- Email: [email protected]
Determination of molybdenum in ores, iron and steel by atomic-absorption spectrophotometry after separation by α-benzoinoxime extraction or further xanthate extraction
Ta|anta. Vol. 27, pp. 79 to 84 Pergamon Press Ltd 1980. Printed in Great Britain
DETERMINATION OF MOLYBDENUM IN ORES, IRON AND STEEL BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY AFTER SEPARATION BY ~-BENZOINOXIME EXTRACTION OR FURTHER XANTHATE EXTRACTION ELSIE M. DONALDSON
Mineral Sciences Laboratories, Canada Centre for Mineral and Energy Technology, Department of Energy, Mines and Resources, Ottawa, Canada
(Received 22 June 1979. Accepted 31 August 1979) Summary--A simple and moderately rapid method for determining 0.001% or more of molybdenum in ores, iron and steel is described. After sample decomposition, molybdenum is separated from the matrix elements, except tungsten, by chloroform extraction of its u-benzoinoxime complex from a 1.75M hydrochloric~ll3M tartaric acid medium. Depending on the amount of tungsten present, molybdenum, if necessary, is back-extracted into concentrated ammonia solution and subsequently separated from coextracted tungsten by chloroform extraction of its xanthate complex from a 1.5M hydrochloric-0.13M tartaric acid medium. It is ultimately determined by atomic-absorption spectrophotometry, at 313.3 nm, in a 15% v/v hydrochloric acid medium containing 1000 /tg/ml of aluminium as the chloride, after evaporation of either extract to dryness with nitric, perchloric and sulphuric acids and dissolution of the salts in dilute ammonia solution.
This paper describes the successful determination of molybdenum in ores, iron and steel, after its separation from the matrix elements, except tungsten, by chloroform extraction of its u-benzoinoxime complex from 1.75M hydrochloric acid containing tartaric acid to keep tungsten in solution. Depending on the tungsten content of the sample, molybdenum, if necessary, is subsequently separated from coextracted tungsten, after back-extraction of both elements into concentrated ammonia solution, by chloroform extraction of its xanthate complex from a 1.5M hydrochloric-0.13M tartaric acid medium.
For use in the Canadian Certified Reference Materials Project and in routine work in the CANMET chemical laboratory, a reasonably simple and reliable atomic-absorption (AAS) method was required for the determination of ~<200 #g/g of molybdenum in diverse ores and mill products. Because the determination of molybdenum by AAS is subject to many interferences caused by the formation of refractory (not easily dissociated) compounds in the flame, 1-5 it was considered that a suitable method would involve a relatively selective solvent-extraction preconcentration step. In recent years, numerous AAS methods based on the separation and preconcentration of molybdenum by extraction of its thiocyanate, 8-hydroxyquinoline, sodium diethyldithiocarbamate, ammonium pyrrolidinedithiocarbamate and dithiol complexes have been reported. ~-s In these methods, molybdenum is determined by direct aspiration of the organic phase into the flame. A method involving its separation by chloroform extraction of its ~-benzoinoxime complex and its ultimate determination in an ammonium chloride-perchloric acid medium has also been reported? Because of the apparent simplicity and relatively high specificity of this extraction procedure, t°-*2 and the greater ease of preparation of aqueous calibration solutions, particularly for routine work, the applicability of this separation procedure to the determination of small amounts of molybdenum in ores was investigated.
EXPERIMENTAL
Apparatus A Varian Techtron Model AA6 spectrophotometer equipped with a 10-cm laminar-flow air-acetylene burner and a molybdenum hollow-cathode lamp was used for the determination of molybdenum. The following instrumental parameters were employed (Note 1). Wavelength: 313.3 nm. Lamp current: 5 mA. Spectral band-pass: 0.10 nm. Height of light-path above burner: 8 ram. Acetylene flowmeter reading: 4.0-4.5 ( ~ 3 l./min). Air flowmeter reading: 6.5 ( ~ 13 l./min). Flame: brightly luminous, fuel-rich. Aspiration rate: 2 mi/min.
Reagents Standard molybdenum solution, lO001~g/ml. Dissolve 1.5000 g of pure molybdenum trioxide in 50 ml of 2% sodium hydroxide solution and dilute the solution to 1 litre with water. Prepare a 100/ag/ml solution by diluting 25 ml of this stock solution to 250 ral with water.
Crown Copyrights reserved. TAL. 27/2--A
79
80
ELSIE M. DONALDSON
Aluminium, 1% solution. Dissolve 10 g of aluminium metal by heating gently with 400 ml of 50% v/v hydrochloric acid. Cool and, if necessary, filter the solution (Whatman No. 42 paper) into a l-litre standard flask containing 300 ml of concentrated hydrochloric acid. Dilute the resulting solution to volume with water. ot-Benzoinoxime, 0.2% solution in chloroform. Potassium ethyl xanthate, 20% solution. Prepare fresh as required. Ferrous ammonium sulphate, 10% solution. Prepare fresh as required. Bromine, 20% v/.v solution in carbon tetrachloride. Procedures Calibration solutions. To eight 1000ml beakers, add, by burette, 1, 3, 5, 10, 15, 20, 25 and 30 ml respectively, of the 100-gg/ml standard molybdenum solution. Add ~ 10 drops of 50% v/v sulphuric acid to each beaker and evaporate the solutions to dryness. Cool, wash down the sides of the beakers with water and evaporate the solutions to dryness again to ensure the complete removal of sulphuric acid. Add 1 ml of 25% ammonia solution, 1 drop of 0.2% phenolphthalein solution and ~ 5 ml of water to each beaker and heat gently until the solutions are colourless. Add 10 rnl each of concentrated hydrochloric acid and 1% aluminium solution and transfer the resulting solutions to 100-ml standard flasks. Add I0 ml each of concentrated hydrochloric acid and 1% aluminium solution to a ninth flask; this constitutes the zero calibration solution. Dilute each solution to volume with water and mix (Note 2). Ores. Transfer 0.1-1 g of powdered sample, containing up to ~ 2.5 rag of molybdenum and 25 mg of tungsten, to a 4000ml Teflon beaker, cover and add 20 ml of 50% nitric acid and 5 ml of 20% bromine solution in carbon tetrachloride (Notes 3 and 4). Allow the solution to stand for ~ 10 rain, then heat gently to remove the bromine and carbon tetrachloride. Add 10 ml of concentrated hydrochloric acid and 20 ml of 50% sulphuric acid and heat until the evolution of oxides of nitrogen ceases. Remove the cover, wash down the sides of the beaker with water, add 5 ml of concentrated hydrofluoric acid and carefully evaporate the solution to ~ 3 ml. Cool, add 25 ml of water, heat to dissolve the soluble salts, then add 10 ml of bromine water (Note 5) and heat gently to remove the excess of bromine. Add 10 ml of 20% tartaric acid solution and a small piece of red litmus paper to the resulting solution, then add 50% sodium hydroxide solution, in ~0.5-ml portions, until the solution is alkaline or, depending on the amount of iron present, a mahogany colour. Allow the solution to stand for several min to ensure the complete dissolution of any tungsten trioxide present, then add concentrated hydrochloric acid, dropwise, until the solution is acidic. Add 15 ml in excess and, if necessary, filter the resulting solution (Whatman No. 40 paper) into a 250-ml separatory funnel marked at 100 mL Wash the beaker and the paper and residue each three times with small portions of water, then discard the paper and residue. Add 5 ml of 10% ferrous ammonium sulphate solution to the funnel, dilute the solution to the mark with water and mix thoroughly. Add 15 ml of 0.2% ,,-benzoinoxime solution, stopper, shake for 1 min and allow several rain for the layers to separate (Note 6). If tungsten is absent or not more than ~ 2 rag is present (Note 7), drain the chloroform phase into a 1500ml beaker. Extract the aqueous phase three more times, in a similar manner, with 15-ml portions of the a-benzoinoxime solution. Add 10 ml of 50% v/v nitric acid to the combined extracts and heat in a hot water-bath to remove the chloroform. Add 3 ml of concentrated perchloric acid and 2 ml of 50% sulphuric acid, cover the beaker and heat until the evolution of oxides of nitrogen ceases. Remove the cover and evaporate the solution to dryness. Cool, wash down
the sides of the beaker with water and evaporate the solution to dryness again to ensure the complete removal of suiphuric acid (Note 8). Add I ml of 25% ammonia solution,. 1 drop of 0.2% phenolphthalein solution and ~ 3 ml of water and heat gently until the solution is colourless. Add sufficient concentrated hydrochloric acid for 1 ml to be present for each 10 ml of final solution, then add the same volume of 1% aluminium solution. Transfer the solution to a standard flask of appropriate size (10-100 ml), dilute to volume with water and mix. Measure the ahsorbance of the resulting solution, at 313.3 nm, in a strongly reducing air-acetylene flame (Note 9). Determine the molybdenum content of the solution by relating the resulting value to those obtained concurrently for calibration solutions of slightly higher and lower molybdenum concentrations. If more than 2 mg of tungsten is present, collect the a-benzoinoxime extracts (Note 10) in a 125-ml separatory funnel (Note 11), then add 10 ml of concentrated ammonia solution, stopper and shake for 4 min. Allow ~ 5 min for the layers to separate, then drain off and discard the chloroform layer. Add, in succession, 5 ml of 20% tartaric acid solution, 20 ml of water and 15 ml of concentrated hydrochloric acid, blow the resultant ammonium chloride fumes out of the funnel and cool the solution to room temperature. Add 2 ml of freshly prepared 20% potassium ethyl xanthate solution (Note 12), stopper and mix thoroughly. Let the solution stand for ~ 1 min to allow the formation of the reddish-purple molybdenum xanthate complex, then add 10 ml of chloroform and shake for 1 rain. Allow several rain for the layers to separate, then drain the chloroform phase into a 150-ml beaker. Extract the aqueous phase three more times, in a similar manner, with 100, 5- and 5-ml portions of chloroform and 1, 0.5 and 0.5 ml, respectively, of xanthate solution (Note 13). Add 10 mi of 50% nitric acid to the combined extracts and heat in a hot water-bath to remove the chloroform. Add 3 ml of concentrated perchloric acid and 2 ml of 50% sulphuric acid and proceed with the evaporation of the solution to dryness, the dissolution of the salts in 25% ammonia solution and the subsequent determination of molybdenum as described above. Iron and steel. Transfer 0.1-1 g of sample, containing up to ~2.5 mg of molybdenum and 25 mg of tungsten (Note 4), to a 4000ml beaker, cover and add 10 nil each of concentrated hydrochloric acid and 50% nitric acid. Heat gently until the sample is decomposed, then remove the cover, add 4 drops of concentrated hydrofluoric acid and evaporate the solution to dryness in a hot water-bath. Add 5 ml of concentrated hydrochloric acid and 10 mi of bromine water and evaporate the solution to dryness again to ensure the removal of most of the nitric acid. If tungsten is known to be absent, add 15 ml of concentrated hydrochloric acid and 10 nd of 20% tartaric acid solution and, if necessary, heat gently to dissolve the salts. If necessary, filter the resulting solution (Whatman No. 40 paper) into a 2500ml separatory funnel marked at 100 ml, add 5 ml of 10% ferrous ammonium sulphate solution, dilute to the mark with water and proceed with the a-henzoinoxime extraction and the subsequent determination of molybdenum as described above. If tungsten is present, add 5 ml of concentrated hydrochloric acid, ~25 ml of water and 10 mi of 20% tartaric acid solution and, if necessary, heat gently to dissolve the soluble salts. Add 50% sodium hydroxide solution, in ~0.5-ml portions, until the solution is a dark mahogany colour, then allow it to stand for several min to ensure the complete dissolution of tungsten trioxide. Add concentrated hydrochloric acid, dropwise, until the solution is acidic (clear yellow), then add 15 ml in excess. Proceed with the filtration of the solution (if necessary), the a-benzoinoxime extraction and, if necessary, the xanthate extraction (depending on the amount of tungsten present) and
Molybdenum in ores, iron and steel the subsequent determination of molybdenum as described above.
Notes 1. A strongly reducing air-acetylene flame is required to obtain the highest sensitivity for molybdenum. The height at which" the beam from the hollow-cathode lamp passes through the flame is also extremely important, l"s Consequently, after all other instrumental parameters have been set, the acetylene flow-rate and the height of the light-path above the burner should be adjusted to give maximum absorbance while a solution containing molybdenum is aspirated into the flame. 2. The calibration solutions should be prepared fresh every day because they are not stable on standing. Evaporation of the molybdenum solutions to dryness with suiphuric acid is required for the molybdenum in the resulting calibration solutions to be present in the same form as in the sample solutions obtained after treatment of xanthate extracts with nitric acid, which produces sulphuric acid. 3. The addition of carbon tetrachloride solution of bromine is not necessary if the sample does not contain sulphitles. 4. For samples of high molybdenum content, up to 1 g can be taken (tungsten content ~<25 mg) if the solution ultimately obtained after the addition of 15 ml of concentrated hydrochloric acid is diluted to 100 ml with water. A suitable aliquot of the resultant solution--to which the recommended volume of ferrous ammonium sulphate solution has been added--can subsequently be diluted to ~ 100 ml in the separatory funnel with 15% hydrochloric acid-2% tartaric acid solution before the a-benzoinoxime extraction step. 5. Bromine wirer is added to ensure that all of the molybdenum is present in the sexivalent state required for its extraction w'ith ,,-benzoinoxime. 6. The u-benzoinoxime cQmplexes of molybdenum and tungsten are not appreciably soluble in chloroform. Consequently, if mg-quantities of these elements are present, the chloroform phase will be cloudy or will contain flocculent white material. This does not interfere with the quantitative separation of molybdenum. 7. If the molybdenum content of the sample is so low that the final solution is to be diluted to 10 ml before the determination of molybdenum, the amount of tungsten that is co-extracted at the 2-mg level may interfere by precipitating in the final solution. In that case, it is recommended that the molybdenum should be stripped from the chloroform phase and separated from the co-extracted tungsten by xanthate extraction as described in the subsequent procedure. 8. If the tungsten content of the sample is not known, its presence will be indicated at this point (see also Note 10) by a yellow compound (WO3) that is insoluble in water. If an appreciable amount is present, add 0.5 ml of concentrated perchloric acid and evaporate the solution to dryness again. Add ~ 25 ml of water, 5 ml of 20% tartaric acid solution and 1 drop of 0.2% phenolphthalein solution, then make alkaline with 50% sodium hydroxide solution added dropwise. Add concentrated hydrochloric acid, dropwise, until the solution is acidic, then add 6 ml in excess and transfer the solution to a 125-ml separatory funnel marked at 50 ml. Dilute the resultant solution to the mark with water and proceed with the separation of molybdenum by extraction as the xanthate. 9. Scale expansion (~ 5-fold) is recommended for the determination of approximately 3/~g/ml or less of molybdenum. 10. If the sample contains an appreciable amount of tungsten, the fourth extract will still be cloudy. 11. The separatory funnel should be drained thoroughly after washing, to prevent dilution of the concentrated ammonia solution used for the subsequent back-extraction of molybdenum.
81
12. The xanthate solution should be added by safety pipette or a graduated or marked medicine dropper, and the extraction should be carried out in a fume hood. Prolonged exposure to xanthate vapour can produce an allergic reaction. 13. Usually a four-stage extraction with a total volume of 4 ml of 20% potassium ethyl xanthate solution is sufficient for the separation of up to 2.5 mg of molybdenum. However, if the aqueous phase is still pink after the fourth addition of xanthate solution, continue the extraction~ using 5-ml portions of chloroform and 0.5 ml of xanthate solution until both the aqueous and chloroform phases are colourless. RESULTS
Calibration solutions In initial tests, the calibration solutions used for comparison purposes were prepared by direct dilution of appropriate volumes of the standard molybdenum solution and the recommended volumes of concentrated hydrochloric acid and 1% aluminium solution. However, in these tests high results ( ~ 2-3 #g/ml at the 25-#g/ml level) were obtained with an air-acetylene flame, when the molybdenum ~-benzoinoxime extracts were treated with nitric, perchloric and sulphuric acids, followed by evaporation of the solution to dryness and dissolution of the salts in dilute hydrochloric acid. This was considered to be due to a change in the oxidation state of the molybdenum because a blue compound is produced under these conditions (though only when sulphuric acid is present). Previously, Hutchison, 9 using calibration solutions prepared from ammonium molybdate, obtained low results when aliquots of the standard solutions were evaporated to dryness with perchloric acid and the salts were dissolved in dilute perchloric acid; this was attributed to the formation of molybdic acid. Hutchison found that complete recovery of molybdenum could be obtained if the molybdic acid was subsequently converted into a m m o n i u m molybdate by dissolving the salts in dilute ammonia solution and evaporating the resultant solution to dryness. However, high results were still obtained in the present work when this procedure involving evaporation with perchloric acid and subsequent dissolution of the salts in dilute ammonia solution was applied to the ~-benzoinoxime extracts. Ultimately it was found that this positive error was due to an increase in the absorbance of molybdenum solutions after evaporation to dryness with sulphuric or perchloric acids and that it can be readily avoided by treating the molybdenum solutions taken for calibration purposes in the same way as the sample solutions. The molybdenum in the resultant calibration solutions will subsequently be present in the same form as in the final sample solutions.
Extraction complex
of the molybdenum(H) ~-benzoinoxime
Previous investigators 1°,11 showed that the molybdenum(VI) ~-benzoinoxime complex can be quantitat-
82
Et.sm M. DONALDSON
ively extracted into chloroform from up to approximately 2.3M hydrochloric acid, and that the extraction step is reasonably specific when chromium(VI) and vanadium(V) are reduced with ferrous ammonium sulphate and thus prevented from reacting with the reagent. ~3 Only niobium, zirconium, tungsten(VI) t2 (and possibly palladium) ~a are partly coextracted from _>IM hydrochloric acid, and the coextraction of niobium and zirconium can be readily prevented by complexing them with hydrofluoric acid. ~2 However, the possible interference of tungsten had to be considered in the present work because it is a common constituent of ores. Hutchison 9 reported that tungsten does not interfere in the determination of molybdenum by AAS after its separation by ~t-benzoinoxime extraction. However, tests showed that in an air-acetylene flame 100 #g/ml suppress the absorbance of 20 #g/ml of molybdenum by about 10%. Tungsten also interferes by precipitating in the final dilute acid solution. Tartaric acid cannot be used in this case to keep tungsten in solution because it strongly suppresses the molybdenum absorbance. Tungsten also causes difficulty before the ~t-benzoinoxime extraction step because of its insolubility in acid media. It has been reported that potassium dihydrogen phosphate complexes tungsten and inhibits the extraction of its a-benzoinoxime complex ~4 but this reagent was found to be completely ineffective. Further work showed that up to ~ 25 mg of tungsten can be kept in solution, in ~ 1.7M hydrochloric acid, with 2 g of tartaric acid and that, under these conditions, up to at least 2.5 nag of molybdenum can be quantitatively extracted as the ~t-benzoinoxime complex in four successive extractions with 15-ml portions of 0.2% solution of the reagent in chloroform. Approximately 50°,/0 of the tungsten initially present in the solution is co-extracted under these conditions. Consequently, a method is required for the subsequent separation of tungsten if more than a few rag are present in the sample taken for analysis.
Separation of molybdenumfrom tungsten by extraction of its xanthate complex Previous work by the author ~ showed that molybdenum [after its reduction to molybdenum(V) by xanthate] can be quantitatively extracted into chloroform as the xanthate from 0.1-~2M hydrochloric acid. In earlier work the extraction of molybdenum xanthate from 1.5M hydrochloric acid containing tartaric acid was used for the separation of molybdenum from tungsten before the spectrophotometric determination of tungsten after its extraction as the thiocyanatediantipyrylmethane ion-association complex. ~6 Consequently, it was considered that this separation procedure could also be used in the present work after back-extraction of the molybdenum and tungsten ~t-benzoinoxime complexes from the chloroform phase.
Preliminary experiments showed that the molybdenum complex cannot be completely back-extracted from the organic phase by shaking it with 4M ammonia solution or 10% sodium hydroxide solution or with 12M hydrochloric acid. However, shaking for ~4 rain with 15M ammonia solution was found to be effective. Under these conditions only negligible amounts of molybdenum ( <4 #g at the 2.5-rag level) remain in the organic phase. Subsequent work showed that, after appropriate treatment of the resultant ammoniacal solution, molybdenum can be readily separated from tungsten by extraction as the xanthate from 50 ml of 1.5M hydrochloric acid containing ~ 1 g of tartaric acid to keep tungsten in solution.
Effect of diverse ions As mentioned above, only niobium, zirconium, tungsten(VI), lz and possibly palladium 13 are partly co-extracted as a-benzoinoxime complexes from hydrochloric acid media containing ferrous ammonium sulphate as a reductant for chromium(VI) and vanadium(V). However, tests showed that zirconium is not extracted in the presence of tartaric acid, and niobium forms an insoluble hydrolysis compound during the sample decomposition step. This compound is removed by filtration before the extraction step. The effect of palladium was not considered because more than #g-quantities are not usually present in ores. The amount of tungsten that is co-extracted, at approximately the 2-rag level., will not interfere in the subsequent determination of molybdenum in an airacetylene flame when aluminium chloride solution ~ is added to the final solution to obviate the interference of tungsten, and if the volume of the solution is > 25 ml. If the final volume is less, tungsten may precipitate. Because the tungsten compound that is formed on evaporation of the ct-benzoinoxime extract to dryness with acids is insoluble in water or dilute acids, dilute ammonia solution is required to dissolve this compound and convert it into ammonium tungstate before the addition of hydrochloric acid and aluminium solution. The excess of ammonia is readily removed by heating the solution.
Applications To test the reliability of the proposed method, it was applied to the analysis of three CCRMP ores that have been certified for molybdenum and to three CCRMP tungsten ores for which only approximate molybdenum values are available. 17 It was also applied to certified reference iron and steel samples. The results of these analyses are given in Table 1. DISCUSSION
Table I shows that the results obtained for the CCRMP reference ores MP-1, HV-1 and PR-I and, where applicable, for the National Bureau of Standards (NBS) and British Chemical Standards (BCS)
Molybdenum in ores, iron and steel
~_,~ o~
~e ~,.~
-
~
o
o
IIIII
I
~
83
~
~
~
~
o. 0
e.
•~.~
a
0
0
< Z
0
,g
,,.-k e-, o ..~
8:
i
i2
~
~
~--~~~
~~
~
~ - ~
-
~
~
~
~
~ es --
0
9 ~o.~. 0
r
y
0
~
o
~
=
.~ o
o
=
o-
._~o-~ ~ ~.~
~'~-~-~ o~
~.__.
Z
Z
u ~ z
~='"
~-~..~
=
.~
~=z~
"7 ,.~
[..
ZZZZZ
Z
Z
Z
ZZ
Z
~
,~
84
ELSm M. DONALDSON
iron and steel samples, after separation of molybdenum by ct-benzoinoxime extraction alone, are in excellent agreement with the certified values. Similarly, the results obtained for the ores, after the addition of tungsten, and for the NBS and BCS samples containing tungsten, after further separation of molybdenum from co-extracted tungsten by xanthate extraction, are also in good agreement with the certified values and, where applicable, with the values obtained after ~t-benzoinoxime extraction alone. The results obtained for the three CCRMP tungsten ores, CT-1, BH-1 and TLG-1 are considered to be more reliable than those given for information purposes. 17 The latter are only approximate values obtained in the CANMET chemical laboratory by a spectrophotometric thiocyanate method after the separation of iron by precipitation with sodium hydroxide. Molybdenum can be separated from tungsten and certain other elements by direct xanthate extraction, ~s but this was not considered in the present work because the extraction step is not sufficiently selective. 15 The co-extraction of iron(Ill)can be avoided by reducing it with ascorbic acid, and that of copper(II) and vanadium(V) by complexing them with thiourea and potassium hydrogen fluoride, respectively, la Tin(II and IV), antimony(Ill), arsenic(IlI) and selenium(IV), which are also co-extracted, :5 can be removed by volatilization as the bromides during the sample decomposition step. However, the co-extraction of certain elements that commonly occur in ores, such as cobalt, nickel, silver, bismuth and, particularly, lead cannot be prevented. ~ Possibly more tungsten can be tolerated during the ct-benzoinoxime extraction step if the initial tartaric acid concentration is increased. However, because tartaric acid slightly inhibits the extraction of the molybdenum ~t-benzoinoxime complex, a more concentrated solution of the reagent in chloroform would be required for the complete extraction of molybdenum. Under these conditions, more tungsten would also be co-extracted. Subsequently, more than 1 g of tartaric acid may be required to keep tungsten in solution before and during the xanthate extraction step. The addition of more is not recommended because too much tartaric acid inhibits the extraction of molybdenum as the xanthate. Up to six extraction stages are required for the complete extraction of 2.5 mg of molybdenum from 50 ml of 1.5-2M hydrochloric acid
containing 2 g of tartaric acid. The inhibiting effect of tartaric acid is greater at lower hydrochloric acid concentrations. A hydrochloric acid medium was chosen for the ultimate determination of molybdenum because this acid has very little effect on the absorbance of molybdenum in either an air-acetylene or a nitrous oxideacetylene flame.4's'19 A 15% hydrochloric acid medium was chosen for convenience when working with 10-ml final sample solution volumes, Probably the addition of I ml of 50% hydrochloric acid, rather than the concentrated acid, would also be sufficient. However, the concentration of hydrochloric acid in the calibration solutions should be adjusted accordingly. An air-acetylene flame was chosen because it was found to be sufficiently sensitive and because interference effects from other elements, anions and acids are considered to be less in this flame than in the nitrous oxide-acetylene flame. 19'2°
REFERENCES
1. D. J. David, Analyst, 1961, 86, 730. 2. ldera, ibid., 1968, 93, 79. 3. R. A. Mostyn and A. F. Cunningham, Anal. Chem., 1966, 38, 121. 4. T. V. Ramah'ishna, P. W. West and J. W. Robinson, Anal. Chirt Acta, 1969, 44, 437. 5. A. Purushottam, P. P. Naidu and S. S. Lal, Talanta, 1972, 19, 1193. 6. C. H. Kim, C. M. Owens and L. E. Smythe, ibid., 1974, 21, 445 (and references therein). 7. C. H. Kim, P. W. Alexander and L. E. Smythe, ibid., 1975, 22, 739. 8. ldem, ibid., 1976, 23, 229 (and references therein). 9. D. Hutchison, Analyst, 1972, 97, 118. 10. P. G. Jeffery, ibid., 1956, 81, 104. 11. G. Goldstein, D. L. Manning and O. Menis, Anal. Chert, 1958, 30, 539. 12. L. Wish, ibid., 1962, 34, 625. 13. P. Y. Peng and E. B. Sandell, Anal. Chirt Acta, 1963, 29, 325. 14. H. J. Hoenes and K. G. Stone, Talanta, 1960, 4, 250. 15. E. M. Donaldson, ibid., 1976, 23, 411. 16. lderrg ibid., 1975, 22, 837. 17. G. H. Faye, W. S. Bowman and R. Sutarno, CANMET Rept. 76-5, Department of Energy, Mines and Resources, Ottawa, 1976. 18. V. Yatirajam and J. Ram, Talanta, 1974, 21,439. 19. S. Dilli, K. M. Gawne and G. W. Ocago, Anal. Chirt Acta, 1974, 69, 287. 20. J. D. Kerbyson and C. Ratzkowski, Can. Spectrosc., 1970, 15, 43.
DETERMINATION OF MOLYBDENUM IN ORES, IRON AND STEEL BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY AFTER SEPARATION BY ~-BENZOINOXIME EXTRACTION OR FURTHER XANTHATE EXTRACTION ELSIE M. DONALDSON
Mineral Sciences Laboratories, Canada Centre for Mineral and Energy Technology, Department of Energy, Mines and Resources, Ottawa, Canada
(Received 22 June 1979. Accepted 31 August 1979) Summary--A simple and moderately rapid method for determining 0.001% or more of molybdenum in ores, iron and steel is described. After sample decomposition, molybdenum is separated from the matrix elements, except tungsten, by chloroform extraction of its u-benzoinoxime complex from a 1.75M hydrochloric~ll3M tartaric acid medium. Depending on the amount of tungsten present, molybdenum, if necessary, is back-extracted into concentrated ammonia solution and subsequently separated from coextracted tungsten by chloroform extraction of its xanthate complex from a 1.5M hydrochloric-0.13M tartaric acid medium. It is ultimately determined by atomic-absorption spectrophotometry, at 313.3 nm, in a 15% v/v hydrochloric acid medium containing 1000 /tg/ml of aluminium as the chloride, after evaporation of either extract to dryness with nitric, perchloric and sulphuric acids and dissolution of the salts in dilute ammonia solution.
This paper describes the successful determination of molybdenum in ores, iron and steel, after its separation from the matrix elements, except tungsten, by chloroform extraction of its u-benzoinoxime complex from 1.75M hydrochloric acid containing tartaric acid to keep tungsten in solution. Depending on the tungsten content of the sample, molybdenum, if necessary, is subsequently separated from coextracted tungsten, after back-extraction of both elements into concentrated ammonia solution, by chloroform extraction of its xanthate complex from a 1.5M hydrochloric-0.13M tartaric acid medium.
For use in the Canadian Certified Reference Materials Project and in routine work in the CANMET chemical laboratory, a reasonably simple and reliable atomic-absorption (AAS) method was required for the determination of ~<200 #g/g of molybdenum in diverse ores and mill products. Because the determination of molybdenum by AAS is subject to many interferences caused by the formation of refractory (not easily dissociated) compounds in the flame, 1-5 it was considered that a suitable method would involve a relatively selective solvent-extraction preconcentration step. In recent years, numerous AAS methods based on the separation and preconcentration of molybdenum by extraction of its thiocyanate, 8-hydroxyquinoline, sodium diethyldithiocarbamate, ammonium pyrrolidinedithiocarbamate and dithiol complexes have been reported. ~-s In these methods, molybdenum is determined by direct aspiration of the organic phase into the flame. A method involving its separation by chloroform extraction of its ~-benzoinoxime complex and its ultimate determination in an ammonium chloride-perchloric acid medium has also been reported? Because of the apparent simplicity and relatively high specificity of this extraction procedure, t°-*2 and the greater ease of preparation of aqueous calibration solutions, particularly for routine work, the applicability of this separation procedure to the determination of small amounts of molybdenum in ores was investigated.
EXPERIMENTAL
Apparatus A Varian Techtron Model AA6 spectrophotometer equipped with a 10-cm laminar-flow air-acetylene burner and a molybdenum hollow-cathode lamp was used for the determination of molybdenum. The following instrumental parameters were employed (Note 1). Wavelength: 313.3 nm. Lamp current: 5 mA. Spectral band-pass: 0.10 nm. Height of light-path above burner: 8 ram. Acetylene flowmeter reading: 4.0-4.5 ( ~ 3 l./min). Air flowmeter reading: 6.5 ( ~ 13 l./min). Flame: brightly luminous, fuel-rich. Aspiration rate: 2 mi/min.
Reagents Standard molybdenum solution, lO001~g/ml. Dissolve 1.5000 g of pure molybdenum trioxide in 50 ml of 2% sodium hydroxide solution and dilute the solution to 1 litre with water. Prepare a 100/ag/ml solution by diluting 25 ml of this stock solution to 250 ral with water.
Crown Copyrights reserved. TAL. 27/2--A
79
80
ELSIE M. DONALDSON
Aluminium, 1% solution. Dissolve 10 g of aluminium metal by heating gently with 400 ml of 50% v/v hydrochloric acid. Cool and, if necessary, filter the solution (Whatman No. 42 paper) into a l-litre standard flask containing 300 ml of concentrated hydrochloric acid. Dilute the resulting solution to volume with water. ot-Benzoinoxime, 0.2% solution in chloroform. Potassium ethyl xanthate, 20% solution. Prepare fresh as required. Ferrous ammonium sulphate, 10% solution. Prepare fresh as required. Bromine, 20% v/.v solution in carbon tetrachloride. Procedures Calibration solutions. To eight 1000ml beakers, add, by burette, 1, 3, 5, 10, 15, 20, 25 and 30 ml respectively, of the 100-gg/ml standard molybdenum solution. Add ~ 10 drops of 50% v/v sulphuric acid to each beaker and evaporate the solutions to dryness. Cool, wash down the sides of the beakers with water and evaporate the solutions to dryness again to ensure the complete removal of sulphuric acid. Add 1 ml of 25% ammonia solution, 1 drop of 0.2% phenolphthalein solution and ~ 5 ml of water to each beaker and heat gently until the solutions are colourless. Add 10 rnl each of concentrated hydrochloric acid and 1% aluminium solution and transfer the resulting solutions to 100-ml standard flasks. Add I0 ml each of concentrated hydrochloric acid and 1% aluminium solution to a ninth flask; this constitutes the zero calibration solution. Dilute each solution to volume with water and mix (Note 2). Ores. Transfer 0.1-1 g of powdered sample, containing up to ~ 2.5 rag of molybdenum and 25 mg of tungsten, to a 4000ml Teflon beaker, cover and add 20 ml of 50% nitric acid and 5 ml of 20% bromine solution in carbon tetrachloride (Notes 3 and 4). Allow the solution to stand for ~ 10 rain, then heat gently to remove the bromine and carbon tetrachloride. Add 10 ml of concentrated hydrochloric acid and 20 ml of 50% sulphuric acid and heat until the evolution of oxides of nitrogen ceases. Remove the cover, wash down the sides of the beaker with water, add 5 ml of concentrated hydrofluoric acid and carefully evaporate the solution to ~ 3 ml. Cool, add 25 ml of water, heat to dissolve the soluble salts, then add 10 ml of bromine water (Note 5) and heat gently to remove the excess of bromine. Add 10 ml of 20% tartaric acid solution and a small piece of red litmus paper to the resulting solution, then add 50% sodium hydroxide solution, in ~0.5-ml portions, until the solution is alkaline or, depending on the amount of iron present, a mahogany colour. Allow the solution to stand for several min to ensure the complete dissolution of any tungsten trioxide present, then add concentrated hydrochloric acid, dropwise, until the solution is acidic. Add 15 ml in excess and, if necessary, filter the resulting solution (Whatman No. 40 paper) into a 250-ml separatory funnel marked at 100 mL Wash the beaker and the paper and residue each three times with small portions of water, then discard the paper and residue. Add 5 ml of 10% ferrous ammonium sulphate solution to the funnel, dilute the solution to the mark with water and mix thoroughly. Add 15 ml of 0.2% ,,-benzoinoxime solution, stopper, shake for 1 min and allow several rain for the layers to separate (Note 6). If tungsten is absent or not more than ~ 2 rag is present (Note 7), drain the chloroform phase into a 1500ml beaker. Extract the aqueous phase three more times, in a similar manner, with 15-ml portions of the a-benzoinoxime solution. Add 10 ml of 50% v/v nitric acid to the combined extracts and heat in a hot water-bath to remove the chloroform. Add 3 ml of concentrated perchloric acid and 2 ml of 50% sulphuric acid, cover the beaker and heat until the evolution of oxides of nitrogen ceases. Remove the cover and evaporate the solution to dryness. Cool, wash down
the sides of the beaker with water and evaporate the solution to dryness again to ensure the complete removal of suiphuric acid (Note 8). Add I ml of 25% ammonia solution,. 1 drop of 0.2% phenolphthalein solution and ~ 3 ml of water and heat gently until the solution is colourless. Add sufficient concentrated hydrochloric acid for 1 ml to be present for each 10 ml of final solution, then add the same volume of 1% aluminium solution. Transfer the solution to a standard flask of appropriate size (10-100 ml), dilute to volume with water and mix. Measure the ahsorbance of the resulting solution, at 313.3 nm, in a strongly reducing air-acetylene flame (Note 9). Determine the molybdenum content of the solution by relating the resulting value to those obtained concurrently for calibration solutions of slightly higher and lower molybdenum concentrations. If more than 2 mg of tungsten is present, collect the a-benzoinoxime extracts (Note 10) in a 125-ml separatory funnel (Note 11), then add 10 ml of concentrated ammonia solution, stopper and shake for 4 min. Allow ~ 5 min for the layers to separate, then drain off and discard the chloroform layer. Add, in succession, 5 ml of 20% tartaric acid solution, 20 ml of water and 15 ml of concentrated hydrochloric acid, blow the resultant ammonium chloride fumes out of the funnel and cool the solution to room temperature. Add 2 ml of freshly prepared 20% potassium ethyl xanthate solution (Note 12), stopper and mix thoroughly. Let the solution stand for ~ 1 min to allow the formation of the reddish-purple molybdenum xanthate complex, then add 10 ml of chloroform and shake for 1 rain. Allow several rain for the layers to separate, then drain the chloroform phase into a 150-ml beaker. Extract the aqueous phase three more times, in a similar manner, with 100, 5- and 5-ml portions of chloroform and 1, 0.5 and 0.5 ml, respectively, of xanthate solution (Note 13). Add 10 mi of 50% nitric acid to the combined extracts and heat in a hot water-bath to remove the chloroform. Add 3 ml of concentrated perchloric acid and 2 ml of 50% sulphuric acid and proceed with the evaporation of the solution to dryness, the dissolution of the salts in 25% ammonia solution and the subsequent determination of molybdenum as described above. Iron and steel. Transfer 0.1-1 g of sample, containing up to ~2.5 mg of molybdenum and 25 mg of tungsten (Note 4), to a 4000ml beaker, cover and add 10 nil each of concentrated hydrochloric acid and 50% nitric acid. Heat gently until the sample is decomposed, then remove the cover, add 4 drops of concentrated hydrofluoric acid and evaporate the solution to dryness in a hot water-bath. Add 5 ml of concentrated hydrochloric acid and 10 mi of bromine water and evaporate the solution to dryness again to ensure the removal of most of the nitric acid. If tungsten is known to be absent, add 15 ml of concentrated hydrochloric acid and 10 nd of 20% tartaric acid solution and, if necessary, heat gently to dissolve the salts. If necessary, filter the resulting solution (Whatman No. 40 paper) into a 2500ml separatory funnel marked at 100 ml, add 5 ml of 10% ferrous ammonium sulphate solution, dilute to the mark with water and proceed with the a-henzoinoxime extraction and the subsequent determination of molybdenum as described above. If tungsten is present, add 5 ml of concentrated hydrochloric acid, ~25 ml of water and 10 mi of 20% tartaric acid solution and, if necessary, heat gently to dissolve the soluble salts. Add 50% sodium hydroxide solution, in ~0.5-ml portions, until the solution is a dark mahogany colour, then allow it to stand for several min to ensure the complete dissolution of tungsten trioxide. Add concentrated hydrochloric acid, dropwise, until the solution is acidic (clear yellow), then add 15 ml in excess. Proceed with the filtration of the solution (if necessary), the a-benzoinoxime extraction and, if necessary, the xanthate extraction (depending on the amount of tungsten present) and
Molybdenum in ores, iron and steel the subsequent determination of molybdenum as described above.
Notes 1. A strongly reducing air-acetylene flame is required to obtain the highest sensitivity for molybdenum. The height at which" the beam from the hollow-cathode lamp passes through the flame is also extremely important, l"s Consequently, after all other instrumental parameters have been set, the acetylene flow-rate and the height of the light-path above the burner should be adjusted to give maximum absorbance while a solution containing molybdenum is aspirated into the flame. 2. The calibration solutions should be prepared fresh every day because they are not stable on standing. Evaporation of the molybdenum solutions to dryness with suiphuric acid is required for the molybdenum in the resulting calibration solutions to be present in the same form as in the sample solutions obtained after treatment of xanthate extracts with nitric acid, which produces sulphuric acid. 3. The addition of carbon tetrachloride solution of bromine is not necessary if the sample does not contain sulphitles. 4. For samples of high molybdenum content, up to 1 g can be taken (tungsten content ~<25 mg) if the solution ultimately obtained after the addition of 15 ml of concentrated hydrochloric acid is diluted to 100 ml with water. A suitable aliquot of the resultant solution--to which the recommended volume of ferrous ammonium sulphate solution has been added--can subsequently be diluted to ~ 100 ml in the separatory funnel with 15% hydrochloric acid-2% tartaric acid solution before the a-benzoinoxime extraction step. 5. Bromine wirer is added to ensure that all of the molybdenum is present in the sexivalent state required for its extraction w'ith ,,-benzoinoxime. 6. The u-benzoinoxime cQmplexes of molybdenum and tungsten are not appreciably soluble in chloroform. Consequently, if mg-quantities of these elements are present, the chloroform phase will be cloudy or will contain flocculent white material. This does not interfere with the quantitative separation of molybdenum. 7. If the molybdenum content of the sample is so low that the final solution is to be diluted to 10 ml before the determination of molybdenum, the amount of tungsten that is co-extracted at the 2-mg level may interfere by precipitating in the final solution. In that case, it is recommended that the molybdenum should be stripped from the chloroform phase and separated from the co-extracted tungsten by xanthate extraction as described in the subsequent procedure. 8. If the tungsten content of the sample is not known, its presence will be indicated at this point (see also Note 10) by a yellow compound (WO3) that is insoluble in water. If an appreciable amount is present, add 0.5 ml of concentrated perchloric acid and evaporate the solution to dryness again. Add ~ 25 ml of water, 5 ml of 20% tartaric acid solution and 1 drop of 0.2% phenolphthalein solution, then make alkaline with 50% sodium hydroxide solution added dropwise. Add concentrated hydrochloric acid, dropwise, until the solution is acidic, then add 6 ml in excess and transfer the solution to a 125-ml separatory funnel marked at 50 ml. Dilute the resultant solution to the mark with water and proceed with the separation of molybdenum by extraction as the xanthate. 9. Scale expansion (~ 5-fold) is recommended for the determination of approximately 3/~g/ml or less of molybdenum. 10. If the sample contains an appreciable amount of tungsten, the fourth extract will still be cloudy. 11. The separatory funnel should be drained thoroughly after washing, to prevent dilution of the concentrated ammonia solution used for the subsequent back-extraction of molybdenum.
81
12. The xanthate solution should be added by safety pipette or a graduated or marked medicine dropper, and the extraction should be carried out in a fume hood. Prolonged exposure to xanthate vapour can produce an allergic reaction. 13. Usually a four-stage extraction with a total volume of 4 ml of 20% potassium ethyl xanthate solution is sufficient for the separation of up to 2.5 mg of molybdenum. However, if the aqueous phase is still pink after the fourth addition of xanthate solution, continue the extraction~ using 5-ml portions of chloroform and 0.5 ml of xanthate solution until both the aqueous and chloroform phases are colourless. RESULTS
Calibration solutions In initial tests, the calibration solutions used for comparison purposes were prepared by direct dilution of appropriate volumes of the standard molybdenum solution and the recommended volumes of concentrated hydrochloric acid and 1% aluminium solution. However, in these tests high results ( ~ 2-3 #g/ml at the 25-#g/ml level) were obtained with an air-acetylene flame, when the molybdenum ~-benzoinoxime extracts were treated with nitric, perchloric and sulphuric acids, followed by evaporation of the solution to dryness and dissolution of the salts in dilute hydrochloric acid. This was considered to be due to a change in the oxidation state of the molybdenum because a blue compound is produced under these conditions (though only when sulphuric acid is present). Previously, Hutchison, 9 using calibration solutions prepared from ammonium molybdate, obtained low results when aliquots of the standard solutions were evaporated to dryness with perchloric acid and the salts were dissolved in dilute perchloric acid; this was attributed to the formation of molybdic acid. Hutchison found that complete recovery of molybdenum could be obtained if the molybdic acid was subsequently converted into a m m o n i u m molybdate by dissolving the salts in dilute ammonia solution and evaporating the resultant solution to dryness. However, high results were still obtained in the present work when this procedure involving evaporation with perchloric acid and subsequent dissolution of the salts in dilute ammonia solution was applied to the ~-benzoinoxime extracts. Ultimately it was found that this positive error was due to an increase in the absorbance of molybdenum solutions after evaporation to dryness with sulphuric or perchloric acids and that it can be readily avoided by treating the molybdenum solutions taken for calibration purposes in the same way as the sample solutions. The molybdenum in the resultant calibration solutions will subsequently be present in the same form as in the final sample solutions.
Extraction complex
of the molybdenum(H) ~-benzoinoxime
Previous investigators 1°,11 showed that the molybdenum(VI) ~-benzoinoxime complex can be quantitat-
82
Et.sm M. DONALDSON
ively extracted into chloroform from up to approximately 2.3M hydrochloric acid, and that the extraction step is reasonably specific when chromium(VI) and vanadium(V) are reduced with ferrous ammonium sulphate and thus prevented from reacting with the reagent. ~3 Only niobium, zirconium, tungsten(VI) t2 (and possibly palladium) ~a are partly coextracted from _>IM hydrochloric acid, and the coextraction of niobium and zirconium can be readily prevented by complexing them with hydrofluoric acid. ~2 However, the possible interference of tungsten had to be considered in the present work because it is a common constituent of ores. Hutchison 9 reported that tungsten does not interfere in the determination of molybdenum by AAS after its separation by ~t-benzoinoxime extraction. However, tests showed that in an air-acetylene flame 100 #g/ml suppress the absorbance of 20 #g/ml of molybdenum by about 10%. Tungsten also interferes by precipitating in the final dilute acid solution. Tartaric acid cannot be used in this case to keep tungsten in solution because it strongly suppresses the molybdenum absorbance. Tungsten also causes difficulty before the ~t-benzoinoxime extraction step because of its insolubility in acid media. It has been reported that potassium dihydrogen phosphate complexes tungsten and inhibits the extraction of its a-benzoinoxime complex ~4 but this reagent was found to be completely ineffective. Further work showed that up to ~ 25 mg of tungsten can be kept in solution, in ~ 1.7M hydrochloric acid, with 2 g of tartaric acid and that, under these conditions, up to at least 2.5 nag of molybdenum can be quantitatively extracted as the ~t-benzoinoxime complex in four successive extractions with 15-ml portions of 0.2% solution of the reagent in chloroform. Approximately 50°,/0 of the tungsten initially present in the solution is co-extracted under these conditions. Consequently, a method is required for the subsequent separation of tungsten if more than a few rag are present in the sample taken for analysis.
Separation of molybdenumfrom tungsten by extraction of its xanthate complex Previous work by the author ~ showed that molybdenum [after its reduction to molybdenum(V) by xanthate] can be quantitatively extracted into chloroform as the xanthate from 0.1-~2M hydrochloric acid. In earlier work the extraction of molybdenum xanthate from 1.5M hydrochloric acid containing tartaric acid was used for the separation of molybdenum from tungsten before the spectrophotometric determination of tungsten after its extraction as the thiocyanatediantipyrylmethane ion-association complex. ~6 Consequently, it was considered that this separation procedure could also be used in the present work after back-extraction of the molybdenum and tungsten ~t-benzoinoxime complexes from the chloroform phase.
Preliminary experiments showed that the molybdenum complex cannot be completely back-extracted from the organic phase by shaking it with 4M ammonia solution or 10% sodium hydroxide solution or with 12M hydrochloric acid. However, shaking for ~4 rain with 15M ammonia solution was found to be effective. Under these conditions only negligible amounts of molybdenum ( <4 #g at the 2.5-rag level) remain in the organic phase. Subsequent work showed that, after appropriate treatment of the resultant ammoniacal solution, molybdenum can be readily separated from tungsten by extraction as the xanthate from 50 ml of 1.5M hydrochloric acid containing ~ 1 g of tartaric acid to keep tungsten in solution.
Effect of diverse ions As mentioned above, only niobium, zirconium, tungsten(VI), lz and possibly palladium 13 are partly co-extracted as a-benzoinoxime complexes from hydrochloric acid media containing ferrous ammonium sulphate as a reductant for chromium(VI) and vanadium(V). However, tests showed that zirconium is not extracted in the presence of tartaric acid, and niobium forms an insoluble hydrolysis compound during the sample decomposition step. This compound is removed by filtration before the extraction step. The effect of palladium was not considered because more than #g-quantities are not usually present in ores. The amount of tungsten that is co-extracted, at approximately the 2-rag level., will not interfere in the subsequent determination of molybdenum in an airacetylene flame when aluminium chloride solution ~ is added to the final solution to obviate the interference of tungsten, and if the volume of the solution is > 25 ml. If the final volume is less, tungsten may precipitate. Because the tungsten compound that is formed on evaporation of the ct-benzoinoxime extract to dryness with acids is insoluble in water or dilute acids, dilute ammonia solution is required to dissolve this compound and convert it into ammonium tungstate before the addition of hydrochloric acid and aluminium solution. The excess of ammonia is readily removed by heating the solution.
Applications To test the reliability of the proposed method, it was applied to the analysis of three CCRMP ores that have been certified for molybdenum and to three CCRMP tungsten ores for which only approximate molybdenum values are available. 17 It was also applied to certified reference iron and steel samples. The results of these analyses are given in Table 1. DISCUSSION
Table I shows that the results obtained for the CCRMP reference ores MP-1, HV-1 and PR-I and, where applicable, for the National Bureau of Standards (NBS) and British Chemical Standards (BCS)
Molybdenum in ores, iron and steel
~_,~ o~
~e ~,.~
-
~
o
o
IIIII
I
~
83
~
~
~
~
o. 0
e.
•~.~
a
0
0
< Z
0
,g
,,.-k e-, o ..~
8:
i
i2
~
~
~--~~~
~~
~
~ - ~
-
~
~
~
~
~ es --
0
9 ~o.~. 0
r
y
0
~
o
~
=
.~ o
o
=
o-
._~o-~ ~ ~.~
~'~-~-~ o~
~.__.
Z
Z
u ~ z
~='"
~-~..~
=
.~
~=z~
"7 ,.~
[..
ZZZZZ
Z
Z
Z
ZZ
Z
~
,~
84
ELSm M. DONALDSON
iron and steel samples, after separation of molybdenum by ct-benzoinoxime extraction alone, are in excellent agreement with the certified values. Similarly, the results obtained for the ores, after the addition of tungsten, and for the NBS and BCS samples containing tungsten, after further separation of molybdenum from co-extracted tungsten by xanthate extraction, are also in good agreement with the certified values and, where applicable, with the values obtained after ~t-benzoinoxime extraction alone. The results obtained for the three CCRMP tungsten ores, CT-1, BH-1 and TLG-1 are considered to be more reliable than those given for information purposes. 17 The latter are only approximate values obtained in the CANMET chemical laboratory by a spectrophotometric thiocyanate method after the separation of iron by precipitation with sodium hydroxide. Molybdenum can be separated from tungsten and certain other elements by direct xanthate extraction, ~s but this was not considered in the present work because the extraction step is not sufficiently selective. 15 The co-extraction of iron(Ill)can be avoided by reducing it with ascorbic acid, and that of copper(II) and vanadium(V) by complexing them with thiourea and potassium hydrogen fluoride, respectively, la Tin(II and IV), antimony(Ill), arsenic(IlI) and selenium(IV), which are also co-extracted, :5 can be removed by volatilization as the bromides during the sample decomposition step. However, the co-extraction of certain elements that commonly occur in ores, such as cobalt, nickel, silver, bismuth and, particularly, lead cannot be prevented. ~ Possibly more tungsten can be tolerated during the ct-benzoinoxime extraction step if the initial tartaric acid concentration is increased. However, because tartaric acid slightly inhibits the extraction of the molybdenum ~t-benzoinoxime complex, a more concentrated solution of the reagent in chloroform would be required for the complete extraction of molybdenum. Under these conditions, more tungsten would also be co-extracted. Subsequently, more than 1 g of tartaric acid may be required to keep tungsten in solution before and during the xanthate extraction step. The addition of more is not recommended because too much tartaric acid inhibits the extraction of molybdenum as the xanthate. Up to six extraction stages are required for the complete extraction of 2.5 mg of molybdenum from 50 ml of 1.5-2M hydrochloric acid
containing 2 g of tartaric acid. The inhibiting effect of tartaric acid is greater at lower hydrochloric acid concentrations. A hydrochloric acid medium was chosen for the ultimate determination of molybdenum because this acid has very little effect on the absorbance of molybdenum in either an air-acetylene or a nitrous oxideacetylene flame.4's'19 A 15% hydrochloric acid medium was chosen for convenience when working with 10-ml final sample solution volumes, Probably the addition of I ml of 50% hydrochloric acid, rather than the concentrated acid, would also be sufficient. However, the concentration of hydrochloric acid in the calibration solutions should be adjusted accordingly. An air-acetylene flame was chosen because it was found to be sufficiently sensitive and because interference effects from other elements, anions and acids are considered to be less in this flame than in the nitrous oxide-acetylene flame. 19'2°
REFERENCES
1. D. J. David, Analyst, 1961, 86, 730. 2. ldera, ibid., 1968, 93, 79. 3. R. A. Mostyn and A. F. Cunningham, Anal. Chem., 1966, 38, 121. 4. T. V. Ramah'ishna, P. W. West and J. W. Robinson, Anal. Chirt Acta, 1969, 44, 437. 5. A. Purushottam, P. P. Naidu and S. S. Lal, Talanta, 1972, 19, 1193. 6. C. H. Kim, C. M. Owens and L. E. Smythe, ibid., 1974, 21, 445 (and references therein). 7. C. H. Kim, P. W. Alexander and L. E. Smythe, ibid., 1975, 22, 739. 8. ldem, ibid., 1976, 23, 229 (and references therein). 9. D. Hutchison, Analyst, 1972, 97, 118. 10. P. G. Jeffery, ibid., 1956, 81, 104. 11. G. Goldstein, D. L. Manning and O. Menis, Anal. Chert, 1958, 30, 539. 12. L. Wish, ibid., 1962, 34, 625. 13. P. Y. Peng and E. B. Sandell, Anal. Chirt Acta, 1963, 29, 325. 14. H. J. Hoenes and K. G. Stone, Talanta, 1960, 4, 250. 15. E. M. Donaldson, ibid., 1976, 23, 411. 16. lderrg ibid., 1975, 22, 837. 17. G. H. Faye, W. S. Bowman and R. Sutarno, CANMET Rept. 76-5, Department of Energy, Mines and Resources, Ottawa, 1976. 18. V. Yatirajam and J. Ram, Talanta, 1974, 21,439. 19. S. Dilli, K. M. Gawne and G. W. Ocago, Anal. Chirt Acta, 1974, 69, 287. 20. J. D. Kerbyson and C. Ratzkowski, Can. Spectrosc., 1970, 15, 43.