The use of chlorine as a disinfectant has been a key feature of drinking water treatment in the UK for well over a century. First deployed in water supplies in the late Victorian era to combat waterborne diseases, there is no doubt that the introduction of chlorine was one of the most significant public health improvements ever made – and its power as a residual disinfectant ensures it remains central to strategies for safe water provision across the country.
Nevertheless, the use of chlorine presents a number of potential issues that have led some in recent years to question its ubiquity. These include taste and odour concerns from customers, and the potential to form so-called disinfection by-products, which, in high concentrations, could have a health impact. From the point of view of water company operations, there are also potential health and safety risks associated with the handling of chlorine, together with consideration of the cost and sustainability of its manufacturing and transportation.
The Netherlands has for many years shown that potable water can be delivered without a residual disinfectant providing conditions are right, and a growing number of other countries have followed suit. Cutting down on the use of chemicals in water treatment is an aspirational goal for many in the sector, with a number of benefits for both consumer and water utilities alike.
So is it time for the UK water sector to consider scaling back the use of chlorine, using alternative methods of disinfection, or even rethinking the role of the disinfectant entirely?
CHEMICAL-FREE WATER TREATMENT
One of the key advantages of chlorine is its ability to persist in treated water, protecting against disease as water passes through often lengthy distribution networks. While alternative disinfection processes such as UV and ozone are well established, the absence of a residual effect means water passing through the distribution network may be prone to microbiological contamination before it reaches the customer’s tap, potentially posing serious health risks.
Despite this, countries such as the Netherlands, Germany, Denmark, Switzerland and Austria have been able to implement systems that operate without chlorine. The approach depends on using the best quality source water possible and maintaining the distribution network to sufficiently high standards; multiple steps are also used to ensure adequate water purification, such as sand filtration, ozone, carbon treatment, membrane filtration and UV treatment.
Should there be any issues with the process that lead to contamination, chlorine may need to be introduced as a temporary measure, and boil-water orders are used as a further backstop where necessary. The system appears to work for the Dutch, with better water quality and fewer water quality incidents than most developed countries.
But while chlorine-free drinking water is gaining ground globally, the prevailing view has been that UK water companies lack the high-quality distribution network required to adopt the approach.
Anglian Water is taking steps to understand what would need to change before the use of chemicals, including chlorine, could be reduced. As part of its Innovation Shop Window, the company has set itself seven long-term, aspirational goals, which include ‘100 per cent compliant drinking water and chemical-free water treatment’.
Chlorine is one of the three most significant chemicals – in terms of cost and volume – used by Anglian Water, together with orthophosphoric acid for lead control, and the salt used as part of ion exchange processes to remove nitrate from groundwater sources. Anglian’s long-term strategy around drinking water quality includes a focus on all three.
With regard to chlorine, Anglian has begun work with the University of Sheffield to determine the viability of chemical-free water treatment, with a PhD researcher, Natalie Lamb, carrying out a study using water from the region on test rigs to understand the effect of reducing chemical use.
As part of her studies, Lamb has carried out a literature review to understand the specific circumstances that have led to other countries making the decision to stop having a residual level of chlorine in their distribution systems. She has looked at water quality, the age and type of mains materials present in the distribution system, together with regulatory and societal factors in each country. The town of Vsetín in the Czech Republic soon plans to join the list of areas to switch off chlorine supplies, and Lamb paid a visit this summer to learn from the innovative pipe flushing programme taking place there in preparation for the project.
The next stage of Lamb’s research will be to understand the biostability of mains water in the Shop Window area in Newmarket, and across the Anglian region, looking carefully at the interaction between orthophosphoric acid and chlorine. This work will involve experiments at the cutting-edge distribution rig facilities at the University of Sheffield, together with pipe rigs at live water treatment works within the Shop Window.
“We’re a very long way off considering reducing the amount of chlorine we use within our water treatment works,” Robin Price, head of water quality at Anglian Water, says. “But we know that gaining a further understanding of the characteristics of a water distribution network with the capability of maintaining safe, clean drinking water right through to every customer without the addition of chemicals is a worthwhile piece of work, and we know that such a network will bring us and our customers many other benefits in terms of reductions in leakage and interruptions to supply.”
But if removing chlorine entirely is a distant prospect, can the amounts used be reduced or optimised, to make the best use of resources and help achieve better tasting water?
The challenge in residual network disinfectant dosing is getting the optimal amount that balances safety with palatability. Companies will often seek to keep free chlorine to no more than 0.5mg/l, so that it does not impact on taste and smell, but they must still ensure there is enough to provide protection. Within that target range, companies avoid using excessive quantities of the chemical and keep Opex costs under control.
However, this is no easy task: the chlorine residual decreases over time and is difficult to predict across a complex network. Water is routinely tested at service reservoirs and customers’ taps and, should the chlorine residual regularly fall beneath the target, secondary doses are often applied at service reservoirs or pumping stations.
Atkins, a member of the SNC-Lavalin Group, has recently carried out water quality modelling work that predicts changes in concentrations much more accurately, giving companies the knowledge on how much to dose and how to optimise secondary dosing systems.
“While chlorination has been practiced for over 100 years, people’s understanding of it and the way it acts on different pathogens is still developing,” Hugh Thomas, technical authority for process at SNC-Lavalin’s Atkins business, says. “Our specialist teams have been focusing on how effective disinfection with chlorine is best achieved at the treatment works, and the different operating conditions it works best under.”
The difficulty in modelling the chlorine residual in distribution systems is that the decay rate is influenced by water chemistry and also the condition of the network. Atkins has undertaken lab-based water quality analysis to establish bulk decay characteristics, and deployed field instruments to establish wall decay.
“If you’ve got a network in poor condition, with a lot of corrosion by-products and biofilm, the chlorine decays much faster than if it’s passing through pipes in good condition, that convey the water without any changes,” Thomas adds.
“It’s this information that allows us to predict accurately how to optimise chlorine dosing over a large water quality zone. We can input the bulk and wall decay parameters into a hydraulic model and say, for example, that if the water left the treatment plant with a chlorine residual of 0.5, it will have a residual of 0.4 when it’s 10 hours old. You can predict what the chlorine is at any point in the network, so it’s moving from a reactive to a proactive model.”
It has already been implemented successfully in a section of United Utilities’ supply area and Atkins is now working on other projects around the country.
“Customer acceptability is a key measure for the water companies and maintaining the chlorine in that target range is something they’ve been trying to do using experience and a trial-and-error basis,” he says. “With the model, you can experiment with different primary and secondary dosing strategies without having to test it with the customer.”
While chlorine remains the universal choice for UK water companies, the form the chlorine takes varies. Chlorine gas was once the mainstay but it poses health and safety risks, leading to increased use of sodium hypochlorite – which contains chlorine in liquid form – as well as onsite electro-chlorination.
Many treatment works are situated near residential areas, and the repercussions of a major unplanned chlorine gas release could be significant: used as a chemical weapon in the First World War and in Syria in recent years, it can have a fatal impact on the respiratory system. While deaths are unlikely in the case of a gas leak from a treatment works, particularly when safeguards are put in place to shut down the system if there is a release, it remains a serious threat to health, and those working onsite are at greatest risk. The larger sites must also contend with the dangers and expense of safely transporting and storing chlorine gas drums of up to 1,000kg, with the COMAH (Control of Major Accident Hazards) regulations understandably onerous.
As a liquid, sodium hypochlorite does not pose the same issues. There are some drawbacks – it is only around 10 per cent chlorine, which means it must be stored in far greater quantities than chlorine gas, and its alkalinity means measures must be taken to prevent limescale build-up – but these can be weighed against its health and safety advantages and relative ease of use.
Over recent years, several companies have started to move away from chlorine gas, particularly at their largest sites, but Yorkshire Water is in the process of becoming the first UK water and sewerage company to eliminate chlorine gas – as well as sulphur dioxide – from its water treatment processes entirely.
“We had a re-evaluation of health and safety and it’s driven quite a lot of changes throughout the business,” Mark Broady, Yorkshire Water project manager, says. “Our CEO, Richard Flint, said we’re going to be the first water and sewerage company to go completely away from gas treatments.
“It’s about removing the risk of having any releases that could affect our guys operating the plant or the public if they’re in the area.”
Galliford Try is the main contractor on a programme to replace the chlorine gas with sodium hypochlorite – as well as sulphur dioxide with sodium bisulfite – at nine sites. Doncaster, York and Beverley are large treatment works that required the 1,000kg drums for chlorine gas; the remaining six sites relied on smaller cylinders of up to 87kg.
At four of the sites, modular kiosks featuring everything required for treatment are being used to directly replace the existing systems. For the rest, Galliford Try is reutilising existing apparatus or buildings that were on site to install new dosing equipment, which requires the use of temporary dosing systems while the upgrades take place.
Broady says there have been some “slight teething problems” with the modular kiosks in terms of ensuring compatibility with the other equipment on the sites, and the weather this summer led to supply and demand issues that held up work and mean the £14 million project is likely to take around two years.
Where systems are up and running, though, staff are already seeing the benefits.
Nigel Taylor, Galliford Try project manager, says: “Keld Head Water Treatment Works is running on a permanent sodium hypochlorite dosing system and it’s running very, very well. It’s less problematic than the gas because they don’t get little leaks when they’re changing things over. Those are planned releases, but you need to have several workers with the training and breathing apparatus to do it, so it takes time. With hypochlorite, you don’t have any of that.”