CHEMICAL NEWS, May 16, 1884. Volumetric Estimation of Iron. we may either separate a small quantity of gallium mixed with much boric acid, or a little boric acid mixed with much gallium. The calcium borate obtained at the end of the analysis is often mixed with a little ferric oxide; it is then redissolved in hydrochloric acid, supersaturated with potassium and sodium carbonates (or hydroxides) in equal equivalents, and ignited. The mass, on treatment with water, deposits iron oxide, whilst the boric acid remains in the alkaline liquid. The latter, after being slightly supersaturated with hydrochloric acid, is again treated according to Ditte's method. 2. The hydrochloric solution, slightly acid, is mixed with an excess of acid, potassium, and sodium acetates in equal equivalents, and a suitable quantity of arsenious acid. A current of sulphuretted hydrogen gas precipitates arsenic sulphide, which carries down with it the gallium. Air is forced through the filtrate in order to expel the greater portion of the hydrogen sulphide, and the liquid is then supersaturated with excess of a mixture of potassium and sodium hydroxides in equal equivalents. The mass is placed in a vessel of gold, evaporated to dryness, and finally ignited with access of air. It is dissolved in water, slightly acidified with hydrochloric acid, and M. Ditte's process is applied. This method is especially advantageous for determining small quantities of gallium mixed with large proportions of boric acid. The author has ascertained that an aqueous, or even hydrochloric, solution of boric acid loses no appreciable trace if evaporated in a vacuum at the ordinary temperature.-Comptes Rendus. NOTES ON THE VOLUMETRIC ESTIMATION OF IRON. By R. W. ATKINSON, B.Sc. Lond., F.C.S., F.I.C. IN my last communication I pointed out that in the volumetric estimation of iron great care has to be taken in choosing the method of standardising the solution of dichromate of potassium, without which care very considerable errors may be introduced into the analysis. In the present note I wish to make some observations upon the process of titration, presuming that the strength of the dichromate solution in terms of iron is accurately known. A fair sample of the ore having been ground sufficiently fine to pass completely through a sieve of 120 meshes to the linear inch, part of it is dried at 100° C., and when cold portions of about 0'5 or o6 grm. are weighed out for analysis. Unless the balance is very rapid in its action it will be found necessary to re-dry the portions first weighed out, if accurate and concordant results are to be expected. The ore is in such a very fine state of division that it greedily absorbs moisture from the air during the operation of weighing out: this applies most forcibly to the soft red ores, like Campanil and Vena Dulce, but is also true of the harder Rubio and other brown ores. The weight of the portion being accurately known, it is transferred to a conical flask, and digested at a gentle heat on a hot plate with from 10 c.c. to 15 c.c. strong hydrochloric acid, the flask being closed with a watch-glass. It has been asserted by K. F. Föhr (Zeitschrift, vol. | xxii., p. 609) that ferric chloride is volatile at about 100 C., but this is contrary to all my experience of the matter. It is true that yellow drops of ferric chloride may sometimes be seen depending from the under surface of the watch-glass, but this is only the case when the solution is accompanied by a boiling of the liquid, and is doubtless due to spurting. If the digestion is carried on so as to avoid bubbling, the drops on the cover are always colour 217 less, and no loss of iron from volatilisation need be feared. When the ore has been completely dissolved the next step is the reduction from the ferric to the ferrous state, to effect which there are three reducing agents mainly employed, viz., zinc, stannous chloride, and sulphurous acid in the form of one of its salts. I will deal with these in the order given. Reduction by means of zinc is capable of giving trustworthy results, provided that pure zinc free from iron is employed; or if the zinc contains iron, provided that the amount thus introduced be allowed for. But the use of zinc is open to other objections: in the first place it dissolves only slowly, and thus unduly retards the operation, which, when a number of analyses are to be carried out, is a matter of no small moment. In the second place the titration is further retarded by the slowness with which the blue colour with potassic ferricyanide is developed in consequence of the presence of zinc chloride in the solution. And lastly, the colour towards the end of the titration becomes so faint, even when fully developed, that it is impossible to distinguish the presence of an amount of iron less than one- or two-tenths per cent of the iron contained in the ore. Consequently, although reduction by means of zinc permits the analyst to obtain uniformly concordant results without the risk of error, its action is too slow and its indications are not sufficiently delicate for the most accurate work. But however much rapidity of work may be an object to be kept in view, there can be no doubt but that reduction by zinc is greatly to be preferred to the second method, viz., reduction by means of stannous chloride. Indeed I have no hesitation in saying that the latter method, as usually performed, is open to the grossest abuses, and ought to be prohibited by any authority which may seek to reform the present methods of analysis. I will justify these remarks presently, after I have described the usual manner of carrying out the process of reduction by this reagent. A strongly acid solution of stannous chloride is used, often of a strength roughly corresponding with that of the potassic dichromate solution. Sometimes precautions are taken to prevent oxidation by preserving it under an atmosphere of nitrogen or carbonic acid; sometimes it is merely kept in a stoppered bottle, and portions taken out with a pipette. The portion of ore having been dissolved in hydrochloric acid, in the conical flask, boiling water is added so as to about half-fill the flask, and the liquid is kept boiling during the addition of the stannous chloride, which is added to the boiling solution until the yellow colour of the ferric chloride has disappeared, showing that an excess of the reducing agent has been employed. In order to destroy this excess an oxidising solution is cautiously added until a drop of the liquid gives a faint red colouration when brought into contact with a solution of potassic sulphocyanide on a porcelain slab. When the operator is satisfied with the reduction, the addition of the standard dichromate solution is proceeded with in the usual manner. It will be quite evident that the accuracy of the determination of iron in the sample is dependent upon the exactness with which the oxidising solution (potassic chlorate or dichromate) is added to destroy the excess of stannous chloride present, and that it is therefore possible to get too high or too low a result at will merely by adding too little or too much of the oxidising solution. In this way an unscrupulous chemist may produce a result which will harmonise with the interests of his client, and because this method opens the way to dishonest dealing do I justify my remark that it ought to be absolutely prohibited. But apart from cases of actual dishonesty, which no doubt are exceptional, it ought to be the determination of all those who wish to earn a reputation for accuracy to abandon a process of which it can be said that its results can be made variable at will-which can be made to give 218 Volumetric Estimation of Iron. either high results to suit the seller or low results to suit, the buyer. I am far from saying that analysts consciously take advantage of this power. Most are doubtless influenced quite unconsciously-not wishing to leave any stannous chloride undestroyed, some perhaps add a little too much of the oxidising agent; others may have a weak perception of red rays, so that before they are able to see the red colour produced by contact with potassic sulphocyanide an excess of the oxidising agent has been added. In this way a permanent, unconscious bias towards low results may exist. In a similar way another chemist may be afraid of adding too much of the oxidising agent, and, as the mixture of ferrous chloride, stannous chloride, and potassic sulphocyanide oxidises very readily, he may observe a red tinge, whilst there is still an excess of stannous chloride in the solution, thus finding too high results. In fact it is easy enough to get results which are too high or too low, but with stannous chloride as a reducing agent it is a matter of the greatest difficulty to get accurate results. One reason why stannous chloride has become a favourite reducing agent is the rapidity with which the analysis can be made :-Starting with a tin of ore the percentage of moisture and iron in the ore can be found by this method (with the above qualification as to accuracy) in from two to three hours; but I hold that it is contrary to the best interests of a chemist to seek rapidity of work at the expense of accuracy, and I therefore strongly recommend the abandonment of this method of reduction. Of all methods reduction of the ferric salt by the use of a concentrated solution of ammonic bisulphite is the most accurate and trustworthy. Sodic bisulphite is sometimes used, but is not nearly so satisfactory as the ammonic salt, as it is more difficult to separate the last traces of sul. phurous acid from the former than from the latter. The mode of manipulation is as follows:-The ore having been dissolved in hydrochloric acid in the conicals, the solution is diluted with acidified water and filtered into pear-shaped flasks, the filters being thoroughly washed with hot acid water. The filtered ferric chloride is next carefully neutralised with ammonia, strong at first and afterwards dilute, until a faint reddish precipitate remains permanent. Two or three drops of strong hydrochloric acid are washed round the inner neck of the flask, and as the acid flows down it spreads out, dissolving any particles of ferric hydrate which may have remained on the sides of the flask. When the solution is quite clear, and of a faint reddish colour, 5 c.c. or 6 c.c. of a strong solution of ammonic bisulphite (sp. gr. 1'06) are added, the flask shaken, and boiling water added. On shaking the flask the colour entirely disappears, and the flask is then put over a burner. A small piece of thick platinum wire is introduced to assist the boiling, and about 15 c.c. or 20 c.c. of dilute sulphuric acid (1 acid to 6 water) are added to acidify the solution and to assist in the expulsion of the excess of sulphurous acid. After the liquid is once in a state of ebullition it is kept boiling briskly for thirty minutes (less time is sufficient, but it is always well to err on the safe side), during which time nearly the required amount of dichromate is run out into the dish. At the end of the half-hour the boiled solution is added to the potassic dichromate, and the titration is carried out as usual. CHEMICAL NEWS, lowed for boiling off the sulphurous acid; but if the conditions as given above are fulfilled, constant and accurate results may be relied upon. The above method has this advantage, viz., that the solution is practically one of ammonio-ferrous sulphate, a salt which is one of the most stable of all the ferrous salts. It is therefore less liable to become oxidised by exposure to the air in transferring to the basin than the acid solution of ferrous chloride obtained by the two previous modes of reduction. A further advantage lies in the fact that the end reaction with potassic ferricyanide is beautifully clear and delicate, so that there is no difficulty in distinguishing the addition of 1-20th c.c. of dichromate (strength I c.c. = 0'005 grm. iron), equivalent to 0.00025 grm. Fe. Numerous experiments have shown that perfectly constant results can be obtained by this method, the same percentages of iron in a given ore having been found with different standard dichromate solutions (standardised as described in my first note) after the lapse of several months. It is rarely the case that three experiments carried out simultaneously give percentages of iron differing by more than o'05 to 0.07 per cent of the ore, but the main advantage which the method of reduction by ammonic bisulphite possesses is that results are found which are perfectly independent of any unconscious bias on the part of the operator, and I feel convinced that were this method constantly and generally used we should hear less of differences of 2 per cent and 3 per cent between two chemists' analyses of the same sample of ore. The existence of "sellers" and "buyers" chemists is a disgrace to the profession, and anything which is likely to put an end to the scandal, even in one trade, ought to be welcomed. One other point in the volumetric estimation of iron remains to be noticed. The solution of potassic ferricyanide slowly decomposes when the bottle containing it is exposed to diffused daylight and a yellowish sediment is deposited. This change is very greatly retarded, if not entirely prevented, by protecting the solution from light by covering the bottle with an inverted tin canister. Connected with this decomposition of the ferricyanide solution is the fact, already well known, that when a mixture of that solution with one of ferric chloride is exposed to daylight reduction takes place, and the solution turns blue. But it is not so generally known how rapidly this takes place, and that if the drops of ferricyanide to which the completely oxidised iron solution has been added be allowed to remain exposed to the light (protected from dust by means of a glass plate) for ten or fifteen minutes, a distinct blue tinge will be developed. It is important to remember this, for, in titrating, the blue colour requires two or three minutes to become fully developed when the amount of ferrous salt remaining in the solution is very small, and in order to prevent the drop turning blue by reduction under the influence of daylight, it is advisable to keep the slab covered with a black cloth or with a flat tin cover. By this means it is possible so to protect the mixture from change that no blueness is perceptible after the lapse of an hour or more, provided that all the iron in the solution to be titrated has been fully oxidised. 44, Loudoun Square, Cardiff. By proceeding as above there is only one loophole for Modifications of Gmelin's Test for Biliary Colours. the introduction of error, viz., in the length of time al-Capranica recommends the use of a 5 per cent solution If an excess of stannous chloride be added to the iron solution, and afterwards an oxidising agent in insufficient amount to completely destroy the excess, a red colour will be observed on bringing a drop of the solution into contact with a drop of potassic sulphocyanide, which colour slowly disappears. It would thus seem that the ferrous salt is first oxidised, and that the excess of stannous chloride only slowly reduces it. In this way a very serious error may be quite unconsciously introduced. To avoid this I have attempted to neutralise the effect of the excess by the addition of mercuric chloride, in which I find that I have been anticipated by Dr. F. Kessler (Zeitschrift, xi., 249); but in my hands this method has given results almost as unreliable as are obtained by the use of other oxidising agents, though I am not able at present to assign a reason for its imperfect action. I may be able to explain it later. of bromine, of chloric and iodic acid. These reagent are applied to the colouring matters, not in an aqueous, but in an alcoholic, ethereal or chloroformic solution. As the most until the green colour appears, and then hydrochloric acid. sensitive procedure Capranica adds solution of bromine On shaking, the green colour passes entirely into the hydrochloric acid. One part of bile pigment can thus be detected in 200,000. A. A. Krehbiel mixes 4 parts of the urine of a person afflicted with jaundice with 1 part hydrochloric acid and adds a saturated solution of chloride of lime, drop by drop. The characteristic colour is produced by 3 to 6 drops.-Zeitschrift fur Analytische Chemie. AFTER Some earlier, unsatisfactory determinations, Berzellius, in 1826, published his final estimation of the atomic weight of antimony. He oxidised the metal by means of nitric acid, and found that 100 parts of antimony gave 124.8 of Sb204. Hence, if O=16, Sb=129'03. The value 129 remained in general acceptance till 1855, when Kessler, by special volumetric methods, showed that it was certainly much too high. Kessler's results will be considered more fully further along, in connection with a later paper; for present purposes a brief statement of his earlier conclusions will suffice. Antimony, and various compounds of antimony, were oxidised partly by potassium anhydrochromate and partly by potassium chlorate; and from the amounts of oxidising agent required, the atomic weight in question was deduced : By oxidation of Sb2O3 from 100 parts Sb Sb with K2Cr2O7.. Sb2O3 with Sb2O3 with K2Cr2O, Sb=123.84 219 having tried to determine the amount of gold precipitable by a known weight of antimony, and having obtained discordant results, finally resorted to the original method of Berzelius. Antimony, purified with extreme care, was oxidised by nitric acid, and the gain in weight was determined. From 1.5 to 3'3 grms. of metal were used in each experiment. The reduction of the weights to a vacuum standard was neglected as being superfluous. From the data obtained we get the following percentages of Sb in Sb204 79'268 tartar emetic.. 123.61 123'72 123.80 123'58 119.80 The figures given are those calculated by Kessler himself. A recalculation with our newer atomic weights for O, K, Cl, Cr, S, and C, would yield slightly lower values, It will be seen that five of the estimates agree closely while one diverges widely from the others. It will be shown hereafter that the concordant values are all vitiated by constant errors, and that the exceptional figure is after all the best. Shortly after the appearance of Kessler's first paper, Schneider published some results obtained by the reduction of antimony sulphide in hydrogen. The material chosen was a very pure stibnite from Arnsberg, of which the gangue was only quartz. This was corrected for, and corrections were also applied for traces of undecomposed sulphide carried off mechanically by the gas stream, and for traces of sulphur retained by the reduced antimony. The latter sulphur was estimated as barium sulphate. From 32 to 10'6 grms. of material were taken in each experiment. The final corrected percentages of S in Sb2S3 were as follows: Mean 79-2830'009 Hence, if O=16, Sb=122'46. The determinations of Dumas were published in 1859. This chemist sought to fix the ratio between silver and antimonious chloride, and obtained results for the atomic weight of antimony quite near to those of Dexter. The SbCl3 was prepared by the action of dry chlorine upon pure antimony; it was distilled several times over antimony powder, and it seemed to be perfectly pure. Known weights of this preparation were added to solutions of tartaric acid in water, and the silver chloride was precipitated without previous removal of the antimony. Here, as Cooke has since shown, is a possible source of error, for under such circumstances the crystalline argento-antimonious tartrate may also be thrown down and contaminate the chloride of silver. But be that as it may; Dumas's weighings, reduced to a common standard, give as proportional to 100 parts of silver, the quantities of SbCl, which are stated in the third of the subjoined columns: 1.876 grms. SbCl3 = 2.660 grms. Ag. 70'526 Hence, if S=32, Sb=120°3. Immediately after the appearance of Schneider's memoir, Roses published the result of a single analysis of antimony trichloride, previously made under his supervision by Weber. This analysis, if Cl=35'5, makes Sb=120'7, a value of no great weight, but in a measure confirmatory of that obtained by Schneider. The next research upon the atomic weight of antimony was that of Dexter, ¶ published in 1857. This chemist, * Smithsonian Miscellaneous Collections. Nature." + Poggend. Annal., 8, 1. 1 Ibid., 95, 215. Mean 70 5120'021 Hence, if Ag=108, and Cl=355, Sb=122. In 1861 Kessler's second paper+ relative to the atomic weight of antimony appeared. Kessler's methods were somewhat complicated, and for full details the original memoirs must be consulted. A standard solution of potassium anhydrochromate was prepared, containing 6*1466 grms. to the litre. With this, solutions containing known quantities of antimony or of antimony compounds were titrated, the end reaction being adjusted with a standard solution of ferrous chloride. In some cases the titration was preceded by the addition of a definite weight of potassium chlorate, insufficient for complete oxidation; the anhydrochromate then served to finish the reaction. The object in view was to determine the amount of oxidising agent, and therefore of oxygen, necessary for the conversion of known quantities of antimonious into antimonic compounds. In the later paper Kessler refers to his earlier work, and shows that the values then found for antimony were all too * Ann. Chim. Phys., (3), 55, 175. + Poggend. Annal., 113, 145. 220 A Recalculation of the Atomic Weights. CHEMICAL NEWS, high, except in the case of the series made with tartar The fourth set of experiments was gravimetric. The solution of SbCl3, mixed with tartaric acid, was first precipitated by hydrogen sulphide, in order to remove the antimony. The excess of H2S was corrected by copper sulphate, and then the chlorine was estimated as silver chloride in the ordinary manner. 100 parts of AgCl corres pond to the amounts of SbCl3 given in the third 2'7437 3.8975 In the paper now under consideration four serics of re- 2.6798 sults are given. The first represents experiments made 5'047 upon a pure antimony trioxide which had been sublimed, and which consisted of shining colourless needles. This was dissolved, together with some potassium chlorate, in hydrochloric acid, and titrated with anhydrochromate solution. Six experiments were made, but Kessler rejects the first and second as untrustworthy. The data for the others are as follows: 3'141 53'588 53'569 53 629 53.696 The volumetric series with SbCl, gave Kessler values for Sb ranging from 121 16 to 121 47. The gravimetric series, on the other hand, yielded results from Sb=124'12 to 124 67. This discrepancy Kessler rightly attributes to the presence of oxygen in the chloride; and, ingeniously correcting for this error, he deduces from both sets combined the value of Sb=122*37. The several mean results for antimony agree so fairly with each other, and with the estimates obtained by Dexter and Dumas, that we cannot wonder that Kessler felt satisfied of their general correctness, and of the inaccuracy of the figures published by Schneider. Still, the old series of data obtained by the titration of tartar emetic with anhydrochromate contained no evident errors, and was not accounted for. This series,* if we reduce all of Kessler's figures to a single common standard, give a ratio between K2Cr2O7 and C4H4KSBO7.1H2O. 100 parts of the former will oxidise of the latter: Mean 10953±0.0075 In the second series of experiments pure antimony was dissolved in hydrochloric acid with the aid of an unweighed quantity of potassium chlorate. The solution, containing both antimonious and antimonic compounds, was then reduced entirely to the antimonious condition by means of stannous chloride. The excess of the latter was corrected with a strong hydrochloric acid solution of mercuric chloride, then, after diluting and filtering, a weighed quantity of potassium chlorate was added, and the titration with anhydrochromate was performed as usual. Calculated as above, the percentages of oxygen given in the last column correspond to 100 parts of antimony :- Per cent O. 336.64 338.01 336.83 337'93 338'59 335'79 Mean 337.30 ± 0·29 From this, if K2Cr2O7=294'64, Sb=119°18. (To be continued.) 13'088 13'050 Mean 13.97900096 This series gave Kessler Sb=122°34. The third and fourth series of experiments were made with pure antimony trichloride, SbCl3, prepared by the action of mercuric chloride upon metallic antimony. This preparation, in the third series, was dissolved in hydrochloric acid, and titrated. In one experiment solid K2Cr2O7 in weighed amount was added before titration: in the other two estimations KCIO, was taken as usual. If, according to Siewert's work, we take Cr=52'009, the percentages of oxygen in the last column correspond to 100 parts of SbC13: HYDRATION OF SALTS AND OXIDES. THIS subject may be approached from several points of view, each with its own experimental method. Thorpe and Watts have studied the hydration of salts in the light of the volumes occupied by the successive molecules of water of hydration; Favre and Valsont have determined the heat constants for the solution of salts of varying degrees of hydration, and thus endeavoured to throw light upon the mode of union of water of crystallisation; Hannay and Ramsay|| have taken up the subject from the dynamic side, applying the time method to the observation of the dehydration of salts and oxides. Lastly, I have endeavoured to follow the phenomena of rehydration, by means of a special method and apparatus. I have shown that when a salt is deprived of its water of crystallisation and exposed to an atmosphere saturated with aqueous vapour, it quickly combines with water up to the original limit; that here a break occurs of greater or less duration * Poggend. Annal, 95, 217. 1 Compt. Rendus., vols. 73, 74, 75. ,, 23.2 7'0321 Mean 7'0294 § Ibid., 35, 796. CHEMICAL NEWS, vol. 44, pp. 101, 209; vol. 47, INCREASE IN WEIGHT (HO) FERIDOPTS ORIGINAL CHEMICAL NEWS, May 16, 1884. Hydration of Salts and Oxides. followed by that further combination of water which is known as deliquescence. An anhydrous salt, or a salt containing its limiting quantity of water of crystallisation, passes immediately into this phase of deliquescence, or incipient solution, and these several phenomena are continuous manifestations of the same attractive energy. The hydration of oxides, although so dissimilar in its visible results, follows a similar course and to demonstrate more clearly these common features of resemblance I have undertaken and completed a series of parallel observations upon 221 the yellow solution of the salt very early making its appearance upon the scale pan. The chromic hydrate remained throughout to all appearances a dry solid, notwithstanding that at the close of the | experiment it contained 60 per cent water and had taken up nearly as much water as the copper sulphate. In a future communication it is my intention to discuss these and previous results, together with the collateral researches of other observers, mentioned at the beginning of this paper, in their bearings on the subject of hydration. three substances selected as types of these various modes of rehydration, viz., CuSO4.H2O, K2Cr2O7, and Cr2O3.4H2O. The results are graphically represented by the accompanying curves, each of which originates at the same 0 point, corresponding to 100 pts. of substance; the ordinates therefore represent the percentage increase due to combination with water. The period over which the observations extended was from May to August, 1883; the limits of temperature observed were 16-21°. The similarity of these curves to one another in their main features is unmistakeable. The course of hydration in the case of the copper sulphate was marked by the gradual contraction of the area of the deliquescing solid: from covering a circular area of 70 m.m. diameter at the beginning of the experiment, it occupied at the close an area equally regular and concentric with the former, but of not more than half that diameter. The potassium dichromate presented throughout a very different appearance; the course of hydration was from the beginning rather one of liquefaction than deliquescence, DAYS PROCEEDINGS OF SOCIETIES. ROYAL INSTITUTION OF GREAT BRITAIN. The DUKE OF NORTHUMBERLAND, D.C.L., LL.D., THE Annual Report of the Committee of Visitors for the year 1883, testifying to the continued prosperity and efficient management of the Institution, was read and adopted. The Real and Funded Property now amounts to above £85, 400, entirely derived from the Contributions and Donations of the Members. Thirty-seven new Members paid their admission fees in 1883. Sixty-three lectures and nineteen evening discourses were delivered in 1883. The books and pamphlets presented in 1883 amounted |