CHAPTER XVIII

DETERMINATION OF MANGANESE OXIDIMETRICALLY AND GRAVIMETRICALLY

285.

The determination of manganese is based on the principle of getting it in the form of manganous ion and then, after removal of the interfering substances, proceeding by some one of the following ways:

Oxidimetrically

286. Volhard's Method. The manganous ion, in a solution which is faintly acid with sulphuric acid and free of chloride or nitrate ions, is oxidized to acid manganite ion (manganese dioxide) by adding dropwise to the hot solution (maintained at a temperature around 95°) a standard solution of permanganate ion until the characteristic permanganate end point is reached. The only allowable cathion in this titration besides manganous ion (and hydrogen ion) is zinc ion, and a decided concentration of the latter is really imperative in order that there be a sensible equivalence between the amount of permanganate solution added and the amount of manganous ion oxidized. The method can be used for amounts of manganese ranging from 10 to 80 mg. Mn. The average deviation of the method for amounts over 40 mg. Mn is about 2 parts per 1000; the constant error is about 12 parts per 1000, the method giving low results (see § 298, Example 1, in this regard). For details of method see § 296.

287. Sodium Bismuthate Method.2 The manganous ion is oxidized to permanganate ion by means of sodium bismuthate in the presence of nitric acid, the excess of sodium bismuthate is filtered off, and the permanganate ion reduced by means of an excess of standard ferrous sulphate solution, the amount of ex

1 A. Guyard, Chem. News, 8, 292 (1863); Zeit. anal. Chem. 3, 373 (1864); J. Volhard, Liebigs Ann. 198, 218 (1879); Chem. News, 40, 207 (1879).

2 Schneider, Ding. poly. J. 269, 224; Reddrop and Ramage, Trans. Chem. Soc. 67, 268 (1895).

cess being determined by back titration with standard potassium permanganate solution, or the permanganate ion may be titrated directly to an end point by means of sodium arsenite solution. Iron does not interfere but chromium and vanadium do unless sodium arsenite is used as the reducing agent. The method can be used only for amounts of manganese ranging from 2 to 20 mg. Mn. It is particularly applicable for the determination of manganese in iron and steel products and manganese alloys (see § 395).

288. Persulphate Method. The manganous ion is oxidized to permanganate ion in sulphuric acid solution by means of persulphate ion in the presence of silver ion as a catalyst. The permanganate ion, after removal of the silver ion as silver chloride, is titrated with sodium arsenite solution. Excess of persulphate ion does not interfere, as there is practically no reduction of persulphate ion by arsenite ion. Chromium and vanadium do not interfere. The method can be used only for amounts of manganese ranging from 2 to 10 mg. Mn. It is not very generally used because for some unknown reason the oxidizing behavior of persulphate ion is sometimes very erratic. For further details see Boyle, J. Ind. & Eng. Ch. 4, 202 (1912).

289. Ford-Williams' Method.4 The manganous ion is oxidized to the acid manganite ion (manganese dioxide) by heating with chlorate ion in the presence of concentrated nitric acid 5 at a temperature of 95°-100°. The manganese dioxide so precipitated will always drag down some iron if this latter element is present. The precipitate of manganese dioxide is filtered off on asbestos and washed with water until the washings give no test for acid when tested with litmus paper. The precipitate and asbestos are placed in the flask in which the precipitation was effected, and a known amount of standard ferrous sulphate solution added sufficient to be in excess. The flask is shaken until the manganese dioxide is completely dissolved, after which

3 G. v. Knorre, Zeit. angew. Chem. 14, 1149 (1901).

4 A. P. Ford, Trans. Amer. Inst. Min. Eng. 9, 397 (1880); F. Williams, ibid., 10, 100 (1881); W. Hampe and M. Ukema, Zeit. anal. Chem. 24, 431 (1885); 32, (369) (1893).

5 Chloride ion must be absent, and sulphate ion preferably so as any quantity of the latter interferes.

6 Care must be taken to add the salt, furnishing the chlorate ion, to the solution only when the solution is cool (not over 60°), as serious explosions will result if the salt (NaClO3) is added to the boiling solution. After the salt has been added the solution may be heated.

the excess of ferrous ion is titrated by means of standard permanganate solution. This method is useful for amounts of manganese ranging from 2 to 20 mg. Mn. It can be employed to advantage in the determination of small amounts of manganese in the presence of large quantities of iron in pig iron, etc. As the precipitate has not the exact composition MnO2 but is slightly deficient in oxygen, a blank determination should be run on a standard steel of known manganese content.

Gravimetrically

290. Gibbs' Method. The manganous ion is precipitated as manganese ammonium phosphate MnNH,PO4-H2O and ignited to manganese pyrophosphate Mn2P2O7. Practically all ions but those of the alkalies must be absent. The method is suitable for amounts of manganese ranging from 2-200 mg. Mn, and is one of the best of all methods that we have for manganese. details see § 297.

For

291. Separations. Because of the fact that the great majority of the determinations of manganese are concerned with its presence either in manganese ores, manganese alloys, or in iron and steel products, and these products contain constituents whose presence is incompatible with the use of certain of the foregoing methods of determination, there are several important separations which have to be accomplished time and again. We will describe the most important four of these separations.

292. The Zinc Oxide Method. Separation of Ti++++ Fe+++ Al+++ Cr+++ PO from Ni++ Co++ Mn++ Ca++ Mg++. This separation is based on the fact that the concentration of hydroxyl ion in equilibrium with zinc oxide, which has been pulped into a creamy consistency with water, is sufficient to cause the precipitation of the hydroxides and phosphates of Ti++++, Fe+++,

7 W. Gibbs, Chem. News, 17, 195 (1868); Amer. J. Science, (2) 44, 216, (1867); Zeit. anal. Chem. 7, 101 (1868).

It might be thought that the precipitate of manganese dioxide which can be obtained readily enough through the oxidation of manganous ion, either in acid or alkaline solution, could be used as the basis of a gravimetric method for manganese, but such is not the case because the manganese dioxide is of variable composition and cannot be dried or ignited so as to furnish a compound of definite composition. Strong ignition of manganese dioxide with access of air at the full heat of the Meker burner gives results which show but a poor approximation to MnO4 as the composition of the ignited product

P. Slawik, Chem. Zig. 34, 648 (1910).

Al+++ and Cr+++ but not sufficient to effect a similar precipitation with respect to Ni++, Co++, Mn++, Ca++ and Mg++. The solution from which the above precipitation has been made, however, must not be allowed to stand any length of time before being used, because the manganous ion will slowly oxidize to manganic ion and the latter precipitate out as MnO.OH. This method can be used to advantage in conjunction with the Volhard method. The procedure is very simple: the solution containing the metals as sulphates is first neutralized with sodium carbonate to a p‡ of about 4.5 (neutral to methyl orange, or until there is formed a slight precipitate which dissolves very slowly when the solution is stirred briskly). The zinc oxide, previously pulped up with water in an agate mortar to a fine cream, is added in small portions with vigorous stirring until it is present in small but unmistakable excess. The solution is then made up to definite volume (usually 1000 c.c.), and after being thoroughly shaken is allowed to stand at room temperature for thirty minutes or so to allow the precipitation to complete itself and to allow the excess of zinc oxide to settle to the bottom of the flask. An aliquot portion (50 or 100 c.c.) of the clear solution is then withdrawn by means of a pipette, and the manganese determined by the Volhard method as described in § 296.

293. The Basic Acetate Method. Separation of Ti++++ Zr++++ Fe+++ Al+++ Th+++ Cr+++ (partially) PO from Ni++ (partially) Co++ (partially) Zn++ Mn++ Ca++ Mg++ Ba++,10 This old and important method is founded on the more pronounced hydrolysis of the salts of Ti++++, Fe+++, Al+++ in a solution faintly acid with acetic acid as compared with the salts of Ni++, Co++, Zn++, Mn++, Mg++ and Ca++. By the establishment of certain conditions as described subsequently in § 296, the hydrolysis can be made to proceed to a point where the Ti++++, Fe+++ and Al+++ are quantitatively thrown out in the form of colloidal gels of the general composition [Fe(CH3COO)3]p[Fe(OH)3]q[H2O]r in which the ratio of p: q:r is variable, whereas Ni++ and Co++ will only be partially thrown out, while Zn++, Mn++, Ca++ and Mg++ will be thrown out to such a very small extent only that a repetition of the process serves to keep them out of the iron and aluminum

10 Members of the copper and tin groups must first be removed by hydrogen sulphide (§ 210).

precipitate entirely. With respect to Cr+++ it is to be mentioned that while salts of chromium undergo pronounced hydrolysis the behavior of this element in the basic acetate separation is a matter of dispute, some workers claiming that practically all the chromium can be precipitated with the iron and aluminum, others claiming that the separation is not at all satisfactory when chromium is present.11 Phosphorus (its presence in the usual form of phosphate ion PO being assumed) will be quantitatively precipitated as ferric or aluminum phosphate, provided the iron and aluminum are in excess.

Theory. When the salt of a weak base and strong acid is dissolved in water, it undergoes hydrolysis according to the scheme which we will represent for the case of ferric chloride:

The ferric hydroxide which is formed does not precipitate out but enters into combination with some of the ferric chloride that is left to form the ferric hydroxide sol, which according to recent work 12 is a colloidal aggregate of the composition [pFeCl3.qFe (OH)3·7H2O], where the ratio of p: q:r is variable according to conditions. It has been further shown that when the ratio of FeCl3 Fe(OH), in the sol, namely, the ratio of p: q, becomes less than 120, the sol becomes unstable and precipitates out as the gel [pFeCl3 ·qFe(OH)3 ·rH2O].

Hence, any agency which will decrease the ratio of FeCl3: Fe(OH)3 beyond the limit 1: 20 will cause precipitation. By reference to the equilibrium scheme just given, it will be seen that, if we reduce the concentration of hydrogen ion, we promote the forward reaction and so cut down the FeCl, and at the same time build up the Fe(OH)3. Now there are two ways by which we can reduce the concentration of hydrogen ion:

(1) by neutralization with hydroxyl ion (addition of an alkali or alkaline salt) (2) by tying up the hydrogen ion in the form of a weak acid (addition of a buffer salt)

11 See Mellor, pp. 362, 469 of citation in § 13.

12 A. W. Thomas and A. Frieden, J. A. C. S. 40, 2522 (1923).