COMPOUNDS.] CHEMISTRY 497 by the formula K 4 I. 2 O 7 , and is obtained on dissolving the former in potassium hydroxide solution. The latter crystallizes with 9 molecules of water, which, however, it loses when exposed over sulphuric acid. Corresponding sodium salts exist. That which has the composition XaIO 4 crystallizes either in the anhydrous state, or with 3 molecules of water, which it readily loses in dry air ; the second salt, the formation of which was above described, separates in crystals of the composition Na 4 I 2 O 9 + 2H 2 O , and is only rendered anhydrous by heating to 220 C. Three silver salts are known, represented by the formulae AgI0 4 , Ag 4 I 2 O 9 + 3H 2 O , and Ag 5 IO 6 ; the second of these loses 2 molecules of water at 10"0 C., but the third only at 200 C. The barium salt Ba 2 I 2 9 separates in the anhydrous state from strongly acid solutions, but when prepared by precipitating a solution of the acid with barium hydroxide, or of an alkali salt with a barium salt, it con tains 6 or 7 molecules of water, which are only completely removed by heating to 300 C. It also furnishes a lead salt of the composition Pb 3 I 2 O 10 + 2H 2 O, and amongst other magnesium salts one of the composition Mg 4 I 2 O u , which crystallizes with 6 and with 9 molecules of water. The question now arises What is the nature of the relation between these various salts 1 By the chemical method alone it is extremely difficult, if not impossible, to decide, but from Thomsen s thermochemical investigation of the acid there can be little doubt as to the answer we should make to this question. When successive molecules of potassium hydroxide are added to a solution of 1 molecule of the acid H 5 IO the amounts of heat evolved are as follows : On the addition of the 1st molecule, 5,150 heat-units. 2d 21,440 3d 3,150 4th and 5th 2,300 The first and second molecules, therefore, together cause the development of 26,590 units of heat, or of 13,295 units per molecule. But we have seen (p. 486) that on neutralizing a large number of acids, between 13,750 and 13,150 units of heat are developed per molecule of hydroxide added ; and as the addition of further quantities of the hydroxide causes a comparatively slight development of heat there is little doubt from these results that the molecule H 5 IO 6 is dibasic. But on account of the existence of salts, such as K 4 I 9 , it appears desirable to double this formula, and to represent the molecule of periodic acid by the formula H 4 . 1 2 O 12 H 6 . Several of the salts above alluded to may be regarded as derived from this molecule by the partial or total displacement of the hydrogen by metals ; and we may term those which are formed by displacing 4 of the 10 atoms of hydrogen normal salts, while those in which 2, 4, or 6 of the remaining atoms of hydrogen are displaced may be called basic salts. Thus we have Normal salts. Xa 4 .I 2 12 H 6 ;A g4 .I 2 12 H 6 Basic salts. Pb 2 " . I 2 12 H 4 Pb"; Mg 2 " . I 2 12 H 2 Mg 2 "; Ag 4 . I 2 O 12 Ag a . That the tendency to form basic salts is slight in the case of the highly positive elements is also shown by the small amount of heat developed on the addition of the third, fourth, and fifth molecules of potassium hydroxide. The anhydrous salts such as K 4 I 2 O 9 are to be regarded as derived from a distinct acid formed from the molecule H 4 . 1 2 12 H G by the withdrawal of the elements of 3 molecules of water. The salts, such as KIO 4 , which on account of its isomorphism with potassium permanganate, K 2 Mn.,O s , is more probably represented by the formula K 2 I 2 O 8 , ate, it may be supposed, also derived from a distinct acid, formed in a similar manner by the with drawal of the elements of a fourth molecule of water; these salts have the same empirical composition as the per- chlorates, and as potassium perchlorate is isomorphous with the compound K 2 I 2 O 8 , it is probable that it corre sponds with it in composition, and that perchloric acid therefore is represented by the formula H 2 C1 2 O 8 . If this conclusion be correct, and chloric acid be correctly repre sented by the formula HC10 3 , we have an explanation of the great difference which is observed in the properties of these two acids. Perchloric acid, we have seen, has a great tendency to combine with water, and its hydrates may be_regarded as distinct acids, the liquid hydrate bear ing the same relation to the acid H 2 C1 2 O S that crystallized periodic acid bears to the hypothetical acid H 2 I 2 O 8 , from which salts such as K 2 I 2 O 8 are derived. Basic per- chlorates corresponding to the basic periodates are not known, but a cuprammonium salt and a lead salt have been obtained which apparently are derived from the crystalline hydrate of perchloric acid, the latter having the composition Pb"Cl 2 O 8 + PbH 2 O 2 or Pb"Cl 2 O 10 H 2 Pb" . It will have been noticed that the amount of heat deve loped on the addition of the first molecule of potassium hydroxide to the solution of periodic acid regarded as H 5 I0 6 is much less than is usually observed with other acids, and the amount developed on the addition of the second molecule much greater ; the two molecules together, however, produce an effect comparable with that observed in the case of other acids. But the salt produced on passing chlorine into a solution of potassium hydroxide has the composition K 2 I 2 O 8 , so that the first action of potassium hydroxide on a solution of the acid H 4 . 1 2 O 12 H 6 apparently does not merely consist in the direct displacement of hydrogen in this acid by potassium ; the elements of four molecules of water are also withdrawn, and since this latter operation involves the absorption of heat, the amount of heat finally developed in the reaction is but small. On the addition of a further quantity of hydroxide, however, change in the reverse order is effected, and hence the normal amount of heat is developed by the combined action of the two molecules of hydroxide. It is interesting also to observe that more heat is developed on adding the third than by the fourth and fifth molecules of hydroxide together, and to contrast this with the circumstance that the silver salt Ag 4 I 2 OpH 6 loses the elements of two molecules of water at 100 C., but the elements of a third only at 200 C. Apparently there is a tendency on the part of potassium hydroxide to enter into reaction svith the acid after the production of the normal salt K 4 I 2 O 12 H 6 ; but salts con taining a relatively larger proportion of potassium evidently cannot exist except in solution, and even then only to a limited extent. We are now in a position also to explain the formation of sodium periodate by the action of chlorine on a solution of sodium iodate and sodium hydroxide. We have learnt that the action of chlorine on the latter is to produce sodium hypochlorite, and that this is a powerful oxidizing agent NaOH + Cl a = NaOCl + HC1 , and there can be little doubt, therefore, that the iodate is at first oxidized by it, and converted into the salt Na s IO 8 ; thus Na 2 I 2 O 6 + 2N aOCl = Na 2 I 2 O 8 + 2NaCl . In presence of sodium hydrate and water, however, this salt is at once converted into the normal periodate Xa 2 I 2 8 + 2NaOH + 2H 2 = Na 4 I 2 12 H 6 The following are the results of Thomsen s thermochemi cal examination of iodic and periodic acids :
V. 63