62 THE ANALYST. INORGANIC ANALYSIS. The Chemistry of Bordeaux Mixture. S. Pickering. (Proc. Chem. Soc., 1907, 23, 261-262.)-The substances formed on. the addition of lime to copper sul- phate, as in the preparation of Bordeaux mixture, are dependent on the proportions of lime used, and may be either (1) 4Cu0,S0,,0*06CaS04 ; (2) 4CuO,SO3,O~25CaSO, ; (3) 10Cu0,S0,,1-3CaS04 ; (4) 10CuO,S0,,4CaO,SO, ; (possibly 5 ) lOCuO,SO,, lOCaO,SO,, or (6) Cu0,3CaO ; that present in most cases probably being (4). The fungicidal action of Bordeaux mixture seems to depend on the liberation of normal copper sulphate by the action of carbon dioxide on the basic sulphate. The action begins only after a certain lapse of time, the basic calcium sulphate having to be decomposed before the basic copper sulphate is attacked.By using only sufficient lime to form 4CuO,SO,, the presence of basic calcium sulphate and the consequent delay in the action is avoided ; in addition to this, the amount of normal copper sul- phate liberated (for a given amount of copper sulphate taken) and the insecticidal value of the mixture will be two and a half times as great as when it consists of 10Cu0,S03,4Ca0, SO,. The basic sulphates, when precipitated by lime in presence of excess of sodium sulphate, contain sodium sulphate as well as calcium sulphate. The formula of the most basic sulphate is 10CuO,SO,,(Na,Ca)SO,. When precipitated by soda, they contain no appreciable amount of sodium sulphate, unless a large excess of the latter is present in the liquid. Nickel sulphate with lime gives a basic sulphate containing a very little calcium sulphate, but, if the lime is at all times in excess, the basic sulphate is deprived of most of its acid, the precipitate consisting essentially of a double oxide.Estimation of Carbon Dioxide in Electrolytic Chlorine. P. Philosophoff. CIzcm. Zczt., 1907, 31, 1256-1257.)-The mixed gases, stored over concentrated sodium chloride solution saturated previously with the gases, are brought into a Bunte burette containing mercury and connected with a graduated levelling tube.THE ANALYST. 63 The diminution due to the absorption of the chlorine by the mercury is measured by readjusting the level of the mercury in the graduated tube. The carbon dioxide is absorbed subsequently by potassium hydroxide.w. P. s. Estimation of Hydrofluosilicic Acid. S. Honig-Szabadka. (Chem. Zed., 1907, 31, 1207-1208.)-The following simple acidimetric method for the estimation of hydrofluosilicic acid, though not theoretically exact, owing to the slight hydrolytic dissociation of sodium fluosili,cate, is sufficiently accurate for industrial purposes, particularly for rapid control of the manufacture of the salts. It consists of a differential titration, first in presence of methyl orange, and then, near the boiling- point, in presence of phenolphthalein. At the neutral point, with methyl orange, all the foreign mineral acids have been neutralised, and the hydrofluosilicic acid has been converted into its sodium salt, thus : H,SiF, + 2NaOH = Na,SiF, + 2H,O. The solu- tion is then heated to boiling, phenolphthalein is added, and normal sodium hydroxide solution is run into the hot liquid until a pink colour is produced.At this stage the following reaction has taken place : Na,SiF, +4NaOH= 6NaF+ Si(OH),. From the result of the latter titration, the quantity of hydro- fluosilic acid may be calculated ; any sodium fluosilicate originally present would, however, be included in the result. J. F. B. A Colorimetric Method for the Determination of Small Percentages of Iron in Copper Alloys. A. W. Gregory. (Proc. Chem. SOC., 1907, 23, 306-307.) -The method is based upon the colour reaction given by salicylic acid and ferric chloride. This reaction is modified by using a solution of salicylic acid in acetic acid, and the test is carried out in the presence of sodium acetate. Under these conditions a deep red colour is produced, which is used as a colorimetric test for iron in small quantities.The interfering action of the blue copper salts is overcome by the addition of a weak solution of potassium cyanide, which results in the formation of the colourless complex cyanide of copper and potassium. Zinc and antimony do not interfere with the reaction, but lead must be removed as sulphate. This method has been found to give very accurate results in cases where the per- centage of iron in the copper alloys is so low that the ordinary gravimetric methods of estimation involve the use of a, large quantity of the alloy. Precipitation of Iron by Ammonia in Presence of Tartaric Acid. W. Strecker. (ClLem. Zeit., 1907,31, 1217.)-The author confirms the statement of Grossmann and Schiick (ANALYST, 1907, 32, 394) that on heating, sodium hydroxide precipitates iron completely from solutions containing tartaric acid.He also shows that animonia causes complete precipitation of the iron on heating, provided that the liquid does not contain more than one molecule of tartaric acid for every 2.5 atoms of iron present. If the quantity of tartaric acid added is increased beyond this pro- portion, precipitation is delayed, becomes incomplete, and finally ceases altogether. Attempts to isolate definite complex compounds of tartaric acid, iron, and ammonia, failed. A. G. L.64 THE ANALYSTo Detection and Estimation of Traces of Manganese. M. Duyk. (Ann. dc Chim. Anal., 1907, 12, 465-466.)-0n adding a trace of a manganese salt to a moderately alkaline 10 per cent.solution of potassium hypochlorite, and then one drop of a 10 per cent;. solution of copper salphate, a precipitate of cupric oxide is produced, and on heating the liquid for a few seconds the manganese dissolves com- pletely, and colours the supernatant liquid violet. This reaction, which is to be attributed to the formation of potassium permanganate, is capable of detecting 0.25 mgrm. of manganese in 4 to 5 C.C. of the hypochlorite solution. As the depth of the colour is proportional to the amount of manganese, the reaction may be made the basis of an exact colorinietric method of estimation. The hypochlorites of sodium or calcium may take the place of the potassium hypochlorite.C. A. M. The Detection of Traces of Mercuric Chloride. K. Kof and H. Haehn. (Arch. der Pharm., 1907, 245, 529-533.)-1f a photographic plate be placed film downwards upon a beaker containing a 2 per cent. solution of mercuric chloride, the surface of which is within 5 to 10 mm. of the film, radiation occurs, which retards the subsequent development of the plate, after at least thirty minutes’. exposure in the dark-room. By placing a strip of glass between the film and surface of the liquid, it is possible to obtain an image of this diaphragm after sufficient exposure. The effect of the mercuric chloride vapour upon the plate may be prevented by adding sodium chloride in excess of saturation to the solubion, but if stannous chloride solution then be added; reaction radiation begins, though it now accelerates the reduc- tion, so that the image of a diaphragm appears as a positive instead of a negative. The radiation reaction is given by solutions containing 0.01 to G per cent.of mercuric chloride, but not by a 0.001 per cent. solution. Solutions of the salt in alcohol, benzene, or toluene also affect the plate, but solid mercuric chloride has no action at a distance of 5 to 10 mm. The vapours from a solution may also be received upon moistened filter-paper, and a colorimetric estimation made by means of hydrogen sulphide. By comparing the stains obtained on a square of filter- paper of equal size (17.3 sq. cm.), impregnated with solutions of mercuric chloride of known strength, it was found that at 13’ C., at a pressure of 763 mm., about 0.00008 gram of mercuric chloride evaporated from a 2 per cent.solution in 165 hours, corresponding to0*000000011 gram per sq. cm. of paper in thirty minutes. If a drop of a 0.01 per cent. solution (0*000005 gram of mercuric chloride) be introduced into a cavity in an object-glass, thin glass strips placed to the right and left to support a photographic plate film downwards, and the whole left for twenty-four hours in the dark-room, the image of the drop will appear as a white spot when the plate is sub- sequently developed. C:A. M. The Use of Nitron for the Estimation of Nitric Acid in Soils and Plants. J. Litzendorff. (Zeits. angcw. Chem., 1907,20, 2209-2213.)-The ‘ I nitron ” method (ANALYST, 1905, 30, 256 ; 1906, 31, 267 ; 1907, 32, 349) is applicable for the estima- tion of nitrates in soil extracts containing about 5 mgms.per 100 C.C. The extracts are prepared by the method of Buhlert and Fickenday by shaking 2 kilogms. of soil with 4 kilogms. of water for half an hour. The extract is then evaporated to aTHE ANALYST. 65 suitable concentration with the addition of magnesium oxide, or else the extract may be filtered cold through a long filter-bag, such as is used for filtering wines, and further quantities of the soil shaken with the same liquid. To 100 C.C. of the solu- tion, heated to boiling, are added 5 to 10 C.C. of a 10 per cent. solution of nitron ” acetate, and the liquid is allowed to stand for several hours in the ice-box; the “ nitron ” nitrate is collected in a Gooch crucible, washed with ice-cold water and dried at 110’ C.I n soils containing not less than 2 to 3 mgms. of nitric acid per 100 grams, the ‘( nitron ” method gives accurate results. With poorer soils, the extracts of which have to be considerably concentrated, difficulties are experienced owing to the accumulation of substances which prevent the crystallisation of the nitron ” nitrate. Similar inhibiting impurities occur in soil extracts which have been sterilised by heat. These substances may be eliminated by means of hydrogen peroxide, which is added repeatedly to the liquid during evaporation. When sufficiently concentrated, the liquid is transferred to a measuring flask and is heated in a boiling water-bath for several hours, further additions of peroxide being made from time to time until the extract is colourless.The (‘ nitron ” method has also been employed for the estimation of nitrates in plants such as mustard. The aqueous extract of the dried plant is clarified with basic lead acetate before precipitating the nitrates. I t would appear that the older methods of estimation, when applied to plant extracts-at any rate, if basic lead acetate be not used-tend to give results which are too high. A correction is made for the solubility of the (( nitron ” nitrate. J. F. B. The Estimation of Phosphorous Acid. C. Marie and A. Lucas. (Comnptes Re?tdzis, 1907, 145, 60-62.)-The method depends upon the oxidation of the phosphorous acid by means of alkaline permanganate solution, the excess of which is titrated back with standard ferrous ammonium sulphate solution.A sufficient quantity of potassium carbonate to give an excess of 3 grams per 100 C.C. of the final solution is heated to 80° C. on the water-bath with 50 C.C. of $; potamium permanganate solution, the solution of the phosphorous acid or phosphite (0.2 to 0.3 gram) introduced, and the temperature maintained at about 80’ C. for fifteen minutes. A standardised solution of ferrous ammonium sulphate containing 10 per cent. of sulphuric acid is then introduced, and the excess of ferrous salt titrated back with the standard permanganate solution. A convenient method of standardising the latter is to heat it on the water-bath with alkali carbonate and a weighed quantity of calcium formate as in the phosphite estimation, the process of oxidation being as follows : 4KMn0, + 3(HC02),Ca = 4Mn02+ 3CaC0, + K2C0, + SKHCO, + 2H,O.C. A. M. The Analysis of Sodium Peroxide. R. Niemeyer. (Chem. Zeit., 1907, 31, 1257.)-Although the gasometric method for estimating the available oxygen in sodium peroxide (ANALYST, 1906, 31, 313) gives results which are about 0.8 per cent. higher than those obtained by the permanganate method, the author considers that the latter method, on account of its simplicity, is to be preferred. The estimationTHE ANALYST. should be carried out as follows : About 0.2 gram of the sodium peroxide is placed in a small capsule (made from the bottom of a test-tube) contained in a stoppered weighing bottle, and the whole is weighed. The weights of the capsule and bottle are ascertained previously.The capsule is then removed from the bottle by means of a pair of forceps and introduced into a beaker containing 500 C.C. of water, the capsule being held in a sloping position, so that it sinks to the bottom of the beaker. Care must be taken that no particles of the peroxide come to the surface of the water. An excess of eulphuric acid is added, with stirring, and the solution is titrated with -& perinanganate solution. w. P. s. Estimation of the Strength of Concentrated Sulphuric Acid. E. Buch- wald. (Chem. Zeit., 1907, 31, 1256.)-In the volumetric estimation of sulphuric acid the accurate standardisation of the alkali solution used is of the utmost importance, especially in the analysis of concentrated sulphuric acid, where any slight error is multiplied to a considerable extent in the subsequent calculation.The author recommends that the alkali solution be standardised on an acid solution the strength of which has been ascertained by titrating the iodine liberated from a mixture of potassium iodide and iodate by a known volume oE the acid. This iodine is titrated with a thiosulphate solution itself titrated against pure iodine. The standardisation of the acid may also be checked by titration on a known weight of Iceland spar, and, furbher, if hydrochloric acid be employed, a known volume may be precipitated by the addition of silver nitrate and the silver chloride weighe'd. I t is preferable to treat the alkali solution with barium hydroxide to remove carbonates, the excess of barium being afterwards precipitated by means of sulphuric acid, aud to use phenolphthalein as indicator in the titration of the sulphuric acid under examination.w. P. s. Separation of Thorium, Titanium, and Zirconium from Iron. M. Dittrich and S. Freund. (Zeits. Anal. Chena., 1907, 56, 348-352.)-The mixed oxides are fused with potassium bisulphate, the melt is dissolved in cold water and saturated with sulphuretted hydrogen. The filtered liquid is diluted to 600 c.c., sodium acetate is added, and the whole boiled for some time in a flask through which carbon dioxide is passed, the water lost by evaporation being replaced from time to time. The precipitated hydroxides are ignited, weighed, and converted into nitrates by fusion with potassium bisulphate, precipitation of the aqueous solution by ammonia, and solution of the hydroxides in nitric acid.In this solution, zirconium and thorium are separated from titanium by means of ammonium salicylate (see p. 67). The precipitated hydroxides of zirconium and thorium are ignited, weighed, and fused with potassium bisulphate. The melt is dissolved in cold water, and ammonium oxalate added until the oxalates precipitated at first are completely redis- solved. The clear solution is heated to boiling, and thorium oxalate precipitated by adding boiling concentrated hydrochloric acid, and heating on a water-bath for another two hours. The precipitate is washed with a 5 per cent. solution of ammonium oxalate in dilute hydrochloric acid (1 : 4). Zirconium may be precipitatedTHE ANALYST. 67 with ammonia in the filtrate, after destruction of the oxalate with sulphuric.acid. Test results quoted are satisfactory for all four metals. A. G. I;. Simultaneous Precipitation of Titanium and Zirconium in Presence of Iron. M. Dittrich and S. Freund. (Zeits. Anal. Chem., 1907, 56, 337-343.)-Four slightly different methods are described, all depending on the precipitation by prolonged boiling of titanic acid and zirconium hydroxide from nearly neutral solutions containing the iron in the ferrous state. In three of the methods the mixture of the three oxides, obtained as usual, is fused with potassium bisulphate, the melt is dissolved in water, and sulphuretted hydrogen is passed through the liquid until any platinum (from the crucible) present is precipitated, and the iron is completely reduced.The liquid is next filtered into- a large flask, neutralised with sodium carbonate, and made slightly acid with sulphuric acid. Sulphuretted hydrogen is passed into the flask, the excess being removed by heating and passing a stream of carbon dioxide through the liquid. A solution of about 5 grams of sodium acetate, sodium thiosulphate, or, preferably, ammonium sulphate, is then added from a tap-funnel, and the boiling in the current of carbon dioxide is continued for another one or two hours, when the precipitate, which should be quite white, is filtered off, washed, ignited, and weighed. Any oxides adhering to the flask may be brought into solution by warming with strong sulphuric acid and hydrogen peroxide ; the solution obtained is diluted, precipitated with ammonia, and the hydroxides added to the main precipitate.The fourth method consists in dissolving the bisulphate melt in water and hydrochloric acid, reducing the iron by means of sulphur dioxide, nearly neutralising the liquid with ammonia, adding an aqueous solution of sulphur dioxide and precipitating the hydroxides of titanium and zirconium by boiling in a current of carbon dioxide as above. In all cases iron can be directly determined in the filtrates as usual. Titanium is determined colorimetrically, and zirconium is taken by diffarence. Test results quoted are satisfactory. A. G. L. A New Separation of Titanium and Zirconium. M. Dittrich and s. Freund. (Zeits. Anal. Chewz., 1907, 56, 344-345.)-The dilute solution of titanium and zirconium nitrates is nearly neutralised with sodium carbonate, and is then added drop by drop to a boiling solution of 10 grams of ammonium salicylate in 50 C.C.of water. Boiling is continued until the volume of the solution has been reduced to 150 to 200 c.c., when the precipitated zirconium salicylate is filtered off hot, washed with a boiling solution of ammonium salicylate until perfectly white, ignited, and weighed as ZrO,. The oxide should be tested for titanium with hydrogen peroxide, and the separation repeated, if necessary. The results quoted are satisfactory. Titanium and thorium can be separated in the 8ame way. A. G. L. Rapid Electrolytic Estimation of Zinc. F. C. Frary. (Zeits. afzgew. Chem., 1907, 20, 2247-2250.)-The author questions the accuracy of the estimations of zinc made by Exner and Ingham, and by Langness (ANALYST, 1907, 32, 269), with the aid of a rotating anode, chiefly on the ground that apparently the liquids left atTHE ANALYST+ the epd of each experiment were not tested for the presence of zinc. Using his own method of electro-magnetically rotating the liquid, and testing for zinc at the end of each experiment, he finds that 0.1 gram of zinc can be completely deposited from alkaline solutions in thirty minutes, with a current of 4.5 a m p h s , at about 4 to 5 volts. For quantities of zinc of 0.2 gram a current of 4 to 5 amperes was used for the first fifteen minutes, and then 1.5 amperes for another twenty minutes, when deposition was complete. I n each case the volume of the electrolyte was 100 to 125 c.c.; it contained eight grams of sodium hydroxide, and was cooled by an external lead coil, through which water circulated. The cathode consisted of a cylinder of nickel gauze, closely fitting the beaker ; the deposited zinc was removed by means of dilute sulphuric acid, the cathode losing up to 0.01 gram (average, 0-001 gram) in weight by this operation. The author noticed the formation of a yellow substance, apparently an oxide of platinum, in very small quantity, on the platinum spiral used 8s anode in his experiments (cf. ANALYST, 1908, 30). A. G. L.