CHAPTER XXI. Chromium, Molybdenum, Tungsten, Uranium, and Manganese

CHROMIUM, MOLYBDENUM, TUNGSTEN, URANIUM, AND MANGANESE

Sulphur, selenium, and tellurium belong to the uneven series of the sixth group. In the even series of this group there are known chromium, molybdenum, tungsten, and uranium; these give acid oxides of the type RO{3}, like SO{3}. Their acid properties are less sharply defined than those of sulphur, selenium, and tellurium, as is the case with all elements of the even series as compared with those of the uneven series in the same group. But still the oxides CrO{3}, MoO{3}, WO{3}, and even UO{3}, have clearly defined acid properties, and form salts of the composition MO,nRO{3} with bases MO. In the case of the heavy elements, and especially of uranium, the type of oxide, UO{3}, is less acid and more basic, because in the even series of oxides the element with the highest atomic weight always acquires a more and more pronounced basic character. Hence UO{3} shows the properties of a base, and gives salts UO{2}X{2}. The basic properties of chromium, molybdenum, tungsten, and uranium are most clearly expressed in the lower oxides, which they all form. Thus chromic oxide, Cr{2}O{3}, is as distinct a base as alumina, Al{2}O_{3}.

Of all these elements chromium is the most widely distributed and the most frequently used. It gives chromic anhydride, CrO{3}, and chromic oxide, Cr{2}O{3}--two compounds whose relative amounts of oxygen stand in the ratio 2 : 1. Chromium is, although somewhat rarely, met with in nature as a compound of one or the other type. The red chromium ore of the Urals, lead chromate or crocoisite PbCrO{4}, was the source in which chromium was discovered by Vauquelin, who gave it this name (from the Greek word signifying colour) owing to the brilliant colours of its compounds; the chromates (salts of chromic anhydride) are red and yellow, and the chromic salts (from Cr{2}O{3}) green and violet. The red lead chromate is, however, a rare chromium ore found only in the Urals and in a few other localities. Chromic oxide, Cr{2}O{3}, is more frequently met with. In small quantities it forms the colouring matter of many minerals and rocks--for example, of some serpentines. The commonest ore, and the chief source of the chromium compounds, is the chrome iron ore or chromite, which occurs in the Urals and Asia Minor, California, Australia, and other localities. This is magnetic iron ore, FeO,Fe{2}O{3}, in which the ferric oxide is replaced by chromic oxide, its composition being FeO,Cr{2}O{3}. Chrome iron ore crystallises in octahedra of sp. gr. 4·4; it has a feeble metallic lustre, is of a greyish-black colour, and gives a brown powder. It is very feebly acted on by acids, but when fused with potassium acid sulphate it gives a soluble mass, which contains a chromic salt, besides potassium sulphate and ferrous sulphate. In practice the treatment of chrome iron ore is mainly carried on for the preparation of chromates, and not of chromic salts, and therefore we will trace the history of the element by beginning with chromic acid, and especially with the working up of the chrome iron ore into potassium dichromate, K{2}Cr{2}O{7}, as the most common salt of this acid. It must be remarked that chromic anhydride, CrO{3}, is only obtained in an anhydrous state, and is distinguished for its capacity for easily giving anhydro-salts with the alkalis, containing one, two, and even three equivalents of the anhydride to one equivalent of base. Thus among the potassium salts there is known the normal or yellow chromate, K{2}CrO{4}, which corresponds to, and is perfectly isomorphous with, potassium sulphate, easily forms isomorphous mixtures with it, and is not therefore suitable for a process in which it is necessary to separate the salt from a mixture containing sulphates. As in the presence of a certain excess of acid, the dichromate, K{2}Cr{2}O{7} = 2K{2}CrO{4} + 2HX - 2KX - H{2}O, is easily formed from K{2}CrO{4}, the object of the manufacturer is to produce such a dichromate, the more so as it contains a larger proportion of the elements of chromic acid than the normal salt. Finely-ground chrome iron ore, when heated with an alkali, absorbs oxygen almost as easily (Chapter III., Note 7) as a mixture of the oxides of manganese with an alkali. This absorption is due to the presence of chromic oxide, which is oxidised into the anhydride, and then combines with the alkali Cr{2}O{3} + O{3} = 2CrO{3}. As the oxidation and formation of the chromate proceeds, the mass turns yellow. The iron is also oxidised, but does not give ferric acid, because the capacity of the chromium for oxidation is incomparably greater than that of the iron.

The working of the Ural chrome iron ore into chromium compounds has been firmly established in Russia, thanks to the endeavours of P. K. Ushakoff, who constructed large works for this purpose on the river Kama, near Elabougi, where as much as 2,000 tons of ore are treated yearly, owing to which the importation of chromium preparations into Russia has ceased.

A mixture of lime (sometimes with potash) and chrome iron ore is heated in a reverberatory furnace, with free access of air and at a red heat for several hours, until the mass becomes yellow; it then contains normal calcium chromate, CaCrO{4}, which is insoluble in water in the presence of an excess of lime.[1 bis] The resultant mass is ground up, and treated with water and sulphuric acid. The excess of lime forms gypsum, and the soluble calcium dichromate, CaCr{2}O{7}, together with a certain amount of iron, pass into solution. The solution is poured off, and chalk added to it; this precipitates the ferric oxide (the ferrous oxide is converted into ferric oxide in the furnace) and forms a fresh quantity of gypsum, while the chromic acid remains in solution--that is, it does not form the sparingly-soluble normal salt (1 part soluble in 240 parts of water). The solution then contains a fairly pure calcium dichromate, which by double decomposition gives other chromates; for example, with a solution of potassium sulphate it gives a precipitate of calcium sulphate and a solution of potassium dichromate, which crystallises when evaporated.

There are many variations in the details of the manufacturing processes, and these must be looked for in works on technical chemistry. But we may add that the chromate may also be obtained by slightly roasting briquettes of a mixture of chrome iron and lime, and then leaving the resultant mass to the action of moist air (oxygen is absorbed, and the mass turns yellow).

Potassium dichromate, K{2}Cr{2}O{7}, easily crystallises from acid solutions in red, well-formed prismatic crystals, which fuse at a red heat and evolve oxygen at a very high temperature, leaving chromic oxide and the normal salt, which undergoes no further change: 2K{2}Cr{2}O{7} = 2K{2}CrO{4} + Cr{2}O{3} + O{3}. At the ordinary temperature 100 parts of water dissolve 10 parts of this salt, and the solubility increases as the temperature rises. It is most important to note that the dichromate does not contain water, it is K{2}CrO{4} + CrO{3}; the acid salt corresponding to potassium acid sulphate, KHSO{4}, does not exist. It does not even evolve heat when dissolving in water, but on the contrary produces cold, i.e. it does not form a very stable compound with water. The solution and the salt itself are poisonous, and act as powerful oxidising agents, which is the character of chromic acid in general. When heated with sulphur or organic substances, with sulphurous anhydride, hydrogen sulphide, &c., this salt is deoxidised, yielding chromic compounds.[2 bis] Potassium dichromate is used in the arts and in chemistry as a source for the preparation of all other chromium compounds. It is converted into yellow pigments by means of double decomposition with salts of lead, barium, and zinc. When solutions of the salts of these metals are mixed with potassium dichromate (in dyeing generally mixed with soda, in order to obtain normal salts), they are precipitated as insoluble normal salts; for example, 2BaCl{2} + K{2}Cr{2}O{7} + H{2}O = 2BaCrO{4} + 2KCl + 2HCl. It follows from this that these salts are insoluble in dilute acids, but the precipitation is not complete (as it would be with the normal salt). The barium and zinc salts are of a lemon yellow colour; the lead salt has a still more intense colour passing into orange. Yellow cotton prints are dyed with this pigment. The silver salt, Ag{2}CrO_{4}, is of a bright red colour.

Ammonium and sodium dichromates are now also prepared on a large scale. The sodium salts may be prepared in exactly the same manner as those of potassium. The normal salt combines with ten equivalents of water, like Glauber's salt, with which it is isomorphous. Its solution above 30° deposits the anhydrous salt. Sodium dichromate crystals contain Na{2}Cr{2}O{7},2H{2}O. The ammonium salts of chromic acid are obtained by saturating the anhydride itself with ammonia. The dichromate is obtained by saturating one part of the anhydride with ammonia, and then adding a second part of anhydride and evaporating under the receiver of an air-pump. On ignition, the normal and acid salts leave chromic oxide. Potassium ammonium chromate, NH{4}KCrO{4}, is obtained in yellow needles from a solution of potassium dichromate in aqueous ammonia; it not only loses ammonia and becomes converted into potassium dichromate when ignited, but also by degrees at the ordinary temperature. This shows the feeble energy of chromic acid, and its tendency to form stable dichromates. Magnesium chromate is soluble in water, as also is the strontium salt. The calcium salt is also somewhat soluble, but the barium salt is almost insoluble. The isomorphism with sulphuric acid is shown in the chromates by the fact that the magnesium and ammonium salts form double salts containing six equivalents of water, which are perfectly isomorphous with the corresponding sulphates. The magnesium salt crystallises in large crystals containing seven equivalents of water. The beryllium, cerium, and cobalt salts are insoluble in water. Chromic acid dissolves manganous carbonate, but on evaporation the solution deposits manganese dioxide, formed at the expense of the oxygen of the chromic acid. Chromic acid also oxidises ferrous oxide, and ferric oxide is soluble in chromic acid.

One of the chromates most used by the dyer is the insoluble yellow lead chromate, PbCrO{4} (Chapter XVIII., Note 46), which is precipitated on mixing solutions of PbX{2} with soluble chromates. It easily forms a basic salt, having the composition PbO,PbCrO{4}, as a crystalline powder, obtained by fusing the normal salt with nitre and then rapidly washing in water. The same substance is obtained, although impure and in small quantity, by treating lead chromate with neutral potassium chromate, especially on boiling the mixture; and this gives the possibility of attaining, by means of these materials, various tints of lead chromate, from yellow to red, passing through different orange shades. The decomposition which takes place (incompletely) in this case is as follows: 2PbCrO{4} + K{2}CrO{4} = PbCrO{4},PbO + K{2}Cr{2}O{7}--that is, potassium dichromate is formed in solution.

When potassium dichromate is mixed with potassium hydroxide or carbonate (carbonic anhydride being disengaged in the latter case) it forms the normal salt, K{2}CrO{4}, known as yellow chromate of potassium. Its specific gravity is 2·7, being almost the same as that of the dichromate. It absorbs heat in dissolving; one part of the salt dissolves in 1·75 part of water at the ordinary temperature, forming a yellow solution. When mixed even with such feeble acids as acetic, and more especially with the ordinary acids, it gives the dichromate, and Graham obtained a trichromate, K{2}Cr{3}O{10} = K{2}CrO{4},2CrO{3}, by mixing a solution of the latter salt with an excess of nitric acid.

Chromic anhydride is obtained by preparing a saturated solution of potassium dichromate at the ordinary temperature, and pouring it in a thin stream into an equal volume of pure sulphuric acid. On mixing, the temperature naturally rises; when slowly cooled, the solution deposits chromic anhydride in needle-shaped crystals of a red colour sometimes several centimetres long. The crystals are freed from the mother liquor by placing them on a porous tile.[4 bis] It is very important at this point to call attention to the fact that a hydrate of chromic anhydride is never obtained in the decomposition of chromic compounds, but always the anhydride, CrO{3}. The corresponding hydrate, CrO{4}H{2}, or any other hydrate, is not even known. Nevertheless, it must be admitted that chromic acid is bibasic, because it forms salts isomorphous or perfectly analogous with the salts formed by sulphuric acid, which is the best example of a bibasic acid. A clear proof of the bibasicity of CrO{3} is seen in the fact that the anhydride and salts give (when heated with sodium chloride and sulphuric acid) a volatile chloranhydride, CrO{2}Cl{2}, containing two atoms of chlorine as a bibasic acid should. Chromic anhydride is a red crystalline substance, which is converted into a black mass by heat; it fuses at 190°, and disengages oxygen above 250°, leaving a residue of chromium dioxide, CrO{2}, and, on still further heating, chromic oxide, Cr{2}O_{3}. Chromic anhydride is exceedingly soluble in water, and even attracts moisture from the air, but, as was mentioned above, it does not form any definite compound with water. The specific gravity of its crystals is 2·7, and when fused it has a specific gravity 2·6. The solution presents perfectly defined acid properties. It liberates carbonic anhydride from carbonates; gives insoluble precipitates of the chromates with salts of barium, lead, silver, and mercury.

The sulphuric acid should not contain any lower oxides of nitrogen, because they reduce chromic anhydride into chromic oxide. If a solution of a chromate be heated with an excess of acid--for instance, sulphuric or hydrochloric acid--oxygen or chlorine is evolved, and a solution of a chromic salt is formed. Hence, under these circumstances, chromic acid cannot be obtained from its salts. One of the first methods employed consisted in converting its salts into volatile chromium hexafluoride, CrF{6}. This compound, obtained by Unverdorben, may be prepared by mixing lead chromate with fluor spar in a dry state, and treating the mixture with fuming sulphuric acid in a platinum vessel: PbCrO{4} + 3CaF{2} + 4H{2}SO{4} = PbSO{4} + 3CaSO{4} + 4H{2}O + CrF{6}. Fuming sulphuric acid is taken, and in considerable excess, because the chromium fluoride which is formed is very easily decomposed by water. It is volatile, and forms a very caustic, poisonous vapour, which condenses when cooled in a dry platinum vessel into a red, exceedingly volatile liquid, which fumes powerfully in air. The vapours of this substance when introduced into water are decomposed into hydrofluoric acid and chromic anhydride: CrF{6} + 3H{2}O = CrO{3} + 6HF. If very little water be taken the hydrofluoric acid volatilises, and chromic anhydride separates directly in crystals. The chloranhydride of chromic acid, CrO{2}Cl{2} (Note 5), is also decomposed in the same manner. A solution of chromic acid and a precipitate of barium sulphate are formed by treating the insoluble barium chromate with an equivalent quantity of sulphuric acid. If carefully evaporated, the solution yields crystals of chromic anhydride. Fritzsche gave a very convenient method of preparing chromic anhydride, based on the relation of chromic to sulphuric acid. At the ordinary temperature the strong acid dissolves both chromic anhydride and potassium chromate, but if a certain amount of water is added to the solution the chromic anhydride separates, and if the amount of water be increased the precipitated chromic anhydride is again dissolved. The chromic anhydride is almost all separated from the solution when it contains two equivalents of water to one equivalent of sulphuric acid. Many methods for the preparation of chromic anhydride are based on this fact.

Berzelius observed, and Rose carefully investigated, this remarkable reaction, which occurs between chromic acid and sodium chloride in the presence of sulphuric acid. If 10 parts of common salt be mixed with 12 parts of potassium dichromate, fused, cooled, and broken up into lumps, and placed in a retort with 20 parts of fuming sulphuric acid, it gives rise to a violent reaction, accompanied by the formation of brown fumes of chromic chloranhydride, or chromyl chloride, CrO{2}Cl{2}, according to the reaction: CrO{3} + 2NaCl + H{2}SO{4} = Na{2}SO{4} + H{2}O + CrO{2}Cl{2}. The addition of an excess of sulphuric acid is necessary in order to retain the water. The same substance is always formed when a metallic chloride is heated with chromic acid, or any of its salts, in the presence of sulphuric acid. The formation of this volatile substance is easily observed from the brown colour which is proper to its vapour. On condensing the vapour in a dry receiver a liquid is obtained having a sp. gr. of 1·9, boiling at 118°, and giving a vapour whose density, compared with hydrogen, is 78, which corresponds with the above formula. Chromyl chloride is decomposed by heat into chromic oxide, oxygen, and chlorine: 2CrO{2}Cl{2} = Cr{2}O{3} + 2Cl{2} + O; so that it is able to act simultaneously as a powerful oxidising and chlorinating agent, which is taken advantage of in the investigation of many, and especially of organic, substances. When treated with water, this substance first falls to the bottom, and is then decomposed into hydrochloric and chromic acids, like all chloranhydrides: CrO{2}Cl{2} + H{2}O = CrO{3} + 2HCl. When brought into contact with inflammable substances it sets fire to them; it acts thus, for instance, on phosphorus, sulphur, oil of turpentine, ammonia, hydrogen, and other substances. It attracts moisture from the atmosphere with great energy, and must therefore be kept in closed vessels. It dissolves iodine and chlorine, and even forms a solid compound with the latter, which depends upon the faculty of chromium to form its higher oxide, Cr{2}O{7}. The close analogy in the physical properties of the chloranhydrides, CrO{2}Cl{2} and SO{2}Cl{2}, is very remarkable, although sulphurous anhydride is a gas, and the corresponding oxide, CrO{2}, is a non-volatile solid. It may be imagined, therefore, that chromium dioxide (which will be mentioned in the following note) presents a polymerised modification of the substance having the composition CrO_{2}; in fact, this is obvious from the method of its formation.

If three parts of potassium dichromate be mixed with four parts of strong hydrochloric acid and a small quantity of water, and gently warmed, it all passes into solution, and no chlorine is evolved; on cooling, the liquid deposits red prismatic crystals, known as Peligot's salt, very stable in air. This has the composition KCl,CrO{3}, and is formed according to the equation K{2}Cr{2}O{7} + 2HCl = 2KCl,CrO{3} + H{2}O. It is evident that this is the first chloranhydride of chromic acid, HCrO{3}Cl, in which the hydrogen is replaced by potassium. It is decomposed by water, and on evaporation the solution yields potassium dichromate and hydrochloric acid. This is a fresh instance of the reversible reactions so frequently encountered. With sulphuric acid Peligot's salt forms chromyl chloride. The latter circumstance, and the fact that Geuther produced Peligot's salt from potassium chromate and chromyl chloride, give reason for thinking that it is a compound of these two substances: 2KCl,CrO{3} = K{2}CrO{4} + CrO{2}Cl{2}. It is also sometimes regarded as potassium dichromate in which one atom of oxygen is replaced by chlorine--that is, K{2}Cr{2}O{6}Cl{2}, corresponding with K{2}Cr{2}O_{7}. When heated it parts with all its chlorine, and on further heating gives chromic oxide.

This intermediate degree of oxidation, CrO{2}, may also be obtained by mixing solutions of chromic salts with solutions of chromates. The brown precipitate formed contains a compound, Cr{2}O{3},CrO{3}, consisting of equivalent amounts of chromic oxide and anhydride. The brown precipitate of chromium dioxide contains water. The same substance is formed by the imperfect deoxidation of chromic anhydride by various reducing agents. Chromic oxide, when heated, absorbs oxygen, and appears to give the same substance. Chromic nitrate, when ignited, also gives this substance. When this substance is heated it first disengages water and then oxygen, chromic oxide being left. It corresponds with manganese dioxide, Cr{2}O{3},CrO{3} = 3CrO{2}. Krüger treated chromium dioxide with a mixture of sodium chloride and sulphuric acid, and found that chlorine gas was evolved, but that chromyl chloride was not formed. Under the action of light, a solution of chromic acid also deposits the brown dioxide. At the ordinary temperature chromic anhydride leaves a brown stain upon the skin and tissues, which probably proceeds from a decomposition of the same kind. Chromic anhydride is soluble in alcohol containing water, and this solution is decomposed in a similar manner by light. Chromium dioxide forms K{2}CrO{4} when treated with H{2}O{2} in the presence of KHO.

The action of hydrogen peroxide on a solution of chromic acid or of potassium dichromate gives a blue solution, which very quickly becomes colourless with the disengagement of oxygen. Barreswill showed that this is due to the formation of a perchromic anhydride, Cr{2}O{7}, corresponding with sulphur peroxide. This peroxide is remarkable from the fact that it very easily dissolves in ether and is much more stable in this solution, so that, by shaking up hydrogen peroxide mixed with a small quantity of chromic acid, with ether, it is possible to transfer all the blue substance formed to the ether.[6 bis]

With oxygen acids, chromic acid evolves oxygen; for example, with sulphuric acid the following reaction takes place: 2CrO{3} + 3H{2}SO{4} = Cr{2}(SO{4}){3} + O{3} + 3H{2}O. It will be readily understood from this that a mixture of chromic acid or of its salts with sulphuric acid forms an excellent oxidising agent, which is frequently employed in chemical laboratories and even for technical purposes as a means of oxidation. Thus hydrogen sulphide and sulphurous anhydride are converted into sulphuric acid by this means. Chromic acid is able to act as a powerful oxidising agent because it passes into chromic oxide, and in so doing disengages half of the oxygen contained in it: 2CrO{3} = Cr{2}O{3} + O{3}. Thus chromic anhydride itself is a powerful oxidising agent, and is therefore employed instead of nitric acid in galvanic batteries (as a depolariser), the hydrogen evolved at the carbon being then oxidised, and the chromic acid converted into a non-volatile product of deoxidation, instead of yielding, as nitric acid does, volatile lower oxides of offensive odour. Organic substances are more or less perfectly oxidised by means of chromic anhydride, although this generally requires the aid of heat, and does not proceed in the presence of alkalis, but generally in the presence of acids. In acting on a solution of potassium iodide, chromic acid, like many oxidising agents, liberates iodine; the reaction proceeds in proportion to the amount of CrO{3} present, and may serve for determining the amount of CrO{3}, since the amount of iodine liberated can be accurately determined by the iodometric method (Chapter XX., Note 42). If chromic anhydride be ignited in a stream of ammonia, it gives chromic oxide, water, and nitrogen. In all cases when chromic acid acts as an oxidising agent in the presence of acids and under the action of heat, the product of its deoxidation is a chromic salt, CrX{3}, which is characterised by the green colour of its solution, so that the red or yellow solution of a salt of chromic acid is then transformed into a green solution of a chromic salt, derived from chromic oxide, Cr{2}O{3}, which is closely analogous to Al{2}O{3}, Fe{2}O{3}, and other bases of the composition R{2}O_{3}. This analogy is seen in the insolubility of the anhydrous oxide, in the gelatinous form of the colloidal hydrate, in the formation of alums, of a volatile chloride of chromium, &c.[7 bis]

As a mixture of potassium dichromate and sulphuric acid is usually employed for oxidation, the resultant solution generally contains a double sulphate of potassium and chromium--that is, chrome alum, isomorphous with ordinary alum--K{2}Cr{2}O{7} + 4H{2}SO{4} + 20H{2}O = O{3} + K{2}Cr{2}(SO{4}){4},24H{2}O or 2(KCr(SO{4}){2},12H{2}O). It is prepared by dissolving potassium dichromate in dilute sulphuric acid; alcohol is then added and the solution slightly heated, or sulphurous anhydride is passed through it. On the addition of alcohol to a cold mixture of potassium dichromate and sulphuric acid, the gradual disengagement of pleasant-smelling volatile products of the oxidation of alcohol, and especially of aldehyde, C{2}H{4}O, is remarked. If the temperature of decomposition does not exceed 35°, a violet solution of chrome alum is obtained, but if the temperature be higher, a solution of the same alum is obtained of a green colour. As chrome alum requires for solution 7 parts of water at the ordinary temperature, it follows that if a somewhat strong solution of potassium dichromate be taken (4 parts of water and 1-1/2 of sulphuric acid to 1 part of dichromate), it will give so concentrated a solution of chrome alum that on cooling, the salt will separate without further evaporation. If the liquid, prepared as above or in any instance of the deoxidation of chromic acid, be heated (the oxidation naturally proceeds more rapidly) somewhat strongly, for instance, to the boiling-point of water, or if the violet solution already formed be raised to the same temperature, it acquires a bright green colour, and on evaporation the same mixture, which at lower temperatures so easily gives cubical crystals of chrome alum, does not give any crystals whatever. If the green solution be kept, however, for several weeks at the ordinary temperature, it deposits violet crystals_ of chrome alum. The green solution, when evaporated, gives a non-crystalline mass, and the violet crystals lose water at 100° and turn green. It must be remarked that the transition of the green modification into the violet is accompanied by a decrease in volume (Lecoq de Boisbaudran, Favre). If the green mass formed at the higher temperature be evaporated to dryness and heated at 30° in a current of air, it does not retain more then 6 equivalents of water. Hence Löwel, and also Schrötter, concluded that the green and violet modifications of the alum depend on different degrees of combination with water, which may be likened to the different compounds of sodium sulphate with water and to the different hydrates of ferric oxide.

However, the question in this case is not so simple, as we shall afterwards see. Not chrome alum alone, but all the chromic salts, give two, if not three, varieties. At least, there is no doubt about the existence of two--a green and a violet modification. The green chromic salts are obtained by heating solutions of the violet salts, the violet solutions are produced on keeping solutions of the green salts for a long time. The conversion of the violet salts into green by the action of heat is itself an indication of the possibility of explaining the different modifications by their containing different proportions (or states) of water, and, moreover, by the green salts having a less amount of water than the violet. However, there are other explanations. Chromic oxide is a base like alumina, and is therefore able to give both acid and basic salts. It is supposed that the difference between the green and violet salts is due to this fact. This opinion of Krüger is based on the fact that alcohol separates out a salt from the green solution which contains less sulphuric acid than the normal violet salt. On the other hand, Löwel showed that all the acid cannot be separated from the green chromic salts by suitable reagents, as easily as it can be from the same solution of the violet salts; thus barium salts do not precipitate all the sulphuric acid from solutions of the green salts. According to other researches the cause of the varieties of the chromic salts lies in a difference in the bases they contain--that is, it is connected with a modification of the properties of the oxide of chromium itself. This only refers to the hydroxides, but as hydroxides themselves are only special forms of salts, the differences observed as yet in this direction between the hydroxides only confirm the generality of the difference observed in the chromic compounds (see Note 7 bis).

The salts of chromic oxide, like those of alumina, are easily decomposed, give basic and double salts, and have an acid reaction, as chromic oxide is a feeble base. Potassium and sodium hydroxides give a precipitate of the hydroxide with chromic salts, CrX{3}. The violet and green salts give a hydroxide soluble in an excess of the reagent; but the hydroxide is held in solution by very feeble affinities, so that it is partially separated by heat and dilution with water, and completely so on boiling. In an alkaline solution, chromic hydroxide is easily converted into chromic acid by the action of lead dioxide, chlorine, and other oxidising agents. If the chromic oxide occurs together with such oxides as magnesia, or zinc oxide, then on precipitation it separates out from its solution in combination with these oxides, forming, for example, ZnO,Cr{2}O{3}. Viard obtained compounds of Cr{2}O{3} with the oxides of Mg, Zn, Cd, &c.) On precipitating the violet solution of chrome alum with ammonia, a precipitate containing Cr{2}O{3},6H{2}O is obtained, whilst the precipitate from the boiling solution with caustic potash was a hydrate containing four equivalents of water. When fused with borax chromic salts give a green glass. The same coloration is communicated to ordinary glass by the presence of traces of chromic oxide. A chrome glass containing a large amount of chromic oxide may be ground up and used as a green pigment. Among the hydrates of oxide of chromium Guignet's green forms one of the widely-used green pigments which have been substituted for the poisonous arsenical copper pigments, such as Schweinfurt green, which formerly was much used. Guignet's green has an extremely bright green colour, and is distinguished for its great stability, not only under the action of light but also towards reagents; thus it is not altered by alkaline solutions, and even nitric acid does not act on it. This pigment remains unchanged up to a temperature of 250°; it contains Cr{2}O{3},2H_{2}O, and generally a small amount of alkali. It is prepared by fusing 3 parts of boric acid with 1 part of potassium dichromate; oxygen is disengaged, and a green glass, containing a mixture of the borates of chromium and potassium, is obtained. When cool this glass is ground up and treated with water, which extracts the boric acid and alkali and leaves the above-named chromic hydroxide behind. This hydroxide only parts with its water at a red heat, leaving the anhydrous oxide.

The chromic hydroxides lose their water by ignition, and in so doing become spontaneously incandescent, like the ordinary ferric hydroxide (Chapter XXII.). It is not known, however, whether all the modifications of chromic oxide show this phenomenon. The anhydrous chromic oxide, Cr{2}O{3}, is exceedingly difficultly soluble in acids, if it has passed through the above recalescence. But if it has parted with its water, or the greater part of it, and not yet undergone this self-induced incandescence (has not lost a portion of its energy), then it is soluble in acids. It is not reduced by hydrogen. It is easily obtained in various crystalline forms by many methods. The chromates of mercury and ammonium give a very convenient method for its preparation, because when ignited they leave chromic oxide behind. In the first instance oxygen and mercury are disengaged, and in the second case nitrogen and water: 2Hg{2}CrO{4} = Cr{2}O{3} + O{5} + 4Hg or (NH{4}){2}Cr{2}O{7} = Cr{2}O{3} + 4H{2}O + N{2}. The second reaction is very energetic, and the mass of salt burns spontaneously if the temperature be sufficiently high. A mixture of potassium sulphate and chromic oxide is formed by heating potassium dichromate with an equal weight of sulphur: K{2}Cr{2}O{7} + S = K{2}SO{4} + Cr{2}O{3}. The sulphate is easily extracted by water, and there remains a bright green residue of the oxide, whose colour is more brilliant the lower the temperature of the decomposition. The oxide thus obtained is used as a green pigment for china and enamel. The anhydrous chromic oxide obtained from chromyl chloride, CrO{2}Cl{2}, has a specific gravity of 5·21, and forms almost black crystals, which give a green powder. They are hard enough to scratch glass, and have a metallic lustre. The crystalline form of chromic oxide is identical with that of the oxide of iron and alumina, with which it is isomorphous.

The blue solution of chromous chloride may also be obtained by the action of metallic zinc on the green solution of the hydrated chromic chloride; the zinc in this case takes up chlorine just as the hydrogen did. It must be employed in a large excess. Chromic oxide is also formed in the action of zinc on chromic chloride, and if the solution remain for a long time in contact with the zinc the whole of the chromium is converted into chromic oxychloride. Other chromic salts are also reduced by zinc into chromous salts, CrX{2}, just as the ferric salts FeX{3} are converted into ferrous salts FeX{2} by it. The chromous salts are exceedingly unstable and easily oxidise and pass into chromic salts; hence the reducing power of these salts is very great. From cupric salts they separate cuprous salts, from stannous salts they precipitate metallic tin, they reduce mercuric salts into mercurous and ferric into ferrous salts. Moreover, they absorb oxygen from the air directly. With potassium chromate they give a brown precipitate of chromium dioxide or of chromic oxide, according to the relative amounts of the substances taken: CrO{3} + CrO = 2CrO{2} or CrO{3} + 3CrO = 2Cr{2}O{3}. Aqueous ammonia gives a blue precipitate, and in the presence of ammoniacal salts a blue liquid is obtained which turns red in the air from oxidation. This is accompanied by the formation of compounds analogous to those given by cobalt (Chapter XXII.). A solution of chromous chloride with a hot saturated solution of sodium acetate, C{2}H{3}NaO{2}, gives, on cooling, transparent red crystals of chromous acetate, C{4}H{6}CrO{4},H_{2}O. This salt is also a powerful reducing agent, but may be kept for a long time in a vessel full of carbonic anhydride.

The insoluble anhydrous chromic chloride CrCl{3} very easily passes into solution in the presence of a trace (0·004) of chromous chloride CrCl{2}. This remarkable phenomenon was observed by Peligot and explained by Löwel in the following manner: chromous chloride, as a lower stage of oxidation, is capable of absorbing both oxygen and chlorine, combining with various substances. It is able to decompose many chlorides by taking up chlorine from them; thus it precipitates mercurous chloride from a solution of mercuric chloride, and in so doing passes into chromic chloride: 2CrCl{2} + 2HgCl{2} = Cr{2}Cl{6} + 2HgCl. Let us suppose that the same phenomenon takes place when the anhydrous chromic chloride is mixed with a solution of chromous chloride. The latter will then take up a portion of the chlorine of the former, and pass into a soluble hydrate of chromic chloride (hydrochloride of oxide of chromium), and the original anhydrous chromic chloride will pass into chromous chloride. The chromous chloride re-formed in this manner will then act on a fresh quantity of the chromic chloride, and in this manner transfer it entirely into solution as hydrate. This view is confirmed by the fact that other chlorides, capable of absorbing chlorine like chromous chloride, also induce the solution of the insoluble chromic chloride--for example, ferrous chloride, FeCl{2}, and cuprous chloride. The presence of zinc also aids the solution of chromic chloride, owing to its converting a portion of it into chromous chloride. The solution of chromic chloride in water obtained by these methods is perfectly identical with that which is formed by dissolving chromic hydroxide in hydrochloric acid. On evaporating the green solution obtained in this manner, it gives a green mass, containing water. On further heating it leaves a soluble chromic oxychloride, and when ignited it first forms an insoluble oxychloride and then chromic oxide; but no anhydrous chromic chloride, Cr{2}Cl{6}, is formed by heating the aqueous solution of chromic chloride, which forms an important fact in support of the view that the green solution of chromic chloride is nothing else but hydrochloride of oxide of chromium. At 100° the composition of the green hydrate is Cr{2}Cl{6},9H{2}O, and on evaporation at the ordinary temperature over H{2}SO{4} crystals are obtained with 12 equivalents of water; the red mass obtained at 120° contains Cr{2}O{3},4Cr{2}Cl{6},24H{2}O. The greater portion of it is soluble in water, like the mass which is formed at 150°. The latter contains Cr{2}O{3},2Cr{2}Cl{6},9H{2}O = 3(Cr{2}OCl{4},3H{2}O)--that is, it presents the same composition as chromic chloride in which one atom of oxygen replaces two of chlorine. And if the hydrate of chromic chloride be regarded as Cr{2}O{3},6HCl, the substance which is obtained should be regarded as Cr{2}O{3},4HCl combined with water, H{2}O. The addition of alkalis--for example, baryta--to a solution of chromic chloride immediately produces a precipitate, which, however, re-dissolves on shaking, owing to the formation of one of the oxychlorides just mentioned, which may be regarded as basic salts. Thus we may represent the product of the change produced on chromic chloride under the influence of water and heat by the following formulæ: first Cr{2}O{3},6HCl or Cr{2}Cl{6},3H{2}O is formed, then Cr{2}O{3},4HCl,H{2}O or Cr{2}OCl{4},3H{2}O, and lastly Cr{2}O{3},2HCl,2H{2}O or Cr{2}O{2}Cl{2},3H{2}O. In all three cases there are 2 equivalents of chromium to at least 3 equivalents of water. These compounds may be regarded as being intermediate between chromic hydroxide and chloride; chromic chloride is Cr{2}Cl{6}, the first oxychloride Cr{2}(OH){2}Cl{4}, the second Cr{2}(OH){4}Cl{2}, and the hydrate Cr{2}(OH){6}--that is, the chlorine is replaced by hydroxyl.

It is very important to remark two circumstances in respect to this: (1) That the whole of the chlorine in the above compounds is not precipitated from their solutions by silver nitrate; thus the normal salt of the composition Cr{2}Cl{6},9H{2}O only gives up two-thirds of its chlorine; therefore Peligot supposes that the normal salt contains the oxychloride combined with hydrochloric acid: Cr{2}Cl{6} + 2H{2}O = Cr{2}O{2}Cl{2},4HCl, and that the chlorine held as hydrochloric acid reacts with the silver, whilst that held in the oxychloride does not enter into reaction, just as we observe a very feebly-developed faculty for reaction in the anhydrous chromic chloride; and (2) if the green aqueous solution of CrCl{3} be left to stand for some time, it ultimately turns violet; in this form the whole of the chlorine is precipitated by AgNO_{3}, whilst boiling re-converts it into the green variety. Löwel obtained the violet solution of hydrochloride of chromic oxide by decomposing the violet chromic sulphate with barium chloride. Silver nitrate precipitates all the chlorine from this violet modification; but if the violet solution be boiled and so converted into the green modification, silver nitrate then only precipitates a portion of the chlorine.

Recoura (1890-1893) obtained a crystallohydrate of violet chromium sulphate, Cr{2}(SO{4}){3}, with 18 or 15 H{2}O. By boiling a solution of this crystallohydrate, he converted it into the green salt, which, when treated with alkalis, gave a precipitate of Cr{2}O{3},2H{2}O, soluble in 2H{2}SO{4} (and not 3), and only forming the basic salt, Cr{2}(OH){2}(SO{4}){2}. He therefore concludes that the green salts are basic salts. The cryoscopic determinations made by A. Speransky (1892) and Marchetti (1892) give a greater 'depression' for the violet than the green salts, that is, indicate a greater molecular weight for the green salts. But as Étard, by heating the violet sulphate to 100°, converted it into a green salt of the same composition, but with a smaller amount of H{2}O, it follows that the formation of a basic salt alone is insufficient to explain the difference between the green and violet varieties, and this is also shown by the fact that BaCl_{2} precipitates the whole of the sulphuric acid of the violet salt, and only a portion of that of the green salt. A. Speransky also showed that the molecular electro-conductivity of the green solutions is less than that of the violet. It is also known that the passage of the former into the latter is accompanied by an increase of volume, and, according to Recoura, by an evolution of heat also.

Piccini's researches (1894) throw an important light upon the peculiarities of the green chromium trichloride (or chromic chloride); he showed (1) that AgF (in contradistinction to the other salts of silver) precipitates all the chlorine from an aqueous solution of the green variety; (2) that solutions of green CrCl{3},6H{2}O in ethyl alcohol and acetone precipitate all their chlorine when mixed with a similar solution of AgNO{3}; (3) that the rise of the boiling-point of the ethyl alcohol and acetone green solutions of CrCl{3},6H{2}O (Chapter VII., Note 27 bis) shows that i in this case (as in the aqueous solutions of MgSO{4} and HgCl{2}) is nearly equal to 1, that is, that they are like solutions of non-conductors; (4) that a solution of green CrCl{3} in methyl alcohol at first precipitates about 7/8 of its chlorine (an aqueous solution about 2/3) when treated with AgNO{3}, but after a time the whole of the chlorine is precipitated; and (5) that an aqueous solution of the green variety gradually passes into the violet, while a methyl alcoholic solution preserves its green colour, both of itself and also after the whole of the chlorine has been precipitated by AgNO{3}. If, however, in an aqueous or methyl alcoholic solution only a portion of the chlorine be precipitated, the solution gradually turns violet. In my opinion the general meaning of all these observations requires further elucidation and explanation, which should be in harmony with the theory of solutions. Recoura, moreover, obtained compounds of the green salt, Cr{2}(SO{4}){3}, with 1, 2, and 3 molecules of H{2}SO{4}, K{2}SO{4}, and even a compound Cr{2}(SO{4}){3}H{2}CrO{4}. By neutralising the sulphuric acid of the compounds of Cr{2}(SO{4}){3} and H{2}SO{4} with caustic soda, Recoura obtained an evolution of 33 thousand calories per each 2NaHO, while free H{2}SO{4} only gives 30·8 thousand calories. Recoura is of opinion that special chromo sulphuric acids, for instance (CrSO{4})H{2}SO{4} = 1/2Cr{2}(SO{4}){3}H{2}SO_{4}, are formed. With a still larger excess of sulphuric acid, Recoura obtained salts containing a still greater number of sulphuric acid radicles, but even this method does not explain the difference between the green and violet salts.

These facts must naturally be taken into consideration in order to arrive at any complete decision as to the cause of the different modifications of the chromic salts. We may observe that the green modification of chromic chloride does not give double salts with the metallic chlorides, whilst the violet variety forms compounds Cr{2}Cl{6},2RCl (where R = an alkali metal), which are obtained by heating the chromates with an excess of hydrochloric acid and evaporating the solution until it acquires a violet colour. As the result of all the existing researches on the green and violet chromic salts, it appears to me most probable that their difference is determined by the feeble basic character of chromic oxide, by its faculty of giving basic salts, and by the colloidal properties of its hydroxide (these three properties are mutually connected), and moreover, it seems to me that the relation between the green and violet salts of chromic oxide best answers to the relation of the purpureo to the luteo cobaltic salts (Chapter XXII., Note 35). This subject cannot yet be considered as exhausted (see Note 7).

We may here observe that with tin the chromic salts, CrX{3}, give at low temperatures CrX{2} and SnX{2}, whilst at high temperatures, on the contrary, CrX{2} reduces the metal from its salts SnX_{2}. The reaction, therefore, belongs to the class of reversible reactions (Beketoff).

Poulenc obtained anhydrous CrF{3} (sp. gr. 3·78) and CrF{2} (sp. gr. 4·11) by the action of gaseous HF upon CrCl{2}. A solution of fluoride of chromium is employed as a mordant in dyeing. Recoura (1890) obtained green and violet varieties of Cr{2}Br{6},6H{2}O. The green variety can only be kept in the presence of an excess of HBr in the solution; if alone its solution easily passes into the violet variety with evolution of heat.

Chromic oxide, Cr{2}O{3}, rarely found, and in small quantities, in chrome ochre, is formed by the oxidation of chromium and its lower oxides, by the reduction of chromates (for example, of ammonium or mercuric chromate) and by the decomposition (splitting up) of the saline compounds of the oxide itself, CrX{3} or Cr{2}X_{6}, like alumina, which it resembles in forming a feeble base easily giving double and basic salts, which are either green or violet.

The reduction of chromic oxide--for instance, in a solution by zinc and sulphuric acid--leads to the formation of chromous oxide, CrO, and its salts, CrX{2}, of a blue colour (see Notes 7 and 7 ^{bis}). The further reduction of oxide of chromium and its corresponding compounds gives metallic chromium. Deville obtained it (probably containing carbon) by reducing chromic oxide with carbon, at a temperature near the melting point of platinum, about 1750°, but the metal itself does not fuse at this temperature. Chromium has a steel-grey colour and is very hard (sp. gr. 5·9), takes a good polish, and dissolves in hydrochloric acid, but cold dilute sulphuric and nitric acids have no action upon it. Bunsen obtained metallic chromium by decomposing a solution of chromic chloride, Cr{2}Cl{6}, by a galvanic current, as scales of a grey colour (sp. gr. 7·3). Wöhler obtained crystalline chromium by igniting a mixture of the anhydrous chromic chloride Cr{2}Cl{6} (see_ Note 7 bis) with finely-divided zinc, and sodium and potassium chlorides, at the boiling-point of zinc. When the resultant mass has cooled the zinc may be dissolved in dilute nitric acid, and grey crystalline chromium (sp. gr. 6·81) is left behind. Frémy also prepared crystalline chromium by the action of the vapour of sodium on anhydrous chromic chloride in a stream of hydrogen, using the apparatus shown in the accompanying drawing, and placing the sodium and the chromic chloride in separate porcelain boats. The tube containing these boats is only heated when it is quite full of dry hydrogen. The crystals of metallic chromium obtained in the tube are grey cubes having a considerable hardness and withstanding the action of powerful acids, and even of aqua regia. The chromium obtained by Wöhler by the action of a galvanic current is, on the contrary, acted on under these conditions. The reason of this difference must be looked for in the presence of impurities, and in the crystalline structure. But in any case, among the properties of metallic chromium, the following may be considered established: it is white in colour, with a specific gravity of about 6·7, is extremely hard in a crystalline form, is not oxidised by air at the ordinary temperature, and with carbon it forms alloys like cast iron and steel.

The reduction of metallic chromium proceeds with comparative ease in aqueous solutions. Thus the action of sodium amalgams upon a strong solution of Cr{2}Cl{6} gives (first CrCl{2}) an amalgam of chromium from which the mercury may be easily driven off by heating (in hydrogen to avoid oxidation), and there remains a spongy mass of easily oxidizable chromium. Plaset (1891), by passing an electric current through a solution of chrome alum mixed with a small amount of H{2}SO{4} and K{2}SO{4}, obtained hard scales of chromium of a bluish-white colour possessing great hardness and stability (under the action of water, air, and acids). Glatzel (1890) reduced a mixture of 2KCl + Cr{2}Cl_{6} by heating it to redness with shavings of magnesium. The metallic chromium thus obtained has the appearance of a fine light-grey powder which is seen to be crystalline under the microscope; its sp. gr. at 16° is 6·7284. It fuses (with anhydrous borax) only at the highest temperatures, and after fusion presents a silver-white fracture. The strongest magnet has no action upon it.

Moissan (1893) obtained chromium by reducing the oxide Cr{2}O{3} with carbon in the electrical furnace (Chapter VIII., Note 17) in 9-10 minutes with a current of 350 ampères and 50 volts. The mixture of oxide and carbon gives a bright ingot weighing 100-110 grams. A current of 100 ampères and 50 volts completes the experiment upon a smaller quantity of material in 15 minutes; a current of 30 ampères and 50 volts gave an ingot of 10 grams in 30-40 minutes. The resultant carbon alloy is more or less rich in chromium (from 87·37-91·7 p.c.). To obtain the metal free from carbon, the alloy is broken into large lumps, mixed with oxide of chromium, put into a crucible and covered with a layer of oxide. This mixture is then heated in the electric furnace and the pure metal is obtained. This reduction can also be carried on with chrome iron ore FeOCr{2}O{3} which occurs in nature. In this case a homogeneous alloy of iron and chromium is obtained. If this alloy be thrown in lumps into molten nitre, it forms insoluble sesquioxide of iron and a soluble alkaline chromate. This alloy of iron and chromium dissolved in molten steel (chrome steel) renders it hard and tough, so that such steel has many valuable applications. The alloy, containing about 3 p.c. Cr and about 1·3 p.c. carbon, is even harder than the ordinary kinds of tempered steel and has a fine granular fracture. The usual mode of preparing the ferrochromes for adding to steel is by fusing powdered chrome iron ore under fluxes in a graphite crucible.

The two analogues of chromium, molybdenum and tungsten (or wolfram), are of still rarer occurrence in nature, and form acid oxides, RO{3}, which are still less energetic than CrO{3}. Tungsten occurs in the somewhat rare minerals, scheelite, CaWO{4}, and wolfram; the latter being an isomorphous mixture of the normal tungstates of iron and manganese, (MnFe)WO{4}. Molybdenum is most frequently met with as molybdenite, MoS{2}, which presents a certain resemblance to graphite in its physical properties and softness. It also occurs, but much more rarely, as a yellow lead ore, PbMoO{4}. In both these forms molybdenum occurs in the primary rocks, in granites, gneiss, &c., and in iron and copper ores in Saxony, Sweden, and Finland. Tungsten ores are sometimes met with in considerable masses in the primary rocks of Bohemia and Saxony, and also in England, America, and the Urals. The preliminary treatment of the ore is very simple; for example, the sulphide, MoS{2}, is roasted, and thus converted into sulphurous anhydride and molybdic anhydride, MoO{3}, which is then dissolved in alkalis, generally in ammonia. The ammonium molybdate is then treated with acids, when the sparingly soluble molybdic acid is precipitated. Wolfram is treated in a different manner. Most frequently the finely-ground ore is repeatedly boiled with hydrochloric and nitric acids, and the resultant solutions (of salts of manganese and iron) poured off, until the dark brown mass of ore disappears, whilst the tungstic acid remains, mixed with silica, as an insoluble residue; it is treated also with ammonia, and is thus converted into soluble ammonium tungstate, which passes into solution and yields tungstic acid when treated with acids. This hydrate is then ignited, and leaves tungstic anhydride. The general character of molybdic and tungstic anhydrides is analogous to that of chromic anhydride; they are anhydrides of a feebly acid character, which easily give polyacid salts and colloid solutions.[8 bis]

Tungsten is sometimes called scheele in honour of Scheele, who discovered it in 1781 and molybdenum in 1778. Tungsten is also known as wolfram; the former name was the name given to it by Scheele, because he extracted it from the mineral then known as tungsten and now called scheelite, CaWO_{4}. The researches of Roscoe, Blomstrand and others have subsequently thrown considerable light on the whole history of the compounds of molybdenum and tungsten.

The ammonium salts of tungsten and molybdic acids when ignited leave the anhydrides, which resemble each other in many respects. Tungsten anhydride, WO{3}, is a yellowish substance, which only fuses at a strong heat, and has a sp. gr. of 6·8. It is insoluble both in water and acid, but solutions of the alkalis, and even of the alkali carbonates, dissolve it, especially when heated, forming alkaline salts. Molybdic anhydride, MoO{3}, is obtained by igniting the acid (hydrate) or the ammonium salt, and forms a white mass which fuses at a red heat, and solidifies to a yellow crystalline mass of sp. gr. 4·4; whilst on further heating in open vessels or in a stream of air this anhydride sublimes in pearly scales--this enables it to be obtained in a tolerably pure state. Water dissolves it in small quantities--namely, 1 part requires 600 parts of water for its solution. The hydrates of molybdic anhydride are soluble also in acids (a hydrate, H{2}MoO{4}, is obtained from the nitric acid solution of the ammonium salt), which forms one of their distinctions from the tungstic acids. But after ignition molybdic anhydride is insoluble in acids, like tungstic anhydride; alkalis dissolve this anhydride, easily forming molybdates. Potassium bitartrate dissolves the anhydride with the aid of heat. None of the acids yet considered by us form so many different salts with one and the same base (alkali) as molybdic and tungstic acids. The composition of these salts, and their properties also, vary considerably. The most important discovery in this respect was made by Marguerite and Laurent, who showed that the salts which contain a large proportion of tungstic acid are easily soluble in water, and ascribed this property to the fact that tungstic acid may be obtained in several states. The common tungstates, obtained with an excess of alkali, have an alkaline reaction, and on the addition of sulphuric or hydrochloric acid first deposit an acid salt and then a hydrate of tungstic acid, which is insoluble both in water and acids; but if instead of sulphuric or hydrochloric acids, we add acetic or phosphoric acid, or if the tungstate be saturated with a fresh quantity of tungstic acid, which may be done by boiling the solution of the alkali salt with the precipitated tungstic acid, a solution is obtained which, on the addition of sulphuric or a similar acid, does not give a precipitate of tungstic acid at the ordinary or at higher temperatures. The solution then contains peculiar salts of tungstic acid, and if there be an excess of acid it also contains tungstic acid itself; Laurent, Riche, and others called it metatungstic acid, and it is still known by this name. Those salts which with acids immediately give the insoluble tungstic acid have the composition R{2}WO{4},RHWO{4}, whilst those which give the soluble metatungstic acid contain a far greater proportion of the acid elements. Scheibler obtained the (soluble) metatungstic acid itself by treating the soluble barium (meta) tetratungstate, BaO,4WO{3}, with sulphuric acid. Subsequent research showed the existence of a similar phenomenon for molybdic acid. There is no doubt that this is a case of colloidal modifications.

Many chemists have worked on the various salts formed by molybdic and tungstic acids. The tungstates have been investigated by Marguerite, Laurent, Marignac, Riche, Scheibler, Anthon, and others. The molybdates were partially studied by the same chemists, but chiefly by Struvé and Svanberg, Delafontaine, and others. It appears that for a given amount of base the salts contain one to eight equivalents of molybdic or tungstic anhydride; i.e. if the base have the composition RO, then the highest proportion of base will be contained by the salts of the composition ROWO{3} or ROMoO{3}--that is, by those salts which correspond with the normal acids H{2}WO{4} and H{2}MoO{4}, of the same nature as sulphuric acid; but there also exist salts of the composition RO,2WO{3}, RO,3WO{3} ... RO,8WO_{3}. The water contained in the composition of many of the acid salts is often not taken into account in the above. The properties of the salts holding different proportions of acids vary considerably, but one salt may be converted into another by the addition of acid or base with great facility, and the greater the proportion of the elements of the acid in a salt, the more stable, within a certain limit, is its solution and the salt itself.

The most common ammonium molybdate has the composition (NH{4}HO){6},H{2}O,7MoO{3} (or, according to Marignac and others, NH{4}HMoO{4}), and is prepared by evaporating an ammoniacal solution of molybdic acid. It is used in the laboratory for precipitating phosphoric acid, and is purified for this purpose by mixing its solution with a small quantity of magnesium nitrate, in order to precipitate any phosphoric acid present, filtering, and then adding nitric acid and evaporating to dryness. A pure ammonium molybdate free from phosphoric acid may then be extracted from the residue.

Phosphoric acid forms insoluble compounds with the oxides of uranium and iron, tin, bismuth, &c., having feeble basic and even acid properties. This perhaps depends on the fact that the atoms of hydrogen in phosphoric acid are of a very different character, as we saw above. Those atoms of hydrogen which are replaced with difficulty by ammonium, sodium, &c., are probably easily replaced by feebly energetic acid groups--that is, the formation of particular complex substances may be expected to take place at the expense of these atoms of the hydrogen of phosphoric acid and of certain feeble metallic acids; and these substances will still be acids, because the hydrogen of the phosphoric acids and metallic acids, which is easily replaced by metals, is not removed by their mutual combination, but remains in the resultant compound. Such a conclusion is verified in the phosphomolybdic acids obtained (1888) by Debray. If a solution of ammonium molybdate be acidified, and a small amount of a solution (it may be acid) containing orthophosphoric acid or its salts be added to it (so that there are at least 40 parts of molybdic acid present to 1 part of phosphoric acid), then after a period of twenty-four hours the whole of the phosphoric acid is separated as a yellow precipitate, containing, however, not more than 3 to 4 p.c. of phosphoric anhydride, about 3 p.c. of ammonia, about 90 p.c. of molybdic anhydride, and about 4 p.c. of water. The formation of this precipitate is so distinct and so complete that this method is employed for the discovery and separation of the smallest quantities of phosphoric acid. Phosphoric acid was found to be present in the majority of rocks by this means. The precipitate is soluble in ammonia and its salts, in alkalis and phosphates, but is perfectly insoluble in nitric, sulphuric, and hydrochloric acids in the presence of ammonium molybdate. The composition of the precipitate appears to vary under the conditions of its precipitation, but its nature became clear when the acid corresponding with it was obtained. If the above-described yellow precipitate be boiled in aqua regia, the ammonia is destroyed, and an acid is obtained in solution, which, when evaporated in the air, crystallises out in yellow oblique prisms of approximately the composition P{2}O{5},20MoO{3},26H{2}O. Such an unusual proportion of component parts is explained by the above-mentioned considerations. We saw above that molybdic acid easily gives salts R{2}OnMoO{3}mH{2}O, which we may imagine to correspond to a hydrate MoO{2}(HO){2}nMo{3}mH{2}O. And suppose that such a hydrate reacts on orthophosphoric acid, forming water and compounds of the composition MoO{2}(HPO{4})nMoO{3}mH{2}O or MoO{3}(H{2}PO{4}){2}nMoO{3}mH{2}O; this is actually the composition of phosphomolybdic acid. Probably it contains a portion of the hydrogen replaceable by metals of both the acids H{3}PO{4} and of H{2}MoO{4}. The crystalline acid above is probably H{3}MoPO{7},9MoO{3},12H{2}O. This acid is really tribasic, because its aqueous solution precipitates salts of potassium, ammonium, rubidium (but not lithium and sodium) from acid solutions, and gives a yellow precipitate of the composition R{3}MoPO{7},9MoO{3},8H{2}O, where R = NH{4}. Besides these, salts of another composition may be obtained, as would be expected from the preceding. These salts are only stable in acid solutions (which is naturally due to their containing an excess of acid oxides), whilst under the action of alkalis they give colourless phosphomolybdates of the composition R{3}MoPO{3},MoO{2},3H{2}O. The corresponding salts of potassium, silver, ammonium, are easily soluble in water and crystalline.

Phosphomolybdic acid is an example of the complex inorganic acids first obtained by Marignac and afterwards generalised and studied in detail by Gibbs. We shall afterwards meet with several examples of such acids, and we will now turn attention to the fact that they are usually formed by weak polybasic acids (boric, silicic, molybdic, &c.), and in certain respects resemble the cobaltic and such similar complex compounds, with which we shall become acquainted in the following chapter. As an example we will here mention certain complex compounds containing molybdic and tungstic acids, as they will illustrate the possibility of a considerable complexity in the composition of salts. The action of ammonium molybdate upon a dilute solution of purpureocobaltic salts (see