some sealed cells which had been filled up for three months were found to give the same results as freshly prepared cells. electromotive force, which is equal to about 139 volts, does not become constant until after some days. This time being necessary for the mercury to acquire a condition of equilibrium. The electromotive force is independent of the concentration of the zinc sulphate solution, provided that its sp. gr. is above 102, and it is not affected by replacing or amalgamating the zinc. It varies very slightly with the temperature, the variations being only 0.0001 per degree between 0° and 30°. If the current does not exceed 0·001 ampère, polarisation rapidly disappears when the circuit is broken, and in practice the circuit is only closed just at the time the observation is to be made. Any effect of accidental closing of the circuit may be avoided by making the internal resistance equal to 1000 ohms. The cell can be used with galvanometers as well as with electrometers.


C. H. B.

Galvanic Polarisation produced by Feeble Electromotive Forces. By C. FROMME (Ann. Phys. Chem. [2], 29, 497-544; 30, 77-95, 320-343, and 503-530).-The author has made a great number of experiments on the polarisation of platinum, gold, and palladium electrodes. The arrangements were such that the polarisation could be measured by means of an electrometer both during the passage of the current and after the circuit had been broken. By comparison with a neutral plate, the polarisation of each electrode could be separately measured, and the rate of its change during the

passage of the current and after breaking circuit, as well as the influence on it of the presence or absence of air in the voltameter, rise of temperature, and so forth, could be studied. In the first three papers, the results obtained respectively with platinum, gold, and palladium, are described; the fourth contains a résumé and discussion. From the latter, the following few observations, bearing specially on the occlusion of gases by these metals, are taken.

Water acidified with sulphuric acid appears to be decomposed even by the weakest currents. Of the liberated gases, one portion condenses on the surface of the electrodes; a second portion penetrates or is occluded by the metal; and a third is either scattered or dissolved in the surrounding liquid. The experiments make it highly probable that oxygen is occluded as well as hydrogen. The electromotive force of polarisation depends solely on the surface-condensed gas; that which has penetrated the electrodes has a great influence on the rate at which polarisation increases or diminishes under different conditions.

The polarisation developed in an air-free voltameter by a feeble current (from one Daniell), approaches the battery electromotive force more nearly the less the total resistance of the circuit. Under no conditions does it actually equal the latter; hence there is always a slight residual current flowing through the cell. The insertion of a resistance into the circuit then diminishes both anode and cathode polarisation. If the charging current has been feeble and of short duration, and the voltameter air-free, the rate of loss of H-polarisation reaches its maximum only after a time; but when air is present part of the hydrogen is at once oxidised, and the polarisation sinks rapidly. When the charging current has been strong, presence of air makes but little difference in the loss of O-polarisation; but with a feeble charging current, presence of air acts so as to increase the original O-polarisation, and hence its rate of change when a resistance is inserted. On the other hand, when the charging current has been long continued, both the electrodes and the surrounding liquid become saturated with gas; the occluded gas slowly leaks out when the charging current is diminished by the resistance, and consequently both polarisations fall away slowly.

The polarisation developed by a current is also affected by the previous condition of the electrodes. The important factor in such cases is occlusion. The occlusion of hydrogen by platinum and palladium is well established; that of oxygen is rendered probable by the fact that when a feeble current is passed successively in opposite directions through a voltameter, and the circuit then broken, the plate which was originally the anode shows at first a slight H-polarisation, which soon gives place to O-polarisation. The occlusion of gases. by platinum appears to occur for forces between 08 and 16 of a Daniell.

A curious phenomenon is observed when two electrodes, previously polarised by a Daniell, are placed in circuit. Both then show equal O-polarisation. This is, partly at least, due to the fact that the original O- was stronger than the H-polarisation. With a stronger charging current (one chromic acid element), the electrodes become

« PreviousContinue »