Jan., 19501 The PART I. VOLUMETRIC DETERMINATION OF CHROMATE Determination of Chromium in Chromite 9 Part 11. Determination of Chromium in a Synthetic Sample of Known Composition BY P. J. HARDWICK SYNOPSIS-Four methods in common use for the determination of chromium in chromite were tested on a synthetic mixture of potassium dichromate and “mixed oxides” equivalent in composition to the mineral. All the methods involved fusion with sodium peroxide and removal of most of the excess of peroxide by boiling with water, and the chromate formed was titrated by adding excess of ferrous ammonium sulphate solution and titrating back with dichromate solution. Two of the methods gave satisfactory results (within 1 part in 2000 of the chromium present) after allowance was made for the small amount of vanadium present; they were selected for trial on a standard sample of chrome ore.The other two were more laborious and required amending for accurate work. FOLLOWING the study of titration methods for the determination of chromate described in Part I,l an examination has been made of four methods in common use in this country for determining chromium in chromite. In all four methods the ore is decomposed and chromium oxidised to chromate by fusion with sodium peroxide, usually in a nickel crucible. Most of the excess of peroxide is then destroyed by boiling with water. In three of the four methods the interference of residual peroxide is eliminated by an auxiliary oxidation in acid solution with ammonium persulphate and silver nitrate2s3 (Method 1) or potassium permanganate in small or large excess4~6 (Methods 2 and 3, respectively). In the fourth method the alkaline solution of peroxide is boiled for a longer time.The interference of manganese as pennan- ganate or manganese dioxide is eliminated in all cases by boiling with hydrochloric acid (for small amounts of manganese) or by filtration. In the present investigation the chromium as chromate was determined by addition of excess of ferrous ammonium sulphate solution and back-titration with standard potassium dichromate solution, using barium diphenvlamine sulphonate as indicator in presence of phosphoric acid. This titration method was chosen for its convenience and for the sharpness of the end-point. Also, provided the ferrous ammonium sulphate solution is standardised against potassium dichromate under the same conditions, particularly with regard to acidity, volume of solution and concentration of dichromate,6,7 the procedure is capable of yielding exact results.It is necessary, however, to correct for vanadium. The main object of the investigation was to test the accuracy of these methods, working with a known amount of chromium. For this purpose a synthetic sample was used containing in typical proportions all the principal constituents of chromite, all as oxides or carbonates except chromium, which was present as potassium dichromate. Unnecessary wastage of a standard sample of chromite was thereby avoided. Moreover, it was an advantage to start with all the chromium in the form of a soluble salt, corresponding with complete decomposition of the ore.10 HARDWICK THE DETERMINATION OF CHROMIUM I N CHROMITE [Vol.75 The results obtained showed that two of the methods (Methods 1 and 2) were suitable for accurate work. The other two methods required amending. EXPERIMENTAL PREPARATION OF A SYNTHETIC SAMPLE OF KNOWN COMPOSITION- The composition of the mineral chromite varies within wide limits, but as commonly purchased in this country it contains on average about 48 per cent. of chromic oxide. The remainder consists mainly of oxides of iron, aluminium, magnesium and silicon, and may include small amounts of calcium, barium, manganese, titanium, vanadium and nickel. For the purpose of this investigation a synthetic representative material (subsequently referred to as “mixed oxides”) was prepared consisting of an intimate mixture of ignited oxides and carbonates in the following proportion by weight: Fe,O, 16; A1,0, 15.5; MgO 14; SiO, 4; CaCO, 1; MnO, 0.5; BaCO,, TiO,, V20, and Ni203, each 0.25.Addition of an accurately known weight of K,Cr,O, to a weighed portion of the “mixed oxides” in the ratio 1-79 to 1 provided a sample of known composition containing all the principal con- stituents of an average chromite in typical proportions with a little K,O in addition. FUSION AND EXTRACTION PROCEDURE- The sample of “mixed oxides” and added potassium dichromate (total weight about 0.7g.) was fused with 5g. of sodium peroxides at dull red heat for 5 minutes in a nickel crucible of about 40-ml. capacity. On cooling, the crucible and lid were transferred to a 600-ml.beaker fitted with a cover glass, 100 ml. of water were added and the solution was boiled for 10 minutes to extract the chromate and decompose most of the excess peroxide. The crucible and lid were then rinsed with about 50ml. of water and removed. In preliminary experiments the melt was maintained at dull red heat for 15 minutes, but attack of the nickel by the peroxide was severe and crucibles of wall-thickness about 0.7 mm. could not be safely used for more than 2 or 3 fusions. Parallel experiments in which a mixture of sodium peroxide and sodium hydroxide (Theobald’s mixtureg) was used as flux showed reduced attack of the nickel and the crucible served for 6 fusions. In subsequent work with sodium peroxide, however, it was found that the time of fusion could be shortened to 5 minutes for complete decomposition of the sample.Under these conditions about 0.4 g. of nickel was dissolved during a fusion and the average life of a crucible was 5 or 6 fusions. Examination of the crucibles spectrographically showed that the nickel contained only a trace of chromium (less than 0.01 per cent.) and approximately 0-3 per cent. of manganese. No vanadium was detected. TREATMENT OF THE EXTRACT BY DIFFERENT METHODS- Method 1-Auxiliary oxidation with amnzonium persdphate and d v e r nitrate2 J-The extract was acidified with 120 ml. of diluted sulphuric acid (1 + 3), warmed to dissolve the precipitate and transferred to a 1000-ml. conical flask. After dilution with water to 500 ml. to decrease the sulphuric acid concentration to about 2 N , the auxiliary oxidation was carried out in the usual manner2 and the small amounts of permanganate and manganese dioxide present were destroyed by boiling with dilute hydrochloric acid.After cooling, a measured volume (100 ml.) of standard ferrous ammonium sulphate solution was added from a pipette and the small excess titrated back with 0.1 N potassium dichromate solution after addition of 25 ml. of phosphoric acid (sp.gr. 1-75) and 5 drops of a 0.3 per cent. solution of barium diphenylamine sulphonate in water as indicator. A t the end-point the indicator was sensitive to 0.02 ml. of 0.1 N potassium dichromate solution added to a total volume of about 600 ml. of solution. Small departures from standard practice2 were the higher initial concentration of hydro- chloric acid used (20 ml.of diluted hydrochloric acid (1 + 3) per 500 ml. of solution) and the longer time of boiling given (15 minutes) to destroy persistent traces of manganese dioxide and remove chlorine. Separate experiments in which known amounts of potassium dichromate were boiled with hydrochloric acid under similar conditions showed that no reduction of dichromate occurred. Method 2-Auxiliary oxidation with Permanganate in small excess-The extract was acidified and diluted to 500 ml. as in Method 1, and 1 to 2 ml. of approximately 0.1 N potassium perrnanganate were added to the hot solution to give a small excess, shown by theJan., 19501 PART 11. DETERMINATION OF CHROMIUM IN A SYNTHETIC SAMPLE 11 deepening amber colour.After boiling for 5 minutes, the excess of permanganate was destroyed by boiling with hydrochloric acid and the chromate titrated as in Method 1. Method 3-Auxiliary oxidation with Permanganate in large excess (VignaZ Method4+- The extract was acidified and diluted to 500 ml. as in Method 1. It was then heated to boiling and a 5 per cent. solution of potassium permanganate was slowly added until a brown precipitate appeared. Approximately 5 ml. of a 10 per cent. solution of manganous sulphate were added and the solution was boiled for 5 minutes, cooled and filtered by suction through a closely packed pad of asbestos (previously boiled with dilute nitric acid) or through a sintered glass crucible (Jena 3 G4). The precipitate of hydrated manganese dioxide was washed with hot water and the filtrate and washings were diluted to 500 ml.and titrated as in Method 1. Method 4-AZkaZine~ltrations-The extract was boiled for a further 5 minutes to destroy most of the remaining peroxide and the precipitate of hydroxides allowed to settle. The solution was then filtered through a hardened paper (Whatman No. 52, 12.5 cm.) a t room temperature to avoid reduction of the chromate, or through sintered glass, and the precipitate washed 8 to 10 times with 30-ml. quantities of water until the washings were colourless. The combined filtrate and washings were acidified with 120ml. of diluted sulphuric acid (1 + 3), cooled to room temperature and titrated as in Method 1. RESULTS The results obtained for the chromium content of samples containing a known weight of potassium dichromate added to 0.26g. of “mixed oxides” are summarised in Table I.The weight of potassium dichromate taken was checked by titration against standard ferrous ammonium sulphate solution of an equal amount of dichromate weighed out at the same time as the test sample. In Methods 1 and 2 the chromic oxide found was corrected for a small consistent “blank” with the “mixed oxides” amounting to 0-0004 g. of chromic oxide, which was chemically equivalent to the known amount of vanadium present (0.00125 g. V,05). TABLE I DETERMINATION OF CHROMIUM IN A SYNTHETIC SAMPLE BY DIFFERENT METHODS Method Test No. 2 3 4 1 1 2 3 4 5 6 7 8 1 2 3 4 1 2 1 2 3 4 Potassium dichromate added g. 0.4643 0-4643 0.4643 0.4643 0.4644 0.4644 0.4643 0.4643 0.4642 0-4642 0.4641 0.4641 0.4644 0.4644 0.4642 0.4642 0.4642 0.4642 Equivalent wt.of chromic oxide g. 0.2399 0-2399 0.2399 0-2399 0.2400 0-2400 0-2399 0-2399 0.2399 0-2399 0-2398 0-2398 0-2400 0-2400 0.2399 0.2399 0.2399 0-2399 Chromic oxide found Error g- 0.2398 0-2398 0.2399 0.2399 0.2399 0-2399 0,2399 0.2400 0.2399 0.2398 0.2399 0.2399 0.2436 0,2429 0.2394 0-2330 0-2353 0-2351 R- - 0~0001 - 0*0001 o*oooo o*oooo - 0~0001 - 0~0001 0~0000 + 0~0001 o*oooo - 0~0001 + 0~0001 + 0-0001 + 0-0036 + 0.0029 - 0.0005 - 0.0069 - 0.0046 - 0.0048 The results, calculated as chromic oxide, show satisfactory agreement with the weights taken in all the tests of Methods 1 and 2. The maximum error is about 1 in 2000 which is admissible in most analytical work.The results obtained by Method 3 are 1 to 2 per cent. too high. “Blanks” with the “mixed oxides” by this method were also high and variable. Separate experiments showed that the error probably arose through peptisation of the hydrated manganese dioxide precipitate; it was reduced by washing the precipitate with a 1 per cent. solution of sulphuric acid instead of water and was completely eliminated10 when a wash solution containing 5 per cent. of potash alum and 1 per cent. of sulphuric acid was used. Amended in this way the method gave satisfactory results but was more laborious than Methods 1 and 2.12 BRYANT AND HARDWICK: THE DETERMINATION OF CHROMIUM IN CHROMITE [VOl. 75 The results obtained by Method 4 are low. The error was found to be due mainly to retention of chromium by the bulky insoluble residue, consisting largely of nickel hydroxide.Separate experiments1O in which the sample was fused with Theobald’s mixture instead of sodium peroxide, to reduce the amount of nickel dissolved from the crucible, showed that the residue still retained a small amount of chromium (about 0.0005 g. of Cr,O,) even after repeated washing with hot water. By fusing the ignited residue with the flux, however, as in the method described by Hillebrand and Lundel12 using sodium peroxide , the chromium was almost completely recovered. The amended method, involving two fusions and extractions followed by an auxiliary oxidation with ammonium persulphate and silver nitrate in acid solution, and removal of traces of permanganate by boiling with hydrochloric acid, gave results comparable in accuracy with those obtained by Methods 1 and 2 but was laborious. Methods 1 and 2 were therefore considered the simplest and most reliable of the methods tested and were selected for further trial with a standard sample of chrome ore.ll I am indebted to the Government Chemist and to the Chief Scientist, Ministry of Supply, for permission to publish this paper. 1. 2. 3. 4. 6 . 6. 7. 8. 9. 10. 11. REFERENCES Furness, W., Analyst, 1960, 75, 2. Hillebrand, W. F., and Lundell, G. E. F., Applied Inorganic Analysis, 1929, pp. 408-415. Cunningham, T. R., and McNeill, T. R., Ind. Eng. Chem., Anal. Ed., 1929, 1, 70. Vignal, H., Bull. SOC. Chim., 1886, 45, 171. Lundell, G. E. F., Hoffman, J. I., and Bright, H. A., Chemical Analysis of Iron and Steel, 1931, Eppley, M., and Vosburgh, W. C., J . Amer. Cl’zem. SOC., 1922, 44, 2148. Willard, H. H., and Gibson, R. C., Ind. Eng. Chew., Anal. Ed., 1931, 3, 88. Mellor, J. W., and Thompson, H. V., Quantitative Inorganic Analysis, 1938, pp. 527-532. Theobald, L. S., Analyst, 1942, 67, 287. Furness, W., private communication. Bryant, F. J., and Hardwick, P. J., AnaZyst, 1950, 75, 12. p. 298. DEPARTMENT OF THE GOVERNMENT CHEMIST GOVERNMENT LABORATORY, W.C.2 28th September, 1949