Remote pooling of trihalomethanes

Drinking water is generally treated with disinfectants to make it safe by killing off any lurking pathogenic organisms which might cause disease in humans. The disinfection process is carried out once at the water treatment plant before distribution, which is quite sufficient to protect the water until it is consumed.

The same principle applies to swimming pools with one notable exception. Here, the water is continually recycled, so must be treated regularly to remove extraneous matter and ensure that it remains healthy. Apart from filtering to remove physical matter such as hair, plasters and leaves, disinfection must be repeated to remove any microorganisms introduced by the bathers.

In drinking water, disinfection with chlorine, often added as hypochlorite, produces a series of disinfection by-products (DBPs) formed by the reaction with small amounts of organic matter. In swimming pool water, the formation of DBPs is exacerbated by the presence of additional organic matter from the bathers, such as sweat, cosmetics and excretion products.

DBPs present a problem due to their links to asthma and cancer in humans, although these associations are not yet fully established. Nevertheless, one group of DBPs, the trihalomethanes, has been classified as Group B carcinogens because they cause cancer in laboratory animals.

It is of prime importance to monitor the levels of the DBPs in water, to ensure that they remain below recommended levels. Under common practice, samples will be withdrawn at regular intervals and transported to the lab for analysis. However, a new approach has been undertaken in Denmark for swimming pool water that involves long-term, continuous on-site monitoring.

The method involved the use of a membrane inlet mass spectrometer working in unsupervised mode and was set up to analyse trihalomethane levels. The experiments were organised by Frants Lauritsen from the University of Copenhagen, Gert Kristensen and Morten Klausen of DHI, Horsholm and Vagn Hansen from Mikrolab Aarhus A/S, Hojbjerg.

The pool consisted of a warm water basin at 31-33°C with a strong massage jet system and a connected wading pool, the total volume being 50 m³. Water was circulated continuously through a water treatment system and sampled from this pipework towards a flow-through membrane inlet fitted with a six-port valve to allow input from up to five sources. The membrane was positioned right against the ion source to minimise interferences from transient species arising from reactions between the vaporised sample and the instrument surfaces.

Mass spectra from the pool water revealed the presence of bromodichloromethane (BDM) and chloroform. They were measured using ions at m/z 129 (BDM) and m/z 83 (chloroform), while also monitoring the ion at m/z 210 to confirm the absence of dibromochloromethane, which would also give an ion at m/z 129. Their concentrations were determined from calibration curves.

The instrument was operated continuously for one year, apart from periods when the mass spectrometer filament was replaced. This occurred every 6-8 weeks and required a period of 1-2 days before the new filament had stabilised. Apart from instrument maintenance, it was operated remotely with off-site supervision, recording signals every 10 seconds. These were averaged over 5 minutes and sent telemetrically to a remote database system for data storage.

The data revealed a daily rhythm for the concentrations of both compounds, which correlated directly with the opening hours of the pool. Levels rose in the evening after pool closure to a steady state level then decreased in the morning, after opening hours. The maximum daily values were 30-100 and 3-10 µg/L for chloroform and BDM, respectively, with minimal values 30-50% lower than these.

This pattern suggested to the researchers that the levels were linked to activity in the pool. To check out this assumption, they left the water massage jets running overnight and found that the levels of the trihalomethanes did not increase but remained at a steady level, which was lower than overnight levels with the jets off. So, agitation of the water increased evaporation of the compounds, reducing their water levels.

During the day, the activity of the bathers as well as the massage jets instigated a greater transfer of trihalomethanes from water to air, so that water concentrations continued to fall, without reaching a steady state.

A second pattern that the researchers observed was linked to the monthly addition of 100 kg of sodium chloride to the water, which was added to aid the in-line electrolytic production of chlorine for disinfection. Coinciding with the salt top ups was a large spike in the levels of BDM to about 100 µg/L and a fall in chloroform levels to almost zero.

This behaviour was caused by trace amounts of bromide in the salt that were oxidised by chlorine in the pool to give bromine, which generated brominated trihalomethanes over 2-3 days until the bromide was exhausted. Thereafter, levels of both THMs returned to normal.

This observation triggered a change in strategy at the pool, with the introduction of more frequent additions of 25 kg salt, which reduced the surge in BDM formation.

This is the first reported use of an on-site membrane inlet mass spectrometer operating over a long period of time in unsupervised mode with off-site surveillance. It has succeeded in uncovering the daily variations in the concentrations of trihalomethane disinfection by-products, which would not have been possible by the normal practice of withdrawing samples at regular intervals from the pool for testing. The results will be used to improve water quality and reduce disinfection by-products in swimming pools.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.