- Email: [email protected]
Indirect determination of tetrahydroborate (BH−4) by gas-diffusion flow injection analysis with amperometric detection
0039-9140/93$6.00+ 0.00
Talanta, Vol. 40, No. 8, pp. 12831287, 1993 rights reserved
Copyright 0
Printed in Great Britain. All
1993 Pergamon Press Ltd
INDIRECT DETERMINATION OF TETRAHYDROBORATE (BH,) BY GAS-DIFFUSION FLOW INJECTION ANALYSIS WITH AMPEROMETRIC DETECTION SNE~ANA D. NIKOLIC and EMIL B. MILOSAVLJEVIC* Faculty of Chemistry, University of Belgrade, P.O. Box 550, 11001 Belgrade, Yugoslavia JAMFS L. HENDRIX and JOHN H. NELSON Departments of Chemistry and Chemical and Metallurgical Engineering, Mackay School of Mines, University of Nevada, Reno, NV 89557, U.S.A. (Received
26 December
1991. Revised 22 April 1992. Accepted
22 April 1992)
Summary-A rapid, indirect gas-diffusion flow injection analysis (FIA) method with amperometric detection has been developed for the selective and sensitive determination of tetrahydroborate (BH; ). The injected analyte reduces arsenic(III) to arsine. The amine formed diffuses through the PTFE (polytetrafluoroethylene) membrane and is quantified amperometrically at a platinum working electrode. The precision of the technique was better than a relative standard deviation of 2.1% at 60 ~44 levels and better than 0.5% at 0.1 mM, with a throughput of 60 samples/hr. The detection limit of the method was found to be 1 +W (1.5 ng BHi) with a linear range up to 1 mM. The dynamic range extends over five orders of magnitude in BH; concentration. The effects of working potential, concentration of As(II1) and HCI in the reagent stream, type and flow rate of the acceptor solution, temperature and interferents on the FIA signals were studied.
Tetrahydroborate (borohydride) salts have been extensively utilized in inorganic and organic syntheses. ‘*’ As was pointed out recently,3 for carrying out these syntheses and in studies of the methods for the synthesis of borohydride, rapid, selective and accurate analytical methods to control its concentration in solution are needed. To the best of our knowledge, there are no FIA methods reported for tetrahydroborate analysis, even though FIA methods are well suited for studying and/or continuously monitoring processes of importance to the chemical and metallurgical industries.4-7 On the other hand, there are only a few FIA publicationP3 that combine separation based on membrane diffusion with amperometric detection, which is surprising considering that the selectivity of the diffusion process and inherent sensitivity of the amperometric detection makes this combination a powerful analytical tool. Extensive discussions of the merits of gas-diffusion FIA methods may be found elsewhere.‘4’6
*Author for correspondence. *AL
m/s-J
The present paper describes a novel approach to the use of gas-diffusion in combination with amperometric detection for the indirect determination of BH;. In the FIA manifold developed, the injected analyte reduces on-line arsenic(II1) to arsine. The arsine formed diffuses from the donor stream through the hydrophobic PTFE membrane into the acidic acceptor solution. The latter carries arsine to the flowthrough amperometric detector, where it is oxidized at a platinum working electrode. The anodic current measured is proportional to the concentration of BH; in the sample or standard injected. It is interesting to note that, to the best of our knowledge, this is the first time FIA amperometry is utilized in an analytical procedure based on the hydride generation principle. This opens up the possibility of utilizing similar FIA schemes for determining the elements which form volatile hydrides, as well as mercury. EXPERIMENTAL
Reagents
All chemicals were of analytical-reagent grade. The aqueous reagent and standard solutions 1283
S. D. NIKOLIC et al.
1284
were stored in glass containers. Demineralized water was used throughout. A stock solution of O.OlM BH; was made from sodium borohydride (Fisher Scientific, Fair Lawn, NJ, U.S.A.) and checked using the published procedure.” Standard BH; solutions, which were stabilized with 1% (w/v) sodium hydroxide, were prepared by diluting aliquots of the stock solution to the appropriate volume. The As3+ solutions were made by dissolving reagent-grade sodium arsenite (Carlo Erba, Milan, Italy) in reagent-grade hydrochloric acid and diluting to the appropriate volume. Apparatus
The FIA manifold is illustrated in Fig. 1. Two peristaltic pumps were used. One was a Model Mini S-840 (Ismatec, Zurich, Switzerland) and the other was a Model MS-4-REGLO 100 (Ismatec). The injection valve was a Model 5020 (Rheodyne, Cotati, CA, U.S.A.) equipped with a loo-p1 sample loop. The gas-diffusion unit, which was obtained from Shenyang FilmProjector Reflector Factory (Shenyang, China), is similar in construction to the Tecator (Hog&as, Sweden) Chemifold V gas diffusion cell. The PTFE membrane used was supplied with the unit. All connections were made with 0.5 mm i.d. tubing. A thin layer flow-through amperometric cell was part of a LC-17A package (BAS, West Lafayette, IN, U.S.A.) and was equipped with a Model MF-1012 dual platinum working electrode (BAS) and a Model MW-2021 Ag/AgCl reference electrode (BAS). A 0.13 mm thick MF-1047 PTFE gasket (BAS) was used to separate the working electrode from the cell
P A
2.0 I
I
I
Fig. 1. FIA Manifold used for indirect determination of BH;: C, carrier; R, reagent (25 m&f As’+ in l.OM HCI); A, acceptor solution (O.OlM H,SO,); P, peristaltic pump; I, injection valve; MC, mixing coil (30 cm x 0.5 mm i.d.); D, diffusion cell; T, constant temperature bath; FC, amperometric flow-through cell; PO, potentiostat; RI?, recorder; W, waste. Flow rates are given in ml/min.
body. The working electrode was polished daily with CF-1050 polishing alumina (BAS). The potential was applied to the flow-through amperometric cell and currents were measured with a Model MA 5450 polarograph (Iskra, Kranj, Yugoslavia); the resulting FIA signals were recorded on a Servograph Model 61 stripchart recorder (Radiometer, Copenhagen, Denmark) equipped with a REA 110 unit. The measurements were made with both donor and acceptor streams flowing continuously and concurrently. Temperature regulation was achieved with a constant temperature bath, type VEB MLW (Priifgerate, Medingen, Germany). RESULTS AND DISCUSSION
The indirect gas-diffusion flow injection amperometric determination of BH; was performed with the manifold illustrated in Fig. 1. The alkaline (1% w/v sodium hydroxide) BH; standard or sample, after injection (I), is washed by the water carrier (C) to a mixing point with a reagent (R) (0.025M As3+ in 1M hydrochloric acid). The mixing coil (MC), positioned downstream ensures thorough reduction of As3+ to arsine by the injected analyte, according to the following reaction: H,AsO, + NaBH, + HCl + ASH, + B(OH), + NaCl + H, . The arsine formed on-line in the FIA manifold diffuses from the donor stream through the hydrophobic PTFE membrane into the acidic (O.OlM sulphuric acid) acceptor solution. The latter carries arsine to the flow-through amperometric detector (FC), where it is oxidized at a platinum working electrode. The anodic current measured is proportional to the BH; concentration in the injected standard or sample. The main reason for choosing hydride generation of arsine us. some other volatile hydride was based on the fact that the number of publications dealing with hydride generation of arsine is larger than for any other volatile hydride, which facilitates comparisons with the present work. The effects of several parameters on the performance of the FIA system were studied. The effect of the applied potential at the working Pt electrode, shown in Fig. 2, was investigated in the range 0.4-1.0 V us. the Ag/AgCl reference electrode. Taking into account the signal-to-noise ratio achieved, the
Indirect determination of tetrahydroborate
1285
2.5 mM to 0.15M. In this series of experiments the concentration of hydrochloric acid in the A 190 reagent solution was kept constant at a l.OM a 5 190 level. It should be pointed out that no statisticA A 5 ally significant change in the blank signal was $I70 A A A 0 observed as As(II1) concentration in the donor 160 stream increased. As a compromise between reI I I I I I I 1501 l agent consumption and sensitivity, the concen0.3 0.4 0.5 0.6 0.7 0.6 0.9 1.0 1.1 tration of As3+ used for most of the subsequent Potential/V experiments was 25 mM. The concentration of Fig. 2. Hydrodynamic voltamperogram for a 100 ~1 hydrochloric acid in the reagent solution chosen injection of a 1.00 mM sodium borohydride standard. was l.OM since an increase in its concentration causes a slow but steady decrease in sensitivity. optimum potential was found to be 0.8V. A At this point in time we do not have a definitive possible reason for the current decrease at explanation for the observed phenomenon. A potentials greater than 0.9 V is the enhanced possible reason for the decrease observed could anodic formation of thin oxide films at the be that the concurrent reaction of BH; with Pt-working electrode. I8 Of the three acceptor H,O+ to form hydrogen is more pronounced at solutions tested (O.OlM sodium hydroxide, higher hydrochloric acid levels. It was suggested potassium nitrate and sulphuric acid), the best in the review process that formation of volatile results were obtained with the latter. For AsCl, and its effects on the membrane pores due example, when sulphuric acid was used as the to possible hydrolysis at the interface might play acceptor stream, the peak current for a 1.OOmM a role. However, in a recent study of utilizing BH; standard was over eight times greater than polymer-bound tetrahydroborate for arsine with sodium hydroxide as the acceptor solution. generation in a FIA system, Tesfalidet and Even though no tendency to form “onium” Irgum” found similar hydrochloric acid concen(MHZ) ions for arsenic has been established, it tration effects on the sensitivity of the system, is plausible that sulphuric acid in the acceptor even though they employed a “classical” gassolution might show a slight trapping effect liquid separator for hydride generation atomic (small ASH, and relatively large hydronium ion spectroscopy and not a microporous membrane. concentrations). Also the highest FIA signals This observation coupled with what we have were achieved with the shortest mixing coil seen probably rules out the notion that, under utilized. Hence, for most of the subsequent our particular experimental conditions, hydroexperiments, a potential of 0.80 V US.Ag/AgCl lysis reactions at the membrane surface play an reference electrode, a O.OlM sulphuric acid important role. acceptor solution and a 30 cm x 0.5 mm i.d. The effect of acceptor flow rate was mixing coil were used. investigated by injecting a 1.00 mM sodium The effect of arsenic(II1) concentration in the borohydride standard, while varying the flow reagent stream, illustrated in Fig. 3, was studied rate in the range 0.5-2.5 ml/min (Fig. 4). As by injecting the same BH; standard, while vary- may be seen, the highest peak current was ing the concentration of As3+ in the reagent from obtained when the acceptor flow rate was 2.0 ml/
0
I
I
I
I
I
20
40
60
60
100
I
I
I
120 140 160
As(lll) ConcentrationlmM Fig. 3. Variation of the peak current as a function of As’+ concentration in the reagent stream.
P
260 250 240 230
3!$
220 210
5 0
200 190 160 170 160
0
A A
A A A
I
I
I
I
I
0.5
1.0
1.5
2.0
I
2.5
3.0
Flow rate/ml/min Fig. 4. Variation of the peak current as a function of the acceptor flow rate.
S. D.
1286
NIKOLIC et al.
min. Bearing in mind that the total donor stream flow rate was also 2.0 ml/min, this finding is in agreement with the previously reported observation’6 that the optimum FIA signals were obtained when the donor and acceptor flow rates were equal. Temperature effects were studied by injecting the same sodium borohydride standard, while varying the temperature in the interval 20-50”. No increase in the sensitivity with increase in temperature was observed. This is contrary to previously described indirect gas-diffusion amperometric FIA methods.“*13 A plausible explanation for the observed phenomenon could be based on the much lower solubility of the diffusing species; arsine in comparison to chlorine” or bromine.‘3 Also, it is obvious that the reduction of As(II1) to arsine is fast on the FIA time scale (longer mixing coils decreased the peak currents), which was not the case with the permanganate oxidation of chloride” and bromide.13 Linearity studies were conducted by injecting in triplicate a total of 12 BH; standards between 8.00 PM and 1Ml mM stabilized with 1% (w/v) sodium hydroxide. The linear calibration equation for a typical calibration run was: i = (- 5.38 + 1.42) + (196.0 + 7.3) x C (i is the peak current expressed in nA and C is the millimolar concentration of BH; ) with a correlation coefficient of 0.9986 (all the statistics were calculated for a 95% confidence level). The relative standard deviations were found to be 2.01% (n = 5) at 60 PM levels and only 0.45% (n = 5) at 0.10 mM. The detection limit, calculated according to the recommended procedure,” was 1 PM which corresponds to only 1.5 ng BH; (the sample loop volume was 100 ~1). Figure 5 illustrates the dynamic range for the method obtained at two different As(II1) concentrations in the reagent 100000
10000
4
s E 5
v
1000E
1
100 10 1
BH; Concentration/mM Fig. 5. Logarithmic calibration graphs illustrating the dynamic range of the method developed; As(III) concentration in the reagent stream: 25 mM (AA& and 100 mM (000).
Table I. Solutions tested for their possible interference* Compound
Cont./M
Compound
Cont./M
NH,NO, KC1 Na,SO, KBrO, Na&Or KCN NarC,O, Na,CO, Na,S
0.1 0.1 2 x 10-e 0.01 0.001 0.04 0.1 0.1 1 x IO-5
Na,EDTAt Na-citrate CH,COONa NaF KH,PO, KBr NaNO, Kl L-cysteine
0.1 0.1 0.1 0.1 0.1 0.1 0.002 0.1 0.1
*All samples contained 1% (w/v) NaOH. The response of the amperometric detector to 100 nl injections of the solutions tested could not be distinguished from the base line. The maximum concentrations tested were O.lM. tNa,EDTA = Disodium ethylenediaminetetraacetate.
stream. As may be seen, the dynamic range (defined as that range of concentrations of the test substance, over which a change in concentration produces a change in detector signal)2’ extends over five orders of magnitude in BH; concentration. Table 1 summarizes the study of possible interferents. It has been established previously that the PTFE membranes used in the FIA gas-diffusion studies are effective barriers for ionic species. 22,23 Hence, in order for a particular species to interfere in the determination of BH; with the manifold described, it has to satisfy two conditions. It has to form a gas or a molecular species with a high vapor pressure when acidified (the donor stream of the FIA manifold contains hydrochloric acid) and the gas or molecular species formed has to be electroactive (oxidizable) at the potential applied to the platinum working electrode. As can be seen from Table 1, the only serious interferents found were the sulphite and sulphide ions. This work demonstrates that it is possible to use gas-diffusion FIA amperometry for the development of novel analytical procedures based on a hydride generation principle. authors acknowledge the financial support of the U.S. Bureau of Mines under the Mining and Mineral Resources Institute Generic Center program (Grant number G1125132-3205, Mineral Industry Waste Treatment and Recovery Generic Center) and the Serbian Republic Research Fund. Acknowledgements-The
REFERENCES 1. H. C. Brown, Organic Synthesis via Boranes, Wiley, New York, 1975. 2. A. Petler, K. Smith and H. C. Brown, Borane Reagents, Academic Press, London, 1988.
Indirect determination of tetrahydroborate 3. M. V. Mirkin and A. J. Bard, Anal. Chem., 1991, 63, 532. 4. J. R&%Eka and E. H. Hansen, Flow Injection Analysis, 2nd Ed., pp. 372-376. Wiley, New York, 1988. 5. N. P. Gallagher, J. L. Hendrix, E. B. Milosavlevic and J. H. Nelson, J. Electrochem. Sot., 1989, 136, 2546. 6. N. P. Gallagher, J. L. Hendrix, E. B. Milosavljevic, J. H. Nelson and Lj. Solujic, Hydrometallurgy, 1990, 25, 305. 7. C. C. Nesbitt, E. B. Milosavljevic and J. L. Hendrix, Ind. Eng. Chem. Res., 1990, 29, 1698. 8. M. Granados, S. Maspoch and M. Blanco, Anal. Chim. Acta, 1986, 179, 445. 9. E. B. Milosavljevic, Lj. Solujic, J. L. Hendrix and J. H. Nelson, Anal. Chem., 1988, 60, 2791. 10. D. Utley, Analyst, 1990, 115, 1239. 11. S. D. Nikolic, E. B. Milosavljevic, J. L. Hendrix and J. H. Nelson, Analyst, 1991, 116, 49. 12. M. NoviE and B. Pihlar, Anal. Chim. Acta, 1991, 251, 261. 13. S. D. Nikolic, T. D. Jankovic, E. B. Milosavljevic, J. L. Hendrix and J. H. Nelson, Fresenius J. Anal. Chem., 1992, 342, 98.
1287
14. G. E. Pacey, D. A. Hollowell, K. G. Miller, M. R. Straka and G. Gordon, Anal. Chim. Acta, 1986, 179, 259. 15. J. Rtieka and E. H. Hansen, Flow Injection Analysis, 2nd Ed., pp. 192-200. Wiley, New York, 1988. 16. B. Karlberg and G. E. Pacey, Flow Injection AnalysisA Practical Guide, pp. 111-118. Elsevier, Amsterdam, 1989. 17. J. M. Kolthoff and P. J. Elving, Treattie on Analytical Chemistry, Part II, Vol. 12, p. 213. Wiley, New York, 1965. 18. D. S. Austin, J. A. Polta, T. Z. Polta, A. P. -C. Tang, T. D. Cabelka and D. C. Johnson, J. Electroanal. Chem., 1984, 168, 227. 19. S. Tesfalidet and K. Irgum, Anal. Chem., 1989,61,2079. 20. Analytical Methods Committee, Analyst, 1987, 112, 199. 21. C. A. Dorshel, J. L. Ekmanis, J. E. Oberholtzer F. V. Warren, Jr. and B. A. Bidlingmeyer, Anal. Chem., 1989, 61, 951A. 22. W. E. Van der Linden, Anal. Chim. Acta, 1983,151,359. 23. D. A. Hollowell, G. E. Pacey and G. Gordon, Anal. Chem., 1985, 57, 2851.
Talanta, Vol. 40, No. 8, pp. 12831287, 1993 rights reserved
Copyright 0
Printed in Great Britain. All
1993 Pergamon Press Ltd
INDIRECT DETERMINATION OF TETRAHYDROBORATE (BH,) BY GAS-DIFFUSION FLOW INJECTION ANALYSIS WITH AMPEROMETRIC DETECTION SNE~ANA D. NIKOLIC and EMIL B. MILOSAVLJEVIC* Faculty of Chemistry, University of Belgrade, P.O. Box 550, 11001 Belgrade, Yugoslavia JAMFS L. HENDRIX and JOHN H. NELSON Departments of Chemistry and Chemical and Metallurgical Engineering, Mackay School of Mines, University of Nevada, Reno, NV 89557, U.S.A. (Received
26 December
1991. Revised 22 April 1992. Accepted
22 April 1992)
Summary-A rapid, indirect gas-diffusion flow injection analysis (FIA) method with amperometric detection has been developed for the selective and sensitive determination of tetrahydroborate (BH; ). The injected analyte reduces arsenic(III) to arsine. The amine formed diffuses through the PTFE (polytetrafluoroethylene) membrane and is quantified amperometrically at a platinum working electrode. The precision of the technique was better than a relative standard deviation of 2.1% at 60 ~44 levels and better than 0.5% at 0.1 mM, with a throughput of 60 samples/hr. The detection limit of the method was found to be 1 +W (1.5 ng BHi) with a linear range up to 1 mM. The dynamic range extends over five orders of magnitude in BH; concentration. The effects of working potential, concentration of As(II1) and HCI in the reagent stream, type and flow rate of the acceptor solution, temperature and interferents on the FIA signals were studied.
Tetrahydroborate (borohydride) salts have been extensively utilized in inorganic and organic syntheses. ‘*’ As was pointed out recently,3 for carrying out these syntheses and in studies of the methods for the synthesis of borohydride, rapid, selective and accurate analytical methods to control its concentration in solution are needed. To the best of our knowledge, there are no FIA methods reported for tetrahydroborate analysis, even though FIA methods are well suited for studying and/or continuously monitoring processes of importance to the chemical and metallurgical industries.4-7 On the other hand, there are only a few FIA publicationP3 that combine separation based on membrane diffusion with amperometric detection, which is surprising considering that the selectivity of the diffusion process and inherent sensitivity of the amperometric detection makes this combination a powerful analytical tool. Extensive discussions of the merits of gas-diffusion FIA methods may be found elsewhere.‘4’6
*Author for correspondence. *AL
m/s-J
The present paper describes a novel approach to the use of gas-diffusion in combination with amperometric detection for the indirect determination of BH;. In the FIA manifold developed, the injected analyte reduces on-line arsenic(II1) to arsine. The arsine formed diffuses from the donor stream through the hydrophobic PTFE membrane into the acidic acceptor solution. The latter carries arsine to the flowthrough amperometric detector, where it is oxidized at a platinum working electrode. The anodic current measured is proportional to the concentration of BH; in the sample or standard injected. It is interesting to note that, to the best of our knowledge, this is the first time FIA amperometry is utilized in an analytical procedure based on the hydride generation principle. This opens up the possibility of utilizing similar FIA schemes for determining the elements which form volatile hydrides, as well as mercury. EXPERIMENTAL
Reagents
All chemicals were of analytical-reagent grade. The aqueous reagent and standard solutions 1283
S. D. NIKOLIC et al.
1284
were stored in glass containers. Demineralized water was used throughout. A stock solution of O.OlM BH; was made from sodium borohydride (Fisher Scientific, Fair Lawn, NJ, U.S.A.) and checked using the published procedure.” Standard BH; solutions, which were stabilized with 1% (w/v) sodium hydroxide, were prepared by diluting aliquots of the stock solution to the appropriate volume. The As3+ solutions were made by dissolving reagent-grade sodium arsenite (Carlo Erba, Milan, Italy) in reagent-grade hydrochloric acid and diluting to the appropriate volume. Apparatus
The FIA manifold is illustrated in Fig. 1. Two peristaltic pumps were used. One was a Model Mini S-840 (Ismatec, Zurich, Switzerland) and the other was a Model MS-4-REGLO 100 (Ismatec). The injection valve was a Model 5020 (Rheodyne, Cotati, CA, U.S.A.) equipped with a loo-p1 sample loop. The gas-diffusion unit, which was obtained from Shenyang FilmProjector Reflector Factory (Shenyang, China), is similar in construction to the Tecator (Hog&as, Sweden) Chemifold V gas diffusion cell. The PTFE membrane used was supplied with the unit. All connections were made with 0.5 mm i.d. tubing. A thin layer flow-through amperometric cell was part of a LC-17A package (BAS, West Lafayette, IN, U.S.A.) and was equipped with a Model MF-1012 dual platinum working electrode (BAS) and a Model MW-2021 Ag/AgCl reference electrode (BAS). A 0.13 mm thick MF-1047 PTFE gasket (BAS) was used to separate the working electrode from the cell
P A
2.0 I
I
I
Fig. 1. FIA Manifold used for indirect determination of BH;: C, carrier; R, reagent (25 m&f As’+ in l.OM HCI); A, acceptor solution (O.OlM H,SO,); P, peristaltic pump; I, injection valve; MC, mixing coil (30 cm x 0.5 mm i.d.); D, diffusion cell; T, constant temperature bath; FC, amperometric flow-through cell; PO, potentiostat; RI?, recorder; W, waste. Flow rates are given in ml/min.
body. The working electrode was polished daily with CF-1050 polishing alumina (BAS). The potential was applied to the flow-through amperometric cell and currents were measured with a Model MA 5450 polarograph (Iskra, Kranj, Yugoslavia); the resulting FIA signals were recorded on a Servograph Model 61 stripchart recorder (Radiometer, Copenhagen, Denmark) equipped with a REA 110 unit. The measurements were made with both donor and acceptor streams flowing continuously and concurrently. Temperature regulation was achieved with a constant temperature bath, type VEB MLW (Priifgerate, Medingen, Germany). RESULTS AND DISCUSSION
The indirect gas-diffusion flow injection amperometric determination of BH; was performed with the manifold illustrated in Fig. 1. The alkaline (1% w/v sodium hydroxide) BH; standard or sample, after injection (I), is washed by the water carrier (C) to a mixing point with a reagent (R) (0.025M As3+ in 1M hydrochloric acid). The mixing coil (MC), positioned downstream ensures thorough reduction of As3+ to arsine by the injected analyte, according to the following reaction: H,AsO, + NaBH, + HCl + ASH, + B(OH), + NaCl + H, . The arsine formed on-line in the FIA manifold diffuses from the donor stream through the hydrophobic PTFE membrane into the acidic (O.OlM sulphuric acid) acceptor solution. The latter carries arsine to the flow-through amperometric detector (FC), where it is oxidized at a platinum working electrode. The anodic current measured is proportional to the BH; concentration in the injected standard or sample. The main reason for choosing hydride generation of arsine us. some other volatile hydride was based on the fact that the number of publications dealing with hydride generation of arsine is larger than for any other volatile hydride, which facilitates comparisons with the present work. The effects of several parameters on the performance of the FIA system were studied. The effect of the applied potential at the working Pt electrode, shown in Fig. 2, was investigated in the range 0.4-1.0 V us. the Ag/AgCl reference electrode. Taking into account the signal-to-noise ratio achieved, the
Indirect determination of tetrahydroborate
1285
2.5 mM to 0.15M. In this series of experiments the concentration of hydrochloric acid in the A 190 reagent solution was kept constant at a l.OM a 5 190 level. It should be pointed out that no statisticA A 5 ally significant change in the blank signal was $I70 A A A 0 observed as As(II1) concentration in the donor 160 stream increased. As a compromise between reI I I I I I I 1501 l agent consumption and sensitivity, the concen0.3 0.4 0.5 0.6 0.7 0.6 0.9 1.0 1.1 tration of As3+ used for most of the subsequent Potential/V experiments was 25 mM. The concentration of Fig. 2. Hydrodynamic voltamperogram for a 100 ~1 hydrochloric acid in the reagent solution chosen injection of a 1.00 mM sodium borohydride standard. was l.OM since an increase in its concentration causes a slow but steady decrease in sensitivity. optimum potential was found to be 0.8V. A At this point in time we do not have a definitive possible reason for the current decrease at explanation for the observed phenomenon. A potentials greater than 0.9 V is the enhanced possible reason for the decrease observed could anodic formation of thin oxide films at the be that the concurrent reaction of BH; with Pt-working electrode. I8 Of the three acceptor H,O+ to form hydrogen is more pronounced at solutions tested (O.OlM sodium hydroxide, higher hydrochloric acid levels. It was suggested potassium nitrate and sulphuric acid), the best in the review process that formation of volatile results were obtained with the latter. For AsCl, and its effects on the membrane pores due example, when sulphuric acid was used as the to possible hydrolysis at the interface might play acceptor stream, the peak current for a 1.OOmM a role. However, in a recent study of utilizing BH; standard was over eight times greater than polymer-bound tetrahydroborate for arsine with sodium hydroxide as the acceptor solution. generation in a FIA system, Tesfalidet and Even though no tendency to form “onium” Irgum” found similar hydrochloric acid concen(MHZ) ions for arsenic has been established, it tration effects on the sensitivity of the system, is plausible that sulphuric acid in the acceptor even though they employed a “classical” gassolution might show a slight trapping effect liquid separator for hydride generation atomic (small ASH, and relatively large hydronium ion spectroscopy and not a microporous membrane. concentrations). Also the highest FIA signals This observation coupled with what we have were achieved with the shortest mixing coil seen probably rules out the notion that, under utilized. Hence, for most of the subsequent our particular experimental conditions, hydroexperiments, a potential of 0.80 V US.Ag/AgCl lysis reactions at the membrane surface play an reference electrode, a O.OlM sulphuric acid important role. acceptor solution and a 30 cm x 0.5 mm i.d. The effect of acceptor flow rate was mixing coil were used. investigated by injecting a 1.00 mM sodium The effect of arsenic(II1) concentration in the borohydride standard, while varying the flow reagent stream, illustrated in Fig. 3, was studied rate in the range 0.5-2.5 ml/min (Fig. 4). As by injecting the same BH; standard, while vary- may be seen, the highest peak current was ing the concentration of As3+ in the reagent from obtained when the acceptor flow rate was 2.0 ml/
0
I
I
I
I
I
20
40
60
60
100
I
I
I
120 140 160
As(lll) ConcentrationlmM Fig. 3. Variation of the peak current as a function of As’+ concentration in the reagent stream.
P
260 250 240 230
3!$
220 210
5 0
200 190 160 170 160
0
A A
A A A
I
I
I
I
I
0.5
1.0
1.5
2.0
I
2.5
3.0
Flow rate/ml/min Fig. 4. Variation of the peak current as a function of the acceptor flow rate.
S. D.
1286
NIKOLIC et al.
min. Bearing in mind that the total donor stream flow rate was also 2.0 ml/min, this finding is in agreement with the previously reported observation’6 that the optimum FIA signals were obtained when the donor and acceptor flow rates were equal. Temperature effects were studied by injecting the same sodium borohydride standard, while varying the temperature in the interval 20-50”. No increase in the sensitivity with increase in temperature was observed. This is contrary to previously described indirect gas-diffusion amperometric FIA methods.“*13 A plausible explanation for the observed phenomenon could be based on the much lower solubility of the diffusing species; arsine in comparison to chlorine” or bromine.‘3 Also, it is obvious that the reduction of As(II1) to arsine is fast on the FIA time scale (longer mixing coils decreased the peak currents), which was not the case with the permanganate oxidation of chloride” and bromide.13 Linearity studies were conducted by injecting in triplicate a total of 12 BH; standards between 8.00 PM and 1Ml mM stabilized with 1% (w/v) sodium hydroxide. The linear calibration equation for a typical calibration run was: i = (- 5.38 + 1.42) + (196.0 + 7.3) x C (i is the peak current expressed in nA and C is the millimolar concentration of BH; ) with a correlation coefficient of 0.9986 (all the statistics were calculated for a 95% confidence level). The relative standard deviations were found to be 2.01% (n = 5) at 60 PM levels and only 0.45% (n = 5) at 0.10 mM. The detection limit, calculated according to the recommended procedure,” was 1 PM which corresponds to only 1.5 ng BH; (the sample loop volume was 100 ~1). Figure 5 illustrates the dynamic range for the method obtained at two different As(II1) concentrations in the reagent 100000
10000
4
s E 5
v
1000E
1
100 10 1
BH; Concentration/mM Fig. 5. Logarithmic calibration graphs illustrating the dynamic range of the method developed; As(III) concentration in the reagent stream: 25 mM (AA& and 100 mM (000).
Table I. Solutions tested for their possible interference* Compound
Cont./M
Compound
Cont./M
NH,NO, KC1 Na,SO, KBrO, Na&Or KCN NarC,O, Na,CO, Na,S
0.1 0.1 2 x 10-e 0.01 0.001 0.04 0.1 0.1 1 x IO-5
Na,EDTAt Na-citrate CH,COONa NaF KH,PO, KBr NaNO, Kl L-cysteine
0.1 0.1 0.1 0.1 0.1 0.1 0.002 0.1 0.1
*All samples contained 1% (w/v) NaOH. The response of the amperometric detector to 100 nl injections of the solutions tested could not be distinguished from the base line. The maximum concentrations tested were O.lM. tNa,EDTA = Disodium ethylenediaminetetraacetate.
stream. As may be seen, the dynamic range (defined as that range of concentrations of the test substance, over which a change in concentration produces a change in detector signal)2’ extends over five orders of magnitude in BH; concentration. Table 1 summarizes the study of possible interferents. It has been established previously that the PTFE membranes used in the FIA gas-diffusion studies are effective barriers for ionic species. 22,23 Hence, in order for a particular species to interfere in the determination of BH; with the manifold described, it has to satisfy two conditions. It has to form a gas or a molecular species with a high vapor pressure when acidified (the donor stream of the FIA manifold contains hydrochloric acid) and the gas or molecular species formed has to be electroactive (oxidizable) at the potential applied to the platinum working electrode. As can be seen from Table 1, the only serious interferents found were the sulphite and sulphide ions. This work demonstrates that it is possible to use gas-diffusion FIA amperometry for the development of novel analytical procedures based on a hydride generation principle. authors acknowledge the financial support of the U.S. Bureau of Mines under the Mining and Mineral Resources Institute Generic Center program (Grant number G1125132-3205, Mineral Industry Waste Treatment and Recovery Generic Center) and the Serbian Republic Research Fund. Acknowledgements-The
REFERENCES 1. H. C. Brown, Organic Synthesis via Boranes, Wiley, New York, 1975. 2. A. Petler, K. Smith and H. C. Brown, Borane Reagents, Academic Press, London, 1988.
Indirect determination of tetrahydroborate 3. M. V. Mirkin and A. J. Bard, Anal. Chem., 1991, 63, 532. 4. J. R&%Eka and E. H. Hansen, Flow Injection Analysis, 2nd Ed., pp. 372-376. Wiley, New York, 1988. 5. N. P. Gallagher, J. L. Hendrix, E. B. Milosavlevic and J. H. Nelson, J. Electrochem. Sot., 1989, 136, 2546. 6. N. P. Gallagher, J. L. Hendrix, E. B. Milosavljevic, J. H. Nelson and Lj. Solujic, Hydrometallurgy, 1990, 25, 305. 7. C. C. Nesbitt, E. B. Milosavljevic and J. L. Hendrix, Ind. Eng. Chem. Res., 1990, 29, 1698. 8. M. Granados, S. Maspoch and M. Blanco, Anal. Chim. Acta, 1986, 179, 445. 9. E. B. Milosavljevic, Lj. Solujic, J. L. Hendrix and J. H. Nelson, Anal. Chem., 1988, 60, 2791. 10. D. Utley, Analyst, 1990, 115, 1239. 11. S. D. Nikolic, E. B. Milosavljevic, J. L. Hendrix and J. H. Nelson, Analyst, 1991, 116, 49. 12. M. NoviE and B. Pihlar, Anal. Chim. Acta, 1991, 251, 261. 13. S. D. Nikolic, T. D. Jankovic, E. B. Milosavljevic, J. L. Hendrix and J. H. Nelson, Fresenius J. Anal. Chem., 1992, 342, 98.
1287
14. G. E. Pacey, D. A. Hollowell, K. G. Miller, M. R. Straka and G. Gordon, Anal. Chim. Acta, 1986, 179, 259. 15. J. Rtieka and E. H. Hansen, Flow Injection Analysis, 2nd Ed., pp. 192-200. Wiley, New York, 1988. 16. B. Karlberg and G. E. Pacey, Flow Injection AnalysisA Practical Guide, pp. 111-118. Elsevier, Amsterdam, 1989. 17. J. M. Kolthoff and P. J. Elving, Treattie on Analytical Chemistry, Part II, Vol. 12, p. 213. Wiley, New York, 1965. 18. D. S. Austin, J. A. Polta, T. Z. Polta, A. P. -C. Tang, T. D. Cabelka and D. C. Johnson, J. Electroanal. Chem., 1984, 168, 227. 19. S. Tesfalidet and K. Irgum, Anal. Chem., 1989,61,2079. 20. Analytical Methods Committee, Analyst, 1987, 112, 199. 21. C. A. Dorshel, J. L. Ekmanis, J. E. Oberholtzer F. V. Warren, Jr. and B. A. Bidlingmeyer, Anal. Chem., 1989, 61, 951A. 22. W. E. Van der Linden, Anal. Chim. Acta, 1983,151,359. 23. D. A. Hollowell, G. E. Pacey and G. Gordon, Anal. Chem., 1985, 57, 2851.