[11 bis] It is easy to demonstrate the acid properties of carbonic anhydride by taking a long tube, closed at one end, and filling it with this gas; a test-tube is then filled with a solution of an alkali (for instance, sodium hydroxide), which is then poured into the long tube and the open end is corked. The solution is then well shaken in the tube, and the corked end plunged into water. If the cork be now withdrawn under water, the water will fill the tube. The vacuum obtained by the absorption of the carbonic anhydride by an alkali is so complete that even an electric discharge will not pass through it. This method is often applied to produce a vacuum.
[12] The reasons for distinguishing the uni-, bi-, tri-, and quadrivalent metals will be explained hereafter on passing from the univalent metals (Na, K, Li) to the bivalent (Mg. Ca, Ba), Chapter [XIV].
[13] Up to the year 1840, or thereabout, acids were not distinguished by their basicity. Graham, while studying phosphoric acid, H3PO4, and Liebig, while studying many organic acids, distinguished mono-, bi-, and tribasic acids. Gerhardt and Laurent generalised these relations, showing that this distinction extends over many reactions (for instance, to the faculty of bibasic acids of forming acid salts with alkalis, KHO or NaHO, or with alcohols, RHO, &c.); but now, since a definite conception as to atoms and molecules has been arrived at, the basicity of an acid is determined by the number of hydrogen atoms, contained in a molecule of the acid, which can be exchanged for metals. If carbonic acid forms acid salts, NaHCO3, and normal salts, Na2CO3, it is evident that the hydrate is H2CO3, a bibasic acid. Otherwise it is at present impossible to account for the composition of these salts. But when C=6 and O=8 were taken, then the formula CO2 expressed the composition, but not the molecular weight, of carbonic anhydride; and the composition of the normal salt would be Na2C2O6 or NaCO3, therefore carbonic acid might have been considered as a monobasic acid. Then the acid salt would have been represented by NaCO3,HCO3. Such questions were the cause of much argument and difference of opinion among chemists about forty years ago. At present there cannot be two opinions on the subject if the law of Avogadro-Gerhardt and its consequences be strictly adhered to. It may, however, be observed here that the monobasic acids R(OH) were for a long time considered to be incapable of being decomposed into water and anhydride, and this property was ascribed to the bibasic acids R(OH)2 as containing the elements necessary for the separation of the molecule of water, H2O. Thus H2SO4 or SO2(OH)2, H2CO3, or CO(OH)2, and other bibasic acids decompose into an anhydride, RO, and water, H2O. But as nitrous, HNO2, iodic, HIO3, hypochlorous, HClO, and other monobasic acids easily give their anhydrides N2O3, I2O5, Cl2O, &c., that method of distinguishing the basicity of acids, although it fairly well satisfies the requirements of organic chemistry, cannot be considered correct. It may also be remarked that up to the present time not one of the bibasic acids has been found to have the faculty of being distilled without being decomposed into anhydride and water (even H2SO4, on being evaporated and distilled, gives SO3 + H2O), and the decomposition of acids into water and anhydride proceeds particularly easily in dealing with feebly energetic acids, such as carbonic, nitrous, boric, and hypochlorous. Let us add that carbonic acid, as a hydrate corresponding to marsh gas, C(HO)4 = CO2 + 2H2O, ought to be tetrabasic. But in general it does not form such salts. Basic salts, however, such as CuCO3CuO, may be regarded in this sense, for CCu2O4 corresponds with CH4O4, as Cu corresponds with H2. Amongst the ethereal salts (alcoholic derivatives) of carbonic acid corresponding cases are, however, observed; for instance, ethylic orthocarbonate, C(C2H5O)4 (obtained by the action of chloropicrin, C(NO2)Cl3, on sodium ethoxide, C2H5ONa; boiling point 158°; specific gravity, 0·92). The name orthocarbonic acid for CH4O4 is taken from orthophosphoric acid, PH3O4, which corresponds with PH3 (see Chapter on Phosphorus).
[14] Long ago endeavours were made to find a measure of affinity of acids and bases, because some of the acids, such as sulphuric or nitric, form comparatively stable salts, decomposed with difficulty by heat and water, whilst others, like carbonic and hypochlorous acids, do not combine with feeble bases, and with most of the other bases form salts which are easily decomposed. The same may be said with regard to bases, among which those of potassium, K2O, sodium, Na2O, and barium, BaO, may serve as examples of the most powerful, because they combine with the most feeble acids and form a mass of salts of great stability, whilst as examples of the feeblest bases alumina, Al2O3, or bismuth oxide, Bi2O3, may be taken, because they form salts easily decomposed by water and by heat if the acid be volatile. Such a division of acids and bases into the feeblest and most powerful is justified by all evidence concerning them, and is quoted in this work. But the teaching of this subject in certain circles has acquired quite a new tone, which, in my opinion, cannot be accepted without certain reservations and criticisms, although it comprises many interesting features. The fact is that Thomsen, Ostwald, and others proposed to express the measure of affinity of acids to bases by figures drawn from data of the measure of displacement of acids in aqueous solutions, judging (1) from the amount of heat developed by mixing a solution of the salt with a solution of another acid (the avidity of acids, according to Thomsen); (2) from the change of the volumes accompanying such a mutual action of solutions (Ostwald); (3) from the change of the index of refraction of solutions (Ostwald), &c. Besides this there are many other methods which allow us to form an opinion about the distribution of bases among various acids in aqueous solutions. Some of these methods will be described hereafter. It ought, however, to be remarked that in making investigations in aqueous solutions the affinity to water is generally left out of sight. If a base N, combining with acids X and Y in presence of them both, divides in such a way that one-third of it combines with X and two-thirds with Y, a conclusion is formed that the affinity, or power of forming salts, of the acid Y is twice as great as that of X. But the presence of the water is not taken into account. If the acid X has an affinity for water and for N it will be distributed between them; and if X has a greater affinity for water than Y, then less of X will combine with N than of Y. If, in addition to this, the acid X is capable of forming an acid salt NX2, and Y is not, the conclusion of the relative strength of X and Y will be still more erroneous, because the X set free will form such a salt on the addition of Y to NX. We shall see in Chapter [X]. that when sulphuric and nitric acids in weak aqueous solution act on sodium, they are distributed exactly in this way: namely, one-third of the sodium combines with the sulphuric and two-thirds with the nitric acid; but, in my opinion, this does not show that sulphuric acid, compared with nitric acid, possesses but half the degree of affinity for bases like soda, and only demonstrates the greater affinity of sulphuric acid for water compared with that of nitric acid. In this way the methods of studying the distribution in aqueous solutions probably only shows the difference of the relation of the acid to a base and to water.
In view of these considerations, although the teaching of the distribution of salt-forming elements in aqueous solutions is an object of great and independent interest, it can hardly serve to determine the measure of affinity between bases and acids. Similar considerations ought to be kept in view when determining the energy of acids by means of the electrical conductivity of their weak solutions. This method, proposed by Arrhenius (1884), and applied on an extensive scale by Ostwald (who developed it in great detail in his Lehrbuch d. allgemeinen Chemie, v. ii., 1887), is founded on the fact that the relation of the so-called molecular electrical-conductivity of weak solutions of various acids (I) coincides with the relation in which the same acids stand according to the distribution, (II) found by one of the above-mentioned methods, and with the relation deduced for them from observations upon the velocity of reaction, (III) for instance, according to the rate of the splitting up of an ethereal salt (into alcohol and acid), or from the rate of the so-called inversion of sugar—that is, its transformation into glucose—as is seen by comparing the annexed figures, in which the energy of hydrochloric acid is taken as equal to 100:—
| I | II | III | |
| Hydrochloric acid, HCl | 100 | 100 | 100 |
| Hydrobromic acid, HBr | 101 | 98 | 105 |
| Nitric acid, HNO3 | 100 | 100 | 96 |
| Sulphuric acid, H2SO4 | 65 | 49 | 96 |
| Formic acid, CH2O2 | 2 | 4 | 1 |
| Acetic acid, C2H4O2 | 1 | 2 | 1 |
| Oxalic acid, C2H2O4 | 20 | 24 | 18 |
| Phosphoric acid, PH3O4 | 7 | — | 6 |
The coincidence of these figures, obtained by so many various methods, presents a most important and instructive relation between phenomena of different kinds, but in my opinion it does not permit us to assert that the degree of affinity existing between bases and various acids is determined by all these various methods, because the influence of the water must be taken into consideration. On this account, until the theory of solution is more thoroughly worked out, this subject (which for the present ought to be treated of in special treatises on chemical mechanics) must be treated with great caution. But now we may hope to decide this question guided by a study of the rate of reaction, the influence of acids and bases upon indicators, &c., all of which are treated fully in works on physical and theoretical chemistry.
[15] Thus, for instance, in the washing of fabrics the caustic alkalis, such as sodium hydroxide, in weak solutions, act in removing the fatty matter just in the same way as carbonate solutions; for instance, a solution of soda crystals, Na2CO3. Soap acts in the same way, being composed of feeble acids, either fatty or resinous, combined with alkali. On this account all such substances are applied in manufacturing processes, and answer equally well in practice for bleaching and washing fabrics. Soda crystals or soap are preferred to caustic alkali, because an excess of the latter may have a destructive effect on the fabrics. It may be supposed that in aqueous solutions of soap or soda crystals, part of the base will form caustic alkali; that is to say, the water will compete with the weak acids, and the alkali will be distributed between them and the water.
[16] Although carbonic acid is reckoned among the feeble acids, yet there are evidently many others still feebler—for instance, prussic acid, hypochlorous acid, many organic acids, &c. Bases like alumina, or such feeble acids as silica, when in combination with alkalis, are decomposed in aqueous solutions by carbonic acid, but on fusion—that is, without the presence of water—they displace it, which clearly shows in phenomena of this kind how much depends upon the conditions of reaction and the properties of the substances formed. These relations, which at first sight appear complex, may be best understood if we represent that two salts, MX and NY, in general always give more or less of two other salts, MY and NX, and then examine the properties of the derived substances. Thus, in solution, sodium silicate, Na2SiO3, with carbonic anhydride will to some extent form sodium carbonate and silica, SiO2; but the latter, being colloid, separates, and the remaining mass of sodium silicate is again decomposed by carbonic anhydride, so that finally silica separates and sodium carbonate is formed. In a fused state the case is different; sodium carbonate will react with silica to form carbonic anhydride and sodium silicate, but the carbonic anhydride will be separated as a gas, and therefore in the residue the same reaction will again take place, and ultimately the carbonic anhydride is entirely eliminated and sodium silicate remains. If, on the other hand, nothing is removed from the sphere of the reaction, distribution takes place. Therefore, although carbonic anhydride is a feeble acid, still not for this reason, but only in virtue of its gaseous form, do all soluble acids displace it in saline solutions (see Chapter [X].)
[17] Hydrogen and carbon are near akin to oxygen as regards affinity, but it ought to be considered that the affinity of hydrogen is slightly greater than that of carbon, because during the combustion of hydrocarbons the hydrogen burns first. Some idea of this similarity of affinity may be formed by the quantity of heat evolved. Gaseous hydrogen, H2, on combining with an atom of oxygen, O = 16, develops 69,000 heat-units if the water formed be condensed to a liquid state. If the water remains in the form of a gas (steam) the latent heat of evaporation must be subtracted, and then 58,000 calories will be developed. Carbon, C, as a solid, on combining with O2 = 32 develops about 97,000 calories, forming gaseous CO2. If it were gaseous like hydrogen, and only contained C2 in its molecule, much more heat would be developed, and judging by other substances, whose molecules on passing from the solid to the gaseous state absorb about 10,000 to 15,000 calories, it must be held that gaseous carbon on forming gaseous carbonic anhydride would develop not less than 110,000 calories—that is, approximately twice as much as is developed in the formation of water. And since there is twice as much oxygen in a molecule of carbonic anhydride as in a molecule of water, the oxygen develops approximately the same quantity of heat on combining with hydrogen as with carbon. That is to say, that here we find the same close affinity (see Chapter II., Note [7]) determined by the quantity of heat as between hydrogen, zinc, and iron. For this reason here also, as in the case of hydrogen and iron, we ought to expect an equal distribution of oxygen between hydrogen and carbon, if they are both in excess compared with the amount of oxygen; but if there be an excess of carbon, it will decompose water, whilst an excess of hydrogen will decompose carbonic anhydride. Even if these phenomena and similar ones have been explained in isolated cases, a complete theory of the whole subject is still wanting in the present condition of chemical knowledge.