Page images
PDF
EPUB

which shall yet exhibit difference of potential and continuous current; for if this is possible, it must follow that chemical affinity alone is capable of creating electromotive force as well as of maintaining a current, and that, in an ordinary cell-series, some part at least of the electromotive force is due to this cause, whilst the remainder is the result of the metallic contacts that may exist. Or (ii) if we can establish directly that the two plates in one cell are not at the same potential, as stated by more than one authority.

With regard to the first point, it will be remembered that an old experiment of Faraday's proved that a current can be maintained and decomposition effected by a single cell where there is no dissimilar contact. It is not easy to see how this experiment can be explained by any form of contact theory; indeed it appears unanswerable. But in order to leave no point unsettled by experiment, it seemed desirable to try and arrange a series of cells in which all dissimilar contact was absent, so that the difference of potential due to chemical action might be separated from that due to the contacts and rendered visible by the electroscope.

It is obvious that we can make no attempt to do this unless we can in some way or other obtain a battery with terminals of the same metals; for otherwise the very junctions with the electroscope introduce what we want to eliminate, viz. dissimilar metallic contact. But the following is a method by which this can be accomplished. If plates of lead and copper be placed in nitric acid the lead is positive to the copper, since it is most acted upon; but if lead and copper be placed in solutions of alkaline persulphides, then the copper is most readily acted upon and is positive to the lead; that is, the positions are reversed.

Now, if we place in a cell A dilute nitric acid and a copper

[blocks in formation]

and a lead plate, we cannot join up another cell of the same sort

in series without introducing contact. But if, instead of using a cell containing acid, we place next A a cell, B, containing sodic Pentasulphide, and bend over the lead plate of A to dip into the liquid in B, and place in B also a copper plate, we shall then have two cells joined up in series without dissimilar contact and with similar metals for terminals; and yet the action of the liquids on the metals is such that in A the lead is positive to the copper Cu, and in B the copper Cu' is positive to the lead. Hence there is a regular rise in potential in passing through the two cells; and on joining Cu Cu' by a copper wire a current flows through both cells in the same direction, and the general effect is to urge round a current in the direction shown by the arrows. It is obvious that we need not limit ourselves to two cells. By forming a pile of alternate cells filled with acid and alkaline persulphide, connected by bent copper and lead plates alternately (fig. 2), we shall be able to accumulate differFig. 2.

[blocks in formation]

NO, H

S, Na2

NO3 H

S, Na2

ence of potential to any extent; and if the number of acid and alkaline cells be equal, we shall always end with a plate similar to that with which we began. Such a battery will exhibit a difference of potential between its two terminals when the circuit is opened, and will give a current when it is closed. In it we have nothing but chemical action to rely upon both for creating electromotive force and for maintaining the current. We have no dissimilar contacts; and as the terminal plates are similar, we can effect the junctions with the electroscope without introducing an unbalanced dissimilar contact. I have constructed such a battery of 60 cells; and by the kindness of Professor Guthrie, to whom my thanks are due, I have been permitted to compare its potential with that of a Daniell's cell, by means of a quadrant electrometer belonging to his laboratory. By this means it is at once seen that the difference of potential increases proportionally to the number of cells, the electromotive force of four cells being about equal to that of one Daniell. Joined up with a galvanometer it indicates a current, which, however, rapidly falls off in strength, owing to the formation of an insoluble cupric sulphide upon the copper plates. Joined up in opposition to a single Daniell's cell, with a galvanometer included in the circuit, I find that it requires from four to five cells to balance the force of the Daniell at first immersion; but after leaving it to work on short

circuit for 2 hours its electromotive force had fallen off 50 per cent.; it then required about 8 cells to bring the needle to zero. This gives for the electromotive force of two cells at first about 5 of a volt; or the whole sixty cells are equal nearly to 15 Daniell's cells. It readily effects the decomposition of many electrolytes, and exhibits therefore every property of an ordinary cellseries. Above all, it will be noticed that since there is a regular rise in potential in passing from cell to cell, and as all parts of each plate must be at the same potential, that rise can only take place at the surfaces where the active metals are in contact with the electrolyte (that is, at the seat of the chemical action), and that therefore two metals in one electrolyte cannot be at exactly the same potential. But I find that more direct evidence still of this fact is to be found in an experiment of Faraday's, which seems to have escaped the notice of the contact theorists.

In his 'Experimental Researches' he gives the following fact. "I took a voltaic apparatus, consisting of a single pair of large plates, namely a cylinder of amalgamated zinc and a double cylinder of copper. These were put into a jar containing dilute sulphuric acid, and could at pleasure be placed in metallic communication by a copper wire connecting the two plates. Being thus arranged, there was no chemical action whilst the plates were not connected; on making the contact a spark was obtained. In this case it is evident that the first spark must have occurred before metallic contact was made, for it passed through an interval of air; and also that it must have tended to pass before the electrolytic action began, for the latter could not take place until the current passed, and the current could not pass before the spark appeared." Hence," he says, "I think there is sufficient proof that the zinc and water were in a state of powerful tension previous to the actual contact"*. It is difficult to reconcile this with the experiment of the half disks and drop of water made by Sir W. Thomson. But, at any rate, a consideration of the whole of the facts would seem to point out that the only safe conclusion is, that in any series of cells of any sort the electromotive force is a complex effect, being due to the algebraical sum of all the differences of potential due to dissimilar contacts plus the algebraical sum of the differences of potential due to the chemical affinities of the metals and electrolytes minus any opposing force due to polarization &c.; and that so far from being the exclusive cause, the contacts can only be said strictly to have a share in producing the difference of potentials between the extremities of a battery +. And, lastly, we may with

* Experimental Researches in Electricity, Series viii. ¶ 956.

† Amounting in a Daniell's cell perhaps to 60 or 70 per cent. of the whole electromotive force.

advantage compare the statements of the contact theory with certain other well-ascertained facts. Such statements, for instance, as these:-"If we close the circuit by connecting the metals by a wire, we then have constant separation of electricities at the point of contact of different metals, and constant recombination attended with decomposition through the electrolyte"*. "The electricities separated at the metallic junctions recombine through the water," ""whilst the current flows the water is decomposed",-which seem based on the assumption that the principal seat of the electrical actions is not to be looked for at the seat of the chemical actions. But, now, how does this fit in with those cases of electrochemical inversions noticed by De la Rive, where the direction of the current in a cell is reversed by simply diluting the electrolyte. Thus zinc is negative to tin in strong nitric acid, and mercury negative to lead; but in weak nitric acid the positions are reversed. Hence, if couples be formed of these metals in strong nitric acid, and the acid be gradually diluted, the current first ceases and then is reversed in direction.

Here, without altering the metallic junctions, we can at pleasure alter the direction of the current, and therefore also the direction of the fall in potential, since the current must flow from high to low potential. This seems conclusive that the chemical electromotive force must be even greater than the contact electromotive force. This reversal of the current, by changing the seat of the chemical activity, may be shown in another way, depending on the application of a very old principle. If plates of copper and clean iron be connected by copper wires with a galvanometer, and the iron rendered passive by immersion for a moment in strong nitric acid, then if these plates are plunged into dilute nitric acid the galvanometer indicates a strong current going through the cell from the copper to the iron. If they be removed for an instant and the iron plate touched, on again immersing the current is found to be reversed. Or we may again change the conditions, and notice that it is not sufficient to have merely two different metals and an electrolyte to form a cell. If plates of pure gold and platinum be placed in nitric acid, the most delicate galvanometer detects no current, and the same for many other pairs of metals and electrolytes.

Here we have contact of different metals producing its difference of potential; yet no current flows round "decomposing the electrolyte," as, according to the contact theory, it should do; but the instant we give play to chemical combination the ordinary results ensue. If the extremities of the copper wires from a galvanometer be attached to iron plates, and these plunged *Tait, Thermodynamics,' § 116.

↑ Jenkin, Electricity and Magnetism,' p. 54.

into separate cups of dilute nitric acid, on making connexion between the two cups by a bent iron plate dipping into each no current is detected. On making one limb of the connecting plate passive and re-immersing, a strong current is visible; and we find that we have the direction of the current completely under command by making any of the four plates more or less acted on than the other three.

If these experiments are to have any importance attached to them, it can scarcely be doubted that they land us in conclusions similar to the others, namely:-that we must look for the principal source of the electrical disturbance at that place where the greatest chemical activity is being brought into play; and that whereas contact of metals is in itself productive of definite electrical separation, there is in the battery another cause assisting in the production of difference of electrical potential between the terminals, viz. the potential chemical combination between the metals and electrolytes existing when the circuit is open-the energy of the current produced when the circuit is closed being, of course, the equivalent of this potential energy which disappears.

LI. The Boundary-Conditions of Reflection and Refraction for the Principal Section of Media in motion. By Professor KETTELER, of Bonn*.

WHILE investigating the intensity of reflected and refracted light, I have arrived at equations which, as the most general, I believe include every possible special case, and therefore seem to deserve a peculiar interest.

Imagine two isotropic media (or even two anisotropic, under the limitation that the planes of symmetry of both coincide with the plane of incidence) divided by a partition, and both moving in space, i. e. in the still æther, with any velocity of translation, provided only it be small in comparison with the velocity of light.

I assume that the light incident on the dividing surface is linearly polarized; and accordingly I distinguish two casesthat its vibrations (1) are perpendicular to the incidence-plane, or (2) coincide with it.

For the first case two pure continuity-conditions are sufficient, namely the equations

PE+PR=PD

CE + CR= CD x =

* Translated from a separate impression, communicated by the Author, from the Monatsbericht of the Berlin Academy of Sciences, Jan. 8, 1874.

« PreviousContinue »