CHAPTER III

CONSTITUENTS OF THE IRON ALLOYS: CARBON-OPINIONS AND RESEARCHES ON COMBINED CARBON.

To supplement the particulars given in the preceding chapter with regard to carbide and hardening carbon, it is necessary to refer to a series of researches performed on iron alloys containing both carbide and hardening carbon.

Eggertz, by treating iron with a solution of iodine in iron iodide, obtained a carbonaceous residue of the empirical composition CI+ 20H2O, with 60 per cent. of carbon.

The carbonaceous residue left behind when iron is dissolved in copper - ammonium chloride was examined by Andrew Blair,2 who found therein—

A second portion of the residue when heated in a current of hydrogen lost 39.25 per cent. in weight, of which, however, only 12:54 per cent. was moisture. No oily or tarry substances were formed. The residue was non-crystalline,

1 Berg- und Hüttenm. Ztg., 1863, p. 373; 1875, p. 440; 1381, p. 264. 2 Amer. Chem. Journ.

and contained 57.84 per cent. of carbon, and 2.91 per cent. of ash.

The first analysis would correspond very closely to the formula 3(C50H21CIO15)NO2.1 Of this there remains, after heating, a residue of 3C45, whilst the liberated gases must contain, in addition to 19.5 H2O, also 3C5,047,3Hg,3Cl and N.

Schützenberger and Bourgeois,2 by repeated treatment with solutions of copper chloride and ferric chloride containing hydrochloric acid, obtained from white pig iron a blackish brown residue, which, when washed and dried at 100° C., gave the formula C11HO after deducting the silica, graphite, and other impurities. This carbohydrate they named graphite hydrate, and the same is closely allied to Brodie's graphitic acid or Berthelot's hydrographite oxide.

3

The compound in question rapidly gives off water at 250° C. without undergoing distension; it is briskly attacked by ordinary nitric acid in the warm, and is converted—with liberation of red fumes into an amorphous red - brown substance that is soluble in nitric acid, alcohol, alkalis, ammonia, and pure water, from which, however, it is precipitable by additions of neutral salts. The ammoniacal solution gives, after boiling off the excess of ammonia, light brown compounds with metallic salts. The average composition of the nitro-product is

On heating, this is decomposed, a smell of hydrocyanic acid

1 Possibly the nitrogen is combined with water, and the chlorine may also be a constituent of the ash.

2 Compt. Rend., 1xxx. p. 911; abstracted in Chem. Centralbl., 1885, p. 387.

being given off, and a black residue left, which latter is converted into a brown product by nitric acid.

The above figures lead to the formula C22H17(NO2)O11 for the substance, which has received the name nitrographitoic acid; Schützenberger and Bourgeois regard this as identical with the substance obtained by Eggertz by the action of nitric acid on iron, which product furnishes peculiarly brownish yellow solutions. On evaporating to dryness the nitric solution containing the chemically combined carbon as soluble nitrographitoic acid, and treating it with alcohol, the silica and graphite remain undissolved, whilst the nitrographitoic acid, etc., passes into solution.

These two observers, in fact, have based on this behaviour a method for estimating graphite and combined carbon.

1

Zabudsky 1 dissolved in sodium-copper chloride a pig iron containing, apart from 4:104 per cent. of combined carbon, neither graphite, sulphur, nor phosphorus, and found that the dark brown residue, which behaved exactly like a carbohydrate, corresponded most nearly to the formula C12H¿О. The same composition was exhibited by the residue obtained by decomposing the same pig iron by silver chloride, or by electrolysis. This carbohydrate can be easily nitrated, and readily furnished halogen derivatives, e.g. the iodine compound CH29IO15

3.

Special interest attaches to the investigations conducted by E. Donath,2 which will therefore be quoted verbatim.

Spiegeleisen in coarse lumps was treated, first with sodium-copper chloride solution at the ordinary temperature until the originally deposited copper had just re-dissolved (a tedious operation with the quantities taken), afterwards in the cold with a solution of ferric chloride slightly acidified

1 Prot. d. Journ. d. russ. phys.-chem. Ges., 1882, iii.; Berichte, xv. p. 946. 2 Oesterr. Zeits. f. Berg- u. Hüttenmesen, 1895, p. 149.

with hydrochloric acid, and finally digested in the warm with very dilute hydrochloric acid, to make sure of the whole of the iron being dissolved. The brownish black residue was then filtered off, washed with hot water acidified with hydrochloric acid, and finally with hot water until the chlorine reaction had entirely disappeared, after which it was dried at 110° C. until constant.

"The ultimate analysis of the residue gave the following values:

nevertheless, it was invariably contaminated with chlorine, although the washing had been continued until the complete disappearance of the silver reaction; but combustion with pure lime in a glass tube gave only 0.03 per cent. of chlorine from the resulting silver chloride; it gave a white incineration residue of 1.28 per cent., which, according to the qualitative examination, consisted almost entirely of silica.

'As can be seen, these results differ from those obtained by Schützenberger and Bourgeois to a not unimportant degree, which is not inexplicable, seeing that it could not be proved in either case whether chemical entities or mixtures were in question. It was, however, proved in both that on treating iron (which, like spiegeleisen, contains chemically combined carbon only) with cupric chloride, or cupric salts generally, the carbon of the dissolved iron remains behind, not in the condition of carbon itself, but as a carbon compound of decidedly organic character, and containing hydrogen and oxygen, the formation of which substance cannot well be explained if it be assumed that the carbon is not combined with the iron in atomic proportions. This brownish black

substance was then converted by means of nitric acid in the manner prescribed by the above-named workers, into the compound to which they have given the name 'nitrographitoic acid,' and exhibited the properties indicated by them."

Donath rightly objected to the name nitrographitoic acid, since it has by no means been proved that the body in question is allied in constitution with graphitic acid, or even that it is a nitro-derivative at all. He therefore proposes to confer on it the less definite name of carbazeic acid.

Attempts to obtain the substance in the form of crystals from alcohol and other solvents, failed. Ultimate analysis

"These values agree considerably better with the results obtained for this compound by Schützenberger and Bourgeois.

"Particularly worthy of note is the high staining power of these solutions of nitrographitoic acid, though this is far surpassed by that of their ammonia salts, prepared by neutralising the acid with ammonia, and driving off the excess of the latter by evaporation in a water bath, a deep, black, lustrous, brittle product being thus obtained. It is undoubtedly to this substance that the characteristic coloration of the steel solutions prepared with nitric acid is due in the application of the 'Eggertz test.' When prepared pure, in the manner described, this substance is indubitably the most suitable for the preparation of normal solutions for the Eggertz test, since it enables comparisons to be made with the same substance, and not merely, as is the case with burnt sugar, coffee, etc., with entirely different substances merely