THE LONDON, EDINBURGH, AND DUBLIN PHILOSOPHICAL MAGAZINE AND JOURNAL OF SCIENCE. [FOURTH SERIES.] MARCH 1869. XXII. Historical Notes on some Phenomena connected with the Boiling of Liquids. By CHARLES TOMLINSON, F.R.S.* 1. WHEN water is heated to that temperature at which its tension equals the whole pressure of both air and vapour on its surface, and it begins to emit steam not only from its surface (as it did before), but from all parts of its depth, it is said to be boiling. The boiling-point of any liquid means, therefore, the temperature at which its evaporating tendency equals the pressure of the atmosphere at the time-or the lowest temperature at which its vapour can have the elasticity of common air. 2. During many years after the invention of the barometer, and the consequent discovery of atmospheric pressure, it was supposed that the above statements contained a sufficient account of boiling. It was not until the irregular shifting of the boiling-point in thermometers under a constant pressure had been insisted on by such men as De Luc, Shuckburg, Horsley, and Cavendish, that steps were taken to determine the conditions on which that instrument should be constructed. The Royal Communicated by the Author. [In preparing the paper "On Boiling Liquids " recently read before the Royal Society, I was led to consult a good many books and memoirs, and to accumulate a number of historical details which, though not adapted to the 'Proceedings' of a Society whose chief object is to assist the progress of living science, may yet, if briefly stated, be acceptable in the Philosophical Magazine, which performs the useful functions both of critic and of observer. The insertion of these historical details may serve to correct some errors that have crept into our text-books.] Phil. Mag. S. 4. Vol. 37. No. 248. Mar. 18C9. M . Society gave the sanction of its high authority to a Report on the subject prepared by some of its most distinguished members, and published in the Transactions for the year 1777. In this Report the want of fixity in the boiling-point of water under a constant pressure is noticed, and the cause is referred to the depth of water in the vessel, which causes it to boil in gusts. It was found, however, that the elasticity of the steam from boiling water fairly represents the atmospheric pressure, and it was recommended that the water be boiled in a metal vessel constructed so as to allow the bulb, and nearly the whole of that part of the stem that contained mercury, to be surrounded by the steam. 3. I am not aware that in this Report, or in the results previously published that led to it, it is anywhere stated that the nature of the vessel influences the boiling-point. That fact was distinctly brought forward by M. Achard in 1785*. Distilled water was boiled in a brass cylindrical vessel by means of a spiritlamp, the thermometer-bulb being within half an inch of the bottom. Under these conditions the temperature was constantly varying, although the water appeared to boil equably. Blowing on the side of the vessel, opening and shutting a door, or anything that produced an agitation in the air caused a fall in the thermometer of 1·12° R. When the water was boiled in a matrass of white glass, there was no variation in the boiling-point as indicated by the thermometer. Achard performed a large number of experiments on the boiling-point of water in vessels of silver, brass, porcelain, earthenware, glass, &c., many of them being articles in domestic use; and he gives drawings of them in three folding plates, with measurements of their dimensions, to justify his conclusion (afterwards found to be erroneous) that in vessels of the same material the temperature of boiling water varies according to the size of the opening. There is no doubt, however, as to the influence of the vessel on the boiling-point. He says: "Le degré de chaleur de l'eau en ébullition dans différens vases est différent pour la même pression atmosphérique quoique les vases soient chauffés de la même manière et qu'ils se trouvent dans le même bain de sable." His idea was, that, metal being a good conductor, the heat readily escaped from it when the air was agitated, while glass retained its heat, and hence the oscillation of the mercury in the one case and its fixity in the other. He concludes his memoir with the remark "that the experiments prove that the degree of heat of boiling water under an equal * Nouveaux Mémoires de l'Académie Royale de Berlin for 1785, published in 1787. The following is the title of the memoir:-"Expériences faites dans la vue de s'assurer si le degré de chaleur de l'eau pure bouillante est un degré fixe et invariable, indépendant de toute autre circonstance que de la pression de l'atmosphère." pressure of the atmosphere is not a fixed term, but that many circumstances cause it to vary; that it is much more inconstant in vessels of metal than in vessels of glass; and that the action, more or less direct, of the external air on the sides of the vessels, especially when of metal, as upon the surface of the water, produce considerable changes in the degree of heat that it may receive in boiling. As the construction of the thermometer depends on the fixity of the boiling-point, it is not surprising that thermometers made on this supposition, with the greatest possible care, should not always agree." 4. I have given these details at some length, because French and English writers of authority assign to Gay-Lussac the merit of Achard's discoveries. Biot seems to have led the way in this respect. Writing in 1816*, he says, in reference to the boiling of water, "Il y a aussi quelques différences dans le degré de l'ébullition selon la nature des vases que l'on emploie, et selon celle des substances qui se trouvent mêlées à l'eau, même quand elle ne peut les dissoudre. Cette remarque est due à M. GayLussac.' 5. The effect of insoluble substances on the boiling-point was also first noticed by Achard, in 1784+. The water was boiled in a glass vessel; and when the mercury in the thermometer was steady, a drachm of the solid to be tried was thrown in, and the effect noted in tenths of a degree on Réaumur's scale. When the temperature had again become steady, a second drachm of the substance was thrown in, and so on until no further effect was apparent. A large number of substances were tried in this way, and the results are given in Tables occupying fourteen quarto pages. Each Table contains six columns, for recording the name of the substance, the height of the barometer, the boiling-point of the water before the addition, the weight of the water, the weight of the substance added, and, lastly, the effect on the thermometer in tenths of a degree Réaumur. Thus 1 drachm of iron-filings lowered the thermometer ten tenths, or -10 as Achard writes it; a second drachm had no further effect; copper-filings -8, tin-filings -13, white sand 0 to 3, calcspar13, quicklime -9, rosewood -11, limestone in powder -13, the same in a lump -3, bismuth in powder -12, the same in fragments -8, and so on. Achard does not pretend to offer any satisfactory explanation of these results, but he distinctly claims the merit of having originated them. 6. The effect of soluble substances on the boiling-point was *Traité de Physique, vol. i. p. 42. Berlin Memoirs, 1784, published 1786. The following is the title of the memoir:-"Sur l'effet produit par l'addition de différens corps à l'eau, relativement au degré de chaleur dont elle est susceptible dans l'ébullition." clearly made out during the fine experiments undertaken by Dalton, Watt, Robison, Southern, and others for determining the pressures of saturated steam at different temperatures above and below the standard boiling-point. It was noticed that if a minute portion of soda, or of some salt soluble in water and not capable of rising in vapour with it, be allowed to ascend to the top of the mercury, the column rises, thereby indicating a diminished pressure of steam, although the soda has not touched it, but remains covered by the layer of water on the top of the mercury. This shows that the elasticity depends not merely on the temperature and the nature of the vapour, which are both unchanged, but on the nature of the liquid. The adhesion of the soda to the water tends to restrain the water from evaporating, and this tendency is a measurable force and here measured; for it partly balances the tension of the water, or its tendency to emit steam, and thus makes the steam-emitting tension of a solution of soda measurably less than that of pure water at the same temperature. As the difference remains at all temperatures, the solution must always be made hotter than pure water in order to give steam of the same elasticity. 7. The effect of air dissolved in the water on the boiling-point was noticed in minute detail by De Luc* in 1803, not indeed for the first time; for in his previous works, published in 1772 and 1786†, he had described the principal experiment on which his remarkable theory was based. He says:-"Le phénomène de l'ébullition est produit par des bulles d'air que la chaleur dégage du liquide. ; quand on a préalablement purgé l'eau de tout l'air qu'elle contenoit, elle ne peut plus bouillir; et la raison en est que les vapeurs ne peuvent se former qu'à des surfaces libres. Les bulles d'air qui se rassemblent dans son sein y produisent des solutions de continuité; c'est-à-dire, ces surfaces libres nécessaires; mais quand l'eau est purgée d'air les vapeurs ne peuvent se former qu'à sa surface extérieure" (Introduction, &c. vol. i. page 247). De Luc had already described, in 1772, with great minuteness of detail an experiment in which a matrass containing water, and also a small thermometer, had the upper part of its tube drawn out into a capillary bore, and, the matrass being heated in hot water, the air as it accumulated in the fine tube was got rid of by an ejection of steam, and the tube was sealed. The tube was also subjected to percussion during a long time, and the process of heating was continued * Introduction à la Physique terrestre par les Fluides expansibles. Paris, 1803. † Recherches sur les Modifications de l'Atmosphère. Geneva, 1772. See chapter 10 of the Supplement to vol. ii. Recherches sur les Variations de la Chaleur de l'Eau bouillante. Idées sur la Météorologie. London, 1786. until more air got into the fine tube, when it was opened and the air expelled. After continuing these operations of shaking and heating the tube in order to get rid of the air, the tube was raised to 212°; and the point being broken so as to reestablish atmospheric pressure, the water was further heated to 2341° F. without boiling*. 8. Proceeding in the order of time, we come to Gay-Lussac†. While engaged in his experiments on solubility, he seems to have rediscovered the two facts respecting the influence of the vessel on the boiling-point, and also of the effect of an insoluble body in lowering the boiling-point. But it is remarkable that in both his papers Gay-Lussac refers to Achard's results, or rather to his faulty conclusions, with a view to contradict them, without giving him credit for those points that were true. It is quite possible that Achard's fame was more injured by his friends than his opponents, since an attempt had been made by Gmelin and others, in answer to Gay-Lussac's paper of 1812, to show that in a number of vessels of different material, all sunk to the same depth in a sand-bath, water boiled at the same temperature‡; while in 1817 Muncke stated§ that copper-filings appeared to have no influence in lowering the boiling-point, and that sand did so only to the extent of one-tenth of a degree. Gay-Lussac in 1817 noticed these papers, and reasserted the original facts with even more decision than he had done in 1812. But even in this his first notice of the subject, his language is not to be mistaken. His paper is on Deliquescence, and he says:-"In determining the boiling-point of saline and acid liquids I observed a very singular phenomenon which deserves to be known. Water or any other liquid boils later in a glass vessel than in a metallic one, except when we put into the former some turnings of iron, copper, or other metal, or carbon in powder, or pounded glass. The difference in the case of water may amount to 1.3° C. and upwards." In 1817 he states || that water boils later in glass and earthenware (faïence) than in metal vessels. He does not give the measure of the difference, but believes it to vary with the nature of each body, and, with the same substance, according to the nature of its surface: "car il est probable qu'elle dépend à * De Luc also dropped water into oil heated to 82° or 90°, and even 100° R.; but he was by no means satisfied that the water ever attained this temperature. He says:-" Ces gouttes d'eau, renfermées dans l'huile, pouvoient être dans un état particulier" (see paragraph 993 of the Recherches). That is, they were probably in what would now be called the "spheroidal state," which De Luc understood and accurately describes in § 1007. † Annales de Chimie, vol. lxxxii. p. 171. Schweigger's Journal, vol. xxvii. p. 27. § Gilbert's Annalen, vol. lvii. Ann. de Chim. et de Phys. vol. vii. p. 307. |