was almost exclusively used as standard of electromotive force.

The rules for constructing it are nearly the same as for the cadmium cell. The following differences are to be noted :zinc (pure, redistilled) is used as anode: it may be amalgamated on the surface, and even dissolved in mercury, but no definite compound is formed as with cadmium. On the contrary, the presence of mercury makes practically no difference to its electromotive behaviour: even when the mercury reaches 98 or 99 per cent. the electro-affinity remains that of pure zinc, to within a fraction of a millivolt. The metal (or amalgam) should not be cast on a platinum wire sealed through glass, as zinc slowly alloys with platinum, causing it to expand and crack the glass. The zinc sulphate is used in the form ZnSO47H2O; it must be free from acidity, and from iron. It must not be heated to 39°, on account of its transition point (p. 220); if raised above that temperature it suffers the change into hexahydrate, and does not always return to the other form on cooling, causing uncertainty in the electromotive force.

Clark cells have been made up in many forms, the most familiar being the "Board of Trade" pattern, shown in Fig. 53. This has the advantage of a low internal resistance, but is defective in certain respects. It is exposed to the danger that fragments of zinc may fall into the mercury below. This would at once destroy the accuracy of the cell. This danger may be avoided by using a sheet of amalgamated platinum instead of liquid mercury, and putting it side by side with the zinc.

Further, Clark cells are always apt

Marine

Glue

-Cork

-Zn rod
Zn SO4
Solution

Hg, SO paste

4

-- Hg

Pt wire

FIG. 53.

to show temperature lag, and the Board of Trade pattern is bad in this respect. The lower part of the zinc rod is surrounded

by crystals of sulphate, the upper by solution merely. If the temperature rises, the solution becomes unsaturated, and as, at the top, it can only get saturated again by the slow process of diffusion from below, the electromotive force takes a considerable time to adjust itself to the new temperature. Hence even if the temperature of the cell is accurately known, by a thermometer sealed into it, its electromotive force cannot be depended on unless the cell has been kept for a long time in a thermostat. The temperature lag may be reduced by filling the cell completely with crystals, or surrounding the zinc with a glass tube except at the tip.

Clark cells have also been made with the zinc in a porous pot, and also in H form.

The disadvantages of the Clark as compared with the cadmium are

(i.) Temperature coefficient is thirty times as great.

(ii.) Temperature lag;

(iii.) Liability to generate hydrogen. Zinc acts—very slowly -even upon pure water; and, of course, if the electrolyte is at all acid the action is much more rapid. There is thus a risk of the negative pole becoming covered with a film of gas, which would polarise it, and choke off the current. Cadmium, from its position on the scale of electro-affinity, is much less likely to liberate hydrogen.

The electromotive force of the Clark cell has often been measured by means of standard resistances, and either an absolute current meter or the silver voltameter (p. 253, infra) Carhart1 gives this table of results :

The temperature variation of the E.M.F. of a Clark cell is, according to Jaeger, given by the formula.

E =

1'4328 – 0·00119 († — 15°) — 0'000007 († — 15°)2

(valid between o° and 30°).

When the zinc sulphate solution in a Clark cell is dilute, the electromotive force is greater than that of the ordinary cell with saturated solution at the same temperature. The temperature coefficient of such a cell is only half that of an ordinary cell. Fig. 54 shows the results of an experiment by Jaeger on this point. A saturated solution of ZnSO...7HO

was prepared at 6'75° (To, Fig. 54), separated from crystals, and used to make a cell.

E = 14422

The E.M.F. was found to be

(line Ev in figure). The solubility of zinc sulphate increases rapidly with rise of temperature: consequently, when a cell without crystals is warmed, the solution becomes unsaturated; its electromotive force must become greater than that of a saturated cell at the same temperature, and the corresponding line in the diagram must lie above that for a normal cell (Ec). But even if the cell without crystals be cooled, crystals do not separate out, so the solution becomes supersaturated, and it was found possible to measure it in this state

down to -10° Ceft of To, Fig. 54). It will be seen, then, that of the temperature coefficient of the ordinary Clark cell, about half is due to change in solubility of the salt.

The other cells that have been proposed as standards are unimportant. Helmholtz used the calomel cell

Zn: ZnCl2: Hg Cl: Hg

The very small solubility of mercurous chloride is against this, as it does not form a sufficiently active depolarising agent. Moreover, zinc chloride is a badly defined substance chemically. By properly adjusting the concentration of the electrolyte, this cell may be made to give exactly 1 volt. According to Ostwald, this requires (at 15°) a density of 1409; according to Schoop, 1391. Similar cells have been made by Hibbert. The temperature coefficient is + 0·00007.

Gouy's cell

Zn ZnSO, HgO: Hg

is unsatisfactory because in course of time mercurous sulphate is formed in it, so that it cannot be regarded as properly reversible. The voltage is about 14, the temperature coefficient about 0'0002.

De la Rue's cell

Zn: ZnCl2 : AgCl : Ag

is comparable to the calomel cell. It gives a fairly steady electromotive force, and has been used to make up high voltage batteries, but has not been precisely studied as a standard.

Finally, a lead accumulator that has been charged and allowed to stand a few hours may be used as a standard for moderately accurate measurements. The density of the acid must be taken; the voltage can then be calculated from the following formula, which can be relied upon to within I per cent.,

where E is the electromotive force, d the density of the acid. The formula holds for temperature 15°, and for acid of density between I'I and 13.

To determine the electromotive force of a standard cell

on which, in turn, all other determinations of electromotive force depend-the method used is to balance it against a measured current flowing through a known resistance. The current is measured either (a) absolutely, i.e. according to its electro-magnetic definition, by means of an electro-dynamometer, "current balance," or other instrument, or (b) by the weight of silver deposited by it in a measured time. The former measurement, as well as the experiments by which the absolute value of the resistance standard is arrived at are beyond the range of this book. We may content ourselves with describing briefly the method of the silver voltameter.

The arrangement of apparatus is shown in Fig. 54. Current is taken from a large battery of accumulators—say 20 cells or

more. The advantage of using a large E.M.F. is that the resistance of the voltameter is then small compared with the other resistances in circuit, so that the inevitable variations in it do not much affect the current. A is an ammeter which is convenient for rough adjustment of the current; V is the silver voltameter; K, a key by which the current may be made and broken as rapidly as possible; S, a standard resistance of, say, 4 ohms, capable of carrying ampere without appreciable rise of temperature; C, the standard cell to be tested; G, a mirror galvanometer; K', a key. If C is a cadmium cell, and S 4 ohms, then for balance (shown by absence of current in G) 10186 the main current must be adjusted to =0'025465 amperes.

4

To do this, using 20 accumulators, a resistance of about 150 ohms must be put in the circuit. This is done by means of R, a substantial wire resistance a trifle greater than 150 ohms,