Anal. Chem. 1997, 69, 4275-4277
An Ion Chromatographic Method for Measuring 40 min) because of the long elution times of common anions such as sulfate ion (SO42-). Although this approach is effective at separating BrO3- from Cl-, it is very time consuming and is used only if other options are not available. The use of chlorpromazine (CLP), a selective reagent that does not react with Cl-, has been reported9 for measuring low-level BrO3- concentrations in a continuous-flow analysis system. Chlorpromazine is a phenothiazine derivative that is readily oxidized10,11 by BrO3- in acid, resulting in the formation of an oxidized species that can be measured spectrophotometrically (530 nm). Because CLP can also react with chlorite ion (ClO2-) and nitrite ion (NO2-), the automated continuous-flow method was not applicable for the direct measurement of BrO3- in drinking water. However, the CLP methodology is an ideal postcolumn detection chemistry following IC separation.12 (9) Gordon, G.; Bubnis, B.; Sweetin, D.; Kuo, C. Ozone Sci. Eng. 1994, 16, 1, 79. (10) Kojlo, A.; Puzanowska-Tarasiewicz, H.; Calatoyud, J. M. Anal. Lett. 1993, 26, 593. (11) Pelizzetti, E.; Mentasti, E. Inorg. Chem. 1979, 18, 583. (12) Gordon, G. Bromates and Water Treatment International Workshop, Paris, France, 22-23 November 1993.
Analytical Chemistry, Vol. 69, No. 20, October 15, 1997 4275
Figure 1. Postcolumn detection schematic.
Under typical IC conditions with CLP postcolumn detection, less than 5 µg/L BrO3- is easily measured. Because CLP does not react with Cl-, IC conditions can be established to measure low-level BrO3- concentrations with analysis cycle times under 15 min. EXPERIMENTAL SECTION Reagents. All solutions and standards were prepared using reagent grade chemicals dissolved in deionized, distilled water (DDW) with no chemical demand for BrO3-. All BrO3- standards were prepared from ACS grade sodium bromate (Fisher Scientific) without further purification. All standard solutions were made daily or refrigerated at 4 °C for a maximum of 14 days. The boric acid eluent used for IC separation was prepared by dissolving 1.60 g of NaOH and 12.3 g of H3BO3 in 2 L of DDW. The CLP postcolumn reagent was prepared by dissolving 0.20 g of ACS reagent grade chlorpromazine (Aldrich) in 1 L of DDW. A 2 M HCl solution was made by diluting 33.1 mL of concentrated reagent grade HCl in 200 mL of DDW. Both postcolumn reagent solutions were degassed with nitrogen prior to use. Instrumentation. A Dionex Model Dx-100 ion chromatograph equipped with a metal-free column (MFC), an AG-9 guard column, and an AS-9SC analytical column was used for all IC measurements. The MFC and guard columns are used to protect the analytical column from degrading due to the presence of transition metal ions and unknown organics that might be present in samples. An anion micromembrane suppressor (AMMS) and a conductivity detector were used for measurements made using USEPA method 300. Peak area calculations were made using a Dionex Model 4400 integrator. The eluent flow rate was 1.0 mL/ min. The suppressor regenerant was 25 mN sulfuric acid flowing at 10 mL/min. The postcolumn measurement system is shown in Figure 1. Spectrophotometric measurements were made using a Rainin Dynamax UV-C detector equipped with a 15 mm flow cell. The CLP postcolumn pumping system was a Tecator 5020 analyzer. The mixing manifolds were assembled using a Tecator Chemifold SR type III manifold. An acrylic double-cone debubbling device (Alpkem) was used to remove any dissolved gases in the postcolumn flow stream. Method Detection Limit Calculations. A long established procedure for determining the method detection limit (MDL) is 4276
Analytical Chemistry, Vol. 69, No. 20, October 15, 1997
Table 1. Low-Level BrO3- Method Comparison method
spike (µg/L)
recovery (µg/L)
MDL (µg/L)
USEPA method 300 CLP postcolumn
4.94 3.04
4.68 3.02
0.50 0.49
to calculate the concentration that is equal to 3 times the ratio of the detector signal to background noise. In recent years, the USEPA has required a more rigorous MDL calculation for methods to be used for measuring components in drinking water. The analysis of variance (ANOVA) procedure13 requires a minimum of seven measurements be made on a sample whose concentration is at or near the detection limit. Each sample is processed through the entire analytical method. The standard deviation and variance are calculated for the data set and used to calculate the MDL at the 99% confidence level. RESULTS AND DISCUSSION Ion chromatography (USEPA method 300) with suppressed conductivity detection is a generally reliable method for measuring >50 µg/L BrO3-. However, it does have practical limitations with regard to Cl- interference problems and the long cycle times accompanying eluent dilution. The CLP postcolumn method overcomes the Cl- problem and greatly reduces analysis cycle times. The postcolumn method also retains the advantage of separating BrO3- chromatographically to eliminate potential interferences prior to detection. Ion Chromatography, Method 300. USEPA method 300 has been demonstrated by a small number of laboratories to readily detect
