whether by its exposure to the air, or what is still better, because of its having been in communication with the positive pole; in the second case, the current is stronger than in the former one. This current is stronger and of longer duration with a plate that has a large surface, with spungy platinum, or with a wire whose surface is pulverulent. In these cases the chemical action takes place on a greater number of points, and as it should not cease until the surface is oxidized, it thus necessarily ceases more tardily. It is probable also that the divided state of the platinum permits a slight dissolution of the oxide, and thus constantly presents to the chemical action a portion of surface unoxidized. The explanation I have just given is a consequence of the oxidation of which I have shown that platinum is susceptible: let us now see if it is in accordance with facts.

I soaked for a long time in concentrated nitric acid, wires, plates, and spongy platinum, after having well cleansed them; I then washed them carefully in distilled water, which I renewed several times, in order that there should be no trace of acid left; I always took care that the surfaces of the metals were not in contact with the air. I fixed to the extremities of a galvanometer first a wire and a plate: the plate was positive in relation to the wire; afterwards a wire and a plate on one side, and spongy platinum on the other: the sponge was constantly positive in relation to the plate and the wire; it was the same with a wire whose surface was pulverulent, but this wire was negative with regard to the sponge. The liquid in which the metals were immersed was water, slightly acidulated with nitric acid, or sulphuric acid, both perfectly pure. These experiments seem to indicate perfectly a chemical action on the cleansed surfaces of platinum, which gives birth to a current in which the greatest surface, which is that on which the most extensive chemical action takes place, is positive in relation to the other, as that passes for the oxidable metals.

I afterwards took two platinum plates, which, after having been. cleansed and washed with the greatest care, were left immersed in distilled water; I then dried it in the air, and the other under a bell receiver, which I exhausted of air as quickly as possible, and in which I had placed a capsule filled with concentrated sulphuric acid, in order to hasten the dessication. The plate that had been dried in vacuo was constantly positive to that which had been dried in the air, when they were both immersed in nitric or sulphuric acid diluted with water; a proof that the latter had been oxidized at the surface, and not the former, or that at least this one had been oxidized in the least degree of the two.

I also dried a plate of platinum which had been cleansed in hydrogen it was positive, like that which had been dried in vacuo, but more strongly so; which is not astonishing, seeing that all trace of oxidation should disappear in this case on the surface of the platinum by the effect of the hydrogen.

By immersing two other plates in concentrated nitric acid, I found

the reverse: that is to say, that the one that had been dried in vacuo, or in hydrogen, was negative with regard to that which had been dried in the air. I therefore conclude that in this case, the plate oxidized by its exposition to the air, was attacked by nitric acid, which combined with the oxide of platinum by dissolving it, whilst that the other plate, none or very little oxidized, was not affected or very little so, by the acid; the direction of the current was perfectly in agreement with this explanation.

In order to assure myself of its correctness, I made the following experiments; I left soaking during some months, a piece of spongy platinum in very pure concentrated nitric acid; I then put it in nitric acid very much diluted, and left it in a capsule exposed to the air, but covered with paper, in order to preserve the surface of the platinum from the dust. A large platinum capsule proper for the purpose, the interior of which, however, had been for a long time in contact with the air, was put in communication with the extremity of a galvanometer; it was filled with nitric acid, pure and concentrated; I then immersed in the acid, after having put it in communication with the other extremity of the galvanometer, by means of a platinum wire, at first a morsel of sponge which had been well deoxidized by its long continuance in nitric acid, then a morsel of sponge which, after having been deoxidized, had been left immersed in water, or exposed to the air; the first morsel was negative, the other two strongly positive, particularly that which had been exposed to the air.

There was then put into the capsule nitric acid very much diluted in place of the concentrated acid; the results were precisely inverse. This difference is explained very well if we remark that the action of nitric acid, pure and concentrated, does not take place on platinum not oxidized, but very well on a surface of platinum covered by a slight coating of oxide, whilst that it is the contrary takes place when the acid is very much diluted.

In order that these actions should produce even small deviations of the needle of the galvanometer, it is necessary to have large surfaces of platinum; this is the reason why the deviations are so sensible with the spongy platinum; it was also for this end that I employed a large capsule of platinum. I have even tried the effect by putting to one of the extremities of the galvanometer a platinum wire immersed two or three centimeters only in the liquid, and to the other end spongy oxidized platinum: I had a current of from 5° to 6° only. Replacing the wire by a morsel of the sponge not oxidized, or by a very extensive surface of platinum, I had a current which made the needle deviate more than 90°.

It is evident that in all these phenomena it is a chemical action which is due to electricity that is there developed, and that the oxidation of the platinum has a great part in it. Still, what is this action? This is a point which appears to me to merit a further examination. It is not impossible that it consists, at least in the case of nitric acid, in a decomposition of that acid produced by the oxygen of the oxidized surface of the platinum, as that takes place

with the peroxides, and remarkably so with those of lead and manganese. I shall return to this point, when studying at the same time the circumstances that take place with the other metals, such as gold or palladium, which have analogous properties to those of platinum.

SECTION III.

Of some other Phenomena which may be attributed to the

Oxidation of Platinum.

One phenomenon that proves the facility with which platinum is oxidized, is the state into which a platinum wire constitutes itself after having been employed for some time for an aphlogistic lamp. We take a platinum wire, formed into a helix, of about a demimillimeter in diameter; it is cleansed carefully by putting it into nitric acid, and then washing it several times in distilled water; it is then dried and placed on a spirit lamp. The lamp is then lighted in such a manner that the wire becomes red hot, the flame is then extinguished, and the wire remains incandescent by the effect of the vapour of the lamp. If we continue the phenomenon during twenty-four, or still better forty-eight hours, the wire, which was perfectly bright before the experiment, presents, after it has ceased, a greyish pulverulent surface perfectly similar to the surface of wires that have been used for the alternate disengagement of oxygen and hydrogen. This effect can only be due to the alternative oxidations and reductions which the wire has undergone by the action of the oxygen of the air and the vapour of the alcohol. This is a phenomenon altogether similar to that presented by copper wires exposed to the action of a spirit lamp, which, by the alternate oxidations and reductions they are submitted to, soon become reduced to powder. I have repeated this experiment several times, and I took care to employ perfectly pure alcohol, and threads of amianthus instead of a wick of cotton; I always succeeded very well. The wires whose surfaces have thus become pulverulent, are afterwards much superior to others for the aphlogistic lamp; they become red hot on a much larger extent of surface, and with much greater facility; a new proof that the phenomenon of the aphlogistic lamp is in a great measure due to a succession of alternative oxidations and reductions which facilitate the disaggregation of the metal.

I have likewise rendered pulverulent the surface of a platinum. wire well cleansed, by making use of it to inflame hydrogen in the air, instead of a piece of spongy platinum. The wire, after having been several times removed, and during from ten to fifteen minutes each time rendered incandescent by hydrogen, has taken a greyish surface, slightly pulverulent. This experiment succeeds still better by raising the temperature of the platinum helix by means of a voltaic current, which heats it without making it red. A current of hydrogen mixed with atmospheric air, by means of a double-mouthed

stop-cock, is then directed on the helix, and at the end of a certain time the surface of the platinum has taken the grey appearance already described.

Will not the preceding experiments prove that the phenomena observed by Dæbereiner are due to a succession of alternate oxidations and reductions? The surface of platinum oxidized in the air will be immediately reduced by the hydrogen, then oxidized anew, again reduced, and so on. These alternations succeed each other very rapidly the result is an elevation of temperature that renders the platinum incandescent, and ends by inflaming the hydrogen. What seems to support this explanation is, that the circumstances that favour the greatest oxidation of platinum are those also which are exacted for the success of the experiment of Dobereiner. Thus, for example, Dalony and Thenard have remarked that the platinum, in order to become incandescent under a current of hydrogen in the air, ought to be perfectly pure, and that it is useful, for this end, to wash it in nitric acid; the extent of surface, its state of division, also favour the production of the phenomenon. Let us carefully read all the works that have been written on this subject, and particularly those of Faraday, and we shall find that the most favourable conditions for determining intermedially by platinum the combinations of oxygen and hydrogen, are those also which are the most proper for facilitating the oxidation of this metal.

The explanation I have just given may be objected to: 1st, That there are cases where the action of certain substances determine the combination of oxygen and hydrogen, without admitting a succession of alternate oxidations and reductions; 2ndly, That platinum in its spongy state itself determines certain combinations, which cannot be accounted for in the same manner as I have just explained in the combinations of oxygen and hydrogen. We ought therefore to conclude, if these objections are well founded, that it is necessary to acknowledge in the experiments of Doebereiner, the action of a new force which Berzelius had named the catalytique force.

As to the first objection, I observe that the cases in which there is a combination of oxygen and hydrogen, by the intermedial effect of certain substances for which it is impossible to admit a series of oxidations and reductions, are those in which we make use of a very high temperature, and of bodies that are non-conductors of electricity. These are fragments of glass or porcelain, which are heated very strongly over mercury, and in which we thus develope an electrical state that ought probably to bring about the combinations of the gases by the small discharges which result from it. In fact, if the experiment be made in darkness, we see faint glimmerings, analogous to those observed in agitating mercury in a capsule of porcelain heated to a high temperature. All the cases of this class cited by Dalony and Thenard in their memoir, appear to me to be explicable in this manner. Thus they have not any relation in their immediate cause with the phenomenon discovered by Dæbereiner.

The second objection is stronger. M. Kuhlmann has described

a series of chemical combinations produced through the intermediation of platinum, which seem to conform with difficulty to the explanation that I have given of the combination of hydrogen and oxygen. However, if we study with care all the facts observed by M. Kuhlmann, we may, I believe, explain them by the ordinary chemical actions, without the necessity of having recourse to a new and special force. We must not lose sight of the fact, that if platinum can become oxidized, as I think I have demonstrated, there results from it two important consequences in the phenomena wherein platinum is brought into contact with different substances. The first is this, that when platinum takes oxygen to a compound that contains it, the other element being in a nascent state, it finds itself by that more disposed to combine with a third substance which is present. The second consequence is, that the oxygen condensed on the surface of the platinum which is ready to abandon it, finds itself by the simple fact of its condensation in a state far more proper to form a combination than when it is in the gaseous form. If we follow these two consequences in their applications, the phenomena observed by M. Kuhlmann may, with only a few exceptions, be easily explained. There is only those in which the oxygen, not intervening in an apparent manner, neither in a free state, nor in a state of combination, which present some difficulties in their explanation. But, in these cases, it will be possible that the atmospheric air, and consequently the oxygen it contains, performs a part of which there can be no doubt; it is possible also that the platinum may combine with the carbon, with sulphur, or with cyanogen, in the same manner and with the same facility that it combines with oxygen. I will not dwell any longer upon this particular point, which merits being treated in a special manner, and with the help of experiment. I shall confine myself to resuming my idea, in saying that the superficial chemical actions, of which the surface of platinum is susceptible, appear sufficient to account in a satisfactory manner for the phenomena of combinations which the presence of this metal determines, and which are generally attributed to the force named catalytique.

I will not terminate this paragraph without speaking of a phenomenon that appears to me likewise to be attached to the oxidation of platinum and to depend upon it. I allude to the peculiar odour remarked in the oxygen gas which is disengaged at the positive pole, when we decompose water by the pile, and when we make use of this liquid by putting it in the circuit of wires of gold or platinum. There is no one who having often operated with the pile, who has not been struck with this odour; as for myself I have often observed it: but M. Schoenbein is the first who has studied it with any particular care, and who has made it the object of very interesting researches.*

M. Schoenbein has been struck with the analogy that exists be

Bibl. Univ. nouvelle séree, t. xxvIII, p. 342 (Août, 1840).

N