310 Analyst March 1983 Vol. 108 $9. 310-315 Determination of Trace Amounts of Molybdenum in Natural Waters by Solvent Extraction -Atomic-a bsorption Spectrometry After C helati ng I on -exc h a ng e Pre-co n cent rat i o n Miguel Ternero and lgnacio Gracia Department of Basic and Applied Chemistry E . T . S . I . I . University of Seville Avda. Reina Ilfercedes sln, Seville-12 Spain A method for the determination of molybdenum in natural waters in the micrograms per litre range is proposed. The method involves the pre-concentration of molybdenum on a Chelex- 100 chelating resin and subsequent elution with ammonia solution followed by extraction with a complexing reagent 1,4-dihydroxyphthalimide dithiosemicarbazone dissolved in NN-dimethylformamide - isoamyl alcohol (1 + 4) and a final direct determination by atomic-absorption spectrometry.The addition of ascorbic acid prior to extraction eliminates the interfering effect of several ions a t the concentra-tion levels normally found in natural waters. The sensitivity of the method is 0.3 pg 1-1 for 1% absorption. The method has been applied to the determin-ation of molybdenum in sea and surface waters. Keywords Molybdenum determination ; natural water analysis ; chelating ion-exchange separation ; atomic-absorption spectrometry ; 1,4-dihydroxy-phthalimide dithiosemicarbazone The determination of certain trace elements has become significant in studies of geochemical, biochemical and industrial processes in natural waters. An investigation of this problem is being carried out in our laboratories and this paper describes the development of a method for the determination of molybdenum in the micrograms per litre range.In order to detect the low levels of molybdenum that occur naturally1 (from less than 1 to 10 pg 1-1 and even larger amounts in sea water) pre-concentration methods are usually necessary. For this purpose c~precipitation,~-~ cocrystallisation,6~7 extraction,s-ll ion exchange12-15 and activated charcoaP have been used the determination being completed by either spectrophotometry or atomic-absorption spectrometry (AAS) or sometimes by other procedures (emission spectroscopy neutron activation analysis). In recent years many papers have described the direct determination of molybdenum by the use of modern tech-niques for example graphite furnace AAS,17 anodic-stripping voltamrnetry,l8 electron para-magnetic resonance spectrometrylg and oscillography.20 However such techniques are not available to all laboratories and may also introduce other problems in achieving the desired accuracy of analysis e.g.the prevalence of matrix interference effects in electrothermal atomisation AAS. This paper reports a sensitive and selective method based on pre-concentration of molyb-denum on a Chelex-100 chelating resin,21-22 followed by extraction with 1,4-dihydroxyphthali-mide dithiosemicarbazone (OH-PDT) in isoamyl alcohol and a final direct determination by AAS. The use of OH-PDT as a complexing reagent for molybdenum has been reported previously2s and applied to the determination of microamounts of molybdenum by an extrac-tion - spectrophotometric method.The proposed method in this paper with an AAS finish, gives a greater sensitivity and eliminates most of the interfering cations. In recent years several systems based on solvent extraction of molybdenum complexes with subsequent AAS determination in the organic phase have been used for enhancement of the sensitivity of molybdenum absorption and for concentrating the samples and eliminating interfering cations. Of these quinolin-8-01- isobutyl methyl ketone (IBMK) ,8 quinolin-8-01 -n-amyl methyl ketone,24 ammonium tetramethylene dithiocarbamate - IBMK,25 dithiol-IBMK26 and thiocyanate - IBMKll are generally used for the determination of molybdenum in waters usually with a preliminary concentration step.However limitations with regard to sensitivity simplicity and freedom from interferences (especially iron) are observed. The proposed method with pre-concentration and separation on Chelex-100 resin is relatively simple and sufficiently versatile for the analysis of sea and surface waters TERNERO AND GRACIA Experimental Reagents 311 All reagents and solvents were of analytical-reagent grade unless specified otherwise. Distilled de-ionised water was used. Chelating resin. Digest Chelex-100 (Rio-Rad Laboratories 100-200 mesh) with 2 N nitric acid and fill 2 cm diameter ion-exchange columns to a depth of 3 cm with it. Wash the columns with 20 ml of 2 N nitric acid and then with water until the pH of the eluate is 5-6. Prepare a 0.05% m/ V solution by dissolving 0.05 g of the reagent in 20 ml of NN-dimethylformamide and dilute to 100 ml with isoamyl alcohol.The synthesis of OH-PDT has been described previ~usly.~' Dissolve 1.500 g of molybdenum(V1) oxide in the minimum volume of 0.1 M sodium hydroxide solution dilute with water make slightly acidic (pH 3-4) with 0.1 M hydrochloric acid and dilute to 1 1 with water. Prepare a working solution containing 10 pg ml-l of molybdenum from this stock solution by appropriate dilu-t ion. Chloroacetic acid - sodium hydroxide bufler solution PH 2.6. Add 65 ml of 0.2 M sodium hydroxide solution to 300 ml of 0.2 M chloroacetic acid solution and dilute to 1 1 with water. Ascorbic acid solution 1% m/V. Dissolve 1 g of ascorbic acid in water and dilute to 100 ml with additional water.Nitric acid 0.1-4 N. Prepare from analytical-reagent grade concentrated nitric acid by appropriate dilution. 1,4-Dihydroxyphthalimide dithiosemicarbazone solution. This solution is stable for 1 week. Standard molybdenum solution 1.000 g 1-I. Apparatus A Perkin-Elmer 103 atomic-absorption spectrophotometer equipped with a dinitrogen oxide - acetylene burner head and a multi-element (cobalt copper iron, manganese molybdenum) hollow-cathode lamp was used. The operating conditions used are summarised in Table I. Digital p H meter. An Orion 701A instrument with glass - calomel electrodes was used for pH measurements. Spectrophotometev. TABLE I ATOMIC-ABSORPTION CONDITIONS FOR MOLYBDENUM DETERMINATION Wavelength . . Slit width . . . . Lamp current .. . . Acetylene pressure . . Acetylene flow-rate . . Dinitrogen oxide pressure Dinitrogen oxide flow-rate Aspiration rate . . . . Burner height . . . . 313.3 nm . . . . 0.7 nm . . . . 10mA . . . . 10 p.s.i.g. . . . . Setting 12 (approximately 8 1 min-l) . . . . 40 p.s.i.g. . . Setting 13 (approximately 16 1 min-') . . . . 3 ml min-' . . . . Adjust for optimum reading Procedure Filter the sample as soon as possible after collection through a 0.5-pm membrane filter. Take 1-5 1 of the filtered water and adjust its pH to 5-5.5 by cautious addition of 0.1 N nitric acid. Allow the sample to flow through the Chelex-100 column at a rate not exceeding 5 ml min-l. Wash the resin with 200 ml of water and elute molybdenum with 20 ml of 4 N ammonia solution.Collect the eluate in a 25-ml beaker and adjust its pH approximately to neutrality with appropriate dilute nitric acid (0.1-4 N). Transfer into a separating funnel add 5 nil of pH 2.6 buffer solution and 5 ml of ascorbic solution. After mixing add 10 ml of OH-PDT solution and shake vigorously for 1 min. Allow the phases to separate and draw off the aqueous layer. Transfer the organic phase into a glass-stoppered tube containing anhy-drous sodium sulphate and aspirate directly into the dinitrogen oxide - acetylene flame. Determine the absorbance at 313.3 nm using isoamyl alcohol as a blank under the speci-fied conditions. Prepare a calibration graph by using standard solutions of molybdenum(V1) treated in the same way 312 TERNERO AND GRACIA DETERMINATION OF Mo IN WATERS BY Analyst Vol.108 Results and Discussion Study of the Pre-concentration of Molybdenum on Chelex-100 Chelating Resin This effect was studied using a series of dilute standard solutions of molybdenum in the pH range 1-9. Quantitative retentions were observed in the pH range 5-6. A pH between 5 and 5.5 was selected for the analysis. These results are in agreement with those reported for retention of molybdenum from sea waters.14 The nature and concentration of the eluting agent was studied under the conditions described under Experimental using a 2 x 3 cm column of Chelex-100. Ammonia solution sodium hydroxide solution and nitric sulphuric and perchloric acids were investigated. Molybdenum was removed from resin only by ammonia solution and sodium hydroxide solution.Fig. 1 shows the elution diagrams at several concentrations. These diagrams were constructed by elution with successive 5-ml volumes of eluting agent. The recovery of molybdenum was assessed by the recommended procedure. A 20-ml volume of 4 N ammonia solution is recom-mended as a suitable amount of eluting agent. In order to establish the possibility of determining trace amounts of molybdenum in waters, the recovery from large volumes of sample was studied by the recommended procedure. Quantitative recoveries were obtained for sample volumes of 50 ml-5 1. The retention of molybdenum by Chelex-100 resin is dependent on pH. Eluate volurne/rnl Fig. 1. Elution diagrams of molybdenum from Chelex-100 resin with ammonia solution and sodium hydroxide solution A B C, D and E with 1 2 3 4 and 5 x ammonia solution respectively; F with 2 s sodium hydroxide solution ; amount of molybdenum added 50 pg; successive volumes of eluate collected 5 ml.Extraction and Atomic-absorption Spectrometry of Molybdenum When a solution of OH-PDT dissolved in NN-dimethylformamide - isoamyl alcohol is shaken with an aqueous acidic solution of molybdenum(VI) a yellow complex is formed immediately in the organic phase. This system has been used for the spectrophotometric determination of m0lybdenum.~3 However it is unsuitable for water analysis because of interferences from certain ions [iron(II) iron(ITI) copper(II) cobalt(I1) and vanadium(T’)] at the levels commonly found in natural waters. The extraction of molybdenum with subsequent AAS determination is dependent on pH.The most favourable pH range is 2-4 identical with that reported for the spectrophotometric determination. A chloroacetic acid - sodium hydroxide buffer solution is added for control of the pH of the extraction. The effect of OH-PDT concentration in A‘N-dimethylformamide - isoamyl alcohol (1 + 4) was investigated in the range 0.002-0.170 rn/V. A 0.006~0 solution was necessary in order to obtain maximum absorbance at 313.3 nm; the latter remained constant with increasing con-centration. The volume of the organic phase was kept constant at 10 nil varying the volume of the aqueous phase. When the phase-volume ratio was higher than 4 the absorbance increased because of A 0.05% solution is recommended as a suitable concentration of reagent.The influence of the phase-volume ratio (aqueous to organic phase) was studied March 1983 313 the appreciable solubility of the organic solvent in water. The absorbance remained constant when smaller ratios were employed. It is concluded that the volume of the aqueous phase should be smaller than 40 ml if a 10-ml volume of organic phase is utilised. The ionic strength of the aqueous phase does not affect the atomic absorption of the extracted complex. Salts such as sodium sulphate potassium perchlorate potassium chloride and potassium nitrate do not affect the absorbance signal even at a concentration of 2%. In order to establish the suitability of this system for the determination of molybdenum the results of a preliminary interference study with an AAS or spectrophotometric finish are reported in Table 11.I t is concluded that the selectivity with AAS is greater than that with spectrophotometry . SOLVENT EXTRACTION - AAS AND CHELATING ION EXCHANGE TABLE I1 TOLERANCE LIMITS FOR THE DETERMINATION OF MOLYBDENUM WITH THE OH-PDT - ISOAMYL ALCOHOL SYSTEM BY AAS AND BY SPECTROPHOTOMETRY Amount of molybdenum present 50 pg. I Ions added AI(lII) Ni(II) Mn(I1) . . Zn(II) Cd(II) Ti(1V) . . . . La(II1) . . . . . . . . . . Bi(III) Cr(II1). . . . . . W(V1) . . . . . . Pb(I1) . . . . . . Co(I1) . . . . . . . . Cu(II) Fe(II1) . . Fe(I1) . . . . . . Hg(II) ViV) . . . . Tolerance limit (mass excess relative to Mo) P Spectrophotometry AAS 15 > 100 15 50 10 15 1 50 2 15 <1 2 <1 2 <1 5 Determination of Molybdenum in Waters Calibration graph sensitivity and precision A calibration graph was prepared by using dilute standard solutions treated in the same way as the samples.A linear Calibration graph was obtained up to 12 mg 1-1 of molybdenum with respect to the organic phase and 24 pg 1-1 with respect to water samples when a 5-1 volume was utilised. The sensitivity was 0.15 nig 1-1 for 194 absorption in the organic phase and 0.3 pg 1-1 with respect to watcr samples wlien a 5-1 volume was utilised. The sensitivity obtained for the determination of molybdenum by AAS in tlie organic phase was about 3.3 times greater than that for aqueous solutions (0.50 mg 1-l). The coefficient of variation calculated from ten replicate analyses of dilute standard solutions containing 25 pg of molybdenum was 3.0%).Interference study The recommended procedure was used to analyse standard molybdenum solutions in the presence of the major constituents of natural waters and of several trace elements that interfere in AAS (tolerance limits smaller than 102 mass excess relative to molybdenum) (see Extraction and Atomic-absorption Spcctrometry of Alolybdenum and Table 11). Determinations in the presence of tlie major constituents were carried out at the levels normally present in sea water.‘ The results for tlie determination of 50 pg of molybdenum are shown in Table 111. From these results it is conclutled that the presence of the main constituents and of most of the trace elements did not affect tlie recovery of molybdenum at levels that occur naturally.The tolerance limits for hismuth(III) cobalt (II) copper( II) iron(II) mercury(II), chromium(II1) and leatl(I1) are greater than those obtained previously (Table 11) because of their smaller retention on the resin. Tungstate(V1) does not interfere in amounts up to a 20-fold excess. 1’anatlium( 1’) anti iron (I I I) depress the molybdenum absorption markedly probably be-cause of preferential extraction of their OH-PDT complexes or flame interferences.2s Th 314 TERNERO AND GRACIA DETERMINATION OF MO IN WATERS BY Analyst vd. 108 addition of ascorbic acid was found to prevent these interferences. When 5 ml of 1% ascorbic acid solution are added to the sample before extraction up to 5 mg of iron(II1) and 500 pg of vanadium(V) could be tolerated.The above limits are not likely to be exceeded in analyses of natural waters. TABLE I11 RECOVERY OF MOLYBDENUM IN THE PRESENCE OF THE MAJOR CONSTITUENTS OF WATER AND SEVERAL TRACE ELEMENTS THAT INTERFERE IN AAS Amount of molybdenum present 50 pg. Ions added Na(1) . . K(1) . . . . Ca(I1) . . . . . . . . s o p . . co,2- . . . . . . Cr(III) Pb(I1) . . . . . . W(V1) . . Fe(II1) . . . . . . V(V) . . . . . . . . . . ;p) - * Br- . . . . Bi(III) Co(II) Cu(II) Fe(II) Hg(I1) . . . . . . . . . . . . . . . . . . * . * . . . Mass excess relative Mo 2.1 x 105 7.6 x 103 8.0 x 103 2.7 x 104 3.8 x 105 1.7 x 104 1.3 x 103 5.6 x lo2 102 102 75 102 20 102 102* 102 10* Amount of Mo recovered/ p g 50.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0 33.5 50.0 20.5 50.0 0.0 50.0 10.5 50.0 * With addition of 5 ml of 1% ascorbic acid solution before extraction.Analyses of natural waters The proposed method was applied successfully to the determination of trace amounts of molybdenum in sea and surface water samples (Table IV). The accuracy of the determinations was checked by carrying out replicate analyses of samples spiked with known amounts of molybdenum. The molybdenum recovery was calculated by comparing the results obtained before and after the addition of molybdenum standard solutions. The results showed that the recovery of molybdenum was satisfactory (Table IV).TABLE IV DETERMINATION OF MOLYBDENUM IN NATURAL WATERS Type Bottled mineral water . . . . Public water supply River water . . . . Dock's river water . . . . Sea water A . . . . Sea water B . . . I Estuary water? . . . . Location of sampling Granada Seville Guadalquivir river Seville Guadalquivir's dock Seville Atlantic Ocean, Huelva Atlantic Ocean, Huelva Huelva hIo found*/pg 1-' 7-Memi Range 0.3 0.2-0.3 0.9 0.8-1.0 2.7 2.7-2.8 6.3 6.2-6.4 7.5 7.3-7.6 7 . 2 7.1-7.4 9.9 9.8-10.0 Recovery (;,, 100.0 99.5 99.5 98.9 99.0 99.2 98.5 * Average of three separate determinations. t This water is affected directly by drainage of waste water from an industrial area, March 1983 SOLVENT EXTRACTION - AAS AND CHELATING ION EXCHANGE Conclusion 315 The suitability of the proposed method for determining small amounts of molybdenum in waters in the range 0.3-24 pg 1-1 has been demonstrated.The use of a chelating resin in a pre-concentration and separation step separates molybdenum from the major components and eliminates the matrix effect on the atomic-absorption signal. The combination of solvent extraction with AAS provides greater sensitivity and eliminates most interfering cations. Finally the method offers significant advantages with respect to relative rapidity simplicity and versatility. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1 1 . 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.27. 28. References Bond I<. G. and Straub C. P. “Handbook of Environmental Control Volume 111 Water Supply and Chan K. M. and Riley K. P. Anal. Chim. Acta 1966 36 220. Kim Y. S. and Zeitlin H. Anal. Chitn. Ada 1970 51 516. Kuznetsov V. I. and Loginova L. G. Z h . Anal. Khim. 1958 13 453. Korob I<. O. Cohen I. M. and Agatiello 0. E. J . Radioanal. Chem. 1976 34 329. Weiss H. V. and Lai M. G. Talanta 1961 8 72. Kulathilake A. I. and Chatt A. Anal. Chem. 1980 52 28. Chau Y. and Lum-Shue-Chan K. Anal. Chim. Acta 1969 48 205. Morgen E. A. Iiossins Kaya E. S. and Vlasov N. A. Zh. Anal. Khim. 1975 30 1384. McLeod C. W. Otsuki A Okanioto K. Haraguchi H. and Fuwa K. Analyst 1981 106 419. Kim C. H. Alexander P. W. and Smythe L. E. Talanta 1976 23 229. Kawabuchi K.and Kuroda R. Anal. Chim. Acta 1969 46 223. Kawabuchi K. and Kuroda I<. Anal. Chim. Acta 1968 40 479. Riley J. P. and Taylor D. Anal. Chim. Acta 1968 41 175. Korkish J . and Krivanec H. Anal. Chim. Acta 1976 85 111. Van der Sloot H. A. Wals G. D. and Das H. A. Anal. Chim. Acta 1977 90 193. Nakahara T. and Chakrabarti C. L. Anal. Chim. Acta 1979 104 99. Monien H. Bovenkork I<. Kringe K. P. and Rath D. Fresenius Z. Anal. Chem. 1980 300 363. Hanson G. Szabo A, and Chasteen N. D. Anal. Chem. 1977 49 461. Demkin A. M. Zh. Anal. Ichim. 1977 32 2389. Rosset I<. I3ull. Soc. Chim. Fr. 1964 2 1845. Kosset I<. Bull. lnf. Sci. Ttchnol. Comm. Energ. Atom. 1964 85 13. Ternero Rodriguez M. Analyst 1982 107 41. Butler L. I<. P. and Mathews P. M. Anal. Chim. Acta 1966 36 310. Mansell R. E . and Emniel H. W. At. Absorpt. Nemsl. 1965 4 365. Delaughter B. At. Absorpt. Newsl. 1965 4 273. Perez-Bendito D. Valcdrcel M. Ternero M. and Pino F. Anal. Chim. Acta 1977 94 405. Daud D. J. Anal. Chem. 1961 86 f30. Treatment,” Second Edition CRC Press Cleveland 1980 pp. 15-127. Received July 7th 1982 Accepted September 9th 198