Using Ion Chromatography to Determine Trace Levels of Bromate in Water

In recent years, it has become increasingly vital to determine volumes of bromate in both mineral and drinking waters, as it has been determined to be a potential carcinogen. Bromate is produced by the oxidation of bromide traces during water disinfection; for instance, in ozonization.

Both the US Environmental Protection Agency (US EPA) and the European Union (EU) currently delineate the limit as a maximum bromate concentration of 10 parts per billion (ppb) in drinking water.

The regulations controlling mineral waters stipulate a limit of 3 parts per billion. The determination of bromate needs very sensitive analytical methods, in order to satisfy these regulatory limits. Anion-exchange chromatography (AEC) is the basis for the most widely used methods for the quantification of bromate. Whilst conductivity detection suffers from relatively high detection limits (0.1–20 ppb), the coupling of ion chromatography (IC) with mass spectrometry (MS) attains an outstanding 6 parts per trillion (ppt) detection limit (1).

The issue herein is that MS detection is both expensive and demanding as a technique. This has seen the investigation of several sensitive postcolumn reactions (PCRs) followed by spectrophotometric detection.  A detection limit of approximately 0.2 ppb is achieved by the post-column derivatization of bromate with o-dianisidine (ODA) per EPA method 317, but a major drawback is the use of potentially carcinogenic ODA.

The alternative EPA method 326 requires post-column reaction of bromate with the less harmful iodide under acidic conditions. The iodide is oxidized to the triiodide ion by bromate. The triiodide ion is detected at 352 nm. Without any sample preparation, other than filtration, the aqueous sample can be directly injected. This article explores the results of the optimization of the PCR with respect to temperature, eluent composition and iodide concentration.

The Triiodide Method

The analysis, as stated, is based on the EPA method 326. Bromate, in this post-column derivatization method, is aided by the catalytic effect of ammonium molybdate, and subsequently oxidizes iodide to triiodide in an acidic medium according to Equations 1–4. As per Equation 4, the bromate anion is “stoichiometrically amplified” by a factor of three.

Figure 1. Schematic illustration of the IC system.

Equation 1’s reaction rate is only high enough with molybdate(VI) catalysis and at high sulfuric acid concentrations. However, despite this, direct acidification of the KI solution increases the oxidation of iodide by oxygen and the result is the formation of interfering yellowish triiodide anions (2).

Though this problem is frequently solved by on-line acidification via a micromembrane suppressor, installed just before the mixing T and the reaction coil,(3,4) here a sulfuric acid eluent with catalytic amounts of ammonium molybdate was used (Figure 1). This means that a derivatization reagent must be added, which here was  just the potassium iodide solution. This method is ideal in the selective determination of bromate and remains unaffected by other drinking water matrix anions.

The presented set up is modified using a carbonate/hydrogen carbonate eluent and two different post-column reagents, as set out in the work of Bogenschütz et al.(5) and in Metrohm Application Note U-9 (6). This permits the concurrent spectrophotometric detection of iodate, chlorite, bromate and nitrite.

In acidic solution, these anions specifically oxidize iodide to the triiodide anion. This is in comparison to other strong oxidants such as chlorate and perchlorate, which do not react with iodide under the given conditions and which therefore are best determined by suppressed conductivity and/or MS detection.

Water matrix anions (chloride, nitrate, sulphate, etc.), oxyhalides (BrO₃^- , IO₃^- , ClO₂^- ) and nitrite can be detected in a single run, if a conductivity detector is combined with post-column reaction and subsequent UV detection.

Experimental

Post-column reaction: The Metrohm IC Post-Column Reactor, based in Metrohm AG, Herisau, Switzerland, accomplished the post-column reaction. The KI solution is transferred by a peristaltic pump with a flow-rate of 0.25 mL/min to the reaction coil of the PCR (with a volume of 0.4 mL).

Here, it is mixed with the acid effluent stream of the column. A stable reagent flow is ensured by a built-in pulsation absorber. As displayed in Figure 1, the generated triiodide is then conveyed to the ultraviolet/visible (UV/vis) detector where it is detected with a molar extinction coefficient of 26400 L/(mol•cm) at a wavelength of 352 nm.

Instrumentation: The equipment used for all experiments was the 844 Compact UV/vis ion chromatograph (Metrohm AG) using the Star Ion A300 HC column (Phenomenex Inc., Torrance, California, USA). For all of the experiments, the flow-rate of the mobile phase was 1 mL/min and the injection volume 1000 L. IC Net software (Metrohm AG) performed all instrument control, data acquisition and processing.

Standard solutions, post-column reagents and eluents: All of the reagents which were used in this experiment were of the highest purity grade (puriss p.a.). Potassium iodide, the potassium bromate standard, the sulfuric acid and ammonium molybdate were all provided by Fluka (Sigma-Aldrich, Buchs, Switzerland). Every solution was prepared with deionized water with a specific resistance higher than 18 MV cm.

Results and Discussion

This article analyses the effect had by molybdate, sulfuric acid concentration of the eluent, potassium iodide concentration and temperature on the performance of the triiodide method. Whilst one of these parameters was varied, a recording was taken of the detector response of a 1000 L direct injection of a 10 ppb bromate standard. A tap water sample from Herisau (Switzerland) was analysed for bromate, by applying the optimized conditions.

Influence of temperature

The variation of the PCR's temperature only mildly affects the “bromate peak” (Figure 2), for the adjusted potassium iodide flow-rate of 0.25 mL/min. As reported by Salhi and von Gunten (3), the post-column derivatization can thus be undertaken at a reaction coil temperature of 25 °C. Higher flow-rates necessitate a higher post-column reactor volume and/or a increased reaction coil temperature, according to Wagner et al. (7).

Influence of eluent composition

a. Ammonium molybdate

There was no significant enhancement in sensitivity noted for concentrations exceeding 45 µmol/L ammonium molybdate. A loss of sensitivity results from lower concentrations.In accordance with this, the remaining tests were performed with an ammonium molybdate concentration of 45 µmol/L.

Figure 3. Detector response for 10 ppb bromate as a function of ammonium molybdate concentration (0, 12.5, 22.5, 45 and 90 μmol/L). Each measuring point corresponds to the mean value of five determinations.

b. Sulfuric acid

The influence on the “bromate peak” can be evaluated, by varying the concentration of sulfuric acid in the eluent (Figure 4). For sulfuric acid concentrations above 31 mmol/L, no sensitivity improvement was obtained. The response of the “bromate peak” decreases rapidly below this threshold value. In addition to this, increasing pH means increased retention times, which results in longer analysis times. A sulfuric acid concentration of 100 mmol/L was used for the remaining tests.

Influence of iodide concentration

The potassium iodide concentration was varied between 0.26 and 0.75 mol/L in order to examine the effect had by potassium iodide concentration on the formation of the triiodide ion. The variation of the iodide concentration has no significant effect on the sensitivity of the triiodide method in the range investigated (Figure 5).

Figure 2. Detector response for 10 ppb bromate as a function of temperature (25, 30, 40, 50, 60, 70, and 80 °C) in the reaction coil of the PCR. Every measuring point corresponds to the mean value of four determinations.

Figure 4. Detector response for 10 ppb bromate as a function of sulfuric acid concentration (8, 31, 68, 122, 144, 185 and 381 mmol/L). Each of the measuring points corresponds to the mean value of three determinations.

Figure 5. The detector response for 10 ppb bromate as a function of potassium iodide concentration (0.26, 0.305, and 0.75 mol/L). In the eluent the ammonium molybdate (45 mol/L) and sulfuric acid concentration (100 mmol/L) were held constant. All of the measuring points correspond to the mean value of three determinations.

Figure 6. UV anion chromatogram of tap water from Herisau (Switzerland).

Analysis of tap water

The optimized conditions which are summarized in Table 1 were applied to bromate determination in a tap water sample taken from Herisau (Switzerland). The peak of the UV chromatogram in Figure 6 corresponds to a bromate concentration of 0.55 ppb.

Table 1. Ion chromatographic and post-column reaction conditions for the determination of bromate.

Conclusions

At trace levels, bromate can be determined using anion-exchange chromatography. This is followed by post-column derivatization with subsequent UV detection, done in accordance with US EPA Method 326. The straightforward set-up requires neither suppression nor sample preparation steps, and uses a sulfuric acid eluent with catalytic amounts of ammonium molybdate. Neither the investigated reaction temperatures (25…80 °C) nor the examined iodide concentrations (0.26…0.75 mol/L KI) determined the bromate response.

Contrastingly, the molybdate and sulfuric acid concentrations significantly influenced method sensitivity. Improvement of sensitivity was noted through increasing sulfuric acid concentrations (. 31 mmol/L) and the shifting of the “bromate peak” to shorter retention times. Better results were yielded by ammonium molybdate concentrations of 45 and 90 mol/L in the eluent. A detection limit for bromate of less than 50 ng/L (50 ppt) is achieved by applying the optimum conditions for the triiodide method.

References

(1) A. Wille and S. Czyborra, IC–MS coupling — Theory, concepts and applications, Technical Paper, Metrohm AG, Herisau, Switzerland (2007)

(2) Y. Bichsel and U. von Gunten, Analytical Chemistry 71, 34–38 (1999)

(3) E. Salhi and U. von Gunten, Water Research 33, 3239–3244 (1999)

(4) H.S. Weinberg and H.Yamada, Analytical Chemistry 70, 1–6 (1998)

(5) G. Bogenschütz et al., Advanced detection techniques in ion chromatography, Metrohm Monograph, Herisau, Switzerland (2007) in press

(6) Metrohm Application Note No. U-9, Iodate, chlorite, bromate and nitrite by suppressed ion chromatography applying post column reaction (PCR) and UV/vis detection

(7) H.P. Wagner et al., Journal of Chromatography A 956, 93–101 (2002)

This information has been sourced, reviewed and adapted from materials provided by Metrohm AG.

For more information on this source, please visit Metrohm AG.