[16 bis] Hannay, in 1880, obtained diamonds by heating a mixture of heavy liquid hydrocarbons (paraffin oils) with magnesium in a thick iron tube. This investigation, however, was not repeated.
[17] The electrical furnace is an invention of recent times, and gives the possibility of obtaining a temperature of 3,500°, which is not only not obtainable in ordinary furnaces, but even in the oxyhydrogen flame, whose temperature does not exceed 2,000°. The electrical furnace consists of two pieces of lime, laid one on the other. A cavity is made in the lower piece for the reception of the substance to be melted between two thick electrodes of dense carbon. On passing a current of 70 volts and 450 ampères a temperature of 3,000° is easily obtained. At a temperature of 2,500° (100 ampères and 40 volts) not only do all metals melt, but even lime and magnesia (when placed in the space between the carbon electrodes, i.e. in the voltaic arc) become soft and crystallise on cooling. At 3,000° lime becomes very fluid, metallic calcium partially separates out and a carbon compound, which remains liquid for a long time. At this temperature oxide of uranium is reduced to the suboxide and metal, zirconia and rock crystal fuse and partially volatilise, as also does alumina; platinum, gold, and even carbon distinctly volatilise; the majority of the metals form carbides. At such a temperature also cast iron and carbon give graphite, while according to Rousseau, between 2,000° and 3,000° the diamond passes into graphite and conversely graphite into the diamond, so that this is a kind of reversible reaction.
[17 bis] Moissan first investigated the solution of carbon in molten metals (and the formation of the carbides) such as magnesium, aluminium, iron, manganese, chromium, uranium, silver, platinum, and silicon. At the same time Friedel, owing to the discovery of the diamond in meteoric iron, admitted that the formation of the diamond is dependent upon the influence of iron and sulphur. With this object, that is to obtain the diamond, Friedel caused sulphur to react upon samples of cast iron rich in carbon, in a closed vessel at a maximum temperature of 500°, and after dissolving the sulphide of iron formed, he obtained a small quantity of a black powder which scratched corundum, i.e. diamond. Moissan's experiments (1893) were more successful, probably owing to his having employed the electrical furnace. If iron be saturated with carbon at a temperature between 1,100° and 3,000°, then at 1,100°–1,200° a mixture of amorphous carbon and graphite is formed, while at 3,000° graphite alone is obtained in very beautiful crystals. Thus under these conditions the diamond is not formed, and it can only be obtained if the high temperature be aided by powerful pressures. For this purpose Moissan took advantage of the pressure produced in the passage of a mass of molten cast iron from a liquid into a solid state. He first melted 150–200 grams of iron in the electrical furnace, and quickly introduced a cylinder of carbon into the molten iron. He then removed the crucible with the molten iron from the furnace and plunged it into a reservoir containing water. After treating with boiling hydrochloric acid, three varieties of carbon were obtained: (1) a small amount of graphite (if the cooling be rapid); (2) carbon of a chestnut colour in very fine twisted threads, showing that it had been subjected to a very high pressure (a similar variety was met with in various samples of the Canon Diabolo), and lastly (3) an inconsiderable quantity of an exceeding dense mass which was freed from the admixture of the lighter modifications by treatment with aqua regia, sulphuric and hydrofluoric acids, and from which Moissan, by means of liquid bromoform (sp. gr. 2·900), succeeded in separating some small pieces, having a greater density than bromoform, which scratched the ruby and had the properties of the diamond. Some of these pieces were black, others were transparent and refracted light strongly. The dark grey tint of the former resembled that of the black diamonds (carbonado). Their density was between 3 and 3·5. The transparent specimens had a greasy appearance and seemed to be, as it were, surrounded by an envelope of carbon. At 1,050° they did not burn entirely in a current of air, so that the imperfectly burnt particles, and a peculiar form of grains of a light ochre colour, which retained their crystalline form, could be examined under the microscope. Similar grains also remain after the imperfect combustion of the ordinary diamond. Moissan obtained the same results by rapidly cooling in a stream of coal gas a piece of cast iron, saturated with carbon obtained from sugar and first heated to 2,000°. In this instance he obtained small crystals of diamonds. K. Chroustchoff showed that at its boiling point silver dissolves 6 p.c. of carbon. This silver was rapidly cooled, so that a crust formed on the surface and prevented the metal expanding, and so produced a powerful pressure. A portion of the carbon which separates out under these conditions exhibits the properties of the diamond.
[18] The existence of a molecule S6 is known (up to 600°), and it must be held that this accounts for the formation of hydrogen persulphide, H2S5. Phosphorus appears in the molecule P4 and gives P4H2. When expounding the data on specific heat we shall have occasion to return to the question of the complexity of the carbon molecule.
[19] The hydrocarbons poor in hydrogen and containing many atoms of carbon, like chrysene and carbopetrocene, &c., CnH2(n-m), are solids, and less fusible as n and m increase. They present a marked approach to the properties of the diamond. And in proportion to the diminution of the water in the carbohydrates CnH2mOm—for example in the humic compounds (Note [5])—the transition of complex organic substances to charcoal is very evident. That residue resembling charcoal and graphite which is obtained by the separation (by means of copper sulphate and sodium chloride) of iron from white cast-iron containing carbon chemically combined with the iron, also seems, especially after the researches of G. A. Zaboudsky, to be a complex substance containing C12H6O3. The endeavours which have been directed towards determining the measure of complexity of the molecules of charcoal, graphite, and the diamond will probably at some period lead to the solution of this problem and will most likely prove that the various forms of charcoal, graphite, and the diamond contain molecules of different and very considerable complexity. The constancy of the grouping of benzene, C6H6, and the wide diffusion and facility of formation of the carbohydrates containing C6 (for example, cellulose, C6H10O5, glucose, C6H12O6) give reason for thinking that the group C6 is the first and simplest of those possible to free carbon, and it may be hoped that some time or other it may be possible to get carbon in this form. Perhaps in the diamond there may be found such a relation between the atoms as in the benzene group, and in charcoal such as in carbohydrates.
[20] When charcoal burns, the complex molecule Cn is resolved into the simple molecules nCO2, and therefore part of the heat—probably no small amount—is expended in the destruction of the complex molecule Cn. Perhaps by burning the most complex substances, which are the poorest as regards hydrogen, it may be possible to form an idea of the work required to split up Cn into separate atoms.
[21] The viscosity, or degree of mobility, of liquids is determined by their internal friction. It is estimated by passing the liquids through narrow (capillary) tubes, the mobile liquids passing through with greater facility and speed than the viscid ones. The viscosity varies with the temperature and nature of the liquids, and in the case of solutions changes with the amount of the substance dissolved, but is not proportional to it. So that, for example, with alcohol at 20° the viscosity will be 69, and for a 50 p.c. solution 160, the viscosity of water being taken as 100. The volume of the liquid which passes through by experiment (Poiseuille) and theory (Stokes) is proportional to the time, the pressure, and the fourth power of the diameter of the (capillary) tube, and inversely proportional to the length of the tube; this renders it possible to form comparative estimates of the coefficients of internal friction and viscosity.
As the complexity of the molecules of hydrocarbons and their derivatives increases by the addition of carbon (or CH2), so does the degree of viscosity also rise. The extensive series of investigations referring to this subject still await the necessary generalisation. That connection which (already partly observed) ought to exist between the viscosity and the other physical and chemical properties, forces us to conclude that the magnitude of internal friction plays an important part in molecular mechanics. In investigating organic compounds and solutions, similar researches ought to stand foremost. Many observations have already been made, but not much has yet been done with them; the bare facts and some mechanical data exist, but their relation to molecular mechanics has not been cleared up in the requisite degree. It has already been seen from existing data that the viscosity at the temperature of the absolute boiling point becomes as small as in gases.
[22] In a number of hydrocarbons and their derivatives such a substitution of metals for the hydrogen may be attained by indirect means. The property shown by acetylene, C2H2, and its analogues, of forming metallic derivatives is in this respect particularly characteristic. Judging from the fact that carbon is an acid element (that is, gives an acid anhydride with oxygen), though comparatively slightly acid (for carbonic acid is not at all a strong acid and compounds of chlorine and carbon, even CCl4, are not decomposed by water as is the case with phosphorus chloride and even silicic chloride and boric chloride, although they correspond with acids of but little energy), one might expect to find in the hydrogen of hydrocarbons this faculty for being substituted by metals. The metallic compounds which correspond with hydrocarbons are known under the name of organo-metallic compounds. Such, for instance, is zinc ethyl, Zn(C2H5)2, which corresponds with ethyl hydride or ethane, C2H6, in which two atoms of hydrogen have been exchanged for one of zinc.
[23] Gaseous and volatile hydrocarbons decompose when passed through a heated tube. When hydrocarbons are decomposed by heating, the primary products are generally other more stable hydrocarbons, among which are acetylene, C2H2, benzene, C6H6, naphthalene, C10H8, &c.