98 DALZIEL AND THOMPSON SOLVENT-EXTRACTION AXD ABSORPTIOMETRIC [ A ? Z d J ' S f , VOl. 91 The Solvent-extraction and Absorptiometric Determination of Iron with 2-Mercaptopyridine-l-oxide BY J. A. W. DALZIEL AND M. THOMPSON* (Department of Chemistry, Chelsea College of Science and Technology, Manresa Road, London, S. W.3) Iron(1r) and iron(II1) can be completely extracted from M sulphuric acid into a 0.02 M solution of 2-mercaptopyridine- 1-oxide and determined absorp- tiometrically at 550 mp. The only significant interference is from copper(I1); the sensitivity of the method is 0-02 pg cm-2. COLORJMETRIC reagents for iron are numerous,l but relatively few are in common use. Those that depend on reaction with iron(1rr) are generally subject to interference from anions such as fluoride and phosphate.Some methods, viz., the thiocyanate and the thioglycollate (mercaptoacetate) methods, have the additional disadvantage of unstable reactions that cause colour fading. Methods that depend on the reaction of reagents such as 2,2'-dipyridyl, 1,lO-phenanthroline and related compounds with iron(I1) are generally preferred because of their good sensitivity. The reagent 2-mercaptopyridine-l-oxide, subsequently referred to as thione, has already been used for the gravimetric determination of iron.2 This paper reports an extension of the use of thione in the determination of smaller amounts of iron by a spectrophotometric method, and one reagent can now be used for determining iron over a wide range of concentrations. In weakly acidic solutions both iron(r1) and iron(II1) are precipitated as the black iron(II1) complex (C,H,NOS),Fe. This complex can be extracted into chloroform, from relatively strong acidic solutions of iron, to give a violet solution that forms the basis of the method.EXPERIMENTAL THE VISIBLE SPECTRA OF SOME THIONE COMPLEXES- Thione complexes of chromium(m), manganese(II), iron(III), cobalt(n), nickel(II), copper(I1) and zinc(I1) were prepared by precipitation from homogeneous solution by the hydrolysis of S-2-pyridyl thiuronium bromide 1-oxide (Thiurone) in the presence of the appropriate metal ion, at a suitable pH. The visible spectra were recorded, on a Unicam SP500 spectrophotometer, of solutions of known concentration of each neutral inner-complex in analytical-reagent grade chloroform. The spectra are reproduced in Fig.1. Their most notable feature is the exceptional absorption of the iron complex compared with the other complexes. Analytically this is important as it imparts an inherent selectivity to the colorimetric method for iron. The molecular extinction coefficient at 550 mp is about 3400, which is greater than the extinction coefficient of the other complexes by a factor of more than ten, e.g., for the copper(rr) complex at 550 mp the value is only about 150. The strong absorption of the iron(II1) complex has been attributed to charge-transfer rather than to d - d transition^.^ THE EXTRACTION OF IRON AND OTHER METALS- The extraction of iron(II1) from aqueous sulphuric acid into a 0.02 M solution of thione in chloroform was studied ; it was found that iron was extracted quantitatively from solutions with acidities varying between 5 M and 0.005 M.Of the other metals studied only copper was extracted over the whole of this range. Nickel was partly extracted from solutions weaker than O ~ M , and cobalt from solutions weaker than 0.01 M in sulphuric acid. Chrom- ium(m) was not extracted at all, even after prolonged mixing of the phases. A molar solution of sulphuric acid was selected for the composition of the aqueous phase in the recommended method, as this concentration would prevent the extraction of nickel and cobalt. Iron can also be extracted with equal efficiency from 2 N solutions of hydrochloric acid, nitric acid (free from nitrous acid), acetic acid and even phosphoric acid and hydrofluoric acid, * Present address : London Transport Research Laboratories, 566 Chiswick High Road, London, W.4.February, 19661 DETERMINATION OF IRON WITH 2-MERCAPTOPYRIDINE-1-OXIDE 99 Wavelength, mp Fig.1. Visible spectra of 1 + 2 complexes of cobalt(II), copper(II), nickel(II), zinc(I1) and 1 + 3 complexes of iron(II1) and chromium(II1) with thione METHOD REAGENT- Thione solution, 0.02 M, in analytical-reagent grade chloroform. Dissolve 10 g of Thiurone, S-2-pyridylthiuronium bromide-1-oxide (Hopkin and Williams Ltd.), in 40 ml of M sodium hydroxide and boil the solution for about 2 minutes to complete the hydrolysis. Acidify the solution, cool it rapidly with continuous agitation, and filter off, wash and dry the freshly prepared thione. Dissolve 2-5 g of thione in 1 litre of chloroform. The preparation should be carried out with as little exposure to light as possible, and the chloroform solution should be stored in a brown bottle. PROCED u RE- Prepare a solution of iron in M sulphuric acid by a method appropriate for the sample; the iron can be in the (11) or (111) oxidation state. Transfer a 25-ml portion, containing up to 0.4 mg of iron, into a 100-ml separating funnel.Add to the solution with a pipette, 25 ml of the thione solution, and mix the phases by shaking the funnel for 3 minutes. Allow the phases to separate, and run off some of the chloroform through a plug of cotton-wool in the stem of the funnel into a 1-cm spectrophotometer cell. Measure the optical density at a wavelength of 550mp, with some thione solution extracted with pure M sulphuric acid as reference solution.CALIBRATION- A calibration graph was prepared by the extraction of various known amounts of iron(m), up to about 0.4 mg, by the recommended procedure. The best straight line through twelve points was calculated and found to be of the form : optical density = 0.002 + 2.06 x mg of iron extracted. This is proof of adherence to Beer’s law within the experimental error. The sensitivity of the method as described is thus 0.02 pg cm-2. REPRODUCIBILITY OF THE METHOD The reproducibility of the method was investigated by performing replicate extractions on constant amounts of iron, to give a mean optical density of about 0.6. The estimate of100 DALZIEL AND THOMPSON : SOLVENT-EXTRACTION AND ABSORPTIOMETRIC [l!%a&Sif, VOl.91 the standard deviation of the observed optical densities was found to be 1.3 per cent. of the mean. The estimate of the standard deviation due to the instrument alone was 0.9 per cent., giving (by subtraction of the variances) the reproducibility of the extraction as 1.0 per cent. It has been previously noted that thione is decomposed by light,2 and therefore the solution of thione in chloroform must be stored away from light. However, the solution is stable for at least 1 week if stored in a brown bottle, and exposure to light during normal extraction manipulations causes no measurable change in the solution. Chloroform solutions containing extracted iron stored in the open laboratory, but not exposed to direct sunlight, showed no change in their optical densities during several days.Exposure to direct sunlight causes rapid fading of the iron colour, presumably due to the continuous photo-decomposition of the small equilibrium concentration of free thione. INTERFERENCES Interference with the method was tested by performing the extraction of a fixed amount (0-3mg) of iron(m) by the recommended procedure, in the presence of known amounts of foreign substances. The optical density of the resulting solution was compared with that of an extract of an aliquot of pure iron solution. If the difference amounted to more than 2 per cent., the experiment was repeated with a smaller proportion of the interfering substance. The proportion of interfering substance causing an error of 2 per cent.was regarded as being a tolerable ratio, and is the figure quoted in Table I. TABLE I INTERFERENCE FROM VARIOUS SUBSTANCES IN THE DETERMINATION OF 0.3mg OF IRON COLOUR STABILITY Interfering substance Mg2+ . . Ti4+ . . Cr3+ . . Mn2+ . . co2+ . . Ni2+ . . cu2+ . . Zn2+ . . Cd2+ . . Hg2+ . . Sn2+ . . A13+ . . Ce4+ . . Th4+ . . Zr02+ . . NH30H+ NO,- . . NO,- . . c1- . . H,PO,- . . F- . . Aceiic acid Thiourea v0,- . . uo22+ . . . . . . .. .. .. .. . . .. .. . . . . . . . . .. . . . . .. . . . . .. . . . . .. . . . . . . .. .. .. . . . I .. .. .. . . .. .. .. . . . . . . . . .. . . .. . . .. .. . . . . Amount added, mg 6.0 6-0 1-5 6.0 6.0 6.0 6.0 6.0 6-0 6.0 6.0 6.0 6.0 6.0 6-0 6.0 6.0 12-0 300 30 150 450 900 300 600 Iron found, per cent. added 100 103 100 99 99 100 100 98 97 95 101 100 99 100 100 100 100 97 100 100 100 100 100 * * Tolerable ratio of other ion > 20 13 >5 PO 40 > 20 > 20 20 13 8 40 > 20 40 > 20 > 20 > 20 > 40 670 0.2 - > 500 > 1500 > 3000 > 1000 > 2000 * Denotes that gross interference occurs.It can be seen that the determination of iron is free from interference of a wide range of substances. The only important exception is copper, which will cause gross interference if it is not separated prior to the determination of iron. Attempts to “mask” the copper by the addition of 2 per cent. of thiourea to the aqueous phase were unsuccessful. DISCUSSION The colorimetric method of determination is inherently more selective for iron than the gravimetric method that we have previously reported.2 This is because of the relatively large molecular extinction coefficient of the iron(II1) complex and the high acidities at whichFebruary, 19661 DETERMINATION OF IRON WITH 2-MERCAPTOPYRIDINE-1-0~1~~ 101 it is extracted.The latter is due to the combination of the stability of the complex with its favourable partition coefficient from the aqueous phase into chloroform. The method can be used to determine both iron@) and iron(rrr), i.e., total iron, and is exceptional among methods of this kind in being unaffected by high concentrations of fluoride and phosphate. The colour is stable, provided that chloroform solutions of the complex are not left in strong sunlight. The method appears to have two disadvantages in comparison with those that use 1,lO-phenanthroline with iron(I1). These are the lower sensitivity of the thione method, although this is, to some extent, compensated by the possibility of concentration of the iron complex in the solvent-extraction stage, and the gross interference of copper which necessitates its prior separation from the iron. We are indebted to London Transport Board for the provision of study leave for one of us (M.T.). REFERENCES 1. 2. 3. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Interscience Publishers Inc., Dalziel, J . A. W., and Thompson, M., Analyst, 1964, 89, 707. Robinson, M. A . , J . Inorg. Nucl. Chem., 1964, 26, 1277. New York, 1959, pp. 522-554. Received July 23rd, 1965