202 TPEEE ANALYST. MONTBLY RECORD OX GENERAL RESEARCHES IN ANALYTICAL CEUZMIBTRY. THE past few weeks have been very productive of research-work in Analytical Chemirs- try ; in Geiman laboratories several new methods and interesting separations have been worked out ; and of these Pi*esenitw’ laboratory at Wiesbaden takes, of course, the lend. Among other resemches which have lately been brought to completion in this laboratory is a new method for the determination o f arsenic. THE DETERMINATION OF ARSENIC*-BP CIARL HOLTHOQ. Difficulties have always attended the estimation of Arsenic, due pmtly to the volatility of ai*senious chloride, in the presence of concentrated hydrochloric acid, and partly to the imperfect precipitation, by aulphmetted hydrogen, of solutions containing arsenic.Again, the only safe method for the separation of arsenic from antimony, is the one proposed by R. Bunsen,t in which the antimony is removed aspentasulphide in the presence of concentrated hydrochloric acid. According to the directions given by him, the arsenic, after the excess of sulphuretted hydrogen has been expelled by a current of: air and the solution heated with chlorine water, is precipitated as As2S,. This does not give as good results for the arsenic as might be wished, and the author, therefore, proposes to determine it in the strongly acid solution, which remains after the pi*ecipitaa tion of the antimony, by a volumetric method, which was suggested by Mob*, but has never been worked out by him. Xt consists, essentially, in ’ reducing the arsenic acid, either before or after removing the hydrochloric acid, and then titrating with standard “Zeitsuhrift fiu.Anal. Chemie., 23,378. ?Liebig’s Annalen A., Chemie. 192, 805.THE ANAILYST. 203 iodine solution. The author assumes that the reduction is complete when.sulphu- rous acid is used (this has been proved by WGhler Ann. A. Chem. u. Pharm. 30,224): A large quantity of the sulphurous acid is used, the, excess of which is removed by a current of air, or by evaporating to Q ; a platinum-spiral being placed in, the solu- tion to facilitate the escape of the gas. According to the author's observations, if the hydro- chloric acid be neutralized before the reduction, the .results obtained are rather high, due to the large quantity of alkaline chlorides which are thus produced.The best results are got by evaporating off the acid, and then reducing. This course is further warranted by the following facts : ftrstly, that arsenic chloride is not reduced, in fihe presence of chlorine, by boiling hydrochloric acid ; and, secondly, that hydrochloric acid of Sp. Gt. 1.10, containing arsenic, distils, after the addition of potassium chlorate, free from arsenic. To test the method, the following experiment, among others, was made : 50 C.C. containing 0.1814 As,O8 were taken and evaporated, .with 200 C.C. chlorine- water and 1 g m . ICCI03, to dryness. The residue was dissolved in 200 G.C. of sulphurous acid, and as much distilled water ; the solution was heated. for half an hour to near boiling, and then evaporated to *, another 200 C.C. of SO, solution were added, and the whole was finally boiled, with the platinum-spiral to Q.I On titrating, 38*9 C;C. iodine solution were required, this was equivalent to 0.1323 grm Ae203, whereas -the theoretical amount was 0.1330 ; a result which leaves little to be desired. Further researches of the author show that it is not necessary that the solution should be boiled, or even kept long near boiling, to insure a complete reduction with sulphurous acid. THE DETERXINATION OF MOI;YBDENUM.-BP OTTO FREIHERR VON DER - PFORDTEN.* The author has been investigating the different gravimetric methods for the deter- mination of molybdenum, and has improved upon the volumetric method, which consists in titrating with potassium permanganate. The following is a short rdszcmd of his researches.I. Gravimetric Methods. Phe author made use of the ordinary ammonium. molybdate, 3(NH4),0, ?MOO, (a) By reduction of molybdic acid to the metallic state. U. Rounnelsberg recommended heating molybdic acid in a platinum tube, though which a current of hydrogen was being passed, to reduce to the metallic state (Pogg. Annal., 127, 281) ; while Debray proposed to reduce first to a lower oxide in a platinum crucible, and then to complete the operation in the plafinum tube. The author has simplified the method by doing away altogether with the platinum tube; his researches &ow that it is possible to complete the reduction in a current of hydrogen, using simply a platinum crucible. The latter is fitted with a perforated platinum cover (porcelain may be used), through which the hydrogen enters by means of a suitable tube.The heat is supplied by the blowpipe flame. The analysis of ammonium molybdate is conducted in * Zeitsohrift fiir And. Chemie., 23, 413. + 4H,O, in his experiments.204 THE ANALYST. the following may :--The crucible is first kept at a temperature of 1'70" C. for several hours in the air-bath; this prevents spitting during the ignition. It is then gently heated in a slow current of hydrogen, tho heat being allowed to rise slowly. The reduction is completed by a strong ignition xith the blow pipe, about half an hour being required for 0-2 grm. of metallic molybdenum. Care must be taken, after the operation, to well clean the crucible by ignition and successive treatments with nitric acid and ammonia. This method is applicable to neutral solutions containing molybdic acid, if they are first-precipitated by mercurous nitrate according to Rose's method.(6) By reduction of molybdenum trisulphide to disulphide. This is suited for the analysis of acicl solutions containing molybdic acid. It was &st described by T. Paul Liechti and B. Hempe. The acid solution is made alkaline with ammonia, ammonium sulphide is added, and it is then allowed to stand €or 12 hours. The molybdenum forms a molybdic trisulphide, which is soluble in excess of ammonium dphide to a deep brown liquid. The solution thus obtained is decomposed by adding sulphuric acid in excess; and the precipitate of molybdic ti*isulphide and sulphur'is collected upon a weighed filter, mashed with mlphuretted hydrogen water, and dried at 100-105° C.till the weight remains constant. A weighed part of the dried precipitate is then converted by ignition, in an atmosphere of hydrogen, to molybdic di- sulphide ; and from this the molybdenum is calculated. The author obtained good results with this method, but recommends heating with a simple Bunsen burner, as a too powerful ignition with the blow pipe causes a partial reduction to the metallic state. 11, The Polumetrio Nethod. The author reduces the molybdic acid completely in hydrochloric acid aolution, and then titrates with potassium permanganate, without excluding the air. The suboxide, first formed, is subsequently transformed into sesquioxide. The analysis is conducted thus :-To the solution of the salt, containing about 0.3 grm.MOO,, is added 50-60 C.C. of a 27 per cent. solution of hydrochloric acid, together with 8-10 grms. of zinc, in which the amount of iron has been previously determined by titration. As soon as the solution assumes a yellow colour the vessel i s cwkjd, before all the zinc has been used; and its contents are washed into a porcelain dish, containing 40 C.C. of a daute sulphurous acid solution and 20 0.0. of a solution of manganous sulphate. An equal quantity of water (about a litre) is now added, and a considerable amount of standard permanganate solution run in, the liquid being, meanwhile, well stirred. The titration is complete when the solution becomes faintly pink. Allowance must be made for the iron contained in the zinc, and the permangate required to colour the mass of liquid.The results are accurate. The mean of 14 analyses was 81.52 per cent., MOO, (the maximum being 81.78 per cent,, the minimum 81.28 per cent.) while the calculated quantity was 81.55 per cent. THE DETERMINATION OF PHOSPHORIC Acm. Otto Freiherr von der Pfordten, the author, by ingeniously combining the volurne- tric estimation of Molybdenum (as described above), Fith the precipitation of pholrrphoric acid by ammonium molybdate, has produced it useful titration method for the determina- tion of phosphoric acid.The phosphoric acid is fir& precipitated by ammonium molybdate, in the usual manner. (Note.-The ammonium rnolybdate solution must be clear, and filtered from any deposit before use).To cause the precipitate to separate out better, the beaker containing the solution should be warmed in the water bath; the precipitate is thus freed from molybdic acid. The ammonium phospho-slolybdate thus obtained, is washed with the filter pump, by a nearly saturated solution of pure ammonium sulphate, till a drop, on the addition of ammonium sulphide and weak acid, gives no dark colouu.. The precipitate on the filter is then dissolved in a small quantity of ammonia, and diluted to a knom volume, of which a measured portion, containing at the most 0.3 grains, but not more than 30 0.0. is used for the reduction, This takes place as described above. It is advisable to make several titrations, and to take the p a n of- these. From the permanganate required for the oxidation, first the Wqlybdenum is calculated, and from this the amount of phosphoric acid is deduced.A Ferric phosphate contained, according to gravimetric methods, 36.99 per cent, P,O,; the volumetric method gave from 35*85 to 35.97. A Guano-phosphate gave The method is always applicable when the phosphoric acid can be separated by ammonium molybdate. The author recommends it egpecially for cases where the pre- sence of other bodies (Fe, Al, &a) has hitherto prevented the use of a titration method, CraVirnetricaUy - - - 21-19 per cent. P,O, Volwnetdcally - I 21.88 THE DETERMINATION OF PHOSPHORIC ACID IN Sorr;s.-P. DE GASPARIN. P. de Gasparin gives in the Comptes Rendus (96, 314) the following method for the estimation of phosphoric acid in soils :-20 grms.of the finely powdered and sifted earth are treated in a porcelain dish, with sulphuric acid (I : 5 ) as long as any effer- vescence takes place ; 80 c.c:of aqua regia (lHN0, : 3HC1) are then added, and the mixture heated on the water bath, till the liquid becomes syrupy, diluted with cold distilled water, and washed on to the filter with hot water. The filtrate is precipitated with ammonia, collected on the filter and dried. The dry precipitate is heated in a plat- inum crucible to redness, digested with cold dilute nitric acid ( I : 40) and filtered. The filtrate contains, according to the author, all the phosphoric acid. It is concentrated on the water bath, p~ecipitated with molybdic acid, and the phosphoric acid finally deter- mined in the ordinary way as magnesium pyrophosphate.T H ~ DETERMINATION OF THE TECHNICAL WORTH OF CALCIVM TARTRATE.-~EI~ERT.~ The principle of this method depends upon the fact, that calcium. tartrate is decom- posed by boiling potassium carbonate (1 to 2 hours on the water bath) into neutral potassium tartrate and calcium carbonate. The filtered solution is evaporated ; enough concentrated acetic acid is added to the warm liquid, to form wine-stone (bitartrate of , ._ - t Zeitsohrift fiir And. Chem. 23, 358.PO6 THE ANALYST, - ~ - - potash), and the whole allowed to stand for some hours. It is then treated with alcohol (90 per cent.), filtered, mashed with alcohol, and h&y.titrated. The correctness of the results depends upon the following precautions :-(1) The potassium carbonate must be added only in slight excess ; (2) Acetic acid must be added only in a corresponding excess ; (3) The washing must be carried on sufficiently long ; (4) The mixture of potas- sium carbonate nnd neutral calcium tartrate, to which the acetic acid has been added, must still contain water on the addition of alcohol. For titration a potash solution is used, which serves for estimating the acidity of wines.Its strength is snch that 1 c,c, will neutralise 0.01 grm. tartaric acid or *02608 bitartrate (mine-stone). The following correction is given by the author for the bitartrate remaining in. solu- tion for every five grms. of calcium tartrate, 0.33 per cent. bitartrate of potaah i s to be added to the quantity which has been found by titration.F. H. H, Bonn, 21st Oct., 1884. IN the British Hedioal Jownzal for 11th October, are found the following :- RENARKS ON TESTS FOR ALBUMEN IN TEE URINE, NEW AND OLD. BY GEORW JOHNSON, M.D., F.B.S. IN a paper on the above subject in the recently published Xanohestw Midioal Chronicle, Dr. William Roberts, referring to the fact that the urine in health contains various forms of albuminoid matter, expresses his belief that the new tests for albumen which have recently been brought into prominence, especially picric acid, tungstate of soda, potassio- mercuric iodide, and the acidulated brine-test, ‘6 produce frequently in the wines of perfectly healthy persons, a reaction which is undistinguishable from the reaction which indicates digease or abnormality.” This point was put to the proof by the examination of the urine of thirty-one healthy persons-students, candidates for insurance, and others, who exhibited no signs of disordered health, and in whose urine heat and nitric acid gave no indication of albumen.Dr. Roberts, of course, needs not to be reminded that albumen, in greater or less abundance, and for long periods of time, may be unquestionably present in the urine of persons who exhibit no signs of disordered health. If this were not so, albuminuria would not be so frequently unsuspected and overlooked as it is. Dr. Roberts proceeds to state that ( I the acidulated brine-test gave a reaction in eleven cases, picric acid in fourteen, the tungstate test in twenty-eight, and the mercuric iodide in tmenty-nine cases.” Deferring for the present what I have to say of picric acid, I should have expected, from observations which I have quite recently made, that the other three tests would give a slight but appreciable reaction in every specimen of normal urine.It is a fact; that all normal urine contains a small but variable proportion of mucus. (See the-article Mucus,” in Watts’s Dictionary of Ohemisty, vol. iii., p. 1059.60,) It is also precipi- tated, as Dr. Oliver has shown ( I Bedside Urinary Testing,’ p. 37), by citric acid. The addition of a small quantity of acetic or citric acid to normal urine gradually renders it Now, mucin is precipitated by dilute acetic acid and mineral acids.THE ANBEPST. 207 slightly but decidedly turbid, by coagulating the mucin ; and Dr. Roberts mentions the fact that, when nitric acid is added to albuminous urine, the albumen is thrown down just above the line of junction of the two liquids, while the miicin is brought into view towards the upper part of the column of urine, where it gradually forms a-diffused haze quite distinct from the opalescent haze at the line of junction.” To this I may add that, when nitric acid is placed at the bottom of a column of normal urine, a diffused ham of coagulated mucia may commonly, after a time, be seen near the upper part of the column.Seeing then that mucin is precipitated by both mineral and vegetable acids, we are at no loss to understand that any test containing one or other of these agents should give a reaction with normal urine. The acidulated brine contains hydrochloric acid, the tungstate of soda and potassio-mercuric iodide require the addition of either citric or acetic acid before they act as albumen-precipitants ; and they one and all, by the reaction with mucin, slowly cause, in most, if not all, normal urines, a cloudiness more decided than that which results from the action of the acids alone.Withpicric acid, however, the oase is entirely different. In the form of a saturated aqueous solutioD, and uncombined with any other agent, it is a most delicate albumen-precipitant, but it gives no precipi- tate in normal urino unless an acid, such as citric or acetic acid, be added to it. This can readily be proved by the following experiment. Take about a drachm of freshly passed normal urine, and add an equal bulk of picric acid solution.The yellow mixture will remain quite clear, unless, as sometimes though rarely happens, some turbidity results from a deposit of urates, which would be at once removed by heat. Now add a few drops of dilute acetic or citric acid, and the mixture will, in aminute or two, become hazy from precipitated niucin, the haziness occurring much more slowly than the immediato opalescence, which results from the presence of a slight trace of albumen, but, like that, being unchanged by heat. Another experiment consists ill adding acetic or citric acid to normal urine, then, after waiting a minute or two to complete the coagulation of the mucin, passing the urine through a filter and adding picric acid to the filtrate; when the mixture will remain quite free from turbidity.I have tested many hundred specimens of normal urine with picric acid, and I confidently assert that in such specimens, no precipitate or haziness occurs when unmixed picric acid is used as the test-agent ; and it may be that the different results with this test obtained by Dr. Roberts are due to his having added acetio or citric acid to the picric acid in his experiments. The only precipitates other than albuminous which may result from pic& acid, employed alone, are urates which rarely occur, except when the mixture is allowed to stand for some time ; peptones which I have met with only twice in as many years; and vegetable alkaloids, such as quinine, *hen lmge doses are being taken. These all differ from an albuminow precipi- tate in the fact that they are readily and completely redissolveii by heat, while thsy may be distinguished from each otlier by the microscope‘ (See the author’s locttires OIL ‘ Albumen tmJ Sugar Testing,’ p.11, Smith, @der and Go.)-____--- 208 THE ANALYST. - ---________ ~ - - It appears, therefore, from very numerous and careful observations, that albumen is the only substance found in the urine which gives with picric acid aprecipitate insoluble by heat. The difference, then, between pic1.c acid and the other new tests for albumen is this-that picric acid, unmixed with other reagents, while it is a most sensitive and trustworthy test for albumen, gives no reaction with mucin. On the other hand, the potassio-mercuric iodide, tungstate of soda, and brine do not precipitate albumen unless when combined with an acid ; and this combination gives a reaction with mucin, which is not distinguishable €rom a minute trace of albumen.I have been in the habit of using the potassio-mercuric iodide only as a check upon the picric acid test, when small quantities of albumen only were present, and, until lately, had not thought of applying it to normal urine. I now find, however, that the test-liquid, when acidulated-as it must be, to act at all-gives a distinct opalescence in most, if not all, normal urines. I find, too, that after the ~nucin has been removed from normal urine by its coagulation with acetic or citric acid, and subsequent filtration, the addition of the potassio-mercuric iodide to the filtrate causes a decided opalescence, which is probably due to the precipitation of some substance other than mucin in the urine.In testing urines which contain a mere trace of albumen, it is important to remove any turbidity that would interfere with the process. Urates would be removed by heat, suspended mucus and other particles by filtration. The addition of the picric acid solu- tion to a twbid specimen might give a fallacious appearance of coagulated albumen, when, in fact, there is nothing more than Bome hicreased opacity, due to the yeUow staining of the suspended particles. Picric acid is itself sufficiently acid, when added iu excess, to dissolve and clear a, phosphatic tuybidity. In the rare case of the urine being so highly alkaline as to prevent the coagulation of the albumen by an excess of picric acid, the plan is to add sufficient citric or acetic acid to neutmlize the alkali, then to filter, and add the picric acid to the fi1ti4at e.It appeai's, then, that picpic acid as a test fop albumen is more free from fallacy than any other, not even excepting heat and nitric acid, which Dr. Roberts expresses his deter- mination to fall back upon. Of course, in a doubtful case, no one would neglect to apply more than one test. That pic& acid is a more sensitive test than heat and nitric acid js eaaily proved by taking a highly albuminous specimen aYrd gradually diluting it up to the point where-though theae testa fail to detect it-picric acid still gives a diatinot reaction. The Main advantages of picfic acid as ft teat for albumen are the following :-It hstantly detects a saall amount of albumen which nitric acid would indicate only slowly or not at all; while, on the one hand, anjnsufficient addition of the test does not, as is the case with nitric acid, prevelit the subsequent coagulation by heat ; neither, on the other hand, does an excess of picric acid redissolve the precipitate, as does an excess uf nitpic acid. Por bedside urinary testing, the portability of the innocuous powder is a great convenience. $he fact that, with caustic potash, it is an infallible qualitative and quantitative twt for augai; may be said to be more than double its value as an uainaryTHE ANALYST. 209 - -------- test. For bedside use, Mi. Hawksloy, 35’7, Oxford Street, makes a waistcoat-pocket test-case, consisting of a test-tube four inches long, in which are packed two smaller tubes, one containing picric acid powder, the other grain-lumps of caustic potash, and also a s m d spirit lamp, These are enclosed in a metal case, not; much larger than a pencil-case. Another small case contains a nipple-pipette, which, amongst other uses, is convenient for conveying urine from .the vessel to the test-tube. The picric acid which is used for sugar-testing should be purified by recrystallisa- tion. The commercial samples usually give a red colour when boiled with liquor potassm; and 1 lately saw an impure sample, which not only gave this red colour, but the liquid was rendered turbid by fine granules. The impurity was removed by solution and recryst allization .