- Email: [email protected]
Extraction—spectrophotometric determination of vanadium with 3,5-dinitrocatechol and Brilliant Green
Tafania, Vol. 35, No. 12, pp. KfWtOO4, 1988 Printed in &cat Britain. All rights n~etwd
0039-9140/88 s3.00 + 0.00
Copyright (B 1988 Pergamon Press plc
EXT~CTION-SPECTROPHOTOMETRIC DETERMINATION OF VANADIUM WITH 3,5-DINITROCATECHOL AND BRILLIANT GREEN Z. MARCZENKO
and R.
ICOBI~SKI
Department of Analytical Chemistry, Technical University, Warsaw, Poland (Received 22 January 1988. Accepted 10 Juiy 1988) Summary-The formation and extraction of the ion-associates of the vanadiumw3,5-dinitrocatechol (DNC) anionic chelate complex with various basic dyes have been studied and a new sensitive
extraetion-speetrophotometrie method for the deter&ration of vanadium based on the system V~~DNC-B~lliant Green has been develored. Beer’s law is obeyed UD to a vanadium ~n~ntration of013 &ml and the molar absorptivity is I:7 x 10s l.m~le-‘.cm~~ at 630 nm. The molar ratios of the components and the form of the vanadium(V) cation in the extracted compound have been determined, and the formula [VO(OH)(DNC)i-][BG+], is proposed. Titanium, molybdenum, tungsten, EDTA and thiocyanate interfere seriously. The method becomes specific after a preliminary separation of vanadium by its extraction as the BPHA complex from H,SO,-HF medium, and is 40 times more sensitive than the spectrophotometric BPHA method. The proposed method has been applied to determination of traces of vanadium (about IO-‘%) in alums.
Many sensitive extraction-spectrophotometric methods for the determination of metals and non-metals are based on extraction of the ion-associates of basic dyes with anionic (mostly halide) complexes. Anionic complexes with chelating reagents (e.g. oxine, benzoic acid, 3,bdinitrocatechol) are seldom applied although they allow a sensitive determination of the metals which do not form sufficiently stable anionic complexes with halides (e.g., Ti, Th, U, V, W, Zr, lanthanides)! There is a lack of very sensitive s~ctrophotome~~ methods for the determination of vanadium and therefore the present work is devoted to the development of such a method based on the extraction of the ion-associate of a basic dye with the doubly charged anionic complex of vanadium(V) with 3,5~nitr~at~hol. Hitherto, the 3,5dinitrocatechol complex/basic dye system has been restricted to the extractive spectrophotometric determination of germanium, tin, titanium and lanthanidess9
N-Benzoyl-N-phenylhydroxyiamine (BPHA) solution, O.l%, in chloroform. Doubly distilled water was used throughout. All other chemicals used were of analytical-reagent grade. Apparatus A Speeord UV-VIS recording spectrophotometer and a VSUZ-P spectrophotometer with IO-mm glass cells were used. Polypropylene separatory funnels were used for the extraction from fluoride medium. Procedure Place the sample solution (PH 5-7), containing not more than 3 pg of vanadium(V), in a separatory funnel, and add 1 ml of 3,S-dinitrocatechol solution, followed by 1 ml of 0.2544 sulphuric acid. Make up with water to about 10 ml, add 5 ml of carbon tetrachloride, then 0.5 ml of Brilliant Green solution and shake for 1 min. After separation of the phases, transfer the organic layer to a IO-ml standard flask containing 1 ml of ethanol and one drop of 0.01 M sulphuric acid. Make up to the mark with carbon tetrachloride, mix, and measure the absorbance at 630 nm against a reagent blank prepared in the same way. RESULTS AND DISCUSSION
Preliminary Reagents Vanadium(V) standard solution (I mg/ml). Dissolve 1.786 g of V,O,, previously ignited at Soo”, in dilute sodium hydroxide solution. Acidify the solution with sulphuric acid and dilute to volume with water in a I-litre- standard flask. Dilute further with water as required. 3,5-~~nitrocatechol @NC) solution, 8 x IO-‘M, in 20% ethanol medium. DNC was synthesized6 and its purity confirmed by melting point measurement, elemental analysis, thin-layer chromatography, and infrared and NMR spectroscopy. Brilliant Green (BG) (Reakhim, USSR) solution, 8 x IO-‘M, in ethanol. The commercial preparation was purified according to Fogs et al.” TAL. 3%12--F
investigations
In an acidic (pH l-3) medium vanadium~) reacts with 3,5dinitrocatechol (DNC) to form an anionic yellow complex (I,, = 428 nm). In the presence of a suitable basic dye an ion-associate is formed. The following basic dyes were examined: Brilliant Green, Malachite Green, Methyl Violet, Crystal Violet (t~pheny~methane dyes), Methylene Blue, Nile Blue A (azine dyes), Rhodamine B and Rhodamine 6G (xanthene dyes). The possibility of the extraction of the vanadium-DNC-basic dye ion-associate from a sulphuric acid medium (pH O-4) into various organic solvents [benzene, toluene, carbon tetra-
2. MARCZENKO
loo2
chloride, chloroform, n-pentyl acetate, cyclohexane, methyl isobutyl ketone (MIBK), di-isopropyl ether and petroleum ether] was studied. Ion-associates are formed with all the basic dyes examined (apart from Methylene Blue) and are either extracted into the organic phase (benzene, toluene, chloroform, MIBK, n-pentyl acetate) or floated (cyclohexane, di-isopropyl ether and petroleum ether). Carbon tetrachloride extracts the Brilliant Green ion-associate but floats the ion-associates of all the other dyes tested. From the analytical point of view the extraction solvent should completely extract the vanadium ion-associate but not the dye itself or its ion-pair with DNC. These conditions were fulfilled only by the Malachite Green-toluene and Brilliant Green-carbon tetrachloride systems. As the latter gave the higher sensitivity, it seemed the more promising and was further investigated in detail. F5r~atio~ and extraction ion -associate
of
the V(V)-DNC-BG
From Fig. 1 it is evident that maximum absorbance is obtained when the sulphuric acid concentration is about 0.025M (pH 1.3 + 0.1). From more acidic media, the vanadium(V) ion-associate is not extracted quantitatively, and with progressively less acidic ones the blank absorbance increases considerably owing to the greater dissociation of DNC, the resulting increasing concentration of the DNC anion, and consequently greater probability of DNC-BG ionpair formation. Because the pH of the aqueous phase is practically constant during the extraction, no buffer solution is needed. The required acidity is established by adding a suitable amount of sulphuric acid to a sample solution of pH 5-7. This way of establishing the acidity gives more reproducible results and is faster than adjustment with pH-meter control. Figure 2 shows the dependence of the absorbance of the extract and of the blank on the DNC concentration. A DNC concentration of at least 8 x IO-‘M (16-fold molar excess with respect to 2.5 pg of vanadium) gives maximal and reproducible absorbance values.
and
R.
LOBIX?SKI
2
4 DNC]
6 lo-’
610 M
Fig. 2. Dependence of the absorbance of the extract on the DNC concentration: curve 1, absorbance of the vanadium ion-associate measured against the blank; 2, blank absorbance measured against CCl,-ethanol (9 + 1) mixture. Figure 3 shows that maximal absorbance readings are obtained with a Brilliant Green concentration of at least 4 x 10-5M (about g-fold molar excess with respect to 2.5 pg of vanadium). Because of the dimerization of Brilliant Green in more concentrated (about I x 10F3A4) aqueous solutions, and then dissociation into monomer after dilution,” an ethanolic solution of the dye was used. It was observed that maximal and well reproducible absorbances were obtained when the Brilliant Green solution was introduced after addition of the extracting solvent. The extraction should then be done as soon as possible because of the protonation of Brilliant Green in acidic (pH < 2) aqueous solutions.Lo The vanadium is completely extracted by shaking the aqueous phase and one S-ml portion of carbon tetrachloride for 1 min. The colour of the extract fades after separation of the layers and therefore the extract itself is useless for analytical purposes, but it has been found that the presence of a small amount of acid prevents the fading of the colour. The addition of even one drop of O.OlM sulphuric acid makes the extract turbid, but subsequent addition of ethanol or methanol clears the turbidity. This addition also causes a significant increase in the absorbance of the extract and a change 1 Cl8
/ 0.6 A 0.4 0.2
0.2-
2
2 I I (101 a02
I 003 Hz=4
Y a04
~ 0.05 Cl06
hf
Fig. 1. Effect of the sulphuric acid concentration on formation and extraction of the vanadium ion-associate: curve I, absorbance of the extract measured against the blank; 2, blank absorbance measured against CCl,-ethanol (9 + I) mixture.
1
2
3
4
5
UK+1 10-S M
Fig. 3. Dependence of the absorbance of the extract on the Brilliant Green concentration: curve I, absorbance of the vanadium ion-associate measured against the blank; 2, blank absorbance measured against CC&-ethanol (9 + I) mixture.
Determination of vanadium in colour from green to blue. A final concentration of 10% ethanol or methanol is optimal. Although the molar absorptivity is higher ( 1.94 x 1051.mole- ’ .cm-‘) for the methanol than the ethanol (1.70 x 105) system, the latter is preferred because the blank value is lower (0.08 instead of 0.12 with methanol). Composition of the V( V)-DNC-BG
ion-associate
The molar ratios of vanadium to DNC and to Brilliant Green were determined by the Bent and French logarithmic methodi and both found to be 1: 2. These results were confirmed by the Job method and are the same for both the methanol and ethanol systems. An extract containing a known amount of vanadium was evaporated and the residue was dissolved in a I : 4 v/v mixture of water and methanol. The absorbance of the solution was almost equal to the absorbance of a solution (in the same solvent) of an amount of Brilliant Green that was in 2: 1 molar ratio to the amount of vanadium in the extract. Attempts were made to determine the form in which vanadium(V) is present in the ion-associate. The molar ratios DNC: V and BG: V found for the ion-associate permit postulation of the following equation for its formation and extraction (DNC is denoted by H,R): VO(OH);-‘+
2H,R+
2(BG)+
+~O(OH):-i(H~_.R)2][(BG):]+2nH+ The equilibrium 1y
_
constant is
[vO(OH)j-'(H,-,R),(BG):l W+l'" pO(OH);-‘1 [H, R]‘[(BG)+]’
Hence -1ogB
= -pK
+ 2npH
where
B = IYWW - ‘1W, R12KBG>+12 ~ofoHt2-‘(H,-.R),tB~):l The relationship -log B = f (pH) was investigated for all possible vanadium cation forms (i = 0, 1,2, 3). All the constants necessary for the calculations are known.s,‘3 The values of -log B plotted vs. pH give a straight line with a slope that is an even number (n should be an integer) only for i = 2, which corresponds to the cation VO(OH)2+. We conclude from this that the ion-associate has the formula
1003
Table 1. Statistical evaluation of the results for determination of vanadium by the proposed method Vanadium, 1118
Standard
Added
Found
deviation,* !Jg
1.00 2.00 3.00
1.01 1.99 2.98
0.032 0.030 0.036
Confidence limits ~robabiiity level 0.95), pg
1.Ol f 0.03 1.98 + 0.03 2.98 + 0.04
*For seven determinations. It must he emphasized that, in contrast to the extraction-spectrophotometric methods based on the ion-associates formed by singly charged anionic metal-halide complexes with basic dyes, in the present method a doubly charged anion is extracted in association with two singly charged dyestufT cations, which increases the sensitivity.
Determination of vanadium The calibration graph obeys Beer’s law over the range 0.005-0.3 pg/ml V. The molar absorptivity is 1.70 x 10’ l.mole-l.cm-i at 630 nm (specific abso~ti~ty a = c/at.wt x 1000 = 3.3). Thus, the proposed method is 3040 times more sensitive than the well-known BPHA and 8-hydroxyquinoline methods and about 5 times more sensitive than the PAR method.3 The statistical evaluation of determination of trace amounts of vanadium at three concentration levels shows the good precision and accuracy of the proposed method (Table 1). The effects of many cations and anions on the determination were examined. Titanium, molybdenum and tungsten interfere at any level. Gallium and indium can be tolerated when present in 1: 1 w/w ratio to vanadium, and iron(III) in l~fold ratio. Aluminium and bismuth give 10% positive error at lOOO-fold w/w ratio to vanadium. Cr(III), Mn, Co, Ni, Cu, Zn, Cd, Sb(V), and alkali and alkaline-earth metal ions do not interfere at all. Anions can interfere either by forming extractable ion-pairs with the dye cation, thus influencing the blank, or by complexing the vanadium cation. Chloride, bromide, nitrate and acetate do not interfere at IOOO-fold molar ratio to vanadium; IO-fold molar ratio of fluoride, iodide, perchlorate, oxalate, tartrate and citrate can be tolerated. Thiocyanate and EDTA interfere at any level. Thus the proposed method, like most of the sensitive methods based on ion-associates with basic dyes,
2-
Z. MARCZENKO
1004
and R.
JGOB&SKI
Table 2. Results for determination of vanadium in the presence of other metals (enumerated in the text; 100 pg of each metal) Vanadium Added, pg
Found, w
Recovery, %
Standard deviation,* pg
Confidence limits (probability level 0.95), frg
1.00 2.00 3.00
0.94 1.89 2.85
94.0 94.3 94.8
0.061 0.066 0.074
0.95 + 0.06 1.89 f 0.07 2.85 f 0.07
*For seven determinations. Table 3. Results for determination
Salt Iron(II1) ammonia alum Aluminium ammonium alum
Sample weight, g
Vanadium, pg Added
Found
Standard deviation,* fig
-
1.18
0.066
1.50
2.60
0.075
-
0.84
0.057
1.50
2.28
0.071
of vanadium in alums Vanadium content, %
Confidence limits (probability level 0.95), fig
2.4 x lo-$
1.18 Ito.
1.7 x 10-s
0.84 + 0.06
5.0
5.0
*For seven determinations.
is poorly selective and a preliminary separation of vanadium is necessary. The method becomes specific after the extraction of vanadium as its complex with BPHA from a mixture of sulphuric and hydrofluoric acids.i4 Because stripping of the vanadium proved difficult, the organic solvent was evaporated, the BPHA complex thermally decomposed, and the vanadium(V) oxide obtained was dissolved in a
small amount of sodium hydroxide solution. If a sulphuri-hydrofluoric acid medium is used, Nb, Ta, MO, W, and Ti at lOOO-fold w/w ratio to V, as well as all other metals which do not form complexes with BPHA in a strongly acidic m~ium, do not interfere. The precision aud accuracy of the method employing separation of vanadium with BPHA were verified for synthetic samples containing a known amount of vanadium (1, 2 or 3 pg) as well as Nb, Ta, MO, W, Ti, Zr, Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd, Al, Ga, In, Sn, Pb, Sb, Bi (100 pg of each element). The results are shown in Table 2. The method was applied to the determination of trace amounts of vanadium (cu. 10w5%) in ferric and aluminium alums, by the following procedure. Dissolve the sample in 2M sulphuric acid plus 0.2 g of ammonium persulphate, and place the solution in a polypropylene separatory funnel. Add 5 ml of concentrated hydrofluoric acid, dilute with water to 50 ml and mix. Shake the solution with three successive portions (10, 10 and 5 ml) of BPHA solution, each time for 2 min. Transfer the combined extracts into a plating crucible. Evaporate the solvent on a water-bath and heat the crucible to ash the residue. After cooling the crucible, add 1 ml of 0.1M sodium hydroxide, dilute with water and heat to boiling. After cooling, adjust the pH to 5-7 and continue as already described. Run a blank in parallel,
Note. As some commercially available ammonium Persulphate and sulphuric acid can contain traces of vanadium, these reagents should be checked for vanadium content before use, and purified if necessary by shaking the mixture (to be used for dissolving the alum sample) with a suitable amount of BPHA solution.
The results for the dete~ination of v~adium in the alums are shown in Table 3. The recovery of known added amounts of vanadium was about 95%. work was supported by Research
Acknowledgemenr-This
Programme CPBP-01.17. REFERENCES
1. E. B. Sandell and H. Onishi, Photomefric Defermination of Traces of Merals, 4th Ed., General Aspects, Wiley, New York, 1978. 2. K. L. Cheng, K. Ueno and T. Imamura, Handbook of Organic Analytical Reagents, CRC Press, Boca Raton, 1982. 3. Z. Marczenko, Separation and Spec~rophoto~~ric Determination of Elemems, Horwood, Chichester, 1986. 4. Idem, Microchim. Acta, 1977 II, 651. 5. V. A. Nazarenko, N. V. Lebedeva and L. I. Vinarova, Zh. Analit. Khim., 1973, 28, 1100. 6. Idem, ibid., 1972, 27, 128. 7. J. Chwastowska and E. Grzegrzolka, Chem. Anal. (Warsaw), 1975, 20, 1065. 8. V. A. Nazarenko, E. A. Biryuk, L. I. Vinarova and K. A. Mukelo, Zh. Analit. Khim., 1982, 27, 252. 9. N. S. Poluektov, R. S. Lauer, L. A. Ovchar and S. F.
Potapova, ibid., 1975, 30, 1513. 10. A. G. Fogg, C. Burgess and D. T. Burns, Analyst, 1970, 95, 1012.
11. A. G. Fogg, A. Willcox and D. T. Burns, ibid., 1976, 101,67. 12. H. Bent and C. French, J. Am. Chem. Sot., 1941, 63, 568. 13. V. A. Nazarenko, V. P. Antonovich and E. M. Nevskaya, Metal Ion Hydrolysis in Diluie Solutions (in Russian), Atomizdat, Moscow, 1979. 14. E. M. Donaldson, Talama, 1970, 17, 583.
0039-9140/88 s3.00 + 0.00
Copyright (B 1988 Pergamon Press plc
EXT~CTION-SPECTROPHOTOMETRIC DETERMINATION OF VANADIUM WITH 3,5-DINITROCATECHOL AND BRILLIANT GREEN Z. MARCZENKO
and R.
ICOBI~SKI
Department of Analytical Chemistry, Technical University, Warsaw, Poland (Received 22 January 1988. Accepted 10 Juiy 1988) Summary-The formation and extraction of the ion-associates of the vanadiumw3,5-dinitrocatechol (DNC) anionic chelate complex with various basic dyes have been studied and a new sensitive
extraetion-speetrophotometrie method for the deter&ration of vanadium based on the system V~~DNC-B~lliant Green has been develored. Beer’s law is obeyed UD to a vanadium ~n~ntration of013 &ml and the molar absorptivity is I:7 x 10s l.m~le-‘.cm~~ at 630 nm. The molar ratios of the components and the form of the vanadium(V) cation in the extracted compound have been determined, and the formula [VO(OH)(DNC)i-][BG+], is proposed. Titanium, molybdenum, tungsten, EDTA and thiocyanate interfere seriously. The method becomes specific after a preliminary separation of vanadium by its extraction as the BPHA complex from H,SO,-HF medium, and is 40 times more sensitive than the spectrophotometric BPHA method. The proposed method has been applied to determination of traces of vanadium (about IO-‘%) in alums.
Many sensitive extraction-spectrophotometric methods for the determination of metals and non-metals are based on extraction of the ion-associates of basic dyes with anionic (mostly halide) complexes. Anionic complexes with chelating reagents (e.g. oxine, benzoic acid, 3,bdinitrocatechol) are seldom applied although they allow a sensitive determination of the metals which do not form sufficiently stable anionic complexes with halides (e.g., Ti, Th, U, V, W, Zr, lanthanides)! There is a lack of very sensitive s~ctrophotome~~ methods for the determination of vanadium and therefore the present work is devoted to the development of such a method based on the extraction of the ion-associate of a basic dye with the doubly charged anionic complex of vanadium(V) with 3,5~nitr~at~hol. Hitherto, the 3,5dinitrocatechol complex/basic dye system has been restricted to the extractive spectrophotometric determination of germanium, tin, titanium and lanthanidess9
N-Benzoyl-N-phenylhydroxyiamine (BPHA) solution, O.l%, in chloroform. Doubly distilled water was used throughout. All other chemicals used were of analytical-reagent grade. Apparatus A Speeord UV-VIS recording spectrophotometer and a VSUZ-P spectrophotometer with IO-mm glass cells were used. Polypropylene separatory funnels were used for the extraction from fluoride medium. Procedure Place the sample solution (PH 5-7), containing not more than 3 pg of vanadium(V), in a separatory funnel, and add 1 ml of 3,S-dinitrocatechol solution, followed by 1 ml of 0.2544 sulphuric acid. Make up with water to about 10 ml, add 5 ml of carbon tetrachloride, then 0.5 ml of Brilliant Green solution and shake for 1 min. After separation of the phases, transfer the organic layer to a IO-ml standard flask containing 1 ml of ethanol and one drop of 0.01 M sulphuric acid. Make up to the mark with carbon tetrachloride, mix, and measure the absorbance at 630 nm against a reagent blank prepared in the same way. RESULTS AND DISCUSSION
Preliminary Reagents Vanadium(V) standard solution (I mg/ml). Dissolve 1.786 g of V,O,, previously ignited at Soo”, in dilute sodium hydroxide solution. Acidify the solution with sulphuric acid and dilute to volume with water in a I-litre- standard flask. Dilute further with water as required. 3,5-~~nitrocatechol @NC) solution, 8 x IO-‘M, in 20% ethanol medium. DNC was synthesized6 and its purity confirmed by melting point measurement, elemental analysis, thin-layer chromatography, and infrared and NMR spectroscopy. Brilliant Green (BG) (Reakhim, USSR) solution, 8 x IO-‘M, in ethanol. The commercial preparation was purified according to Fogs et al.” TAL. 3%12--F
investigations
In an acidic (pH l-3) medium vanadium~) reacts with 3,5dinitrocatechol (DNC) to form an anionic yellow complex (I,, = 428 nm). In the presence of a suitable basic dye an ion-associate is formed. The following basic dyes were examined: Brilliant Green, Malachite Green, Methyl Violet, Crystal Violet (t~pheny~methane dyes), Methylene Blue, Nile Blue A (azine dyes), Rhodamine B and Rhodamine 6G (xanthene dyes). The possibility of the extraction of the vanadium-DNC-basic dye ion-associate from a sulphuric acid medium (pH O-4) into various organic solvents [benzene, toluene, carbon tetra-
2. MARCZENKO
loo2
chloride, chloroform, n-pentyl acetate, cyclohexane, methyl isobutyl ketone (MIBK), di-isopropyl ether and petroleum ether] was studied. Ion-associates are formed with all the basic dyes examined (apart from Methylene Blue) and are either extracted into the organic phase (benzene, toluene, chloroform, MIBK, n-pentyl acetate) or floated (cyclohexane, di-isopropyl ether and petroleum ether). Carbon tetrachloride extracts the Brilliant Green ion-associate but floats the ion-associates of all the other dyes tested. From the analytical point of view the extraction solvent should completely extract the vanadium ion-associate but not the dye itself or its ion-pair with DNC. These conditions were fulfilled only by the Malachite Green-toluene and Brilliant Green-carbon tetrachloride systems. As the latter gave the higher sensitivity, it seemed the more promising and was further investigated in detail. F5r~atio~ and extraction ion -associate
of
the V(V)-DNC-BG
From Fig. 1 it is evident that maximum absorbance is obtained when the sulphuric acid concentration is about 0.025M (pH 1.3 + 0.1). From more acidic media, the vanadium(V) ion-associate is not extracted quantitatively, and with progressively less acidic ones the blank absorbance increases considerably owing to the greater dissociation of DNC, the resulting increasing concentration of the DNC anion, and consequently greater probability of DNC-BG ionpair formation. Because the pH of the aqueous phase is practically constant during the extraction, no buffer solution is needed. The required acidity is established by adding a suitable amount of sulphuric acid to a sample solution of pH 5-7. This way of establishing the acidity gives more reproducible results and is faster than adjustment with pH-meter control. Figure 2 shows the dependence of the absorbance of the extract and of the blank on the DNC concentration. A DNC concentration of at least 8 x IO-‘M (16-fold molar excess with respect to 2.5 pg of vanadium) gives maximal and reproducible absorbance values.
and
R.
LOBIX?SKI
2
4 DNC]
6 lo-’
610 M
Fig. 2. Dependence of the absorbance of the extract on the DNC concentration: curve 1, absorbance of the vanadium ion-associate measured against the blank; 2, blank absorbance measured against CCl,-ethanol (9 + 1) mixture. Figure 3 shows that maximal absorbance readings are obtained with a Brilliant Green concentration of at least 4 x 10-5M (about g-fold molar excess with respect to 2.5 pg of vanadium). Because of the dimerization of Brilliant Green in more concentrated (about I x 10F3A4) aqueous solutions, and then dissociation into monomer after dilution,” an ethanolic solution of the dye was used. It was observed that maximal and well reproducible absorbances were obtained when the Brilliant Green solution was introduced after addition of the extracting solvent. The extraction should then be done as soon as possible because of the protonation of Brilliant Green in acidic (pH < 2) aqueous solutions.Lo The vanadium is completely extracted by shaking the aqueous phase and one S-ml portion of carbon tetrachloride for 1 min. The colour of the extract fades after separation of the layers and therefore the extract itself is useless for analytical purposes, but it has been found that the presence of a small amount of acid prevents the fading of the colour. The addition of even one drop of O.OlM sulphuric acid makes the extract turbid, but subsequent addition of ethanol or methanol clears the turbidity. This addition also causes a significant increase in the absorbance of the extract and a change 1 Cl8
/ 0.6 A 0.4 0.2
0.2-
2
2 I I (101 a02
I 003 Hz=4
Y a04
~ 0.05 Cl06
hf
Fig. 1. Effect of the sulphuric acid concentration on formation and extraction of the vanadium ion-associate: curve I, absorbance of the extract measured against the blank; 2, blank absorbance measured against CCl,-ethanol (9 + I) mixture.
1
2
3
4
5
UK+1 10-S M
Fig. 3. Dependence of the absorbance of the extract on the Brilliant Green concentration: curve I, absorbance of the vanadium ion-associate measured against the blank; 2, blank absorbance measured against CC&-ethanol (9 + I) mixture.
Determination of vanadium in colour from green to blue. A final concentration of 10% ethanol or methanol is optimal. Although the molar absorptivity is higher ( 1.94 x 1051.mole- ’ .cm-‘) for the methanol than the ethanol (1.70 x 105) system, the latter is preferred because the blank value is lower (0.08 instead of 0.12 with methanol). Composition of the V( V)-DNC-BG
ion-associate
The molar ratios of vanadium to DNC and to Brilliant Green were determined by the Bent and French logarithmic methodi and both found to be 1: 2. These results were confirmed by the Job method and are the same for both the methanol and ethanol systems. An extract containing a known amount of vanadium was evaporated and the residue was dissolved in a I : 4 v/v mixture of water and methanol. The absorbance of the solution was almost equal to the absorbance of a solution (in the same solvent) of an amount of Brilliant Green that was in 2: 1 molar ratio to the amount of vanadium in the extract. Attempts were made to determine the form in which vanadium(V) is present in the ion-associate. The molar ratios DNC: V and BG: V found for the ion-associate permit postulation of the following equation for its formation and extraction (DNC is denoted by H,R): VO(OH);-‘+
2H,R+
2(BG)+
+~O(OH):-i(H~_.R)2][(BG):]+2nH+ The equilibrium 1y
_
constant is
[vO(OH)j-'(H,-,R),(BG):l W+l'" pO(OH);-‘1 [H, R]‘[(BG)+]’
Hence -1ogB
= -pK
+ 2npH
where
B = IYWW - ‘1W, R12KBG>+12 ~ofoHt2-‘(H,-.R),tB~):l The relationship -log B = f (pH) was investigated for all possible vanadium cation forms (i = 0, 1,2, 3). All the constants necessary for the calculations are known.s,‘3 The values of -log B plotted vs. pH give a straight line with a slope that is an even number (n should be an integer) only for i = 2, which corresponds to the cation VO(OH)2+. We conclude from this that the ion-associate has the formula
1003
Table 1. Statistical evaluation of the results for determination of vanadium by the proposed method Vanadium, 1118
Standard
Added
Found
deviation,* !Jg
1.00 2.00 3.00
1.01 1.99 2.98
0.032 0.030 0.036
Confidence limits ~robabiiity level 0.95), pg
1.Ol f 0.03 1.98 + 0.03 2.98 + 0.04
*For seven determinations. It must he emphasized that, in contrast to the extraction-spectrophotometric methods based on the ion-associates formed by singly charged anionic metal-halide complexes with basic dyes, in the present method a doubly charged anion is extracted in association with two singly charged dyestufT cations, which increases the sensitivity.
Determination of vanadium The calibration graph obeys Beer’s law over the range 0.005-0.3 pg/ml V. The molar absorptivity is 1.70 x 10’ l.mole-l.cm-i at 630 nm (specific abso~ti~ty a = c/at.wt x 1000 = 3.3). Thus, the proposed method is 3040 times more sensitive than the well-known BPHA and 8-hydroxyquinoline methods and about 5 times more sensitive than the PAR method.3 The statistical evaluation of determination of trace amounts of vanadium at three concentration levels shows the good precision and accuracy of the proposed method (Table 1). The effects of many cations and anions on the determination were examined. Titanium, molybdenum and tungsten interfere at any level. Gallium and indium can be tolerated when present in 1: 1 w/w ratio to vanadium, and iron(III) in l~fold ratio. Aluminium and bismuth give 10% positive error at lOOO-fold w/w ratio to vanadium. Cr(III), Mn, Co, Ni, Cu, Zn, Cd, Sb(V), and alkali and alkaline-earth metal ions do not interfere at all. Anions can interfere either by forming extractable ion-pairs with the dye cation, thus influencing the blank, or by complexing the vanadium cation. Chloride, bromide, nitrate and acetate do not interfere at IOOO-fold molar ratio to vanadium; IO-fold molar ratio of fluoride, iodide, perchlorate, oxalate, tartrate and citrate can be tolerated. Thiocyanate and EDTA interfere at any level. Thus the proposed method, like most of the sensitive methods based on ion-associates with basic dyes,
2-
Z. MARCZENKO
1004
and R.
JGOB&SKI
Table 2. Results for determination of vanadium in the presence of other metals (enumerated in the text; 100 pg of each metal) Vanadium Added, pg
Found, w
Recovery, %
Standard deviation,* pg
Confidence limits (probability level 0.95), frg
1.00 2.00 3.00
0.94 1.89 2.85
94.0 94.3 94.8
0.061 0.066 0.074
0.95 + 0.06 1.89 f 0.07 2.85 f 0.07
*For seven determinations. Table 3. Results for determination
Salt Iron(II1) ammonia alum Aluminium ammonium alum
Sample weight, g
Vanadium, pg Added
Found
Standard deviation,* fig
-
1.18
0.066
1.50
2.60
0.075
-
0.84
0.057
1.50
2.28
0.071
of vanadium in alums Vanadium content, %
Confidence limits (probability level 0.95), fig
2.4 x lo-$
1.18 Ito.
1.7 x 10-s
0.84 + 0.06
5.0
5.0
*For seven determinations.
is poorly selective and a preliminary separation of vanadium is necessary. The method becomes specific after the extraction of vanadium as its complex with BPHA from a mixture of sulphuric and hydrofluoric acids.i4 Because stripping of the vanadium proved difficult, the organic solvent was evaporated, the BPHA complex thermally decomposed, and the vanadium(V) oxide obtained was dissolved in a
small amount of sodium hydroxide solution. If a sulphuri-hydrofluoric acid medium is used, Nb, Ta, MO, W, and Ti at lOOO-fold w/w ratio to V, as well as all other metals which do not form complexes with BPHA in a strongly acidic m~ium, do not interfere. The precision aud accuracy of the method employing separation of vanadium with BPHA were verified for synthetic samples containing a known amount of vanadium (1, 2 or 3 pg) as well as Nb, Ta, MO, W, Ti, Zr, Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd, Al, Ga, In, Sn, Pb, Sb, Bi (100 pg of each element). The results are shown in Table 2. The method was applied to the determination of trace amounts of vanadium (cu. 10w5%) in ferric and aluminium alums, by the following procedure. Dissolve the sample in 2M sulphuric acid plus 0.2 g of ammonium persulphate, and place the solution in a polypropylene separatory funnel. Add 5 ml of concentrated hydrofluoric acid, dilute with water to 50 ml and mix. Shake the solution with three successive portions (10, 10 and 5 ml) of BPHA solution, each time for 2 min. Transfer the combined extracts into a plating crucible. Evaporate the solvent on a water-bath and heat the crucible to ash the residue. After cooling the crucible, add 1 ml of 0.1M sodium hydroxide, dilute with water and heat to boiling. After cooling, adjust the pH to 5-7 and continue as already described. Run a blank in parallel,
Note. As some commercially available ammonium Persulphate and sulphuric acid can contain traces of vanadium, these reagents should be checked for vanadium content before use, and purified if necessary by shaking the mixture (to be used for dissolving the alum sample) with a suitable amount of BPHA solution.
The results for the dete~ination of v~adium in the alums are shown in Table 3. The recovery of known added amounts of vanadium was about 95%. work was supported by Research
Acknowledgemenr-This
Programme CPBP-01.17. REFERENCES
1. E. B. Sandell and H. Onishi, Photomefric Defermination of Traces of Merals, 4th Ed., General Aspects, Wiley, New York, 1978. 2. K. L. Cheng, K. Ueno and T. Imamura, Handbook of Organic Analytical Reagents, CRC Press, Boca Raton, 1982. 3. Z. Marczenko, Separation and Spec~rophoto~~ric Determination of Elemems, Horwood, Chichester, 1986. 4. Idem, Microchim. Acta, 1977 II, 651. 5. V. A. Nazarenko, N. V. Lebedeva and L. I. Vinarova, Zh. Analit. Khim., 1973, 28, 1100. 6. Idem, ibid., 1972, 27, 128. 7. J. Chwastowska and E. Grzegrzolka, Chem. Anal. (Warsaw), 1975, 20, 1065. 8. V. A. Nazarenko, E. A. Biryuk, L. I. Vinarova and K. A. Mukelo, Zh. Analit. Khim., 1982, 27, 252. 9. N. S. Poluektov, R. S. Lauer, L. A. Ovchar and S. F.
Potapova, ibid., 1975, 30, 1513. 10. A. G. Fogg, C. Burgess and D. T. Burns, Analyst, 1970, 95, 1012.
11. A. G. Fogg, A. Willcox and D. T. Burns, ibid., 1976, 101,67. 12. H. Bent and C. French, J. Am. Chem. Sot., 1941, 63, 568. 13. V. A. Nazarenko, V. P. Antonovich and E. M. Nevskaya, Metal Ion Hydrolysis in Diluie Solutions (in Russian), Atomizdat, Moscow, 1979. 14. E. M. Donaldson, Talama, 1970, 17, 583.