Nov. , 19581 IN MOLYBDENUM - URANIUM AND MOLYBDENUM - NIOBIUM MIXTURES 615 The Colorimetric Determination of Aluminium in Steel with Solochrome Cyanine BY P. H. SCHOLES AND D. VALERIE SMITH (Metallurgy (General) Divisional Laboratories, The British Iron and Steel Research Association, Hoyle Street, Shefield, 3) Eriochrome cyanine R has been used extensively in Germany for the colorimetric determination of aluminium in ferrous materials. An equivalent British dye, Solochrome cyanine R, has been examined in order to establish the best conditions for colour development. The application of Solochrome cyanine R to the analysis of steel has also been studied. A method incorporat- ing a preliminary mercury-cathode separation and then treatment with sodium hydroxide solution is proposed for the determination of aluminium in the range 0.001 to 0.035 per cent.A precision of &O.OOl per cent. is claimed. The range of the method can be extended to cover any concentration of aluminium in plain carbon and low alloy steel. IN present-day metallurgical research it is becoming increasingly important to examine the influence of small amounts of aluminium on the properties of steel. An accurate knowledge of the aluminium content of experimental test samples is an essential pre-requisite for investigations into such problems as grain-size control, the influence of aluminium on the nitrogen content and its effect on the low-temperature properties of mild steel. A volumetric oxine method for determining aluminium in steel has been recommended by the Methods of Analysis Committee of the British Iron and Steel Research Association1 and has now been adopted as a standard method by the British Standards I n ~ t i t u t i o n .~ ~ ~ Although this procedure-in the hands of a competent analyst-is capable of giving accurate results, it is less satisfactory a t aluminium levels below about 0.03 per cent. A tendency towards high results is noticeable at this concentration, which, to some extent, may be due to difficulties associated with the removal of excess of oxine from the aluminium quinolate precipitate. Non-reproducible blank values equivalent to an aluminium content of from 0.0002 to 0.002 per cent, on a 10-g sample seem to confirm this observation. The unreliability of this method at low levels is demonstrated by divergencies in the results obtained by five co-operating laboratories for aluminium in a series of mild-steel616 SCHOLES AND SMITH : THE COLORIMETRIC DETERMIKATIOK OF [Vol.83 spectrographic standards. For example, on two samples with contents of 0.01 and 0.02 per cent. of aluminium, the results ranged from 0.008 to 0.015 and 0.015 to 0.026 per cent., respectively. The Methods of Analysis Committee considered that this unreliability might be caused by contamination of the aluminium quinolate precipitate by manganese. The procedure has recently been re-examined by the Mercury Cathode Study Group of the Committee, and results obtained by the co-operating laboratories now show much closer agreement: A modified British Standard for the volumetric determination of aluminium will shortly be issued.Small amounts of aluminium are more conveniently determined by a colorimetric pro- cedure, and the literature contains many references to the use of lake-forming reagents, such as aluminon, Eriochrome cyanine R, stilbazo and alizarin. The reaction between Eriochrome cyanine R and aluminium has been studied exiensively by German worker^,^?^ and between 1938 and 1940 it was applied to the determination of aluminium in steel by Koch and his co-worker~.~?s The dye has also been used in the United States of A m e r i ~ a , ~ ? ~ ~ but does not seem to have found favour in this country. Eriochrome cyanine R is a brick-red powder that is readily soluble in water to give an orange-red solution; in the presence of an acetate buffer, a red-violet lake or complex is formed with aluminium.Werz and Neubergerll recently established the superiority of this reagent over aluminon, the reagent most widely used for determining small amounts of aluminium in steel. Essentially, the main advantage of Eriochrome cyanine R is its greater stability towards temperature change; formation of the complex between aluminium and the dye takes place at room temperature, whereas the complex with aluminon must be heated under controlled conditions to ensure reproducible colour development. In addition, the need for accurate control of pH is slightly less critical. Eriochrome cyanine R is the trade name atssigned by Geigy and Co. to the sodium salt of 2 -sulpho-3 : 3’-dimethyl-4-hydroxyfuchsone-Ei : 5 -dicarboxylic acid.In the Colour Index,12 the dye is listed under usage number Mordant blue 3 and has the constitution number 43820. The dye is also marketed under such trade names as Alizarol cyanine R, Omega Chrome cyanine R, Chromoxane cyanine RA, Solochrome cyanine R, Pontachrome blue ECR and Fenakrom blue XR. Solochrome cyanine R, available from the British Drug Houses Ltd., has been used in the work described. CONDITIONS FOR COLOUR DEVELOPMENT The literature suggests that the most import ant factors in the formation of the aluminium - dye complex are (a) pH value, (b) concentration of dye and the effect of “ageing” the dye solution, (c) time required for full development of the colour, and ( d ) effect of temperature. Since there appears to be a divergence of opinion on the optimum conditions for colour development, each variable has been examined in detail, an aqueous solution of Solochrome cyanine R being used.SELECTION OF OPTIMUM pH- Previous investigators have studied the selection of an optimum pH value and have found that complex formation is satisfactory in the pH range 5 to 7 with maximum colour development between pH 5.7 and 6.5. In order to study the effect of pH, a series of calibration graphs in the concentration range 0 to 80 pg of ,aluminium per 100 ml was prepared at various pH values. The test solutions contained 5 mg of dye per 100 ml and were buffered with a mixture of ammonium and sodium acetates. In common with all other measurements in this paper, the optical densities of the test solutions were measured against a reagent blank at 535 mp with a Unicam SP600 spectrophotometer.The calibration graphs were found to be linear in the pH range 5.7 to 6.3, but not at pH 6.4 to 6.5. Fig. 1 shows the variations in optical density of a calibration test containing 60 pg of aluminium and of the reagent blank solution. The optical density of the aluminium complex is fairly stable in the pH range 5.7 to 6.1, although that of the reagent blank diminishes with increasing pH, but is relatively stable in the pH range 5.9 to 6.2. A value of 6.0 has therefore been selected. At this pH, a variation of up to k0-l optical-density units will not affect the optical density of the aluminium complex. DYE COXCENTRATION- In order to determine the optimum concentration of dye in the final solutions, different amounts of dye were added to test solutions containing 70 pg of aluminium.A concen- tration of 5 mg of dye per 100 ml was adequate to give maximum optical density for solutionsNov., 19581 ALUMINIUM IN STEEL WITH SOLOCHROME CYANINE R 617 containing up to 70 pg of aluminium. Some workers advocate that the dye solution should be set aside for 1 hour or even overnight before use, whereas others apparently consider this I , I 0 10 20 30 40 50 60 Time, minutes Fig. I . Effect of pH on the aluminium - Solochrome cyanine R complex: curve A, Fig. 2. Effect of time for formation of the aluminium - test solution containing 60 p g of aluminium Solochrome cyanine R complex: curve A, test solution con- per 100 ml; curve B, test solution corrected taining 60 pg of aluminium per 100 ml; curve B, test solution for reagent blank value; curve C, reagent corrected for reagent blank value; curve C, reagent blank blank solution solution 0 Wavelength, mp Fig.3. Absorption spectra for the alu- minium - Solochrome cyanine R complex : curve A, test solution containing 60 p g of aluminium per 100 ml measured against the reagent blank solution; curve B, reagent blank solution to be unnecessary. effect on either the optical density or stability of the complex. This factor was examined, and tests indicate that "ageing" has little A decrease of about 0.03618 SCHOLES AND SMITH: THE COLORIMETRIC DETERMINATION OF y01. 83 optical-density units was, however, noticed in the blank value for dye solutions that had been set aside for more than 1 hour.A minimum "ageing" time of 1 hour is therefore recommended. TIME FOR NAXIMUM COLOUR DEVELOPMENT- Concurrent with the pH tests, the time required to attain maximum colour development was also studied. Optical-density measurements were made a t intervals of 10 minutes for a period of 1 hour after dilution to volume. The results (see Fig. 2 ) show a slight increase with time for the aluminium complex and an iritial decrease followed by a slight fading for the reagent blank. Hence, it is necessary to set the solutions aside for a definite period; 30 minutes was chosen. A reasonable margin in timing is permissible, but, to avoid wide variation in the time allowed for colour development of individual tests, it is advisable to restrict the number of samples in a batch to six (including a reagent blank).Approximately the same time interval should elapse before measurement of both the reagent blank and the test solution, and it is therefore recommended1 that the spectrophotometer should be set against the reagent blank solution before each individual test solution is measured. EFFECT OF TEMPERATERE- The effect of temperature in the normal laboratory range of 16" to 23" C is not critical. On heating to 30" C, however, a slight decrease in optical density was noticed, which did not exceed 1 per cent. of the optical density at 20°C. The optimum conditions for colour development having been established, a comparison was made between the absorption spectra of the aluminium complex and the reagent blank (see Fig.3). The curve for the Solochrome cyanine R - aluminium complex is in good agreement with the findings of Hill, who used the dye Eriochrome cyanine R.I0 The optical density of the aluminium complex is a maximum at 535 mp. SEPARATION OF ALUMINIUM Koch and his co-worker~~>~ separated aluminium from iron and most other heavy metals by mercury-cathode electrolysis. More recently, Werz and Neuberger,ll in a comparative study of separation methods, stated that as much as 25 per cent. of the aluminium content may be lost through adsorption of aluminium or1 ferric hydroxide during the classical separa- tion with sodium hydroxide. A loss of 11 per cent. by diethyl ether extraction was reported, and the authors concluded that mercury-cathode electrolysis was the most accurate method of separation without loss of aluminium.The separation of aluminium from iron on a cellulose column has recently been described by Bishop.13 Iron is removed from the column, and aluminium together with any nickel present is retained quantitatively. The aluminium is subsequently determined polaro- graphically by means of Solochrome violet RS. Recent work by Rooneyl4 has, however, shown that removal of titanium, vanadium, chromium, cobalt and zirconium is not complete and small amounts of these elements may remain on the column with the aluminium, MERCURY-CATHODE ELECTROLYSIS- An elaborate apparatus is not required for mercury-cathode electrolysis. The elec- trolytic cell recommended by us is a 150-ml squat beaker containing about 20 ml of mercury; a spiral of heavy-gauge platinum wire forms the anode, a platinum wire contact sealed in glass makes the electrical connection to the mercury cathode and a split clock-glass serves as a cover for the cell to prevent loss by spraying.A device for stirring the mercury cathode is not required. At a current density of 0-16 ampere per sq. cm of mercury surface, the electrolysis of a sample of plain carbon steelis complete in 1 to 14 hours. This time may be considerably reduced by using a higher current density together with a water-cooled coil to maintain the temperature of the solution below 40" C, as recommended in the British Standard procedure.2 In low alloy steels, complete deposition of chromium is often slow; chromium removal can be facilitated by neutralising most of the free acid content of the solution before electrolysis.Mercury-cathode electrolysis in dilute sulphuric acid removes iron as well as such elements as nickel, copper, chromium and molybdenum. In addition, manganese is incom- pletely deposited in the mercury and on the a.node; lead may also be partially deposited on the anode, Zirconium, titanium, vanadium and silicon, together with a small amountNov., 1958: ALUMINIUM IN STEEL ~ I T H SOLOCHROME CYANINE R 619 of mercury that enters the solution during electrolysis, remain with aluminium in the elec- trolyte. Traces of bivalent iron that have escaped separation may also be found in the electrolyte. Previous workers have shown that manganese, titanium, zirconium and residual iron interfere with complex formation in the determination of aluminium with Eriochrome cyanine R and give rise to serious negative errors.Treatment with sodium hydroxide and hydrogen peroxide, as suggested by Werz and Neuberger,ll removes zirconium, manganese and iron, but mercury and silicon are only partly precipitated. In the presence of hydrogen peroxide, which must be added to ensure complete precipitation of iron and manganese, titanium is partly converted to pertitanic acid and is therefore incompletely precipitated by sodium hydroxide. Interference by titanium can be masked by adding hydrogen peroxide to the solution before colour development. Vanadium, which accompanies aluminium into the filtrate, does not interfere with the formation of the aluminium - dye complex. At pH 6.0, the yellow colour of the combined peroxide complexes of titanium and vanadium is so pale that it does not interfere with absorption measurements.Any silicon that accompanies aluminium into the filtrate after the sodium hydroxide separation will normally be present in soluble form, but, in steels with high silicon contents, metasilicic acid may be present in the solution and would prove to be troublesome during colour development. Hence, for samples containing more than 1 per cent. of silicon, it is advisable to remove silicon by evaporating the sulphuric acid solution until fumes are evolved before insoluble matter is removed by filtration. Some workers have recommended that mercury, which is only partly precipitated by sodium hydroxide, should be removed with hydrogen sulphide before colour development. To examine the effects of mercury contamination, additions of 5 and 10mg of mercury were made to solutions containing aluminium after acidification of the filtrate from the sodium hydroxide separation; the colour was then developed in the usual way.The results indicated that the presence of mercury in the electrolyte does not affect complex formation; treatment with hydrogen sulphide is therefore unnecessary. The effects of lead have also been studied and it has been found that this element is without effect in concentrations of up to 1 mg per 100 ml (equivalent to 0.5 per cent.). CHROMATOGRAPHIC SEPARATION- The cellulose-column method of separation described by Bishop13 has been examined in our laboratory as a possible alternative to mercury-cathode electrolysis. The method was adapted as follows.After separation on the cellulose column, the eluate containing aluminium, nickel and traces of other metals was evaporated with sulphuric and nitric acids until fumes were evolved to remove organic matter. The interfering elements were then removed by a sodium hydroxide separation in the presence of 1 mg of tervalent iron to act as carrier. The filtrate was acidified and the colour developed as described on p. 621. Trials with this method of separation were promising, but the time required to prepare and regenerate the cellulose column was a disadvantage. Mercury-cathode electrolysis is more rapid and requires little or no attention during separation. The chromatographic method is, however, suitable for laboratories that have no source of direct current.METHOD The method consists in small-scale mercury-cathode electrolysis and subsequent separa- tion by sodium hydroxide in stainless-steel beakers to remove interfering elements. After careful neutralisation, a measured excess of dilute acid is added and the aluminium - SoIo- chrome cyanine R complex is developed under optimum conditions. REAGENTS- be stored in polythene bottles. All solutions must be prepared from reagents of the highest purity obtainable and should Hydrochloric acid, diluted (1 + 1). Hydrochloric acid, 0.2 N-Dilute 20 ml of hydrochloric acid, sp.gr. 1.16, to 1 litre with Sulphuric acid, dilute (1 + 4). Sul9huric acid, dilute (3 + 20). water.620 SCHOLES AND SMITH : THE COLORIMETRIC DETERMINATION OF [Vol.83 Sodium carbonate solution, saturated. Sodium hydroxide, 10 N-Transfer 400 g of sodium hydroxide pellets to a 1-litre polythene Shake Sodium hydroxide, 2 N-Dilute 20 ml of 10 N sodium hydroxide to 100 ml. Hydrogen peroxide, 5-volume. Phenolphthalein indicator solzdion-Dissolve 0.1 g of phenolphthalein in 50 ml of methy- lated spirit and dilute to 100 ml with water. Solochrome cyanine R solution, 0.1 per cent. w/v-Dissolve 0.1 g of B.D.H. Solochrome cyanine R in water, dilute to 100m1, and filter through a Whatman No. 541 filter-paper. This solution should be prepared daily and set aside for a t least 1 hour before use. Glassware that has been in contact with this solution should be rinsed with nitric acid and washed thoroughly with water before use.Buffer solution, concentrated-Dissolve 275 g of ammonium acetate and 110 g of hydrated sodium acetate in 1 litre of water. Add 10ml of glacial acetic acid, and mix thoroughly. Buffer solution, dilute-To one volume of concentrated buffer solution, add five volumes of water and adjust the pH to exactly 6.1 by adding acetic acid or sodium hydroxide solution. Standard aluminium solution-Dissolve 1.3192 g of analytical-reagent grade aluminium potassium sulphate in water, and dilute to 1 litre in a calibrated flask. bottle, add 900 ml of water and partly immerse the bottle in cold running water. until dissolution is complete, cool, and dilute to 1 litre. 1 ml = 75 pg of aluminium. PROCEDURE- Preparation of sample-Transfer 1 g of sample to a 125-ml conical beaker, add 15 ml of dilute sulphuric acid (3 + 20) and heat gently until dissolution is complete.(If the sample contains more than 1 per cent. of silicon, evaporate the solution until fumes are evolved, re-dissolve the salts by heating with 15 ml of water, and filter.) Cool the solution, and filter through a small pad of filter-paper pulp. Wash the pad with water and reserve the filtrate. (If separate results are required for acid-soluble and acid-insoluble aluminium, determine the former by treating this filtrate as described under “Electrolysis.”) Ignite the pad in a platinum dish, moisten the residue with 3 or 4 drops of dilute sulphuric acid (1 + 4), add 2 ml of hydrofluoric acid and evaporate under an infra-red lamp until fumes are evolved, Continue evaporation until all sulphuric acid has been removed and then heat the dish for a few minutes a t about 800” C.F ~ s e the residue with 0.5 g of sodium hydrogen sulphate, cool, add 10ml of water and warm the dish until the fused mass has dissolved. (If required, determine acid-insoluble aluminium as follows. Extract the fused mass with 10 ml of dilute sulphuric acid (3 + 20), transfer the extract to a small beaker and heat to ensure complete dissolution. Evaporate the solution to about 5 m l and then treat with 10 N sodium hydroxide as described under “Preparation for Colour Development.”) Transfer the aqueous extract to the beaker containing the filtrate, and heat the mixture to ensure complete dissolution of the fused residue. Electrolysis-Cool the solution, and, if the sample contains more than 0.5 per cent.of chromium, neutralise free acid by adding saturated sodium carbonate solution until the first appearance of a permanent precipitate. lie-dissolve this precipitate by adding dilute sulphuric acid (3 + 20), dropwise, and add 1 rril in excess. (If this procedure is necessary, treat the reagent blank solution in a similar manner.) Transfer the solution to a 150-ml squat beaker containing 20 ml of mercury, and dilute to 60 ml. Cover the beaker with a split clock-glass, electrolyse the solution at a current of 2 amperes for 1 hour and then test for completeness of iron removal by a ferricyanide spot-test. When iron has been completely removed, wash the walls of the beaker and the clock-glass with water, and electrolyse for a further 15 minutes, or, in presence of chromium, until the green colour of the solution has disappeared.Without delay, filter the electrolyte through a Whatman Yo. 541 filter-paper and wash the mercury with a minimum of water. (Procedures for the separation of the amalgam after electrolysis and for recovery of mercury from the amalgam are described in British Standard l121C.2) Preparation for colour development-Evaporate the filtered electrolyte to about 5 ml, cool and pour slowly into a 200-ml stainless-steel beaker containing 10 ml of 10 W sodium hydroxide (add this reagent from a polythene measuring cylinder). Cautiously add 5 ml of 5-volume hydrogen peroxide from a dropping pipette, cover the beaker with a clock-glass,Nov., 19581 62 1 heat to boiling-point and boil gently for 10 minutes on a hot-plate.Remove the beaker from the hot-plate, add a little Whatman ashless floc, and set aside for 5 minutes. Filter the solution through a pad of filter-paper pulp (in a polythene funnel) into a 200-ml polythene squat beaker and wash the pad with warm water. Add 2 or 3 drops of phenolphthalein indicator solution to the filtrate and neutralise by adding diluted hydrochloric acid (1 + 1) from a dropping pipette. Stir the solution during neutralisation with a polythene rod, and add 0.5 ml of acid in excess. Cool the solution and dilute to 100 ml in a calibrated flask. (If it is necessary to evaporate the solution somewhat before dilution, this can be done in a glass beaker. However, if care has been taken with washing during the treatment with 10 N sodium hydroxide, the volume a t this stage should be about 90 ml.) Colow development-Transfer a suitable aliquot of the solution, i.e., one containing from 2 to 70 pg of aluminium, to a 250-ml conical beaker.For samples containing up to 0.035 per cent. of aluminium, take a 20-ml aliquot, and for samples containing from 0.035 to 0.07 per cent. of aluminium, take a 10-ml aliquot and add 10 ml of water before colour development. (The range of the method can be extended by taking a 5-ml aliquot or by reducing the sample weight. For batch analysis, the number of aliquots taken for colour development should be restricted to six, including a reagent blank.) To the aliquot for colour development add 5 ml of 5-volume hydrogen peroxide, and mix well.Carefully neutralise by adding 2 N sodium hydroxide from a polythene dropping bottle and add 1 drop in excess. Titrate immediately with 0-2 N hydrochloric acid until the solution is colourless, add exactly 1.0 ml of acid in excess, and mix thoroughly. Add exactly 5.0 ml of Solochrome cyanine R solution, mix, add 50 ml of dilute buffer solution and dilute to 100 ml in a calibrated flask without delay. Set the solution aside for 30 minutes and then measure the optical density against a reagent blank solution in 5-mm cells with a spectro- photometer a t wavelength 535 mp. Re-set the instrument against the blank solution before each sample solution is measured. The optical density of the blank solution is approximately 0.17. Optical-density measurements can also be made with an absorptiometer ; an instrument with a mercury-vapour lamp is suitable, if Ilford No.605 and Calorex H503 filters are used. However, when an absorptiometer is used, the sensitivity is reduced by about half and it is necessary to measure optical density in 1-cm cells; in these circumstances, the optical density of the blank solution is approximately 0.30. Each batch of test solutions must be accompanied by a blank solution. Tests have shown that the reagents used may result in the introduction of up to 4pg of aluminium. Comparative tests over several months have also indicated that the pick-up of aluminium during electrolysis does not exceed 0.5 pg, and hence the blank determination can be simplified by omitting the electrolysis.The aliquot of blank solution should always be equal in volume to that of the test solution. Calibration-Prepare six solutions, each containing 1 g of aluminium-free iron dissolved in 15 ml of dilute sulphuric acid (3 + 20). To these solutions, add, respectively, 0, 1, 2, 3, 4 and 5 ml of standard aluminium solution. Transfer each solution to a mercury-cathode electrolysis cell, dilute to 60 ml and continue as for a sample solution. Measure the optical density of each solution against the solution containing no aluminium. For the instrument used by us, it was found that the optical density (measured in 5-mm cells) multiplied by a factor of 73 was equal to the number of micrograms of aluminium in the aliquot of solution taken for colour development. A factor of 0.0365 gave the percentage of aluminium in a 20-ml aliquot and a factor of 0.073 gave the percentage of aluminium in a 10-ml aliquot. When optical-density measurements were made in 1-cm cells with an absorptiometer, it was found that optical density multiplied by 77 was equal to the number of micrograms of aluminium in the aliquot taken for colour development.These factors are intended as a guide only; specific factors must be determined for individual instruments. ALUMINIUM I N STEEL WITH SOLOCHROME CYANINE R RESULTS In Table I, the results obtained by the proposed method for standard samples of mild and low alloy steel are compared with those obtained by Members of the British Iron and Steel Research Association Methods of Analysis Committee by a standard volumetric r n e t h ~ d .~ ~ ~ In Table 11, a series of results for samples of vacuum-melted mild steel is com- pared with results obtained in our laboratory by the standard volumetric method. These622 SCHOLES AND SMITH THE COLORIMETRIC DETERMINATION OF [Vol. 83 results confirm the observation made earlier on the tendency towards high results when the standard method is used in routine operation. TABLE I COMPARISON BETWEEN VOLUMETRIC ASD PROPOSED METHODS FOR DETERMINING ALUMINIUM Results by the proposed method are given t3 the nearest 0.0005 per cent. except for B.C.S. samples 272, 273 and 255, which are given to the nearest 0.001 per cent. Aluminium r- \ by volumetric Range of aluminium Sumber of Mean aluminium Aluminium by proposed method Sample No.method, % content, (x determinations content, % Santples containing 0.001 to 0.035 per cent. of alumintwn- B.C.S. 271.. . . 0.008" 0.0074 to 0.0095 10 0.0085 B.C.S. 274.. . , 0.033a 0.0315 t o 0.0337 10 0.0325 B.C.S. 275. . . . 0.020b 0.0204 to 04220 6 0,021 B.C.S. 276.. . . O.02Eib 0.0222 to 04239 10 0.023 B.C.S. 2 7 7 . . . . O.O1fib 0.0171 to 0.0198 6 0.0185 M.G.S. 183 . , 0.018C 0.0195 to 04225 6 0.021 B.C.S. 272.. . . 0.06Sa 0.0656 to 0,0703 6 0.068 B.C.S. 273.. . . 0*060* 0.0550 to 0,0598 6 0.067 B.C.S. 255.. .. ~ 0 . 0 5 7 ~ 0.0455 to 0.0503 10 0.048 (I Results obtained by B.I.S.R.A. Methods of Analysis Committee (modified British Standard method). Results obtained by B.I.S.R.A. Methods of -2nalysis Committee (British Standard method) to the c See Reference 1, Table 11.d B.C.S. certificate value. Samples containing 0.035 to 0.07 per cent. of alumzniunt- nearest 0.005 per cent. TABLE I1 COMPARISON BETWEEN VOLUMETRIC, POLAROGRAPHIC AND PROPOSED METHODS FOR DETERMIKISG ALUMINIUM Results by the proposed method are given to the nearest 0.0005 per cent. The polarographic method used was that of Rooney14 Aluminium Sample by volumetric Sample No. method, yo B1 0.002, 0.005 0.002, 0.004 0.002, 0.005 B3 B4 0.004, 0.004 0.001, 0.002 0.003, 0.003 0.006, 0.007 0.003, 0.003 0.012, 0.016 r B2 1 :: Vacuum-melted mild steel B7 _. 1 E r 1 Cast iron . . .. . I Aluminium by polarographic method, yo - 0.0049 - - 0.0022 0.0055 0.0084 < - Aluminium by proposed method, 7; 0.001, 0.001 0.0015, 0.0015 0.0015, 0.0025 0.005, 0.0055 0401.0~0015 . . . ~ . 0.001: 0.0015 0.0015, 0.0015 .0.001, < 0.001 0.005, 0.0065 0.009. 0.0095 0.0012 0.0015, 0.002 0.0061 o m j , 0.005 0.0025 0.003, 0.0035 A comparison has also been made between the proposed method and a polarographic method of exceptional sensitivity developed by 130oney.l~ The aluminium contents of three of the vacuum-melted mild-steel samples have been determined by Rooney, and, in addition, four samples of cast iron have been analysed by both methods. In general, the polaro- graphic and colorimetric results show good agreement, and, together with the results in Table I, are considered to provide satisfactory evidence of the reliability of the proposed procedure. The approximate compositions of the test samples are shown in Table 111. Sufficient results have been obtained for BI-itish Chemical Standards 271, 274 and 276 to provide an estimate of the precision of the method; the standard deviation is &0.0007 per cent.for the first sample, +0.0009 per cent. for the second and k0.0008 per cent. for theNov., 19583 ALUMINIUM IN STEEL WITH SOLOCHROME CYANINE R 623 third. In the range 0.035 to 0.070 per cent, of aluminium, the standard deviation is 10.0017 per cent. for British Chemical Standard 255. The precision of the method is limited by the reagent blank value, which includes the background colour of the dye and the pick-up of aluminium from reagents and glassware. In the tests reported here, reagent blank values were equivalent to an aluminium content of from 0.0055 to 0.0070 per cent. TABLE I11 APPROXIMATE COMPOSITION OF TEST SAMPLES Present in the sample Man- Chrom- Molyb- Vana- Tung- A I 1 Sample Silicon, ganese, Nickel, ium, denum, Copper, dium, sten, Cobalt, O/ % YO % % % YO % % /O - - - - - - - B1 t o B 9 ,.. . 0-3 0.8 B.C.S. 253 . . . . 0.2 0.4 2.9 0 4 1.0 0.5 0.2 B.C.S. 255 . . . . 0.6 1.1 0.6 1.0 1.4 0.2 0.3 B.C.S. 271 to 277* . , 0.4 0.5 0.2 0.2 0-2 0.2 0.1 0.2 0.2 Cast irons 1 to 4* . , 2 ~ 0 0.5 * The figures for these samples are the maximum amounts present. - - - - - - - - - - - CONCLUSIONS A method has been developed in which formation of a coloured lake or complex between aluminium and Solochrome cyanine R can be applied to the determination of aluminium in steel over the range 0.001 to 0.035 per cent. By taking smaller aliquots, it is possible to extend this range to cover any concentration of aluminium in carbon and low alloy steel. ik considerable reduction in the amount of mercury used is an important economic advantage of the proposed method over the standard volumetric procedure. This feature, ease of operation and the achievement of increased speed and sensitivity are the main advantages of the Solochrome cyanine R method. We thank Mr, N. McGowan for assistance in examining the chromatographic separation method and for many helpful suggestions, and Miss C. J. Arlidge for editorial assistance. We also thank Mr. R. C. Rooney of the British Cast Iron Research Association for valuable co-operation in providing a comparison between his polarographic method and our colorimetric method. This paper is published by permission of The British Iron and Steel Research Association. 1. 2. 3. 4. 5. 6. 7 . 8. 9. 10. 11. 12. 13. 14. REFEREXES R.I.S.R.A. Methods of Analysis Committee, J . Iroiz & Steel Iizst., 1954, 176, 263. “Mercury Cathode Electrolysis,” British Standard 1121C : 1955. “Aluminium in Iron, Steel and Ferro-alloys (after Mercury Cathode Separation),” British Standard B. I. S. R. A. Restricted Report MG/DC/281/57. Eegriwe, E., Z . anal. Chenz., 1929, 76, 438. Alten, F., Weiland, H., and Krippenberg, E., Ibid., 1934, 96, 91. Koch, W., Arch. Eisenhuttenw., 193S, 12, 74. ICoch, W., IClinger, P., and Blaschczpk, G., Angew. Chew., 1940, 53, 537. Ikenberry, L. C., and Thomas, A., Anal. Chew., 1951, 23, 1806. Hill, U. T., Ibid., 1956, 28, 1419. Werz, W., and Neuberger, A., Arch. Eisenhuttenw., 1955, 26, 205. “Colour Index,” Second Edition, Parts I and 11, Society of Dyers and Colourists, Bradford, 1957. Bishop, J. R., Analyst, 1956, 81, 201. Rooney, R. C., Ibid., 1958, 83, 546. 1121 : Part 35 : 1965. Received April SLh, 1958