For zinc in sulphuric acid, and copper in sulphate of copper, these two Russian physicists found the electro-motive force 2.17. This gives us a point of reference for reducing the numerical values, given above, for polarization and electro-motive force to our (the chemical) unit. We have found for the electro-motive force of a Daniell element the value of 470 (section 9); and to reduce the values given by Lenz and Saweljev to chemical measure they must be multiplied by 470

2.17

217.

=

For the electro-motive force of a Grove's element, (platinum in nitric acid, zinc in sulphuric acid,) they found the electro-motive force 4.17; consequently, in chemical measure, it is 4.17 x 217 905. Hence, for the polarization of different metallic plates, we get the following values expressed in chemical measure:

for the entire polarization of the two platinum electrodes in dilute sulphuric acid

1185,

while for this case I found the number (section 32)

1000.

§ 36. Polarization in platinized platinum plates.-Poggendorff observed, accidentally, that in an element of the Grove gas column, which was inserted in the closing are of a Grove element, a considerable development of gas took place unexpectedly, while a simple Grove element, closed by a voltameter with uncoated platinum plates, produced a very inconsiderable decomposition of water. (Pogg. Ann., LXX 183.)

For making comparative measurements, he constructed a voltameter with platinized platinum plates, which he compared with an ordinary voltameter. The voltameter with uncoated plates yielded in the closing arc of a Grove element, in thirty minutes,

0.89 cubic centimetres of explosive gas;

while the voltameter with platinized platinum plates, under the same circumstances; yielded

77.68 cubic centimetres;

thus nearly 87 times as much.

This is due simply to the fact, that the polarization in platinized plates is considerably less than in uncoated plates. Poggendorff has proved this by direct measurements.

The electro-motive force of a battery of two Grove's elements was 64; after inserting the platinized plates it was 31; hence the polarization of the platinized plates was

6431 33.

When, instead of the voltameter with platinized plates, that with uncoated platinum plates was substituted, the electro-motive force of the whole battery was equal to 22; therefore the polarization on the uncoated plate was

It is shown, in section 18, that the electro-motive force of a Grove element, as a mean of the observation of different physicists, is 777 in chemical measure; hence the electro-motive force of two elements equals 1554; therefore the value of the polarization of the uncoated plates which Poggendorff found, reduced to chemical measure, is

which accords very closely with the value of the polarization given above in section 32.

Hence the polarization for platinized plates, in chemical measure, is

Poggendorff also found, as mentioned already in section 34, that the magnitude of the polarization diminishes with the strength of the current; when, by the increase of the accidental resistance, the current was so weakened that the needle of the sine compass, inserted in the closing arc, receded from 47° 49' to 5° 44', the polarization in the voltameter diminished from 42 to 38, or, in chemical measure, from 1020 to 922.

According to Poggendorff's experiments, the magnitude of the polarization with platinized plates is but little dependent upon the changes of the strength of the current, so that it may be considered constant, without sensible error..

Svanberg also has instituted many experiments in galvanic polarization, and with great care and accuracy. (Pogg. Ann., LXXIII, 298.) For the polarization which the current of four Daniell elements produced in a voltameter with uncoated platinum plates, he found, reduced to chemical measure, the value

1072.

Svanberg observed, that the polarization in the voltameter increases gradually, and requires some time to attain a maximum. Therefore, to determine the maximum polarization accurately, he made his meas

urements only after the current had been passing for some hours through the voltameter.

Metal plates with rough surfaces appeared from his measurement to be polarized less than polished ones, which accords well with Poggendorff's observation, that the polarization on platinum plates is less than on naked ones. The polarization of polished copper plates by hydrogen, Svanberg found in the ratio of 12 to 8 less when they were made rough with a file, or still better when rendered granular by galvanic precipitated copper.

§ 37. Buff's researches on galvanic polarization.—Single results of these researches have been already mentioned above, but we must here present a few more extracts from Buff's Memoir. (Pogg. Ann., LXXIII, 497.)

He found that a deflection of 45 degrees in his tangent compass corresponded to a development of hydrogen of 21.08 cubic centimetres per minute, (reduced to the temperature of 0° and 760 millitres pressure?), which is equivalent to a development of explosive gas of 31.6 centimetres; hence the strength of the current was reduced to chemical measure by multiplying the tangent of the angle of deflection by 31.6.

In the course of this investigation, Buff found the electro-motive force of a Daniell element equal to 4.207. Since, in establishing our unit we have taken the electromotive force of this element at 470, Buff's data of electro-motive force, as well as his value of polariza

470

tion, must be multiplied by 4.207111 to make the results comparable with ours. Buff's comparison of the strength of current and magnitude of polarization in a voltameter with naked platinum plates, (referred to our unit), gave the following results:

In these experiments the platinum electrodes formed the opposite sides of a trough; the above numbers relate to the case where the trough was filled to a height of 45 millimetres.

Filled to a height of 10 millimetres, the following respective values of strength of current and polarization were obtained:

Thus, under circumstances otherwise equal, the polarization appeared somewhat greater than when the trough was filled higher, as already mentioned in section 35.

Buff also remarked that the maximum polarization required a considerable time to elapse before taking place.

For one decomposing cell formed of two zinc plates in a solution of sulphate of zinc, he found the value of polarization,

in our unit

p=0.85;

p= 94.

From this result he is led to the following conclusions:

"I regard p=0.85 as the electrical difference of zinc and hydrogen, or as an approximation to it. In like manner I regard the polarization resistance of the platinum plates in dilute sulphuric acid as an approximate value for the electrical difference between oxygen and hydrogen. By the stratum of hydrogen at the negative platinum plate, and the stratum of oxygen at the positive plate, the same effect is produced as though not two platinum strips, but a strip of solid hydrogen and one of solid oxygen, were placed in the acid. * *The electro-motive action developed by the immediate contact of hydrogen and oxygen, or the electrical difference of these substances, indicates the extreme limits of the resistance, which can take place by the polarization of two metals in the decomposing cell. This limit will be approached the more nearly, the more perfectly the immersed plates can be coated with the gases, and the more perfectly the immediate contact of the metallic with the liquid conductors is prevented."

*

In the same memoir we find other experiments proving the absence of polarization in all cases, in which the deposition of the gases on the electrodes is prevented, which has been previously mentioned. (Section 35.)

§ 38. Diminution of polarization by heating the liquid.-De la Rive describes the following experiment in the Biblioth. Univers., February, 1837, p. 388: In the closing arc of a battery of four elements, he inserted a galvanometer and a decomposing cell, composed of two platinum plates, immersed in a glass of water; the galvanometer indicated a deflection of 12°. He then placed under the posi tive pole-plate where oxygen was developed, a large spirit-flame, so that the plate began to glow, and the part immersed in the liquid being gradually heated by conduction, raised it to the boiling point. (The platinum plate was probably bent at right-angles.) No change was perceptible in the deflection; the same was done at the negative plate, but now the needle advanced to 30°. After removing the lamp, the deflection returned to 12°.

When the water was replaced by dilute sulphuric acid, the original deflection was 45°; by heating the negative plate it rose to 80°, while heating the positive plate had no effect whatever.

Hence De la Rive concludes, that heat has no influence on the pas

sage of the electrical current from a metal into a liquid, but that it perceptibly favors the passage of the current from a liquid to a metal. Vorsselman de Heer opposed this singular opinion. He ascribed the action not directly to heat, but to the motion of the liquid produced by boiling, and by which the polarizing gases were removed from the electrodes. He supported his view by the fact that the same effect can be produced without heat, by merely agitating the plate slightly in the liquid, or causing motion in the liquid near the plates by a glass rod.

He took a voltaic pile of five pairs charged with pure water. Two platinum wires dipped in a glass of distilled water, forming the poles of the battery, the galvanometer placed in the circuit indicated 45°; this deflection, however, rapidly decreased on account of the increasing polarization, but it always increased again when the negative wire was shaken. The following results were obtained:

After 15' 34°; the negative wire being shaken, 40°

do.

do.

Similar results were obtained with copper wires.

38°

32°

Vorsselman's explanation is certainly the correct one, yet he leaves unexplained the circumstance of the positive pole being unaffected by heating or shaking. Is it because oxygen adheres more firmly to platinum plates than hydrogen?

According to a notice in the "Jahresbericht über die Fortschritte der Chemie, Physik u. s. w. von Liebig und Kopp, Giessen 1849, s. 297," Becker of Giessen has investigated more minutely the decrease of polarization at increasing temperatures of the decomposing fluid; but his labors have not yet been published.

§ 39. Cause of galvanic polarization.-One of the first who opposed the hypothesis of resistance to transition, and endeavored to establish the existence of an electro-motive opposing force in the voltameter, was Schönbein. While all the researches on this subject, hitherto considered, rested upon the relation of the passage of the current through electrolytes, to Ohm's law, and while they were in this way led indirectly to the view that galvanic polarization was to be ascribed to the strata of gas covering the electrodes, Schönbein regarded the subject from an entirely different point of view, and sought to prove directly the polarizing influence of gases on metallic plates.

The most important of Schönbein's memoirs on this subject are the following:

Observations on the electrical polarization of solid and liquid conductors. (Pog. Ann. XLVI, 109.)

New observations on voltaic polarization of solid and liquid conductors. (Pog. Ann. XLVII, 101.)

On voltaic polarization of solid and liquid bodies. (P. A. LVI, 135.) I will here state the essential results of Schönbein's researches, without reporting upon the contents of these separate papers.

The following experiment is mentioned on page 199 of the second vol