Sep. 05, 2019
ScienceEnvironment

Back to the Roots in Disinfection

Ion Chromatography Helps to Understand the Details

  • Fig. 1: Optimization of the separation of chlorooxo anions on a Metrohm IC 883 Plus with MSM anion suppressor. The concentration of anions was 1 mmol L-1 each. 1) Cl-, 2) ClO2-, 3) ClO3-, 4) ClO4-. The arrows and the bar indicate the position of the ClO4-. While a simple increase of temperature, a typical strategy to reduce retention times, leads to a extreme broading of the ClO4- peak, a multi-parameter optimization led to excellent peak separation in a significantly shorter time [7].Fig. 1: Optimization of the separation of chlorooxo anions on a Metrohm IC 883 Plus with MSM anion suppressor. The concentration of anions was 1 mmol L-1 each. 1) Cl-, 2) ClO2-, 3) ClO3-, 4) ClO4-. The arrows and the bar indicate the position of the ClO4-. While a simple increase of temperature, a typical strategy to reduce retention times, leads to a extreme broading of the ClO4- peak, a multi-parameter optimization led to excellent peak separation in a significantly shorter time [7].

In times of growing water scarcity and the formation of bacterial resistance against standard antibiotics [1], classical chemical disinfectants, such as chlorine dioxide, elemental chlorine and hypochlorite, experience a renaissance in water treatment. A problem of these disinfectants is the formation of so-called disinfection by-products (DBPs), such as chlorate (ClO3-) or perchlorate (ClO4-), which are undesirable and need to be monitored. Ion chromatography (IC) is best suited for this type of analysis as it can be applied to various systems, such as drinking water, pools or waste water, combined with excellent detection limits (LOD) for important contaminants.

IC is often used in addition to other methods such as ICP-MS or ESI-MS to monitor anions in aqueous systems [2]. Most commonly the EPA methods 300.0 and 300.1 are applied [3,4], which use a sodium carbonate eluent, a column separating the seven standard anions (F-, Cl-, Br-, NO2-, NO3-, SO42-, PO43-) as well as the halooxo anions, and a suppressor. In recent years, numerous improvements have been made to further optimize these methods and to improve the detection limits for inorganic DBPs [2]. In classical method development, not only the correct column material must be selected, but also parameters such as temperature, flow rate and eluent composition must be optimized in order to achieve ideal selectivity, resolution and detection limit for the respective analytical problem. In addition to the continuous optimization of existing methods and the combination of several measuring techniques, sampling conditions are crucial to guarantee good comparability of the individual measurements. For this purpose, the compliance with standardized sampling protocols is an ideal instrument to obtain reproducible results from highly representative samples and to document their alteration during the measurement period.

Sampling

When it comes to sampling and storage for water analysis, best practice can be found in ISO 5667-1 and ISO 17025 [5,6].

Disinfection-related samples are prone to change quickly upon storage because of the high chemical reactivity. Several reaction paths can be followed depending on temperature, pH value and the presence of other compounds. The most frequent sources of error occurring during sampling are contamination of the sample by dirty sample vessels (especially traces of organic compounds or Cl-), cross-contamination during dilution and the loss of volatile substances due to incorrect storage. These volatile compounds can either be the disinfection reagents themselves (Cl2, HOCl, ClO2) or break-down products, such as O2. This typically occurs at low pH values and can lead to strong deviations in the analytical results. At higher pH values, the disproportionation to Cl- and ClO3- and ClO4- becomes fast, what not only reduces the disinfection power but can render the solutions harmful for humans and the environment. To prevent or at least reduce these changes, cooling of the samples and in particular exclusion of light are necessary. Also, the adjustment of the pH value by basic stable buffers such as phosphate-based systems can be effective.

Disinfectants

IC is ideally suited for the analysis of disinfectants and DBPs, as special sample preparation for injection into the system is usually not necessary. The chlorine-based disinfectants (Cl2, ClO2, HOCl) and their chemistry have been intensively investigated in Frankfurt in recent years and more effective methods for the detection of the individual chlorine-derived species (Cl-, ClO-, ClO2-, ClO3-, ClO4-) have been developed. By optimizing various parameters (Fig. 1), it was possible to reduce the total IC run time for the detection of ClO2-, ClO3- and ClO4- to ≤ 30 minutes, whereby the individual detection limits could be improved significantly [7].

Remaining Challenges

A major challenge remains the detection of OCl-, as the corresponding acid, which is formed in the suppressor, does not readily dissociate (pKa = 7.54) and thus is undetectable in the conductivity sensor of the IC set-up. The authors are currently exploring methods, such as other kinds of detection principles as well as pre- and post-column derivatization reactions. This would open the opportunity for a complete anion analysis in a single IC run, what is also important for the automation of the developed analyses. Thus, at production sites or disinfection plants, samples should be taken at different stages of the processes and measured in- or on-line for continuous quality control.

A question that has not really been addressed yet is the analysis of bromine- and iodine-based disinfectants, with the latter being frequently used in the medical and pharmaceutical sector. The oxo anions of bromine and iodine are much more reactive than the respective chlorooxo anions and thus are prone to decomposition during sample transport and storage. A severe problem is bromate, which has been identified as cancerogeneous and nephrotoxic [8]. The detection of bromate is usually carried out by HPLC or photometrically [8]. Analysis of iodide and iodate can be accomplished by IC [9], but the detection of periodate species still needs further improvement, as periodates are undesirable by-products during production and are highly dangerous for the environment and health. In order to be able to exclude contamination by such substances, the development of a new detection method is indispensable. The problem is much more complex than in the case of the chlorine-based systems, because in addition to metaperiodate (IO4-) further periodates like mesoperiodate (IO53-) have to be detected and characterized individually. To solve this problem, the development of new column materials and detection methods for periodates is necessary. The ultimate goal must be to find a qualified and validated IC method for all halogen species in disinfectant-containing samples.

Authors

Michael Rudolph1, Sebastian Schneider2, Mathias Rößling1, Andreas Terfort1

Affiliations:

1Institute of Inorganic and Analytical Chemistry, Goethe University Frankfurt, Frankfurt, Germany

2Umicore AG, Hanau, Germany

References:

[1] G. Schminke, A. Seubert, Journal of Chromatography A 890, 295-301 (2000) DOI:10/1016/S0021-9673(00)00606-3.

[2] L.Barron, B. Paull, Talanta 69,621-630(2006) DOI:10.1016/j.talanta.2005.10.032.

[3] EPA 300.0 Determination of inorganic anions by ion chromatography.

[4] EPA 300.1 Determination of inorganic anions in drinking water by ion chromatography.

[5] Water quality - Sampling - Part 1: Guidance on the design of sampling programmes and sampling techniques (ISO 5667-1:2006); German version EN ISO 5667-1:2006.

[6] General requirements for the competence of testing and calibration laboratories (ISO/IEC 17025:2017); German and English version EN ISO/IEC 17025:2017.

[7] M. Rudolph, S. Schneider, A. Terfort, Manuscript in preparation.

[8] Y. Kurokawa, A. Maekawa, M. Takahashi, Y. Hayashi, Environmental Health Perspectives, 87, 309-335(1990).

[9] Y. Bihel, U. von Gunten, Analytical Chemistry 71 34-38 (1999) DOI: 10.1021/ac980658j.

Register now!

The latest information directly via newsletter.

To prevent automated spam submissions leave this field empty.