- Email: [email protected]
A chemical enhancement method for the spectrophotometric determination of trace amounts of arsenic
Tahta, Vol. 39, No. 5, pp. 555-561, 1992 Printed in Great Britain. All rights reserved
0039-9140/92 S5.00 + 0.00 Copyright 0 1992 Pcrgamon Press plc
A CHEMICAL ENHANCEMENT METHOD SPECTROtiHOTOMETRIC DETERMINATION AMOUNTS OF ARSENIC
FOR THE OF TRACE
K. PAL.ANIVELU,N.BALASIJBRAMANUN and T.V. RAMAICRKWNA* Department of Chemistry, Indian Institute of Technology, Madras-600 036, India (Received 3 May 1991. Revised 1 August 1991. Accepted 1 August 1991) !Iammary-A highly sensitive spectrophotometric method for the determination of 0.03-1.0 /Ig of arsenic is described. After extraction as AsI, into benxene, it is selectively stripped into water. Both the arsenic@) and iodide present in the aqueous phase are made to react with iodate in acidic medium in the presence of chloride to form the anionic chloro complex, ICI,. The determination is completed after extraction of ICI; species as an ion-pair with Rhodamine 6G into benzene and measuring the absorption of the extract at 535 nm. The coefficient of variation is 1.5% for 10 determinations of 0.5 pg of arsenic. The method has been applied to the detetmination of arsenic content in plant materials, high purity iron, copper base alloys and inorganic arsenic levels of natural waters.
The widely used methods for the spectrophotometric determination of trace amounts of arsenic are based on the formation of molybdenum blue’ and the reaction of arsine with silver diethyldithiocarbamate.2 Methods based on the formation of an ion-pair between arsenomolybdate and a large dye cation have been described to improve the sensitivity and selectivity of the determination.ss None of these methods, however, are sufficiently sensitive to establish the arsenic levels of natural waters without resorting to a preconcentration step. The ability of arsenic(II1) to react with potassium iodate in the presence of chloride in an acidic medium to form the anionic chlorocomplex is well known6 and has been applied to the spectrofluorimetric determination of 0.31.5 pg of arsenic(II1) after extraction of the anionic complex as an ion-pair with Rhodamine B into benzene.’ The presence of ions that are oxidized by iodate however interferes seriously in the determination. Since iodide can also react with potassium iodate to form the anionic chlorocomplex’ similar to arsenic(III), it seemed worthwhile to examine whether the formation of the anionic complex and its interaction with a cationic dye could be put to advantage for the spectrophotometric determination of both arsenic(II1) and arsenic(V) after extractive separation as arsenic triiodide into benzene. Such *Author for correspondence.
an approach, besides improving the selectivity (since relatively few ions extract into benzene from iodide media), would also provide considerable enhancement in sensitivity arising due to reaction of iodide, in addition to arsenic(II1) present in arsenic triiodide with potassium iodate to form the anionic chloro-complex. Detailed examination disclosed that the reaction with potassium iodate in acidic chloride containing medium after selectively stripping the arsenic triiodide into water and subsequent interaction of the anionic iodine complex with Rhodamine 6G can form the basis for the spectrophotometric determination of arsenic. The resulting ion-pair, after extraction into benzene, had maximum absorption at 535 nm, and arsenic concentration as low as 0.03 j.4gin a final volume of 25 ml can be determined with this procedure. The method was found suitable for the rapid determination of arsenic concentration of natural waters and applicable to the determination of arsenic in plant materials, high purity iron and copper base alloys without prior separation of arsine.
EXPERIMENTAL Apparatus
Absorbances were measured with a Carl Zeiss PMQ II spectrophotometer with IO-mm quartz cells. 555
556
K. PALANIVELUet al.
Reagents Standard arsenic(III) solution, I.Omglml. Prepared by dissolving 0.1321 g of arsenious oxide in 10 ml of 1M sodium hydroxide, adding 7.5 ml of 1M sulphuric acid and diluting to volume with water in a loo-ml standard flask. A suitable volume of this solution was diluted to obtain working standards. The following solutions were prepared by dissolving appropriate amounts of the reagents in distilled water. Potassium iodate solution, 0.01%. Rhodamine 6G solution, 0.02%. Sodium chloride solution, 15%. Saturated sodium chloride solution, N 5NL Potassium iodide solution, 2M. Sodium hypophosphite, monohydrate solution, 2M. Sulphuric acid, 2.5 and 12.5M. Benzene (thiophene free’). Used for extrac-
tion.
a 25-ml standard flask. Add 2 ml each of 2.5M sulphuric acid, 0.01% potassium iodate solution and 0.02% Rhodamine 6G solution followed by 4 ml of 15% sodium chloride solution. Dilute to volume with water and allow to stand for 5 min. Transfer the made up solution into a 60-ml separating funnel and extract with 5 ml of benzene for one minute. Separate the organic layer into a dry test tube and add about 1 g of anhydrous sodium sulphate. Measure the absorbance of the extract at 535 nm in lo-mm cells against a reagent blank run through the entire procedure. Establish the concentration of arsenic by reference to a calibration graph prepared with 0.03-l .O pg of arsenic, following the above procedure. RESULTS
AND DISCUSSION
Extractive separation of arsenic as arsenic triiodide
Separation of arsenic. Transfer a portion of the sample solution into a 125-ml separating funnel. Add enough sulphuric acid and sodium chloride to make their concentration 3M (4M for sample volumes greater than 30 ml) and 2M respectively. Add 1 ml of 2M sodium hypophosphite solution followed by sufficient potassium iodide to provide an overall concentration of 0.3M. Shake the solution for 2 min with 10 ml of benzene (separate the aqueous layer and repeat the extraction with 10 ml of benzene if the initial sample volume used is >30 ml). Collect the organic phase(s) in a 60-ml separating funnel, treat with 2 ml of di-isopropyl ether solvent (4 ml if the extraction is performed twice) containing 0.015% peroxide, mix well and allow to stand for 15 min. Then extract with 10 ml of water for one minute and discard the organic layer.
Although arsenic can be extracted from an iodide medium into a variety of solvents, the extraction into hydrocarbon solvents as arsenic triiodide from a highly acidic medium (> 8N) has been shown to be highly selective.‘0-‘3 It was found that from solutions of lower acidity, extraction becomes quantitative in the presence of sodium chloride as a salting-out agent. Using simplex optimization,‘4 it was established that for quantitative extraction the overall concentration with respect to sulphuric acid, potassium iodide and sodium chloride should be 3,0.3 and 2M respectively. A single equilibration for two minutes with 10 ml of benzene was adequate for quantitative extraction which remained unaffected up to an aqueous phase volume of 30 ml. Beyond this volume, the extraction was incomplete and repetition of the extraction on the aqueous phase was not effective. Increasing the concentration of potassium iodide up to 0.6M showed no marked improvement in the recovery of arsenic. For quantitative extraction of arsenic from aqueous volume up to 100 ml, it was found necessary to equilibrate the solution twice with 10 ml of benzene each time, after raising the acidity to >4M with respect to sulphuric acid. A single equilibration for one minute with 10 ml of water was found adequate for stripping the arsenic from the organic layer.
Determination
Determination with Rhodamine 6C
Transfer an aliquot of the stripping agent containing not more than 1.0 pg of arsenic into
Preliminary studies under conditions found optimal (hydrochloric acid-l .5M, IO;-
Di-isopropyl ether solvent. Di-isopropyl
ether containing 0.015 weight per cent peroxide was prepared by equilibrating 50 ml of the solvent rendered free from peroxide9 with 50 ml of 0.5M sulphuric acid containing 0.3 weight per cent of hydrogen peroxide for 5 min. It was prepared fresh every week, stored in a brown bottle and kept in a refrigerator. Procedure
Chemically enhanced spectrophotometric
0.8
0.1
$
8 g
557
determination of trace As
f 0.4 v) 2
0.
u)
2 I
0
lR
_d-.+-.-.--.--.I
I
I
I
1
0.1
0.2
0.3
0.4
0.5
Concentration
of sulphuric acid M
0.025mM and Rhodamine B-0.025mM) for the spectrofluorimetric determination of arsenic’ showed that the colour when extracted into benzene gradually faded on standing. This was overcome by using Rhodamine 6G instead of Rhodamine B, the use of which stabilized the colour of the extract that absorbed maximally at 535 nm for 75 min. Acidification with sulphuric acid in the presence of sodium chloride was preferred to hydrochloric acid because in the former the blank was low and reproducible, due to lower levels of impurities. Variation of the acidity of the reaction medium as shown in Fig. 1, revealed that the ion-pair extraction remained maximal and constant when the overall acidity was greater than 0.15M for both arsenic(II1) and iodide. The
0.6
[
2J 0.3
1
3
8
influence of potassium iodate, Rhodamine 6G and sodium chloride on the extraction of the ion-pair into benzene are shown in Figs. 2,3 and 4, respectively. Based on these experiments an overall concentration of 0.2M sulphuric acid, 0.04mM (0.0008%) potassium iodate, 0.035mM (0.0016%) Rhodamine 6G and 0.04M (2.4%) sodium chloride were chosen as optimum for the determination of both arsenic(II1) and iodide. Although the presence of sodium chloride and Rhodamine 6G in excess of the recommended concentration caused a slight increase in blank value, the net absorbance due to reaction with arsenic(II1) and iodide remained unaffected. The formation of the ion-pair in both instances was found to be almost instantaneous and mixing the phases for about 30 set was found to be sufficient for its quantitative extraction into benzene. Benzene, toluene,
r--t .-.-.-../
_._._._.-
2
6
Fig. 3. Effect of Rhodamine 6G concentration. A. 5 pg of As(II1) with “x” ml of 0.01% Bhodamine 6G, 2 ml each of 2.5M H,SO, and 0.01% KIO,, 5 ml of 15% NaCl, total volume 25 ml, extracted with 5 ml of benzene, measured against respective reagent blank. B. As in (A) with 5 )~g of I- instead of As(II1).
r--f /.-._.
4
2
ml of 0.01% Rhodamine 6G
Fig. 1. Effecf o/acidity. A. 2 ml of 0.01% KIO,, 5 ml each of 0.01% Bhodamine 6G and 15% NaCl, “x” ml of 2SM H2SOdr diluted to 25 ml, extracted with 5 ml of benzene measured against benzene at 535 nm in IO-mm cells. B. As in (A) with 5 pg of As(II1). C. As in (B) but measured against (A). D. As in (A) with 5 pg of I-. E. As in (D) but measured against (A).
p:
0
4
5
ml of 0.01% potassium iodate Fig. 2. EffGct of potassium iodate concentration. A. 5 pg as As(II1) with “xl’ ml of 0.01% KIO,, 2 ml of 2.5M Hr SO,, 5 ml each of 0.01% Bhodamine 6G and 15% NaCl solution, total volume 25 ml, extracted with 5 ml of benzene measured against respective reagent blank. B. As in (A) with 5 pg of I- instead of As(II1).
I
I
I
I
2
4
8
8
ml of 15% NaCl Fig. 4. As(II1) H,SO,, volume against
Effect of sodium chloride concentration. A. 5 pg of with “x” ml of 15% NaCl, 2 ml each of 2.5M 0.01% KIO, and 0.02% of Bhodamine 6G, total 25 ml, extracted with 5 ml of benzene, measured respective reagent blank. B. As in (A) with 5 pg of I- instead of As(II1).
K. PALANNELU et al.
558
xylene, hexane, cyclohexane, carban tetrachloride, chloroform, and isoamyl acetate were investigated as extraction solvents. However, only benzene proved to be satisfactory, as the extraction of the ion-pair was found to be maximum in this solvent. When independently checked, linear calibration curves passing through the origin were obtained for arsenic(II1) and iodide in the range 0.15-l 1.O pg and 0.1-6.0 pg, respectively. The molar absorptivities were found to be 2.6 x lo4 1.mole-’ .cm-’ for arsenic(II1) and 7.6 x lo4 1.mole-’ . cm-’ for iodide. Extractive arsenic
separation
and
determination
of
As the optimal conditions for the determination of arsenic(II1) and iodide were found to be identical, it was decided to apply the method for the determination of arsenic, after its extraction as arsenic triiodide into benzene and stripping into water. Difficulties were encountered because the blank was intensely coloured, evidently due to co-extraction of hydroiodic acid. As washing the benzene layer with acid solution was found to be ineffective in removing the hydroiodic acid, the use of a dilute solution of hydrogen peroxide was considered for washing purposes. The use of a 0.3 weight per cent solution of hydrogen peroxide in 5iU sulphuric acid was effective in minimizing the blank value, but the results were not reproducible. Blank values comparable to the normal blank resulted only when the hydrogen peroxide concentration in the wash liquid was increased above l.O%, but this resulted in a considerable decrease in the sample absorbance also. In this instance the absorbance of the sample extract corresponded to that of arsenic alone, indicating the oxidative loss of the iodide that remained associated with arsenic. Addition of organic solvents containing peroxide to the benzene layer after the separation of the aqueous layer was examined next. The solvents examined included, isoamyl acetate, cyclohexanone, di-isopropyl ether, isobutyl methyl ketone and mesityl oxide. After adding 1 ml of the solvent, the benzene layer was mixed and allowed to stand for 15 min before stripping with water. In all instances, there was a considerable reduction in the blank. As the addition of di-isopropyl ether containing 0.022 weight per cent peroxide was found to be most effective among the solvents examined it was used, with an optimal amount of peroxide,
for destroying the co-extracted hydroiodic acid. In order to establish the optimal range of peroxide in di-isopropyl ether that would be effective for produding low and reproducible blanks, the solvent was first rendered free from peroxide9 and then varying amounts of peroxide were introduced by equilibrating for five min with an equal volume of 0.5M sulphuric acid containing 0.1-l .O weight per cent hydrogen peroxide. After equilibration the solvent was found to contain peroxide in the range 0.005-0.035 weight per cent. The effect of the addition of 2 ml of each peroxide containing ether to the benzene extract was examined separately. After the addition of ether, the extracts were allowed to stand for 15 min before stripping into water for completing the determination. The results, shown in Fig. 5, clearly indicate that the addition of 2 ml of ether containing peroxide in the range 0.012-0.026 weight per cent effectively reduces the blank value without affecting the sample. When the recommended procedure was followed for the preparation of peroxide containing ether, it was found to contain 0.015 weight per cent. After the addition of 2 ml of this solvent to the benzene extract, a standing time of 10 min was found to be adequate (Fig. 6) to destroy the co-extracted hydroiodic acid, and the recovery
0.6 -
0: 0.4{ 3 0.2 -
I 0
0.01
0.02
0.03
0.04
Weight % of peroxide Fig. 5. Effect of variation ofperoxide content of di-iso-propyl ether. A. 10 ml of benzene extract, 2 ml of di-iso-propyl ether solvent, standing time 15 min, before stripping and colour development. Measured against benzene. B. As in (A) using benzene extracts obtained with 1 pg of arsenic, measured against (A).
Chemically enhanced spectrophotometric determination of trace As
559
5Asq-+210,+2H+=12+5As0:-+H20
(1)
51- + 10, + 6H+=312 + 3H2G
(2)
In the presence of chloride, the excess iodate” present in the solution can further oxidize the generated iodine and stabilize it as ICI; species in accordance with 212+ 10; + 6H+ + 5Cl-+5ICl+
3H2G (3)
ICI + Cl-*ICI, 0
10
20
30 Time,
40
50
60
min
Fig. 6. Influence of standing time after the addition of di-iso-propyl ether solvent on the stabiliiy of Asl, in the benzene extract. A. 10 ml of benzene extract, 2 ml of di-iso-propyl ether solvent (0.015%). standing time x min before stripping and colour development. Measured against benzene. B. As in (A) with benzene extracts obtained with 1 pg of arsenic, measured against (A).
of arsenic remained unaffected up to a standing time of 25 min before stripping into water. When the procedure described in the Experimental section was followed, 1.0 c(g of arsenic(III) gave an absorbance of 0.62 f 0.02 compared to the expected value of 0.66 [0.07 for 1 pg of arsenic@) and 0.59 for 5.0 pg of associated iodide]. The calibration graph was linear over the range 0.03-1.0 pg of arsenic. From the slope of the calibration plot the molar absorptivity was established to be 2.37 x 10’ 1.mole-‘. cm-‘. The Sandell sensitivity was found to be 8 x lo-’ pg/cm2. The precision of the proposed procedure was checked by establishing the concentration of 10 samples containing 0.50 pg of arsenic. The mean recovery was found to be 0.497 pg with a coefficient of variation of 1.5%. Nature of the extracted species The equilibrium shift method” was employed to establish the ratio of arsenic to iodide. The slope of the straight line obtained by plotting log D us. log [I-] was found to be 2.98, indicating that arsenic was extracted into the benzene layer as arsenic(II1) triiodide, which conforms to the findings reported in the literature.13 When the benzene layer was equilibrated with water, both arsenic(II1) and the associated iodide were readily stripped into water. The stripping, when reacted with iodate in acidic medium, leads to the formation of iodine as shown in the following equations: TAL 39,5-H
(4)
The overall reaction for the formation of ICI, from AsO:- and iodide can be represented as 2AsO;- + IO; + 2H+ + 2Cl=2AsO:-
+ ICI; + H,O
(5)
21- + IO, + 6H+ + 6Cl+3ICl,
+ 3H2G
(6)
While the reaction with iodide can give rise to only ICI;, the possibility of the formation of ICI;, due to the reduction of the iodate to 13+ state and stabilization as ICI; can also be expected in the case of arsenic(II1) in accordance with AsO:- + IO, + 4H+ + 4Cl*AsO:-
+ ICl,- + 2H,O
(7)
As both ICI; and ICI; can form ion-pairs with Rhodamine 6G and extract into benzene, investigations were carried out to establish the nature of species responsible for the extraction of the ion-pair in the determination of arsenic(II1) with iodine solution in benzene and completing the determination under conditions optimal for the determination of arsenic(II1). Though this experiment would not give direct evidence for the formation of ICI;, it would provide indirect evidence of whether the formation of ICI, was responsible for the observed colour. Experiments with iodine solution in benzene showed a linear relationship with an increasing amount of iodine and the absorbance of 0.20 produced by 2.0 pg of iodine was identical to that produced by 2.94 pg of arsenic(II1). Since, 2.0 pg of iodine would result from 2.94 pg of arsenic(II1) [equation (l)], it was felt reasonable to conclude that the ICI; species was responsible for the observed absorbance with arsenic(II1). Had ICl; been the species, as evident from equation (7), 1.47 pg of arsenic(II1) would have been adequate to produce the observed absorbance.
K. PALANIVELU et al.
560
Table 1. Determination of arsenic in various materials
S. No.
Sample
1
Brass (0.5 g/50 ml) BCS No. 20712 Gun metal (Cuba& (0.1 g/50 ml) BCS No. 26013 High purity iron (0.05 g/25 ml) Plant materials (1 g/25 ml) a. Grass
2 3 4
Arsenic present 0.0057%*
Arsenic added, IJg 2.50
0.006% 2.50 0.026%
10.0
c. Plantain leaf Water (50 ml) a. Tap water b. Sea water (Bay of Bengal, Madras)
64.5 f 0.87 88.7
0.64
12.0 + 0.10 22.2
0.024
2.6 f 0.10 7.5 1.8 f 0.05 6.8 4.0 f 0.17 8.9
b. Mango leaf
5
Arsenic found,t (Mean value + a), % LJg 27.0 + 0.57 0.0054 51.5
2.3 pg/l*
-
2.8 /lg/1*
0.5 0.5
Recovery, % 98.0 96.8 102.0
98.0 100.0 98.0
0.125 k 0.005 (2.; :;$l) 99.0 0.145 i 0.003 (2.9 pg/l) 0.65
101.0
*Molybdenum blue method. tAverage value of three determinations.
Another evidence in support of the formation of ICI; species in the case of arsenic(II1) was obtained by establishing the stoichiometry of the species extracted as an ion-pair into benzene. The combined ratio of the anionic chloro-complex and the dye cation was established by determining the ratio of arsenic(II1) to Rhodamine 6G by the continuous variation and mole ratio methods. Both methods showed that for maximum extraction of the ion-pair, the medium should contain 2 moles of arsenic(II1) and 1 mole of Rhodamine 6G. Since 2 moles of arsenic(II1) would yield 1 mole of ICl; species [equation (5)] and 2 moles of ICI; species [equation (7)], it was concluded that ICI; species and the monovalent Rhodamine 6G cation interact in a 1: 1 ratio for the formation of the ion-pair. If ICI; were to be the species, the arsenic(II1) and Rhodamine 6G would have reacted in a 1: 1 ratio. EJJct
of interfering metal ions
The presence of one milligram amounts of various ions which are known to react with iodide were examined in the determination of 0.5 fig of arsenic by the proposed procedure. The ions examined included Ag+, BP, Cd*+, Cu*+, Fe3+, Ge4+, In3+, MOO:-, Tl+, TeOi-, Sb3+, SeO:-, Sn4+ and Zn*+. The determinations were carried out in the presence of
sodium hypophosphite solution to prevent the liberation of iodine. Although 1 ml of 2M sodium hypophosphite was adequate, the addition of larger volumes would be necessary in the presence of excessive amounts of iron(II1) and copper(I1). The excess hypophosphite had no effect on the recovery of arsenic. Only antimony(II1) at all concentration levels and germanium(IV) in excess of 100 pug were found to cause positive interference in the determination of arsenic. The influence of antimony(II1) could be overcome by extracting off as antimony(II1) triiodide into a 1: 1 mixture of cyclohexanone-isobutyl methyl ketone from 0.5M sulphuric acid and 0.25M potassium iodide. The inclusion of this step also eliminated the difficulties arising due to dispersion of copper(I) iodide in the organic layer when copper(I1) in excess of 1 mg was present with arsenic. Application of the method to real samples
Table 1 presents the results of the analysis of several arsenic containing samples with and without added standard by the proposed procedure. The water samples were filtered immediately after collection and acidified with sulphuric acid to -pH 1. The solid samples were brought into solution with 1: 1 sulphuric acid and hydrogen peroxide.‘6*‘7After evaporating almost to dryness to decompose excess
Chemically enhanced spectrophotometric determination of trace As
peroxide, the residue was dissolved in 10 ml of water, filtered if necessary, and made up to known volume with water. Suitable aliquots were subjected to determination following the procedure given under ‘Experimental’. As no standard samples were available, the arsenic content of brass,” plant materials” and waterI was also established following the standard molybdenum blue method after separation of arsenic. The results summarized in Table 1 clearly show that the method developed works satisfactorily for the analysis of traces of arsenic in various materials. CONCLUSION
The method described provides a simple and reliable means of determining trace amounts of arsenic by spectrophotometry. It is more sensitive (6 = 2.37 x 10’ l.mole-‘.cm-‘) than those based on the formation of Molybdenum Blue (6 = 2.5 x lo4 l.mole-’ .cm-‘) and reacwith silver diethyldithiocarbamate tion (C = 1.4 x lo4 l.molee’.cm-‘). It is superior to the fluoresence method based on the use of Rhodamine B in that the colour system is quite stable and can find use for the determination of both arsenic(II1) and arsenic(V) in the range 0.03-l .O pg. The method also has the advantage of virtual freedom from interferences from extraneous ions and can therefore find use for the rapid and precise determination of trace amounts of arsenic in natural and synthetic samples. Acknowledgement-One of us (KP) is grateful to CSIR, New Delhi for the award of a Research fellowship.
561
REFERENCES
1. K. Sugawara, M. Tanaka and S. Manamori, Bull. Chem. Sot. Japan, 1956, 29, 670. 2. G. Stratton and H. C. Whitehead, J. Am. War. Wks. Ass., 1962, 84, 172. 3. C. Matsubara, Y. Yamamoto and K. Takamura, Analyst, 1987, 112, 1257. 4. T. Nasu and K. Kan, ibid, 1988, 113, 1683. 5. T. V. Ramakrishna and K. Palanivelu, Indian J. Environ. Protection, 1989, 9, 265. 6. Vogel’s Quantitative Inorganic Analysis Including Elementary Instrumental Analysis (revised by J. Bassett, R. C. Denney, G. H. Jefferey and J. Mendnan), 4th Ed., p. 386. ELBS, Longman, London, 1978. 7. D. Yamamoto and K. I. Kisu, Japan Analyst, 1974,X3, 638; Anal. Abstr., 1975, 29, lB134. 8. Volgel’s Text book of Practical Organic Chemistry including Qualitative Organic Analysis (B. S. Furniss, A. J. Hannford, V. Rogers, P. W. G. Smith and A. R. Tatchell), 4th Ed., p. 266. ELBS and Longman, London, 1978. 9. Ibid., B. S. Fur&s, A. J. Hannford, V. Rogers, P. W. G. Smith and A. R. Tatchell), 4th Ed., p. 272. ELBS and Longman, London, 1978. 10. A. R. Byrne and D. Gorenc, Anal. Chim. Acta, 1972,59, 81. 11. E. M. Donaldson and M. Want, Taianta, 1986, 33, 35. 12. K. Tanaka, Japan Analyst, 1960, 9, 700, Anal. Abstr., 1961, 9, 3714. 13. N. Suzuki, K. Satoh, H. Shoji and H. Imura, Anal. Chim. Acta, 1986, 185, 239. 14. D. E. Long, ibid, 1969, 46, 193. 15. S. Motomizu and K. Toei, ibid., 1977, 89, 167. 16. Report of the analytical methorls committee on oxidation of organic matter, Analyst, 1967, 92, 403. 17. F. D. Snell, Photometric and Fluorimetric Metho& of Analysis of Metals, Part I, p. 354. Wiley-Interscience, New York, 1978. 18. Idem, ibid., Part I, p. 343. Wiley-Interscience, New York, 1978. 19. F. J. Welcher, Standard Method of Chemical Analysis, Vol. 2, Part B, 6th Ed., p. 2402. D. Van Nostrand Co., 1963.
0039-9140/92 S5.00 + 0.00 Copyright 0 1992 Pcrgamon Press plc
A CHEMICAL ENHANCEMENT METHOD SPECTROtiHOTOMETRIC DETERMINATION AMOUNTS OF ARSENIC
FOR THE OF TRACE
K. PAL.ANIVELU,N.BALASIJBRAMANUN and T.V. RAMAICRKWNA* Department of Chemistry, Indian Institute of Technology, Madras-600 036, India (Received 3 May 1991. Revised 1 August 1991. Accepted 1 August 1991) !Iammary-A highly sensitive spectrophotometric method for the determination of 0.03-1.0 /Ig of arsenic is described. After extraction as AsI, into benxene, it is selectively stripped into water. Both the arsenic@) and iodide present in the aqueous phase are made to react with iodate in acidic medium in the presence of chloride to form the anionic chloro complex, ICI,. The determination is completed after extraction of ICI; species as an ion-pair with Rhodamine 6G into benzene and measuring the absorption of the extract at 535 nm. The coefficient of variation is 1.5% for 10 determinations of 0.5 pg of arsenic. The method has been applied to the detetmination of arsenic content in plant materials, high purity iron, copper base alloys and inorganic arsenic levels of natural waters.
The widely used methods for the spectrophotometric determination of trace amounts of arsenic are based on the formation of molybdenum blue’ and the reaction of arsine with silver diethyldithiocarbamate.2 Methods based on the formation of an ion-pair between arsenomolybdate and a large dye cation have been described to improve the sensitivity and selectivity of the determination.ss None of these methods, however, are sufficiently sensitive to establish the arsenic levels of natural waters without resorting to a preconcentration step. The ability of arsenic(II1) to react with potassium iodate in the presence of chloride in an acidic medium to form the anionic chlorocomplex is well known6 and has been applied to the spectrofluorimetric determination of 0.31.5 pg of arsenic(II1) after extraction of the anionic complex as an ion-pair with Rhodamine B into benzene.’ The presence of ions that are oxidized by iodate however interferes seriously in the determination. Since iodide can also react with potassium iodate to form the anionic chlorocomplex’ similar to arsenic(III), it seemed worthwhile to examine whether the formation of the anionic complex and its interaction with a cationic dye could be put to advantage for the spectrophotometric determination of both arsenic(II1) and arsenic(V) after extractive separation as arsenic triiodide into benzene. Such *Author for correspondence.
an approach, besides improving the selectivity (since relatively few ions extract into benzene from iodide media), would also provide considerable enhancement in sensitivity arising due to reaction of iodide, in addition to arsenic(II1) present in arsenic triiodide with potassium iodate to form the anionic chloro-complex. Detailed examination disclosed that the reaction with potassium iodate in acidic chloride containing medium after selectively stripping the arsenic triiodide into water and subsequent interaction of the anionic iodine complex with Rhodamine 6G can form the basis for the spectrophotometric determination of arsenic. The resulting ion-pair, after extraction into benzene, had maximum absorption at 535 nm, and arsenic concentration as low as 0.03 j.4gin a final volume of 25 ml can be determined with this procedure. The method was found suitable for the rapid determination of arsenic concentration of natural waters and applicable to the determination of arsenic in plant materials, high purity iron and copper base alloys without prior separation of arsine.
EXPERIMENTAL Apparatus
Absorbances were measured with a Carl Zeiss PMQ II spectrophotometer with IO-mm quartz cells. 555
556
K. PALANIVELUet al.
Reagents Standard arsenic(III) solution, I.Omglml. Prepared by dissolving 0.1321 g of arsenious oxide in 10 ml of 1M sodium hydroxide, adding 7.5 ml of 1M sulphuric acid and diluting to volume with water in a loo-ml standard flask. A suitable volume of this solution was diluted to obtain working standards. The following solutions were prepared by dissolving appropriate amounts of the reagents in distilled water. Potassium iodate solution, 0.01%. Rhodamine 6G solution, 0.02%. Sodium chloride solution, 15%. Saturated sodium chloride solution, N 5NL Potassium iodide solution, 2M. Sodium hypophosphite, monohydrate solution, 2M. Sulphuric acid, 2.5 and 12.5M. Benzene (thiophene free’). Used for extrac-
tion.
a 25-ml standard flask. Add 2 ml each of 2.5M sulphuric acid, 0.01% potassium iodate solution and 0.02% Rhodamine 6G solution followed by 4 ml of 15% sodium chloride solution. Dilute to volume with water and allow to stand for 5 min. Transfer the made up solution into a 60-ml separating funnel and extract with 5 ml of benzene for one minute. Separate the organic layer into a dry test tube and add about 1 g of anhydrous sodium sulphate. Measure the absorbance of the extract at 535 nm in lo-mm cells against a reagent blank run through the entire procedure. Establish the concentration of arsenic by reference to a calibration graph prepared with 0.03-l .O pg of arsenic, following the above procedure. RESULTS
AND DISCUSSION
Extractive separation of arsenic as arsenic triiodide
Separation of arsenic. Transfer a portion of the sample solution into a 125-ml separating funnel. Add enough sulphuric acid and sodium chloride to make their concentration 3M (4M for sample volumes greater than 30 ml) and 2M respectively. Add 1 ml of 2M sodium hypophosphite solution followed by sufficient potassium iodide to provide an overall concentration of 0.3M. Shake the solution for 2 min with 10 ml of benzene (separate the aqueous layer and repeat the extraction with 10 ml of benzene if the initial sample volume used is >30 ml). Collect the organic phase(s) in a 60-ml separating funnel, treat with 2 ml of di-isopropyl ether solvent (4 ml if the extraction is performed twice) containing 0.015% peroxide, mix well and allow to stand for 15 min. Then extract with 10 ml of water for one minute and discard the organic layer.
Although arsenic can be extracted from an iodide medium into a variety of solvents, the extraction into hydrocarbon solvents as arsenic triiodide from a highly acidic medium (> 8N) has been shown to be highly selective.‘0-‘3 It was found that from solutions of lower acidity, extraction becomes quantitative in the presence of sodium chloride as a salting-out agent. Using simplex optimization,‘4 it was established that for quantitative extraction the overall concentration with respect to sulphuric acid, potassium iodide and sodium chloride should be 3,0.3 and 2M respectively. A single equilibration for two minutes with 10 ml of benzene was adequate for quantitative extraction which remained unaffected up to an aqueous phase volume of 30 ml. Beyond this volume, the extraction was incomplete and repetition of the extraction on the aqueous phase was not effective. Increasing the concentration of potassium iodide up to 0.6M showed no marked improvement in the recovery of arsenic. For quantitative extraction of arsenic from aqueous volume up to 100 ml, it was found necessary to equilibrate the solution twice with 10 ml of benzene each time, after raising the acidity to >4M with respect to sulphuric acid. A single equilibration for one minute with 10 ml of water was found adequate for stripping the arsenic from the organic layer.
Determination
Determination with Rhodamine 6C
Transfer an aliquot of the stripping agent containing not more than 1.0 pg of arsenic into
Preliminary studies under conditions found optimal (hydrochloric acid-l .5M, IO;-
Di-isopropyl ether solvent. Di-isopropyl
ether containing 0.015 weight per cent peroxide was prepared by equilibrating 50 ml of the solvent rendered free from peroxide9 with 50 ml of 0.5M sulphuric acid containing 0.3 weight per cent of hydrogen peroxide for 5 min. It was prepared fresh every week, stored in a brown bottle and kept in a refrigerator. Procedure
Chemically enhanced spectrophotometric
0.8
0.1
$
8 g
557
determination of trace As
f 0.4 v) 2
0.
u)
2 I
0
lR
_d-.+-.-.--.--.I
I
I
I
1
0.1
0.2
0.3
0.4
0.5
Concentration
of sulphuric acid M
0.025mM and Rhodamine B-0.025mM) for the spectrofluorimetric determination of arsenic’ showed that the colour when extracted into benzene gradually faded on standing. This was overcome by using Rhodamine 6G instead of Rhodamine B, the use of which stabilized the colour of the extract that absorbed maximally at 535 nm for 75 min. Acidification with sulphuric acid in the presence of sodium chloride was preferred to hydrochloric acid because in the former the blank was low and reproducible, due to lower levels of impurities. Variation of the acidity of the reaction medium as shown in Fig. 1, revealed that the ion-pair extraction remained maximal and constant when the overall acidity was greater than 0.15M for both arsenic(II1) and iodide. The
0.6
[
2J 0.3
1
3
8
influence of potassium iodate, Rhodamine 6G and sodium chloride on the extraction of the ion-pair into benzene are shown in Figs. 2,3 and 4, respectively. Based on these experiments an overall concentration of 0.2M sulphuric acid, 0.04mM (0.0008%) potassium iodate, 0.035mM (0.0016%) Rhodamine 6G and 0.04M (2.4%) sodium chloride were chosen as optimum for the determination of both arsenic(II1) and iodide. Although the presence of sodium chloride and Rhodamine 6G in excess of the recommended concentration caused a slight increase in blank value, the net absorbance due to reaction with arsenic(II1) and iodide remained unaffected. The formation of the ion-pair in both instances was found to be almost instantaneous and mixing the phases for about 30 set was found to be sufficient for its quantitative extraction into benzene. Benzene, toluene,
r--t .-.-.-../
_._._._.-
2
6
Fig. 3. Effect of Rhodamine 6G concentration. A. 5 pg of As(II1) with “x” ml of 0.01% Bhodamine 6G, 2 ml each of 2.5M H,SO, and 0.01% KIO,, 5 ml of 15% NaCl, total volume 25 ml, extracted with 5 ml of benzene, measured against respective reagent blank. B. As in (A) with 5 )~g of I- instead of As(II1).
r--f /.-._.
4
2
ml of 0.01% Rhodamine 6G
Fig. 1. Effecf o/acidity. A. 2 ml of 0.01% KIO,, 5 ml each of 0.01% Bhodamine 6G and 15% NaCl, “x” ml of 2SM H2SOdr diluted to 25 ml, extracted with 5 ml of benzene measured against benzene at 535 nm in IO-mm cells. B. As in (A) with 5 pg of As(II1). C. As in (B) but measured against (A). D. As in (A) with 5 pg of I-. E. As in (D) but measured against (A).
p:
0
4
5
ml of 0.01% potassium iodate Fig. 2. EffGct of potassium iodate concentration. A. 5 pg as As(II1) with “xl’ ml of 0.01% KIO,, 2 ml of 2.5M Hr SO,, 5 ml each of 0.01% Bhodamine 6G and 15% NaCl solution, total volume 25 ml, extracted with 5 ml of benzene measured against respective reagent blank. B. As in (A) with 5 pg of I- instead of As(II1).
I
I
I
I
2
4
8
8
ml of 15% NaCl Fig. 4. As(II1) H,SO,, volume against
Effect of sodium chloride concentration. A. 5 pg of with “x” ml of 15% NaCl, 2 ml each of 2.5M 0.01% KIO, and 0.02% of Bhodamine 6G, total 25 ml, extracted with 5 ml of benzene, measured respective reagent blank. B. As in (A) with 5 pg of I- instead of As(II1).
K. PALANNELU et al.
558
xylene, hexane, cyclohexane, carban tetrachloride, chloroform, and isoamyl acetate were investigated as extraction solvents. However, only benzene proved to be satisfactory, as the extraction of the ion-pair was found to be maximum in this solvent. When independently checked, linear calibration curves passing through the origin were obtained for arsenic(II1) and iodide in the range 0.15-l 1.O pg and 0.1-6.0 pg, respectively. The molar absorptivities were found to be 2.6 x lo4 1.mole-’ .cm-’ for arsenic(II1) and 7.6 x lo4 1.mole-’ . cm-’ for iodide. Extractive arsenic
separation
and
determination
of
As the optimal conditions for the determination of arsenic(II1) and iodide were found to be identical, it was decided to apply the method for the determination of arsenic, after its extraction as arsenic triiodide into benzene and stripping into water. Difficulties were encountered because the blank was intensely coloured, evidently due to co-extraction of hydroiodic acid. As washing the benzene layer with acid solution was found to be ineffective in removing the hydroiodic acid, the use of a dilute solution of hydrogen peroxide was considered for washing purposes. The use of a 0.3 weight per cent solution of hydrogen peroxide in 5iU sulphuric acid was effective in minimizing the blank value, but the results were not reproducible. Blank values comparable to the normal blank resulted only when the hydrogen peroxide concentration in the wash liquid was increased above l.O%, but this resulted in a considerable decrease in the sample absorbance also. In this instance the absorbance of the sample extract corresponded to that of arsenic alone, indicating the oxidative loss of the iodide that remained associated with arsenic. Addition of organic solvents containing peroxide to the benzene layer after the separation of the aqueous layer was examined next. The solvents examined included, isoamyl acetate, cyclohexanone, di-isopropyl ether, isobutyl methyl ketone and mesityl oxide. After adding 1 ml of the solvent, the benzene layer was mixed and allowed to stand for 15 min before stripping with water. In all instances, there was a considerable reduction in the blank. As the addition of di-isopropyl ether containing 0.022 weight per cent peroxide was found to be most effective among the solvents examined it was used, with an optimal amount of peroxide,
for destroying the co-extracted hydroiodic acid. In order to establish the optimal range of peroxide in di-isopropyl ether that would be effective for produding low and reproducible blanks, the solvent was first rendered free from peroxide9 and then varying amounts of peroxide were introduced by equilibrating for five min with an equal volume of 0.5M sulphuric acid containing 0.1-l .O weight per cent hydrogen peroxide. After equilibration the solvent was found to contain peroxide in the range 0.005-0.035 weight per cent. The effect of the addition of 2 ml of each peroxide containing ether to the benzene extract was examined separately. After the addition of ether, the extracts were allowed to stand for 15 min before stripping into water for completing the determination. The results, shown in Fig. 5, clearly indicate that the addition of 2 ml of ether containing peroxide in the range 0.012-0.026 weight per cent effectively reduces the blank value without affecting the sample. When the recommended procedure was followed for the preparation of peroxide containing ether, it was found to contain 0.015 weight per cent. After the addition of 2 ml of this solvent to the benzene extract, a standing time of 10 min was found to be adequate (Fig. 6) to destroy the co-extracted hydroiodic acid, and the recovery
0.6 -
0: 0.4{ 3 0.2 -
I 0
0.01
0.02
0.03
0.04
Weight % of peroxide Fig. 5. Effect of variation ofperoxide content of di-iso-propyl ether. A. 10 ml of benzene extract, 2 ml of di-iso-propyl ether solvent, standing time 15 min, before stripping and colour development. Measured against benzene. B. As in (A) using benzene extracts obtained with 1 pg of arsenic, measured against (A).
Chemically enhanced spectrophotometric determination of trace As
559
5Asq-+210,+2H+=12+5As0:-+H20
(1)
51- + 10, + 6H+=312 + 3H2G
(2)
In the presence of chloride, the excess iodate” present in the solution can further oxidize the generated iodine and stabilize it as ICI; species in accordance with 212+ 10; + 6H+ + 5Cl-+5ICl+
3H2G (3)
ICI + Cl-*ICI, 0
10
20
30 Time,
40
50
60
min
Fig. 6. Influence of standing time after the addition of di-iso-propyl ether solvent on the stabiliiy of Asl, in the benzene extract. A. 10 ml of benzene extract, 2 ml of di-iso-propyl ether solvent (0.015%). standing time x min before stripping and colour development. Measured against benzene. B. As in (A) with benzene extracts obtained with 1 pg of arsenic, measured against (A).
of arsenic remained unaffected up to a standing time of 25 min before stripping into water. When the procedure described in the Experimental section was followed, 1.0 c(g of arsenic(III) gave an absorbance of 0.62 f 0.02 compared to the expected value of 0.66 [0.07 for 1 pg of arsenic@) and 0.59 for 5.0 pg of associated iodide]. The calibration graph was linear over the range 0.03-1.0 pg of arsenic. From the slope of the calibration plot the molar absorptivity was established to be 2.37 x 10’ 1.mole-‘. cm-‘. The Sandell sensitivity was found to be 8 x lo-’ pg/cm2. The precision of the proposed procedure was checked by establishing the concentration of 10 samples containing 0.50 pg of arsenic. The mean recovery was found to be 0.497 pg with a coefficient of variation of 1.5%. Nature of the extracted species The equilibrium shift method” was employed to establish the ratio of arsenic to iodide. The slope of the straight line obtained by plotting log D us. log [I-] was found to be 2.98, indicating that arsenic was extracted into the benzene layer as arsenic(II1) triiodide, which conforms to the findings reported in the literature.13 When the benzene layer was equilibrated with water, both arsenic(II1) and the associated iodide were readily stripped into water. The stripping, when reacted with iodate in acidic medium, leads to the formation of iodine as shown in the following equations: TAL 39,5-H
(4)
The overall reaction for the formation of ICI, from AsO:- and iodide can be represented as 2AsO;- + IO; + 2H+ + 2Cl=2AsO:-
+ ICI; + H,O
(5)
21- + IO, + 6H+ + 6Cl+3ICl,
+ 3H2G
(6)
While the reaction with iodide can give rise to only ICI;, the possibility of the formation of ICI;, due to the reduction of the iodate to 13+ state and stabilization as ICI; can also be expected in the case of arsenic(II1) in accordance with AsO:- + IO, + 4H+ + 4Cl*AsO:-
+ ICl,- + 2H,O
(7)
As both ICI; and ICI; can form ion-pairs with Rhodamine 6G and extract into benzene, investigations were carried out to establish the nature of species responsible for the extraction of the ion-pair in the determination of arsenic(II1) with iodine solution in benzene and completing the determination under conditions optimal for the determination of arsenic(II1). Though this experiment would not give direct evidence for the formation of ICI;, it would provide indirect evidence of whether the formation of ICI, was responsible for the observed colour. Experiments with iodine solution in benzene showed a linear relationship with an increasing amount of iodine and the absorbance of 0.20 produced by 2.0 pg of iodine was identical to that produced by 2.94 pg of arsenic(II1). Since, 2.0 pg of iodine would result from 2.94 pg of arsenic(II1) [equation (l)], it was felt reasonable to conclude that the ICI; species was responsible for the observed absorbance with arsenic(II1). Had ICl; been the species, as evident from equation (7), 1.47 pg of arsenic(II1) would have been adequate to produce the observed absorbance.
K. PALANIVELU et al.
560
Table 1. Determination of arsenic in various materials
S. No.
Sample
1
Brass (0.5 g/50 ml) BCS No. 20712 Gun metal (Cuba& (0.1 g/50 ml) BCS No. 26013 High purity iron (0.05 g/25 ml) Plant materials (1 g/25 ml) a. Grass
2 3 4
Arsenic present 0.0057%*
Arsenic added, IJg 2.50
0.006% 2.50 0.026%
10.0
c. Plantain leaf Water (50 ml) a. Tap water b. Sea water (Bay of Bengal, Madras)
64.5 f 0.87 88.7
0.64
12.0 + 0.10 22.2
0.024
2.6 f 0.10 7.5 1.8 f 0.05 6.8 4.0 f 0.17 8.9
b. Mango leaf
5
Arsenic found,t (Mean value + a), % LJg 27.0 + 0.57 0.0054 51.5
2.3 pg/l*
-
2.8 /lg/1*
0.5 0.5
Recovery, % 98.0 96.8 102.0
98.0 100.0 98.0
0.125 k 0.005 (2.; :;$l) 99.0 0.145 i 0.003 (2.9 pg/l) 0.65
101.0
*Molybdenum blue method. tAverage value of three determinations.
Another evidence in support of the formation of ICI; species in the case of arsenic(II1) was obtained by establishing the stoichiometry of the species extracted as an ion-pair into benzene. The combined ratio of the anionic chloro-complex and the dye cation was established by determining the ratio of arsenic(II1) to Rhodamine 6G by the continuous variation and mole ratio methods. Both methods showed that for maximum extraction of the ion-pair, the medium should contain 2 moles of arsenic(II1) and 1 mole of Rhodamine 6G. Since 2 moles of arsenic(II1) would yield 1 mole of ICl; species [equation (5)] and 2 moles of ICI; species [equation (7)], it was concluded that ICI; species and the monovalent Rhodamine 6G cation interact in a 1: 1 ratio for the formation of the ion-pair. If ICI; were to be the species, the arsenic(II1) and Rhodamine 6G would have reacted in a 1: 1 ratio. EJJct
of interfering metal ions
The presence of one milligram amounts of various ions which are known to react with iodide were examined in the determination of 0.5 fig of arsenic by the proposed procedure. The ions examined included Ag+, BP, Cd*+, Cu*+, Fe3+, Ge4+, In3+, MOO:-, Tl+, TeOi-, Sb3+, SeO:-, Sn4+ and Zn*+. The determinations were carried out in the presence of
sodium hypophosphite solution to prevent the liberation of iodine. Although 1 ml of 2M sodium hypophosphite was adequate, the addition of larger volumes would be necessary in the presence of excessive amounts of iron(II1) and copper(I1). The excess hypophosphite had no effect on the recovery of arsenic. Only antimony(II1) at all concentration levels and germanium(IV) in excess of 100 pug were found to cause positive interference in the determination of arsenic. The influence of antimony(II1) could be overcome by extracting off as antimony(II1) triiodide into a 1: 1 mixture of cyclohexanone-isobutyl methyl ketone from 0.5M sulphuric acid and 0.25M potassium iodide. The inclusion of this step also eliminated the difficulties arising due to dispersion of copper(I) iodide in the organic layer when copper(I1) in excess of 1 mg was present with arsenic. Application of the method to real samples
Table 1 presents the results of the analysis of several arsenic containing samples with and without added standard by the proposed procedure. The water samples were filtered immediately after collection and acidified with sulphuric acid to -pH 1. The solid samples were brought into solution with 1: 1 sulphuric acid and hydrogen peroxide.‘6*‘7After evaporating almost to dryness to decompose excess
Chemically enhanced spectrophotometric determination of trace As
peroxide, the residue was dissolved in 10 ml of water, filtered if necessary, and made up to known volume with water. Suitable aliquots were subjected to determination following the procedure given under ‘Experimental’. As no standard samples were available, the arsenic content of brass,” plant materials” and waterI was also established following the standard molybdenum blue method after separation of arsenic. The results summarized in Table 1 clearly show that the method developed works satisfactorily for the analysis of traces of arsenic in various materials. CONCLUSION
The method described provides a simple and reliable means of determining trace amounts of arsenic by spectrophotometry. It is more sensitive (6 = 2.37 x 10’ l.mole-‘.cm-‘) than those based on the formation of Molybdenum Blue (6 = 2.5 x lo4 l.mole-’ .cm-‘) and reacwith silver diethyldithiocarbamate tion (C = 1.4 x lo4 l.molee’.cm-‘). It is superior to the fluoresence method based on the use of Rhodamine B in that the colour system is quite stable and can find use for the determination of both arsenic(II1) and arsenic(V) in the range 0.03-l .O pg. The method also has the advantage of virtual freedom from interferences from extraneous ions and can therefore find use for the rapid and precise determination of trace amounts of arsenic in natural and synthetic samples. Acknowledgement-One of us (KP) is grateful to CSIR, New Delhi for the award of a Research fellowship.
561
REFERENCES
1. K. Sugawara, M. Tanaka and S. Manamori, Bull. Chem. Sot. Japan, 1956, 29, 670. 2. G. Stratton and H. C. Whitehead, J. Am. War. Wks. Ass., 1962, 84, 172. 3. C. Matsubara, Y. Yamamoto and K. Takamura, Analyst, 1987, 112, 1257. 4. T. Nasu and K. Kan, ibid, 1988, 113, 1683. 5. T. V. Ramakrishna and K. Palanivelu, Indian J. Environ. Protection, 1989, 9, 265. 6. Vogel’s Quantitative Inorganic Analysis Including Elementary Instrumental Analysis (revised by J. Bassett, R. C. Denney, G. H. Jefferey and J. Mendnan), 4th Ed., p. 386. ELBS, Longman, London, 1978. 7. D. Yamamoto and K. I. Kisu, Japan Analyst, 1974,X3, 638; Anal. Abstr., 1975, 29, lB134. 8. Volgel’s Text book of Practical Organic Chemistry including Qualitative Organic Analysis (B. S. Furniss, A. J. Hannford, V. Rogers, P. W. G. Smith and A. R. Tatchell), 4th Ed., p. 266. ELBS and Longman, London, 1978. 9. Ibid., B. S. Fur&s, A. J. Hannford, V. Rogers, P. W. G. Smith and A. R. Tatchell), 4th Ed., p. 272. ELBS and Longman, London, 1978. 10. A. R. Byrne and D. Gorenc, Anal. Chim. Acta, 1972,59, 81. 11. E. M. Donaldson and M. Want, Taianta, 1986, 33, 35. 12. K. Tanaka, Japan Analyst, 1960, 9, 700, Anal. Abstr., 1961, 9, 3714. 13. N. Suzuki, K. Satoh, H. Shoji and H. Imura, Anal. Chim. Acta, 1986, 185, 239. 14. D. E. Long, ibid, 1969, 46, 193. 15. S. Motomizu and K. Toei, ibid., 1977, 89, 167. 16. Report of the analytical methorls committee on oxidation of organic matter, Analyst, 1967, 92, 403. 17. F. D. Snell, Photometric and Fluorimetric Metho& of Analysis of Metals, Part I, p. 354. Wiley-Interscience, New York, 1978. 18. Idem, ibid., Part I, p. 343. Wiley-Interscience, New York, 1978. 19. F. J. Welcher, Standard Method of Chemical Analysis, Vol. 2, Part B, 6th Ed., p. 2402. D. Van Nostrand Co., 1963.