Page images

combining with their own sub-salts, so that the reduction which begins with the sub-salt quickly extends to the normal salt with which it is combined. Other silver salts have not this power, and therefore are more slowly and irregularly attacked by developing solutions.

When paper prepared with silver tartrate, oxalate, &c., is exposed to light, treated with hydrochloric acid and developed, the effect of a short exposure is the same as that of a long exposure. A short exposure produces sufficient alteration to serve as a nucleus for development; a long exposure does no more. C. H. B.

Spectrum Researches on the Energy of the Action of Bromine on Aromatic Hydrocarbons. By J. SCHRAMM and I. ZAKRZEWSKI (Monatsh., 8, 299-309).-Sunlight was reflected through a vertical slit by a heliostat on to a condensing lens of great focal length. By means of a bisulphide of carbon prism, a spectrum 30 cm. long was produced, in which Fraunhofer's lines were distinctly visible. Nine test-tubes were fastened at equal distances between the lines B and H, since preliminary experiments showed that the chemical action was confined to the space between these lines. A few drops of bromine were added to the solution of the carbon-compound, and the mixture was equally distributed among the test-tubes. The time occupied from first illumination to total disappearance of the bromine was observed in each tube. To exhibit graphically the energy of the reaction, curves are plotted, so that the abscissæ indicate the part of the spectrum and the ordinates the reciprocals of the duration of the reaction. In this way experiments were made on toluene, ethylbenzene, and metaxylene. All the researches showed that the maximum action takes place in the yellow or yellow-green. The blue and violet rays exert a slight action, and the dark-red scarcely any at all. The curves are therefore similar in character to that for the intensity of light in the spectrum. C. S.

Increase of Photo-electric Currents. By J. MOSER (Monatsh., 8, 373). The author finds that the electromotive force produced by the action of sunlight on chloride, iodide, or bromide of silver plates can be considerably increased by immersing them in a bath of a dye, say, erythrosin. C. S.

Electric Properties of Rock-salt. By F. BRAUN (Ann. Phys. Chem. [2], 31, 855-872).-Although this substance is isotropic as regards light, its modulus of elasticity according to Voigt is 4170, 3400, and 3180 kilo. per sq. cm. in directions parallel respectively to normals to the cubical, dodecahedral, and octahedral faces. On Maxwell's theory, high insulating properties might be expected in it, since it is highly diathermanous. It is in fact so good an insulator, although not perfect, that its dielectric constants could be well determined.

Much of the paper is devoted to describing the methods of determining the constants of the instruments used. The ends of a rocksalt column were cut parallel to cubical faces, a hole drilled length

wise nearly through its axis, and the vertical edges planed away until the section of the column was octahedral. Alternate faces were then cubical and dodecahedral. Plates of tinfoil were attached to two of these, and, the inner chamber being filled with mercury charged to a high potential, the charges on these were compared by the electrometer. Within the limits of experimental error, the charges were found to be equal. Similar experimens with two little condensers constructed from plates parallel to cubical and octahedral faces showed the dielectric constants of these also to be the same. Absolute measurements were not made.

The conductivity was also measured by the electrometer. One plate of a little rock-salt condenser was joined to a battery of 16 large Leyden jars, the potential of which (1500 to 3000 volts) was observed from time to time, so as to calculate the rate at which its charge was dissipated through leakage. The other plate being joined to the electrometer, the indications of the latter afforded a measure of the conductivity. The method is fully described: the results are as follows (specific resistance of mercury = 1).

Specific resistance perpendicular to cubical face = 1.33 × 1021.

[ocr errors]


octahedral face = 2·63 × 1021. Ratio of resistance perpendicular to cubical face to resistance perpendicular to dodecahedral face = 3 : 2.

Specific resistance of paraffin = 3.02 × 1022.

The dielectric constants were also found to be independent of the time of electrification.

The provisional conclusion is, that rock-salt is dielectrically isotropic, but anisotropic as regards conductivity. Ch. B.

Validity of Joule's Law for Electrolytes. By H. JAHN (Ann. Chem. Phys. [2], 31, 925-940; see also ibid., 25, 49).-The object of these experiments was to ascertain whether the portion of the current energy expended on polarisation during electrolysis contributes to the development of heat or not. The method of inquiry consisted in measuring the heat developed during the electrolysis of antimony chloride between antimony electrodes, in a cell enclosed in a Bunsen's ice-calorimeter, and the current flowing through the circuit, and calculating Joule's constant a from the equation

W., aJ(iR - Jp),

where Wheat developed per second in the calorimeter, iR = difference of potential between two points in the circuit including the cells, p = resistance of the part of the intervening circuit not included in the cell, and J = strength of current. Antimony chloride was selected on account of the high polarisation generated during its electrolysis.

Since antimony separates from its chloride sometimes in the ordinary form, sometimes in the explosive form, experiments were made with electrodes of both kinds. During the rapid rise of the polarisation (about five minutes) the strength of the current E was assumed to vary according to the law E = Eo+at+bt2. The total

current was then determined by summation from observations at intervals of five minutes. From various experiments it was found,

a = 0·2367; 0·2362; 0·2375; 0·2376; 0·2383; 0·2382. The author shows that if the polarisation energy did not contribute to heat development the result should have been a = 0·2693.

Ch. B. Conductivity of Pure Water, and its Temperature Coefficients. By E. PFEIFFER (Ann. Phys. Chem. [2], 31, 831-855). -This paper refers chiefly to the conductivity of water contained in glass vessels, and hence rendered more or less impure by dissolution of the glass substance. Kohlrausch's method was used. One of the electrolytic cells has been already described (ibid., 25, 232). A second form consisted of two platinum plates (3 cm. square) held at a fixed distance apart by glass rods fused to their faces, and suspended by the conducting wires in a glass vessel. Since the conductivity changes rapidly from the moment of filling, observations were made at regular intervals, and the initial conductivity calculated by extrapolation.

The solvent action on glass is always apparent, but increases greatly with the temperature. When successive charges of water are poured into a new glass vessel, the resulting time-rate of increase of conductivity is at first rapid, but after a time reaches a nearly constant minimum value which does not change for months. When the same mass of water is allowed to remain in a glass vessel at constant temperature for a long time, the time-rate of increase of conductivity grows gradually greater; from this the author concludes that the conductivity is not proportional to the amount of dissolved glass substance. He thinks that the suitability of different kinds of glass for chemical purposes might be usefully tested in this way.

The rapid fall in conductivity which takes place immediately after filling (Kohlrausch, Ann. Phys. Chem [5], 26, 220; Pfeiffer, Abstr., 1886, 115) is attributable to the diffusion into the fresh liquid of a layer of the previous charge adhering to the glass and electrodes.


Temperature coefficients have been investigated by Kohlrausch, Vicentini, and Arrhenius. For different solutions, these tend towards equality with increasing dilution; for a particular electrolyte, the coefficient is nearly constant for highly dilute solutions of different strengths; for an excessively dilute solution, it at first decreases as the strength increases. Owing to the constantly changing strength of the solution, its determination for the particular electrolyte furnished by the soluble part of the containing glass vessel was laborious. results are given in lengthy tables. The initial conductivity of the purest water used was λ = 0·65 (X k18 1010) at 14°; and observations were continued until A about equalled 20. Curves were then constructed showing the variations of the temperature coefficients z = At18. In one vessel, neglecting minor variations, as A increased from unity & diminished to a minimum 0·024 at λ = 3, then increased to a maximum 00269 at λ = 5, and from λ = 6 onwards became constant at 0.026. In the second vessel (presumably of a different kind of glass), a reached a minimum for λ = 2, a maximum for A 35, and became constant at 0.023 when λ = 6. For water



which had previously remained for some time in a third vessel, the variations were similar, and a became constant when λ = 9. The initial variations amounted to about 12 per cent.

The law that the coefficient is constant for very dilute solutions appears to hold only so long as the electrolyte exceeds in amount the impurities (possibly organic) originally contained in the water.

Ch. B.

Formation of Hydrogen Peroxide at the Anode during the Electrolysis of dilute Sulphuric Acid. By F. RICHARZ (Ann. Phys. Chem. [2], 31, 912-924; see also Abstr., 1885, 624, and McLeod, Trans., 1886, 591).-The author has repeated some of Berthelot's experiments (Abstr., 1878, 372; 1886, 607) on this subject, and in the main confirms his results. Sulphuric acid of 70 per cent. appears to be most suitable for developing hydrogen peroxide by electrolysis. In a 40 per cent. solution, persulphuric acid alone is formed. As a test for hydrogen peroxide in presence of persulphuric acid, the author recommends a solution of titanic acid in sulphuric acid. This gives an intensely yellow precipitate, which decolorises the same amount of permanganate as the peroxide which goes to form it. The peroxide may be estimated by permanganate in presence of persulphuric acid; and the latter subsequently estimated by adding ferrous sulphate in excess, and titrating back with permanganate.

Theories have been proposed by Hoppe-Seyler (Zeit. physiol. Chem., 2, 25) and Traube (Abstr., 1886, 660), who have shown that hydrogen peroxide can be formed at the cathode by reduction of molecolar oxygen. The author accounts for its appearance at the anode. When a solution containing only sulphuric and persulphuric acids is allowed to remain, the latter gradually disappears, and hydrogen peroxide is at the same time formed. The author has studied this reaction quantitatively, but does not confirm Berthelot's statement that as decomposition goes on the ratio of hydrogen peroxide to persulphuric acid tends to become 2. In a dilute sulphuric acid solution, the decomposition is very slow, but it is greatly accelerated by adding strong sulphuric acid; from this the author concludes that the appearance of hydrogen peroxide at the anode in a strongly acid solution is always the result of this purely chemical reaction. Contact with a platinum plate almost entirely prevents the formation of hydrogen peroxide in a 40 per cent. sulphuric acid solution, although by catalysis it hastens the disappearance of persulphuric acid; especially when the platinum has been ignited, and consequently become charged with flame gases.

When a 68 per cent. acid solution is electrolysed by a current of 11 ampères, the quantity of hydrogen peroxide increases for about three hours, and then becomes stationary. The persulphuric acid, however, goes on increasing. On breaking the current, the peroxide at first noticeably increases, but the persulphuric acid rapidly dis


The fact that water is under certain circumstances oxidised by active oxygen, in conjunction with the fact that the peroxide is not directly formed at the anode during electrolysis, is a proof that the molecules of water cannot be electrolysed. The real anion is SO1,

and from this oxygen can only be liberated as O2 or O3. Traube supposes that peroxides which contain an even number of oxygenatoms can alone form H2O,; and that those formed at the positive pole during electrolysis, and which contain an uneven number of such atoms, cannot form H2O2. The decomposition of persulphuric acid clearly contradicts this latter view. Сн. В.

Electromotive Dilution Constants. By J. MIESLER (Monatsh. 8, 365—372).—The electromotive force of polarisation alone was measured at the beginning and end of each experiment (to test its constancy) by means of a Siemens' universal galvanometer and a capillary electrometer. Next the concentration cell was introduced, so that the electromotive force due to difference of concentration aided or opposed the electromotive force of polarisation. By subtracting or adding the electromotive force of polarisation, the electromotive force due to differences of concentration was determined from the observed total electromotive force. Thus the following values of dilution constants in millivolts were obtained :


[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors]

Taking any two rows in the table, it will be seen that the numbers in corresponding columns have a constant difference (Moser's law), whilst in the case of the three haloid salts of zinc the differences of values of the dilution constants have an obvious relation to the atomic weights of the halogens.

C. S.

Influence of Ultra-violet Light on the Electric Discharge. By H. HERTZ (Ann. Phys. Chem. [2], 31, 983-1000).-The author has discovered that ultra-violet radiation favours the electric discharge between two conductors in a remarkable way. As sources of such radiation, the sun, burning magnesium, or even ordinary flame, may be used; but by far the most effective are the electric arc and an induced electric discharge. To produce the phenomenon, the primary circuits of two induction coils, a large one (10 cm.) and a smaller one (1 cm.), are joined in circuit with the same battery (six Bunsens) and interruptor. Perfect synchronism in the induced discharge is thus secured. The terminals of the large coil being arranged to give a good spark 1 cm. in length, the two coils are placed close together, and an opaque screen interposed. The terminals of the small coil are then drawn apart until sparks just cease to pass. On now removing the screen the discharge is re-established.

The author describes many experiments to test the nature of the effect. The influence is not electrical, since non-conducting screens

« PreviousContinue »