Analysing Inorganic Disinfection By-products by Ion Chromatography

Water treatment by disinfection processes is considered to have been the major public health achievement of the twentieth century. Many drinking water utilities have changed their primary disinfectant from chlorine to alternative disinfectants, which reduce regulated organic by-products, such as trihalomethane levels, but at the same time often increase the level of other potentially toxicologically important compounds. The hazardous inorganic oxyhalide by-products are bromate, chlorite and chlorate. The most important of these is bromate, which is formed when source waters containing bromide are ozonated. Bromate is considered a possible human carcinogen. The article is a short review of separations of chlorite, chlorate and bromate in water using ion chromatography (IC) and their determination using various modes of detection.

In the 19th century, chlorine was introduced to water treatment as a chemical disinfectant, followed by remarkable reductions in cholera, dysentery and typhoid worldwide. The importance of drinking water quality and its influence on human health was well known, but the specific contaminations would not be identified for centuries because the analytical methods available were not appropriate for specific applications. This situation has changed when modern instrumental analytical techniques and methods were introduced into routine analyses. The number of chemicals determined in drinking water has grown exponentially; however, out of hundreds of them, only very few have been studied or have documented proof of their health effects.

In the 1970s, it was discovered that chlorination of drinking water produced carcinogens, such as trihalomethanes and haloacetic acids. Since 1974, the presence of more than 500 disinfection by-products have been determined in drinking water.¹ Since that time, environmental regulatory agencies and drinking water treatment technologists have performed extensive research for alternative disinfection methods that reduce the generation of organic by-products with significant health risks.

Many drinking water utilities are replacing the most popular chlorine, with alternative disinfectants, such as ozone, chlorine dioxide and chloramines. Ozonation has emerged as one of the most promising alternatives to chlorination. In the early 1980s, it became obvious that the application of ozonization in drinking water treatment not only resulted in the formation of oxygenated compounds but also in the formation of bromide-containing water; brominated organic compounds and bromate as well.

The identification of new, possibly hazardous compounds in drinking water has become an important task for water suppliers. In an ideal situation, where standards for different intake routes of exposure are fully adjusted to each other, regular monitoring of these compounds should only be necessary when these compounds are carcinogens, or if the relative contribution of drinking water to total exposure or to the tolerable daily intake is high. If formation, toxicity and methods of analysis of metals and selected organic compounds in drinking water were well known, determination of inorganic disinfection by-products at the required low levels was a new challenge for researchers in the end of the 20th century.

Recently chlorite, chlorate and bromate are the most important inorganic oxyhalide by-products and their concentrations in drinking water have to be controlled. Chlorite is a disinfection by-product that is formed when chlorine dioxide is used for disinfecting drinking water, while chlorate is formed when chlorine dioxide or chloramine is used.² Chlorination using hypochlorite acid solutions, which contain some ClO₃^- as a product of HClO disproportionation, may also contribute to ClO₃^- contamination in disinfected water. Even low levels of chlorite may lead to hemolytic anemia and can have dangerous effects on the nervous system in infants and young children. There is no limit for chlorate because of this limited knowledge about its toxicity; however, the World Health Organization (WHO) recommends minimizing the level of chlorate as much as possible for as long as there is no reliable toxicological data.

Figure 1: Pretreatment steps for samples, calibration and blank solution according to ISO 15061.10

The most dangerous inorganic disinfectant by-product — bromate, is formed when water containing bromide is ozonated. Bromate formation can be restricted by careful adaptation of the ozone dosage to disinfectant demand. Other options include lowering of the pH, the use of ozone and hydrogen peroxide, addition of ammonia, removal of bromide before ozonization, and the use of membrane filtration or anaerobic processes.³

Figure 2: Separation of F-, ClO2-, BrO3-, Cl-, NO2-, ClO3-, NO3-, PO43- and SO42- in standard solution.

Bromate has been identified as an animal, and possibly human, carcinogen. The International Agency for Research on Cancer (IARC) has classified bromate in group B-2 (the agent is possibly carcinogenic for humans).⁴ In 1993, bromate was judged by the WHO as a potential carcinogen, initially at a 25 µg/L level, which was associated with an excessive lifetime cancer risk of 7 × 10^-5 on account of the limitations in the available analytical and treatment methods. Soon, health effects research indicated that it can be a human carcinogen that poses a potential 10^-4 risk of cancer after a lifetime exposure in drinking water at a 5.0 µg/L level and a potential 10^-5 risk at a 0.5 µg/L level.⁵

The US Environmental Protection Agency (EPA).⁶ and the Commission of the European Communities⁷ has recently issued new rules that require public water suppliers to control some of the previously unregulated microorganisms and cancer-causing disinfection by-products in drinking water. According to these regulations, the maximum admissible level (MAL) is 10 µg/L for bromate and 1000 µg/L for chlorite.

Bromate and chlorite are two of the chemicals parameters included in the EU Directive with a maximum allowed limit of 10 µg/L and 1000 µg/L, respectively. That Directive specifies under Annex III, 2.1: "performance characteristics are that the method of analysis used must, as a minimum, be capable of measuring concentrations equal to the parametric value with a trueness, precision and limit of detection specified". Both trueness and precision criteria were set at ± 25% of the parametric value.

Determination of Chlorite, Chlorate and Bromate by using Ion Chromatography

The idealized method for bromate, chlorite and chlorate determinations should meet the following requirements:

• detect the target ions in drinking water with a limit of determination at 25% of the maximum acceptable concentration

• no sample pretreatment

• short analysis time

• low cost analysis

• widely available method.

The most useful analytical technique to determine inorganic anions and cations seems to be IC, which is a significant addition to the ever-expanding field of chromatographic analysis. IC has seen phenomenal growth in most areas of analytical chemistry since it was introduced in 1975. It has become a versatile and powerful technique for the analysis of a vast number of ions present in the environment and in biological tissues and fluids.⁸ IC is now considered a well-established, mature technique for the analysis of ionic species. Many organizations, such as the International Standardization Organization (ISO) and the US EPA have standards or regulatory methods of analysis of anions and cations using IC.⁹

In the mid-1990s, the ISO worked on Method 15061 for bromate¹⁰ and Method 10304-4¹¹ for chlorite, chloride and chlorate determination. The ISO 15061 standard specifies a method for the determination of dissolved bromate in drinking water, raw water, surface water, partially ozonated water and swimming pool water. Measurement of bromate is performed in the range from 0.5–1000 µg/L, with or without sample preconcentration. If preconcentration is necessary, sample is passed through three cation-exchange cartridges, in the Ba²⁺ , Ag⁺ and H⁺ forms, to reduce the total ionic strength. This standard consists of two parts — a direct method and a preconcentration method, but suffers from the existence of interferences such as chloride and chlorite, which require timeconsuming steps for elimination, clean-up and separation before the instrumental measurements.

The ISO 10304-4 standard specifies a method for the determination of dissolved chlorite, chloride and chlorate anions in water with low contamination (e.g., drinking water, raw water, swimming pool water). An appropriate sample pretreatment and the use of conductivity, UV or amperometric detectors make the working ranges from 0.03–10 mg/L (chlorate) to 0.01–1000 mg/L (chlorite).

The methods of bromate, chlorite and chlorate determination using IC can be generally divided into:¹²

• direct methods (suppressed conductivity detection)

• indirect methods (UV/Vis detection after post-column derivatization)

• hyphenated techniques (ICP–MS and MS detection).

The direct methods are based on the selective separation of ClO₂^- , BrO₃^- and ClO₃^- ions in the presence of other anions in the sample and their detection in the conductivity detector after suppression.

These methods are relatively simple and inexpensive but the main disadvantage is the proper resolution of ClO₂^- /BrO₃^- /Cl^- and ClO₃^- /NO₃^- /Br^- ions, whose concentrations in real samples differ significantly.

Better bromate limits of quantification are obtained with indirect methods (derivatization methods) consisting of converting the determined substance, before or after its separation in the analytical column, into derivatives which can then be detected with a UV/vis detector. Reagents such as fuchsine, o-dianisidine, chlorpromazine, potassium bromide or potassium iodide are applied for the bromate determination with the derivatization methods.¹³

Presently, the technique based on the triiodides formation seems to be the most promising one amongst all the derivatization methods for bromate determination by IC.¹⁴ In this connection, in 2009, the Work Group 33 of Technical Committee 147 of the ISO completed the development of the new bromate determination method based on this post-column derivaization reaction with the use of triiodides reaction. It was thoroughly described in 2009 in the ISO standard 11206 entitled Water quality - Determination of Dssolved Bromate — Method using Liquid Chromatography of Ions and Post Column Reaction (PCR).¹⁵

When its verification process within the proficiency testing framework is concluded, it should become the most important standard used for the bromate determination in water. Moreover, it may be expected that it will gradually replace the already used direct method described in the ISO Method 15061. A chromatogram of bromate determination using an indirect method according to reference 15 is shown in Figure 3.

Figure 3: Chromatogram of ozonated drinking water.

The best results in terms of bromate detection and quantification limits in water can be obtained with the hyphenated techniques. Among them, the most popular ones are IC–ICP-MS (Ion Chromatography with Inductively Coupled Plasma Mass Spectrometry) and IC–MS (Ion Chromatography with Mass Spectrometry).¹⁶ The obvious disadvantages of MS-based detection techniques are that they each add considerable complexity and significant cost to the analysis.

The use of hyphenated techniques requires an in-depth understanding of the analytical methodologies and knowing the apparatus in minute detail. Such systems are expensive and for this reason, they are currently mainly used for scientific research rather than routine analyses. Additionally, up till now no international IC standard-based ion MS or ICP-MS detection has been declared for the regulatory monitoring of bromate or any other disinfection by-products anions.

Conclusions

Global and national agencies are continually striving to monitor bromate, chlorite and chlorate levels in drinking water to establish appropriate regulatory limits. Depending on the results of further research, a risk model could indicate a more definitive guideline value for oxyhalides in drinking water.

Compared with non-ionic chromatographic methods for inorganic disinfection by-products determination, IC is usually faster, more accurate and reliable, less susceptible to sample matrix effects and gives much more information about sample composition.

Furthermore, for the moment IC is the only accepted standard method for inorganic disinfectant oxyhalide by-product analysis.

The problem of inorganic oxyhalides also concerns swimming pool waters¹⁷ and bottled water which has become a healthier choice than tap water for many people because they believe that bottled water contains fewer contaminants, or dislike the taste of chlorinated tap water.¹⁸ The next serious problem which has to be solved is using a KBrO₃ in the baking industry and detecting the presence of bromate residues in food samples.¹⁹

The selection of a proper bromate determination method using IC depends on many factors and is usually affected by the availability of appropriate equipment, the number of analysed samples, the expected analyte content and the required detection and quantification limits.²⁰ Among the three main groups of IC methods recently available (suppressed conductivity, UV/vis and MS detection modes) in the majority of routine laboratories all over the world, methods based on suppressed conductivity detection are used. They are good enough for chlorite and chlorate determination, but are criticized because of costs, and difficulties in bromate determination (< 2.5 µg/L in samples with chloride content over 50 mg/L).

Hyphenated techniques such as IC–ICP-MS and IC–MS are still very sophisticated and not popular in routine laboratories. In this connection, derivatization methods for bromate analysis based on UV detection seems to be the most attractive, alternative, reliable, robustness and low operational cost option.

References

1. S.D. Richardson, Drinking Water Disinfection By-products,The Encyclopedia of Environmental Analysis and Remediation, R.A. Meyers, Ed.; Wiley: New York, 1398–1421 (1998).

2. E. Veschetti et al., Microchem. J., 79, 165–170 (2005).

3. V. Camel and A. Bermond, Wat. Res., 32, 3208–3216 (1998).

4. IARC. Some Naturally Occurring and Synthetic Food Components, Furocoumarins and Ultraviolet Radiation: Potassium Bromate, Monographs on the Evaluation of the Carcinogenic Risk to Human, Lyon, 1986; Vol. 40, 207–220.

5. G.A. Boorman et al., Environ. Heal. Persp., 107, 207–213 (1999).

6. US EPA, Stage 1: Disinfectants and Disinfection By-Products Rule. A Quick Reference Guide, EPA 816-F-01-010 (1998).

7. Council Directive Concerning the Quality of Water Intended for Human Consumption Directive 98/89/CE, Commission of the European Union: Brussels. 1998.

8. J. Weiss, Handbook of Ion Chromatography, Vol. 1&2, Wiley-VCH (2004).

9. R. Michalski, Crit. Rev. Anal. Chem., 36(2), 107–127 (2006).

10. ISO 15061: Water Quality — Determination of Dissolved Bromate — Method by Liquid Chromatography of Ions, 2001.

11. ISO 10304-4: Water Quality — Determination of Dissolved Anions by Liquid Chromatography of Ions — Part 4: Determination of Chlorate, Chloride and Chlorite in Water with Low Contamination (1997).

12. R. Michalski, Inorganic Oxyhalide By-Products in Drinking Water: Ion Chromatographic Methods, Encyclopedia of Chromatography, Ed. J. Cazes, Taylor & Francis, CRC Press, Third Edition, 2010, Vol. 2, 1212–1217.

13. R. Michalski, Trends in Chromatography, 5, 27–46 (2009).

14. R. Michalski and A. Lyko, J. Environ. Sci. Health Part A, 45, 1275–1280 (2010).

15. ISO 11206: Water Quality — Determination of Dissolved Bromate — Method Using Liquid Chromatography of Ions and Post Column Reaction (PCR) (draft) 2009.

16. G. Schminke and A. Saubert, Fresen. J. Anal. Chem., 366, 387–391 (2000).

17. R. Michalski and B. Mathews, Pol. J. Environ. Stud., 16(2), 237–241 (2007).

18. Y. Liu and S. Mou, Microchem. J., 75, 79–86 (2003).

19. T. Akiyama et al., J. Food Hyg. Soc. Japan, 43(6), 348–351 (2002).

20. M. Laubli et al., LCGC Europe, 6, 17–22 (2003).

Rajmund Michalski works in the Institute of Environmental Engineering of Polish Academy of Science in Zabrze (Poland). He is engaged in research on the application of ion chromatography for the determination of inorganic disinfection by-products in water samples. He is the author of several books and a few hundreds articles about theory and applications of ion chromatography in environmental research.