This equality deranged my law of latent heat in a singular manner, for d becoming equal to zero, the formula produced a false result. We cannot escape from this difficulty as in the case of wax, for in all the metals the specific heat in the solid state is taken very far from the point of fusion, and the variation with the temperature is also insignificant. As the specific heats to be measured did not extend beyond a few hundredths I thought at first that it might be only an apparent exception, arising from very supposable errors in the temperatures, which reached as far as 440° C. But, besides, that, the same exception was shown in fusi ble alloys, an inquiry into the errors of the processes showed me that it was impossible to admit errors of sufficient magnitude to cause their correction to make the results of the experiment agree with those of calculation. I was then convinced that the specific heat of metals was nearly the same in a solid as in a liquid state. One might have been able to have foreseen this by observing that an atom of mercury, notwithstanding its liquidity, has a specific heat scarcely greater than that of other metals in a solid state. One would believe, after this, that it is impossible to bring the metals within the rule, but the impossibility will disappear when, instead of taking the formula empiri. cally, we have discovered the physical interpretation it includes. I met with a difficulty in the case of wax, which at first checked my progress: I found its specific heat in a solid state was higher than when liquid, so that & becoming negative, the formula would produce an absurd result. But the difficulty vanished by following the progress of the specific heat connected with the act of softening. The specific heat of wax, between 58 and 12, is at least as great as that of water; between 26 and 6; it is still no more than 0.52, and only 0.39 between 2 and 20° below zero; besides, a graphic construction shows that it has a tendency to become constant. If, now, we deduct from the specific heat found between 2 and 20° below zero, as the latent heat of fusion, all that exceeds 0.39, we shall see that the formula is applicable. Now, it is very evident that the enormous increase For this interpretation I remarked first, of specific heat connected with the softening that if we agree to understand by the term ought to be included in the latent heat. In degree the variation of temperature prothe case of phosphorus the effect is analo- duced by the same amount of heat, it will gous, but not so strongly marked; we may happen that the same temperature will be even say that, theoretically, it is the general indicated by different numbers in the case of rule: the latent heat of fusion is spread over different substances, but that the difference, a certain extent of the scale, an extent which in general, will be very small. For instance, for most bodies is perceptibly reduced to a according to Dulong and Petit's experiments, point. The change in the thermometer during the difference will never be more than 4 dethe action of cooling, marks these differences.grees in an interval of 300 for various subThus, for certain compound bodies there is no really fixed point during solidification: there is but a slower change in the thermo stances of different natures, solid, metallic, or non-metallic, such as glass, copper, mercury, and platina. We, therefore, take no notice of this little difference. Now, therefore, let c and C be the specific heats in a solid and in a liquid state, (160+t); c represents the heat existing from 160° to t deg. in unity of weight in a solid state, let us add the latent heat and we shall have the heat contained in the liquid within the same limits. Therefore, since C-c=d, it follows that we shall have, as the result of their point of fusion, is strikingly greater When we consider the result of the phenomenon of superfusion, it becomes natural to believe that the liquid state is in reality compatible with the lowest temperatures, and that solidification is only an accidental accompaniment ; water and phosphorus remain liquid at from 10 to 20° below their melting point; sulphur descends even lower still. I have seen drops of sulphur remain liquid on thermometers which had returned to the ordinary temperature; the contact of a feather will determine the solidification. In general, the liquid state continues to a lower point in proportion to the care we take to avoid the influence of molecular attraction; and it is natural to conclude, by induction, that, if this attraction be overcome, by isolating, for example, the last molecular groups, the liquid state would continue indefinitively. 1 kilogramme of water would continue to lose a portion of caloric at each lowering of the temperature, by a degree; and not half a portion, as in the case of ice. We may, therefore, conceive without difficulty this result of the experiment, that the heat lost from the melting point down to 160° below zero, is equal to the specific heat that agrees with the liquid state, taken as many times as there are degrees in the reduced heat. We have seen that the formula (160+t)d=7 is not verified in the case of the metals, it is easy, however, to see the reason. We place for the specific heat, in a liquid state, the value found above the melting point, while in reality we ought to take the value below the point of fusion. In the case of many bodies, particularly those easily subjected to superfusion, these two values are confounded; but it is not the same with metals, for which we know superfusion is scarcely sensible, probably on account of the enormous power of the molecular action. Provisionally, we cannot say that the metals form an exception to the law represented by the formula (160+t) (C-c) = l, because the exception only manifests itself when we employ, in the place of C, any other value than that which is really represented by that letter. To confirm this explanation it is only necessary to show, at least by some indirect process, that the specific heat of metals, in a liquid state, below | than in the solid state. Now, the fusible alloys offer phenomena which are in favour of this supposition. One consequence of the formula is a very natural determination of the absolute zero; it is evident that the latent heat of fusion is the difference between the total heats of the solid and the liquid, at the melting heat; now the formula has shown us that it is also the difdifference of the continuous heats from 160° below zero. This result is simple if the absolute zero is 160° below the ordinary zero ; while, if we imagine that it is lower, we are forced to admit, without seeing any reason for it, that at 160° below zero the heat in the liquid state is precisely the same as in the solid state. Besides, when we see that the continuous heat in the liquid can already be represented, to 160° below zero, by (160+) C, we can scarcely refuse to extend this formula further, so that, if, instead of 160° we place the distance of the ordinary zero at the absolute zero, we have (x + 1) C, for the total heat of the liquid we have even (x + 1) c for that of the solid, and, consequently, (x + 1) d = 1, because the latent heat, 7, is the difference of the two total heats. Now, this equation compared with the result of the experiment represented by (160+t) d=l gives x=160°. It is generally believed that the latent heat of fusion is a constant quantity; it is, on the contrary, infinitely more probable that it varies with the latent heat of evaporation, according to the temperature or change of state produced. Let us imagine 1 kilogramme of ice at 20°, it is necessary to raise it to zero to impart to it 10 of caloric, and afterwards 79.2 to melt it—in all 89.2. Act how we will, it is always necessary to restore it to its primitive state, to remove all the heat we have bestowed upon it. If, now, we operate by superfusion, the water in cooling 20° loses 20 of caloric; it can, therefore, only disengage 69.2 in becoming solid such would be the latent heat of water at - 20°. It is also evident that this would give the formula, (160 + t) d 7, when we make t = 20°. In order that the latent heat should remain constant, it would be necessary that the water, still preserving its liquid state, should suddenly experience a reduction of one-half in its capacity for heat, which is not at all probable. Besides, the study of fusible alloys furnishes a fresh proof of this variation of the latent heat. On this account, I consider latent heat as the difference between the continuous heats in the solid and liquid, at the tempe rature, be it what it may, at which it becomes solid. This is expressed by the formula (x+t) d=l, where x indicates the distance of the absolute zero from the ordinary zero. Now, since experiment gives (160+t) d=l, and, as l is the same in both cases, that should be equal to 160°, is a necessary consequence. The distance of the absolute zero from the zero of melting ice, being given in degrees of equal capacity, and the ordinary degrees being nearly of equal capacity, through a great portion of the scale, it follows that we have a measure of the total heat of a solid or liquid body, by the very simple formula (160+t) c; t being the temperature of the body, and c the specific heat; in the case of metals, in a liquid state, we have to add the latent heat 7. This formula will, I hope, be applied in researches into the heat produced by chemical action, we cannot measure the heat really produced, if we do not know the total heat, excepting in the very peculiar case where the capacity of the compound would be the mean of the capacity of its constituents. The total heat of vapours at the boiling temperature, may be calculated by adding the latent heat 7. The latent heat of evaporation may also be considered as the difference of the total heats of the liquid and the vapour, at the temperature when the conversion into vapour takes place; so that, if we represent by d, the mean difference of the specific heats of the liquid and the vapour, we have (160+t) d=l. The following table will give the results of calculation for bodies whose specific heat is known when in a liquid state, and the heat when converted into vapour, at least approximatively, at the temperature t of ebullition, under a pressure of 0.76m.; we refer here to atomic weights. To sum up all, the present paper submits to the judgment of the Academy 1. The correction of several melting points, and a fresh determination of them. 2. The measure of the latent heat of fusion, for thirteen substances. 3. Twenty-five determinations of specific heat, at temperatures extending from 30 to 440°. 4. The rule (160+t) d=l. That is to say, that the latent heat of fusion, is equal to the difference of the two specific heats, taken as many times as there are degrees between the temperature of fusion, and 160° below zero. 5. The observation, that the heat necessary to melt metals is nearly in proportion to the force required to separate or divide their molecules. 6. The result, that the specific heat of metals is nearly the same in the solid as in the liquid state. 7. The principle that the heat contained in a liquid, from its melting point to 160° below zero, is expressed by (160+) C, as if there had been no change in its state. 8. The determination of the absolute zero, 160° below the ordinary zero. 9. The principle that latent heat is a quantity varying with the temperature at which solidification takes place. 10. The other principle, that solidification, at the time of cooling is a contingent, and not a necessary phenomenon, so that the liquid state, and even the gaseous state would be compatible with the lowest temperature. 11. A measure of the total heat of a body which is applicable to solids, liquids, and vapours. 12. The result, that the water of crystallisation in salts, would possess nearly the same specific heat, and the same melting heat as ice, so that, to melt a hydrated salt, there would be required at least as much heat as would be necessary to melt a weight of ice equal to that of its water of crystallisation. LATENT HEATS OF FUSION. Tin Bismuth Lead Zinc Arcet's alloy, Pb2 Sn2 Bi3 Latent heat of the unity of wt. 7.63 4.71 9.175 62.98 46.18 24 H2O..... 36.4 54.65 Chloride of lime, 45.79 43.51 Specific heat. 340 ON THE EMPLOYMENT OF THE BY T. KIPP. THE extreme sensitiveness of the sulphocyanide of potassium as a test for peroxide of iron, led to its being employed for detecting iron in nitric acid. Tromsdorff, however, observed that sometimes a nitric acid, prepared from the purest materials, exhibited the reaction of iron. To ascertain more closely the reason of this phenomenon, the author prepared nitric acid,-1st, from ordinary aqua regia, freed from muriatic acid by nitrate of silver containing copper, and then slowly distilled; 2nd, according to the method described by Wackenroder; and 3rd, by employing pure sulphuric acid and pure nitre. The acid so obtained had the specific gravity 1.260-1.30; it was clear, and was not affected by sulphuretted hydrogen. A single drop of a solution of sulphocyanide of potassium gave, however, a more or less reddish colour to all three preparations, the weakest being the least coloured. The acids were now saturated with carbonate of ammonia, when they remained perfectly clear; and in none of them was any alteration perceptible on the addition of prussiate of potash, sulphuretted hydrogen, or sulphocyanide of po0.039 tassium. Upon this the author prepared a nitric acid which contained sooo peroxide of iron, and treated as above. On saturation, the liquid assumed a light yellow colour, and presented a very slight trace of sediment. Prussiate of potash coloured the suspended particles of peroxide of iron blue; sulphocy0.212 anide of potassium, however, did not produce 0.235 the slightest alteration, although the presence of the iron had been rendered perfectly distinct by it before saturation. Rose states, in his Manual of Analytical Chemistry," that the red colour which is produced by sulphocyanide of potassium in liquids containing peroxide of iron again disappears, after a time, on the addition of more nitric acid. and 240 0.061 370 280 0.035 440 340 300 136 .... Arcet's alloy 0.036 0.046 0.413 0.344 0.454 60 0.562 31 0.358 Nitrate of potash Phosphate of soda 79 44 Idem 2 20 Chloride of lime... 127 100 Idem 100 Idem In Berzelius' work we find, that when a solution of sulphocyanide of potassium is mixed with nitric acid and heated, a yellow body is formed, which has great similarity to the so-called sulphuret of cyanogen, but contains 1 equiv. sulphur more. On observing more closely the decomposing action of nitric acid upon the sulphocyanide of potassium, the author was not only able to confirm the statement of Rose, but he found, on the first addition of the acid, that the red colour of the sulphocyanide increased in in* Archiv der Pharm., and Chem. Gazette. tensity, and, on its slow disappearance, an evolution of gas took place, upon which the liquid assumed a light greenish tint. When the nitric acid was mixed with a little more sulphocyanide of potassium, the liquid remained almost colourless for some time; subsequently, however, it became red, and bubbles of gas were given off, upon which the liquid again exhibited a green colour. On the addition of a persalt of iron, the red colour could not be reproduced; but on the addition of a further quantity of sulphocyanide, the liquid not merely became bright red, but a violent evolution of nitrous vapours resulted, so that both the sulphocyanide of potassium and the nitric acid must have been decomposed. The author now mixed a few drops of the first hydrate of nitric acid with some water until the specific gravity was 1.07, and then added some sul phocyanide of potassium, when it immediately assumed a bright red colour. After the author had boiled a portion of the acid, he again added some sulphocyanide, but no colouring ensued. It is evident, therefore, that the decomposition of the sulphocyanide of potassium must have been produced by nitrous acid, which was also confirmed by the fact, that when the nitric acid had been previously freed entirely from nitrous acid by chromate of potash, peroxide of lead, &c., it required a considerable length of time before it became coloured by sulphocyanide of potassium. What the red body may be cannot yet be determined, but from the above experiments it results, that the acid must not be too concentrated for the peroxide of iron to to be detected in it; that it must be in slight excess, but free from nitrous acid. II. CHEMICAL MANUFACTURES AND AGRICULTURAL CHEMISTRY. ON THE MANUFACTURE OF ARTIFICIAL ULTRAMARINE.* BY M..C. P. PRUCKNER. MM. GUINET and ROBIQUET were the first persons who manufactured artificial ultramarine on a large scale, for commercial purposes in 1830, Levercus established a manufactory in the environs of Cologne, and in 1841, MM. Leykauf Heine and Co., set up works for the manufacture of artificial ultramarine, where ultramarine of every quality and price is produced for commercial purposes. The process employed by these gentlemen has not been published; in consequence however of our personal dealings with them, and the experiments which they suggested to us, we are enabled to give an idea of the manufacture, by which we hope to throw some light upon this branch of science. We will first say a few words upon the choice of the primary substances to be employed, which are argil or potter's clay, sulphate of soda, sulphur, coal, or charcoal, and a salt of iron, generally green vitriol (protosulphate of iron). The argil employed in manufacturing artificial ultramarine has the greatest influence upon the colour to be produced; and probably in a great many instances failure is to be *Erdman's Practical Journal of Chemistry. attributed to the employment of clay of too ferruginous a nature. I use a white clay which is not coloured by fire, and which consequently contains but very little iron. It is a species of kaolin, of a dead colour, biting to the tongue, and forming a very short paste with water, and is to be met with in the principality of Reuss, in the environs of Roschitz, being generally used for making porcelain. This clay contains from 42 to 43 per cent. of alumina, and it will readily be conceived that the preference ought to be given to the most aluminous clay. In the manufactory at Nuremberg, a white terra sigillata (called bolus alba in pharmacy), which comes from Tischeureuth, is employed. At Nuremberg impure sulphate of soda is employed, which is the residuum from the manufacture of hydrochloric acid, and which may either be refined in the works, or purchased in a refined state; this operation, which will be referred to hereafter, is intended principally to separate the free hydrochloric acid and the salts of iron, which would injure and might even entirely destroy the blue colour of the ultramarine obtained: roll or stone brimstone is too well known to require any notice. With regard to the coal or charcoal to be employed, dry wood charcoal appears to be that best adapted to the purpose. |