UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of the estimation of iron and the separation of manganese from iron by phenyl-nitroso-hydroxylamine… Fulton, Ruth Vivian 1919

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1919_a8_f8_s8.pdf [ 15.44MB ]
JSON: 831-1.0062363.json
JSON-LD: 831-1.0062363-ld.json
RDF/XML (Pretty): 831-1.0062363-rdf.xml
RDF/JSON: 831-1.0062363-rdf.json
Turtle: 831-1.0062363-turtle.txt
N-Triples: 831-1.0062363-rdf-ntriples.txt
Original Record: 831-1.0062363-source.json
Full Text

Full Text

A STUDY OF THE ESTIMATION OF IRON AND THE SEPARATION OF MANGANESE FROM IRON 3Y PHENYLNITROSO-HYDROXY1AMINE AMMONIUM (CUPFERRON! THESIS Submitted as part of the requirements for the degree of Master of Arts in the University of British Columbia by Ruth V. Fulton, B. A. University of British Columbia Department of Chemistry 1919. 1. INTRODUCTION The use of cupferron as a reagent for the separation of iron from aluminium has "been the subject of a limited number of investigations. Several of the conditions which ensure a complete precipitation of the iron have, however, received little, if any, attention. The results thus far obtained for the separation of iron from manganese would indicate that a complete separation can be attained here as well, but the evidence on this point is not sufficient to warrant the conclusion that this method may be substituted for the old and ^exj troublesome methods of separating iron from man-ganese. In view of these conditions, it seemed ver-y desir-able to make a study of the precipitation of the iron and its separation from manganese, under varying conditions, by means of cupferron. SEPARATION OF MANGANESE AND IRON BY OTHER METHODS In considering other methods used for this separ-ation, we find that two of the best standard methods are:-£. 1. Basic ace ta te method as described in S c o t t ' s ' 'Standard Methods for Chemical Analysis" . 2 2. Bothe ' s "Ether Separation of Iron" as described in Olsen's "Quantitative Analysis". 1. By the first method, iron, aluminium, titanium, zir-conium and vanadium may he separated from manganese, zinc, cobalt and nickel. By treatment with the proper reagents, these several metals are converted into their acetates, the solution being slightly acid. .Solutions of the acetates of the first group of metals decompose on heating and insoluble basic acetates are precipitated. The other acetate solutions remain the same when boiled a short time. Sodium acetate is added to reduce the ionization (and thus the dissolving power) of acetic acid. The procedure is as follows:- To the cold acid solution of the chlorides of the metals, an aqueous solution of sodium carbonate is added until the precipitate formed just disappears. A few drops more of a dilute solution of sodium carbonate is then added until the opalescence is permanent (i.e. until all free acid is neutralized). The color will be dark red if much iron is present, idd 3 c.c. acetic acid to dissolve slight precipitate. Dilute to 1. Quantitative Chemical Analysis by J.C.Olsen, A.M., Ph.D. (D.Van llostrand Co.) P.157 2. Standard Methods of Chemical Analysis by W.W.Scott 3. 500 c.c., boil, and add 6 c.c. of a ZCfl sodium acetate solution. Boil 1 minute. Let the precipitate settle er<? filter while hot. Wash with hot 50$ sodium acetate three times. Wash the precip-itate into the original beaker with hot 1:1 hydrochloric acid solution. A second precipitation must be made in order to free the iron hydroxide from manganese, zinc, etc. These details are sufficient to show that while good results may be obtained by this method, it requires much care and is veiy^ tedious. 2. The second method is more rapid and convenient than the first. Iron may be separated from chronium, aluminium, manganese, cobalt, nickel and copper. Ferric chloride can be completely extracted by ether from a hydrochloric acid sol-ution containing the metals mentioned, nitric acid, chlorine and more than small amounts of sulphuric acid must be absent. The solution should be cold lest ferric chloride be reduced by ether. All suspended matter must be removed by filtration. Ford's method is carried out by converting mangan-ese present in strong nitric acid solution to manganese di-oxide by potassium chlorate. This is filtered off from the iron which remains in solution. The manganese seperated in this way may be determined graviraetrioally or volumetrically. 1. Quantitative Chemical Analysis by -T.C.Olsen, i.M., Ph.L. (D.Van Nostrand Co.) P.285 4. .Another separation of iron end manganese is given "by U.B.Moore and Ivy Miller . It is "based on the fact that if pyridine is added in slight excess to a solution of ferric chloride containing free hydrochloric acid, the iron is com-pletely precipitated as hydroxide. If manganese is present, it remains in solution. Pyridine added to a neutral solution is warned. Then manganese is slowly oxidized and a precipitate results. The rate of oxidation is approximately one-third as rapid as when the solution has "been treated with ammonium chloride end ammonium hydroxide. If hydrochloric acid is present, pyridine in slight excess does not bring down a precipitate even when the solution is heated for 10 minutes. Pyridine is therefore a better reagent than ammonium hydroxide for this separation, especially when a large amount of mangan-ese is present. Ferric hydroxide,thus precipitated may be washed with pyridine water since it does not carry through any iron. The source of the iron in this experiment was pure iron wire. To this were added 10 c.c. of normal hydrochloric acid and a few drops of nitric acid. The solution was warmed to oxidize the iron. After diluting the solution to 100 c.c, it wes again warmed and pyridine was added from a burette. 1. Jour.Am.Chem.Soc., 20,593, (1908). Iff 5. Iron and manganese were "both estimated to see what effect the presence of pyridine has on the precipitation of manganese. It was found to have very little if any effect. PREVIOUS WORK It will he of interest to "briefly review the several cases in which cupferron hes "been used for quanti-tative separations. Baudisch found thet cupferron forms complex salts with iron and copper and that it may therefore he used in the quantitative separation of copper and iron from nearly all metals and, under certain conditions, copper from iron. £ This was tested out by Biltz and Hodke with quite satisfactory results. A 6?o solution of cupferron was used. Iron was precipitated quantitatively from solutions contain-ing as much as 20^ "by volume of concentrated hydrochloric acid and thus it could he separeted from aluminium, nickel and chromium to fifty times the amount of iron. Copper was precipitated from slightly acid solutions, such as acetic acid, even in the presence of zinc or cadmium. Copper and iron are precipitated end separated by dissolving the copper salt in ammonium hydroxide. Maximum errors were .7 mg. (1 part in 200! except in the last mentioned separation of 1. Chem. Abstracts. 4,557, 1910. Chem.-Ztg., 23,1298-1300. B. Chem. Abstracts. 4,2783,1910. 2. Amorg.Chem., 66,426-430, 6. iron and copper, in which there were large errors. Hanus end Soukup worked on the separation of copper from cadmium end zinc by cupferron. They found that copper could be precipitated from acid solution if the reagent was used in great excess and filtering was carried out at once. On standing, copper goes into solution. Cadmium end zinc were thus separated from copper and precipitated after excess of the reagent was destroyed with nitric aced. But this was found to be no advance over older methods. In 1911, Fresenius published an investigation of the uses of cupferron in quantitative analysis. The cost of cupferron was §£.25 per 100 gms. It dissolved readily in cold water and solution remained the same for weeks. If sdded to a not very acid ferric solution slowly, at room temperature, iron gave a red flocky precipitate. Excess of the reagent brought down 8 white finely crystalline precip-itate of nitrosophenylhydroxylemine (except when acetic acid was used). On stirring, the precipitate turned to dense crystalline masses which crushed easily. After 15 minutes, this was filtered, washed free from acid with ammonium hydroxide to remove nitrosophenylhydroxylamine. The pre-cipitate wss heated cautiously in a covered platinum crucible, then ignited. Hydrochloric acid, sulphuric acid, 1. Chem. Abstracts. 4,2175,1910. Z. Amorg.Chem., 68, 55-56. 2. Chera. Abstracts. 5,1718,1911. Z. Anal.Chem., 50, 25-42. 7. acetic acid or even nitric acid solutions were used and a definite excess of acid was not necessary (£0 c.c. excess dilute acid to 150 c.c. colution was "best;. Also aluminium, chromium, zinc, nickel, cobalt, "barium, potassium and ammonium salts and sodium acid phosphate did not interfere. It was found that iron in manganese ores and in ferromangan-ese could he precipitated and washed free from manganese in one precipitation. The filtrate, however, contained decom-position products of nitrosophenylhydroxylamine which hastened the precipitation of manganese "by ammonium sulphide hut retarded washing and filtering of manganese sulphide and made its incineration less simple. Another difficulty was that organic compounds present interfered with the separation of aluminium "before precipitating manganese in ores as man-ganese sulphide. It was found that mercury, lead, tin, "bismuth and silver were precipitated "by cupf err on. In the separation of copper from iron, the results for iron vere a little high. Nissenson separated iron from nickel and cohalt. He prepared a solution of 1 gm. of substance to he analyzed, dissolved it in concentrated hydrochloric acid saturated with bromine. The solution was evaporated and the residue was treated with dilute sufchuric acid and evaporeted till sulphur 1. Chem. Abstracts. 5,3024,1911. Z. Angew. Chem., 23,969. 8. ! • ; -trioxide fumes appeared. Hydrogen sulphide was passed in and the solution v;as filtered. The filtrate was heated to expel hydrogen sulphide, oxidized with hydrogen peroxide or persulphate and a solution of 8 gms. cupferron to 100 gms. water was added slowly. The iron precipitate was filtered off. The filtrate was evaporated and taken up with water and electrolyzed to get the nickel and cobalt, after adding ammonium hydroxide. 1 Bellucci separated titanium from weakly acid solution (hydrochloric acid or sulphuric acid) "by a cold 6^ solution of cupferron, as j^E/lo) NO^Tij . Cupferron was added till a white precipitate in addition to the yellow precipitate formecl. This was filtered after several minutes, washed with cold water, dried in a steam oven, ignited and weighed as TiOj,. It was found that titanium might thus be separated from aluminium; also iron, chromium, manganese, nickel, cobalt, zinc and alkaline earths might be separated out in the presence of aluminium. In separ-ating titanium from aluminium, it was not necessary to wash the precipitate with ammonium hydroxide as in iron determinations since an excess of cupferron did no harm. Also it was not necessary to filter at once. Separations using molecular proportions of 1 Ti to 50 Al gave good 1. Chem. Abstracts. 7,1688,1312 and 7,2582,1912. Atti accad. Tincei, 22, 1, 20-24. Gazz. chim. ital. 42,1, 570-576. j. results. Thornton , in his work on the separation of titan-ium from iron, precipitated iron as FeS. Titanium was precipitated by cupferron added to the filtrate. He also found that zirconium could be quantitatively separated from aluminium by cupferron, but phosphoric acid comes down too. i This was removed by fusion with sodium carbonate. The iron wss reduced and precipitated with hydrogen sulphide. Then 40 - 60 c.c. 1:1 sulphuric acid was added, hydrogen sulphide was boiled out and the solution diluted with ice water to 400 c.c. The cupferron solution was added slowly, while stirring. The precipitate was washed with dilute hydro-chloric scid, dried, ignited and weighed as ZrOx. Rodeja" separated vanadium from phosphorus, using vanadyl salt, since the precipitation was not com-plete with vane die salt. A 5$ cupferron solution was added until white nitrosophenylhydroxylamine formed and a few c.c.'s more. Phosphorus which remained in the filtrate was determined by evaporating, dissolving the residue in acid and precipitating as magnesium ammonium phosphate. In the separation1' of thorium from iron, the iron was precipitated first.with hydrogen sulphide in the 1. Chem. Abstracts 8,1250,1914 end 8,2215,1914. An. G. Sci., 27,172-178., An. G. <5ci. , 27, 407-414. £. Chem. Abstiacts 9, 2202,1915. Anales soc. espan. fis. quim., 12, 279-282 (1914). •»-A — A -M M 0- 1- 1 1 C *Tt 10. presence of tartaric acid. Jjfter sulphuric acid rae added, hydrogen sulphide wes removed "by "boiling. Thorium was pre-cipitated with cupferr on after adding ammonium acetate. Precipitation of thorium was incomplete with small amounts of sulphuric acid present. Results were usually satisfactory hut the method was not given as a substitute to the usual oxalate separation. The preceding methods were combined with the necessary changes and additions to analyze zircon and haddeleyite by James Brown . These minerals contained zirconium, titanium, iron, aluminium, silica and mangan-ese. To test the accuracy of his method, he carried out the following procedure:- 7/eighed amounts of standard sol-utions of zirconium, titanium, iron, manganese and alumin-ium, as the sulphates or chlorides were mixed. The mixtures were treated with ammonium hydroxide in faint excess, sul-phuric acid (1:1) was added in slight excess, then 25 c.c. additional sulphuric acid and solution was made up to 150 c.c. It wes cooled with ice water. One hundred c.c. of a 6f aqueous solution of cupferron was added slowly with constant stirring. This was a large excess. The precip-itate was filtered immediately and washed et least 20 times with hydrochloric ecid (100 c.c. acid of sp. gr. 1.20 diluted 1. Jour. Am. Chem. Soc. vol.?9,no.ll,Eov.l917, P.£758. I 11. to one liter). Filtrate ana washing were saved for the determination of aluminium and manganese. The cupferr on precipitate was washed five times with a dilute solution of ammonium hydroxide. The precipitate with the filter paper was partly dried at 80C, placed in weighed platinum crucible, heated gently at first until the paper and pre-cipitate no longer gave off gases, then ignited with a Meker "burner. Ignition wes continued till constent weight was obtained, usually for a half hour after the carbon of the paper was removed. The precipitate gave the oxides of ferric iron, zirconiun and titanium. After fusing it with potassium bisulphate, the iron was precipitated as the sul-phide and thus separated out. The iron sulphide was dis-solved in nitric acid, precipitated as ferric hydroxide, ignited and weighed as the oxide. The filtrate from the cupferron precipitate was evaporated to small volume, 100 to 200 c.c. concentrated nitric acid being added during the evaporation in portions of 50 c.c. Tifhen large amounts of manganese were present, it was precipitated out as the oxide by treating with potassium chlorate and nitrio acid. The precipitate was dissolved in acid and manganese determined as pyrophos-phate. Aluminium was determined in the filtrate. When 12. small amounts of manganese were present, the aluminium was precipitated first with ammonium chloride and ammonium hydroxide and the manganese wes precipitated from the filtrate by the use of ammonium persulphate and ammonium hydroxide, ignited and weighed as Hn^On. EXPERIMENTAL WORK The liter flasks and pipettes used in this work were calibrated on the "basis of the true liter (U.S. Bureau of Standards). The weights used were also compared the one with the other and all expressed as multiples of the 10 mg. weight. The nitrosophenyl-hydroxylamine ammonium used in the work described below was material supplied for analytical work by Eimer & Amend. This reagent may be prepared accord-ing to the following method. Formula -- Cfc H^NOjONH,,. To 60 gms. nitrobenzene, add 1 liter of distilled water and 50 gms. ammonium chloride. Mix until a milky emulsion is formed. Add 80 gms. zinc dust, with constant stirring, keeping the temperature between 15° and 18 G, by adding ice from time to time to the rapidly whirling liquid. Continue the reduction until the odor of nitrobenzene 13. vanishes. About one-half hour is necessary. Filter off the white zinc hydroxide with a filter pump. Cool the solution to 0 C. with ice and add enough common salt -to saturate the solution. In a short time, a thick mass of white crystals separates out. Filter immediately and dry the crystals for one hour between filter paper. The yield is 70$ to 85% of theory. Dissolve in 300 to 500 c.c. commercial ether. Filter through a dry filter, cool to 0° C. and pass dry ammonia gas into the ether solution for 10 minutes, Add more than the theoretical amount (.5 gm. molecules) of fresh amyl nitrite. The clear solution will suddenly get hot and the vessel is filled with snow white crystals of nitrosophenylhydroxylamine ammonium. Preserve in a stoppered "bottle with a lump of ammonium carbonate. A liter of ferric chloride solution, approx. l/30 IT, was made up and standardized in the following way.-.About 9 gms. ferric chloride (FeClj6Ht0) were dissolved in a liter of water. One to five c.c. concentrated hydro-chloric acid were added to keep the iron from hydrolyzing. Twenty five c.c. were taken out, diluted to 150 c.c. and warmed. Ammonium hydroxide was added in excess and the liquid was heated to "boiling and filtered. The precip-itate was washed with ah out 60 c.c. water containing a 14. few drops of ammonium nitrate and ammonium hydroxide in excess. Then it was ignfted in a weighed porcelain cruoible and weighed as Fe^Oj. Weight of precipitate - .061? gms. Therefore there were 4.064 gms. FeCl, to the liter A liter of manganic chloride (about l/lO II) was made up and standardized. About 19 gms. manganic chloride (MnClx4Hx0) were dissolved in a liter of water, adding 1-5 | c.c. concentrated hydrochloric acid as before. To 25 c.c. solution was added ammonium chloride. The solution was then heated and 10 c.c. cold saturated solution of mioro-cosmic salt (UaNH^HPO^) were added. A white precipitate formed. This was diluted to 200 c.c. and neutralized with ammonium hydroxide. A large excess was avoided. The liquid was heated to boiling and stirred constantly until all the precipitate was crystalline. Then it was filtered out, washed with water containing ammonium hydroxide, ig-nited and weighed in a porcelain crucible as pyrophosphate. The average of the two estimations resulted as follows:-Weight of precipitate (MnJP^O, ) = .2522 gms. Therefore there were 12.496 gms. I&Cl^to the l i t e r . The f i r s t s e r i e s of determinat ions was made for the purpose of t e s t i n g the completeness of the separa t ion of 15. iron from manganese as well as the influence of dilution upon the weight of the resulting precipitate. Twenty-five c.c. of each of the two standard sol-utions just descrihed were mixed. The solutions were diluted as set forth in the table "below. Twenty c.c. concentrated hydrochloric acid was added. About 2 l/4 gras. cupferron were made into a 6^ aqueous solution. It dissolved with difficulty and the lumps had to be broken up with a glass rod and stirred. The solution turned dark almost immediately end was filtered before using. Fresh solutions were prepared for each deter-mination. The solutions of the chlorides were cooled each time in cold water and the cupferron solution was added slowly with const and, stirring. The solution first turned dark red, then a reddish-brown precipitate came clown which was partly amorphous, partly crystalline. The crystals were very bulky. They soon turned dark and could be easily crushed with the stirring rod. The precipitate was filtered immediately, washed with 75 c.c. cold water, twice with water containing ammonium hydroxide and twice with water. It was then heated slowly in a weighed porcelain crucible until no more vapor came off. Then it was ignited with a Meker burner until the precipitate was constant in weight, usually for 1 - lfc hours. The white precipitate of nitrosophenyl-hydroxylamine 16. was noticed. Most of it disappeared on stirring. The iron precipitate was found to he soluble in strong hydrochloric acid, but it could be precipitated by diluting the solution. The results obtained at four different dilutions are set forth in Table 1. The first column shows the volume to whioh the sol-utions were diluted before adding the precipitating reagent. TABLE 1. Volumes of Weight of Fe^.03 Found Solutions. taken 1 2 75 c.c. .0611 gms. .0601 gms. .0597 gms. 100 c.c. .0611 " .0612 " .0604 " 125 c.c. .0611 " .0612 " .0605 " 150 c.c. .0611 " .0609 " .0612 " These results would seem to show that the better determinations would be obtained at the greater dilutions, although it is not certain that the precipitate did not con-tain some manganese. The cupferron filtrate soon turned turbid, due to the oxidation of cupferron. The next series of determinations was made, using standard solutions of the two chlorides of approximately twice the strength of those used in the preceding series. In preparing these, 10 c.c. concentrated hydrochloric scid were added to each solution. These solutions were standardized as before. 17. Weight of Fe^Os from 25 c.c. ferric chloride solution - .1373 gms. Hence there were 11.154 gms. ferric chloride (FeCl5 ) to the liter. Weight of IlnvPx01 from 25 c.c. manganese chloride solution = .6284 gms. Hence there were 82.304 gms. manganese chloride (MnCljJ to the liter. For these determinations, 25 c.c. iron solution and 50 c.c. manganese solution were taken. Three gms. cupferron made into a 6°l solution were used. The iron precipitates, after igniting and weighing, were tested for the presence of manganese in the following way :- The ferric oxide precip-itate was fused in a platinum crucible with potassium acid sulphate and kept melted at red heat for about one-half hour. After cooling, this wes dissolved in hot water. Fifty c.c. nitric acid (sp. g. 1.135) were added. One-half gra. of sodium bismuthate was added and the solution was heated and stirred. Sodium thiosulphate wss added to clear the solution. This was heated until all the brown fumes came off and fumes of sulphur dioxide were noticed. The solution was cooled to 15*C. in cold water. Excess of sodium bismuthate was added. The solution was filtered through a Gooch crucible with an asbestos mat to get rid of the excess of bismuthate and the precipitate was washed with 50 c.c. of a ^ solution of nitric acid. To the filtrate was added an excess of ferrous 18. ammonium sulphate solution, the exact volume being noted. The excess of ferrous iron was then titrated against standard pot-assium permanganate solution, the latter having been standard-ized against pure sodium oxalate. The strength of the perman-ganate solution was 1.271 gms. KMnO^ to the liter. Ten c.o. ferrous ammonium sulphate solution were titrated against the permanganate solution from time to time, since the former may change, but the latter will remain constant for some time if kept in the dark. The two solutions were of almost equal strength. The number of c.c.'s of permanganate solution used to titrate the excess of ferrous ammonium sulphate was sub-tracted from the number corresponding to the volume of ferrous ammonium sulphate solution used. This gave the volume of per-manganate solution containing a weight of manganese equal to the manganese brought down with the iron precipitate by cup-ferron. In a number of cases, in order to check the results, lead peroxide was used for the oxidation of the manganese in place of the sodium bismuthate. The weight of lin30^ , weighed with Fev03, was thus calculated. As in the previous estim-ations, (Table 1), the solutions were made up to different volumes before adding the precipitating reagent. 19. Volume of fl J l) S o l u t i o n 1 100 c . c . Weight of p r e c i p i t a t e (Fe^O^ mo ) " " Mn30H " " F e ^ O j 126 c.c. " precipitate (Ff^03 Mns0+) " " Mn3 0 t " " Fe^0 3 150 c . c . " p r e c i p i t a t e ( F e ^ Mn30M) " Mns04 " » Fe^0 3 175 c.c. " " precipitate (Fe A 0 3 Mna0u ) " " Mn30H it n F e o 200 c.c. " " precipitate (Fe^03 Mn30u) " " Mn^04 " Fe^Oj .12£3 gms. .0002 n .1220 " .1216 " .00025 " .12125 " .1280 " .0018 " .1262 " .1251 " .0019 " .1222 " .1267 " .0029 " .1228 * CM 373 I .1329 gns. .0007 " .1222 " .1216 " .0006 " .1210 " .1275 " .0022 " .1252 n .1244 " .0017 " 1227 " .1260 " .0028 " 1222 M In these results, it is seen that the weight of ic oxide wes too low as compared with the weight found r precipitating the iron with ammonium hydroxide in stand-zing the solutions. Therefore a number of determinations made with iron alone to see if the low values of Table 2 £0. were due to the solubility of the iron precipitate in the mother liquor and wash water. Twenty-five c.c. of iron sol-ution with 20 c.c. concentrated hydrochloric acid were in every case diluted to 150 c.c. "before adding the precipitating reagent. The precipitate was washed first with a constant volume (75 c.c.-) of water containing a varying amount of hydrochloric acid as set forth in the table below; and finally with water (25 c.c.) containing a constant amount of ammonium hydroxide. The results were as follows: TABLE 3. Hydrochloric acid Weight of F e i 0 3 — Found concent ra t ion of frekon wash water X ir Hj_0 .1528 gms. .1320 gms. 10 c.c. HC1 to 100 c.c. H^O .1231 " .1324 " 20 c.c. " " 100 c.c. " .1343 " .1324 " 30 c.c. " " 100 c.c. " .1304 " .1310 " A fresh solution of ferric chloride was now pre-pared of such a concentration that when standardized by precipitating with ammonium hydroxide as before, the average of two estimations showed that '25 c.c. contained a weight of iron equivalent to .140&5 gms. Fe^O^. This solution was used in a series of determinations similar to the above series in which the wash water was evaporated to a small volume and 21. the iron in it determined "by precipitating with ammonium hydroxide and igniting. Also one-half the filtrate was evaporated to dryness to get rid of the excess of cupferron and the residue was taken up with concentrated hydrochloric acid, and a few drops of nitric acid. This was diluted and the iron determined as in the wash water. Thus the amount of iron contained in the filtrate was determined. Other methods were tried for finding the amount of iron in the wash waters and filtrates hut the presence of cupferron interfered with any colorimetric determinations. One method tried was to take one-half the filtrate, dilute, add 10 c.c. sulphuric acid and a piece of zinc and to heat till colorless (light green in these determinations). Then the solution was titrated against potassium permanganate solution hut no color reaction could he obtained since the solution turned "brown on adding a few drops of potassium permanganate. Also the test*.with potassium thiocyanate and potassium ferrocyanide were not satisfactory. The results found on determining the iron in the filtrates and wash waters were as follows:-Wash water Hx0 Ta l rU 1+ 22. 1. height of Fe103from cupferron p r e c i p i t a t e 2. Weight of Fe^O^from wash water 3 . Weight of Fe>03frora f i l t r a t e 100 c.c. H o 4. Total weight of Fex0„ Wash water 5 c^c. cone. HCl to 1.Weight of Fe^03from cupferron p r e c i p i t a t e 2. Weight of Fe^O^from wash r a t e r 3. Weight of Fex03from f i l t r a t e 4. Total weight of FexO, Wash water 10 c . c . cone. HCl 1. Weight of Fe^Oj frora to 100 c , c , HO cupferron p r e c i p i t a t e 2. Weight of Fex03from wash water 3. Weight of F e ^ f r o m f i l t r a t e 1 .1354 .0018 .0028 .1400 gras. M it 11 2 .1361 .0009 .0028 .1398 gms M " n Wash water 20 c . c . cone. HCl to 100 c . c , Hx0 4. Total weight of Fe^ O-^  1. Weight of Fea03from cupferron p r e c i p i t a t e 2. Weight of Fe^O^from wash water 3. Weight of Fe^O^from .1362 .0004 .0034 .1400 1 '^Q .0015 .0024 .1398 •i 'i n •• " n '! '• .1372 .0009 .0024 .1405 .1361 .0014 .0026 .1401 n n ii -I II '• n •i .1355 gms. .1354 " .C016 " .0016 " .0032 " 0032 filtrate 4. Total weight of Fex03 .1403 " .1402 ' In these determinations, 3 gras. cupferron were used. A determination was made using 5 gms. cupferron in a 6$ solution. 22. The wash water was 75 c.c. hydrochloric acid and water (5c.c. HC1 to 100 c.c. H^O) end 25 c.c. water containing ammonium 1. .1548 gms. .0007 " .0052 " .1407 " .1274 ,5ms. .0004 " .0022 TT .1400 " hydroxide as h e f o r e . 1.Weight of Fe^Oj from cupfe r ron p r e c i p i t a t e 2.Weight of F e ^ from wash water 5.Weight of F e x 0 3 from f i l t r a t e 4 . T o t a l weight of Fe^0 3 Another de t e rmina t i on was made, adding 15 c . c . c o n c e n t r a t e d s u l p h u r i c a c i d i n s t e a d of h y d r o c h l o r i c a c i d t o the f e r r i c c h l o r i d e s u l u t i o n . Five gms. cupfe r ron were used . The wash water c o n s i s t e d of 75 c . c . water c o n t a i n i n g hydro-c h l o r i c a c i d (5 c . c . cone. HCl bo 100 c . c . ilx0) and 25 c . c . water c o n t a i n i n g ammonium h y d r o x i d e . 1. 2. 1.Weight of Fe^03from cupfe r ron p r e c i p i t a t e 2.Weight of F e ^ f r o m wash water . 2. Weight of Fe<t03from f i l t r a t e 4 . T o t a l weight of F e 1 C 3 .1250 gms. .1257 gms. .0007 " .0007 " .0024 " .0026 " .1291 .14C0 ' It was decided to make a few determinations cooling the solution in ice water "before adding the cupferron. Twenty c.c. cone, sulphuric acid were added to the iron solution and 24. ahout 4£r gms. cupferron were used. The strength of the hydrochloric acid wash water used was 5 c.c. concentrated hydrochloric acid to 100 c.c. water. This was used in different amounts. Volume of hydrochloric acid wash water 75 c . c . TABLE 5. 2. 100 • O 1. Weight of re i 0 3 f rom cupferron p r e c i p i t a t e 2. Weight of Fej_03from wash water 3. Weight of FeA03from f i l t r a t e 4. Total weight of Fe^Oj 1. Weight of Fe^Oafrora cupferron p r e c i p i t a t e 2. Weight of F e ^ f r o m • wash water 3. Weight of Fe^O^from f i l t r a t e 4. Total weight of Fe^O^ 1. Weight of Fe^Oafrora cupferron p r e c i p i t a t e 2. Weight of F e ^ f r o m wash water 3. Weight of Fe,.03from f i l t r a t e 4. Total weight of Fe^Oj I t was thought t ha t possibly the addi t ion of ammonium chlor ide might help in keeping manganese from coming 125 c.c . 1364 .0008 .0036 .1408 .1354 .0010 .0030 .1394 • -i- tj %J <-.0016 .0032 .1401 gms. •i n it •T TT >( tl 1! II < T II .1356 .0007 .0032 .1395 .1350 .0015 .0028 .1360 .0012 .0028 .1400 gms n •! It 11 rT II 11 I' IT n n down with the iron precipitate. Therefore a determination v/as made with 25 c.c. iron solution end 50 c.c. manganese sol-ution mixed together. Twenty c.c. concentrated hydroohloric acid was added and the solution was diluted to 1EE c.c. Five gns. ammonium chloride were added, then four gms. cup-ferron in e 6$ solution. Wash water was 75 c.c. hydrochloric acid solution (5 c.c. cone. HCl to 100 c.c. K^O) and 25 c.o. water and ammonium hydroxide. 1. 2. 7/eight of cupferron precipitate .1362 gms. .1268 gire. The precipitates were fused and tested by the lead peroxide method. Test showed only .0004 gms. ISnf^ which came down with the iron, which is en almost unweighable amount. To dispel eny doubt as to whether any of the cup-ferron precipitate of iron turned to metallic iron, several of the precipitates were tested in the following way:-The precipitate wes transferred to a weighed platinum crucible and crucible was weighed again. About 2 c.c. concentrated nitric acid were added and the precipitate W8s heated to dryness. After cooling, it wcs weighed again. In each case it was found that the weight had not changed. It would appeer from the results obtained above, that in the precipitation of iron by means of cupferron, an appreciable amount of iron remains in the mother liquor, 26. this amount, apparently, "being practically constant, -within i1 reasonable limits as regards the volume of the mother liquor. iIt would therefore appear necessary and legitimate to apply a correction to the weight of iron obtained from the analysis, this correction having teen determined once for all under any given set of conditions. What is true of the filtrate is true, to a less extent, of the wash water, the equivalent of some-what more than a milligram of iron oxide "being carried along with the washing liquid. With regard to the separation of the iron from the jmanganese, the results of Table 2 show that the amount of jmanganese carried down increases with the dilution of the ;acid. With a concentration of hydrochloric acid, correspond-ing to 20 c.c. concentrated hydrochloric acid to 100 c.c. of gsolution, the separation seems to be as satisfactory as the ;!majority of separations in quantitative analysis. The experiment with ammonium chloride shows that jprobably less manganese is carried down in the cupferron jprecipitete when this rather weak electrolyte is present. The best conditions, therefore, for the separation of iron from manganese as well as for the determination of iron would appear to be obtained when the solution from which precipitation takes place contains 20 c.c. concen-27. trated hydrochloric acid to 1EE e.c. solution end enough ammonium chloride to yield a 5$ solution. SUMMASY. 1. I t has been shown in the above work, f i r s t , t h a t the cupferron p r e c i p i t a t e of i ron ie appreciably soluble in the mother l i quor . The solubi l i t j? corresponds c lose ly to 1 mmg. of Fe^Oj in 100 e . c . of the f i l t r a t e . 2. The cupferron p r e c i p i t a t e of i ron a lso d isso lves to some extent in the wash water. Under the usual cond i t ions , t h i s amounts to about one-third the amoiint ca r r i ed down in the f i l t r a t e . Z. The amount of manganese ca r r i ed down by the cupferron p r e c i p i t a t e of i ron increases as the so lu t ion from which p r e c i p i t a t i o n takes place i s d i lu ted , the best separa t ion being obtained when the so lu t ion contains approximately 7$ hydrochloric acid and Ly ammonium ch lo r ide . -47 Before t h i s reagent can be used for the determination of i ron or i t s separa t ion from aluminium or manganese, i t would seem tha t i t s cost would have to be appreciably reduced as a t present the pr ice of §1.50 per oz. would mi t iga te agains t i t s use on a la rge scale 29. Bibliography. Quantitative Chemical Analysis "by J.C.Olsen, A.M., Ph.D. (D.Van Uostrand Co.) P. 157, 285 Standard Methods of Chemical Analysis by 77.W.Scott CD.Tan Uostrand Co.) P.260 Journal of American Chemical Society,50, 593, (1908) Chem. Abstracts. 4,557, 1910. Chem.-Ztg., 23,1298-1300 Chem. Abstracts. 4,2783, 1910. Z. Amorg.Chem.. 66,426-450 Chem.Abstracts. 4,2175,1910. Z. Amorg.Chem., 68, 55-56 Chem. Abstracts. 5,1718,1911. Z. Anal.Chem., 50, 35-43 Chem. Abstracts. 5,3024,1911. Z. Angew. Che-., 23,969. Chem. Abstracts. 7,1688,1913 and 7,3583,1913 Atti accad. Tineei, 22, 1,20-24 Gazz. ohira. itsl. 42, 1,570-576 Chem. Abstracts 8.1250,1914 and 8,2215,1914 Arc. G. Sci., 57,172-178. Am. G. Sci., 27,407-414 Chem. Abstracts 9, 2*202, 1915 Anales soc. espan. fis. quim., 12,579-382 (1914) Chem. Abstracts 10,2442,1916. W.M.Thornton, Jr. Am. J. Sci. 42,151-154 (1916) Chem. iTev.'s, 114,12-14 (1916) Jour. Amer. Chem. Soc. vol.59, no.11, Uov.1917, p.2558 28. particularly as a considerable excess of the reagent must "be used for each preoipitation. Chemical Laboratory, University of British Columbia, Vancouver. B. C. 28. particularly as a considerahle excess of the reagent must he used for each precipitation. Chemical laboratory-, University of British Columbia, Vancouver , B. C. 28. particularly as a considerable excess of the reagent must he used for each preoipitation. Chemical Laboratory, University of British Columbia, Vancouver, B. C. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items