Advances in Disinfection of Potable Water
14 July 2016
Mike Waite, FWR Water Supply Co-ordinator

I was one of around 60 delegates attending this meeting in the impressive Victorian surroundings of The Leeds Club in Yorkshire. The meeting was clearly aimed at industry professionals well versed in the legislation and practical aspects of disinfection.

The first presentation by Mike Newberry (MWH) reminded the meeting of the DWI approach to UV disinfection, setting out why validation of UV equipment is necessary and the various different systems in use for validating, with specific reference to the UVDGM (USEPA), DVGW (Germany) and ÖNORM (Austria) validation protocols. For any installation it is important to consider the potential pathogen challenge and tables are available for the minimum fluence required for any particular pathogen and required log reduction. While the UVDGM validation is based on achieving the necessary fluence, the other protocols are based on biodosimetry using B. subtilis.

Richard Joshi (atg UV technology) then talked about the practical application of validation to reactor installations. He pointed out that even the best designed reactors may have uneven flow distribution and to take account of this, the concept of Reduction Equivalent Dose (RED) bias has been developed. This compares the reduction achieved in the reactor to the reduction achieved in a bench-scale apparatus and enables the target fluence to be increased to give a suitable safety margin. Considering costs and UV system design, he advocated not going for more than the minimum UV transmittance necessary and setting the maximum flow rate to be delivered to the reactor without overproviding.

Richard Lake described Affinity Water's long experience of, and heavy reliance on UV. It had a UV unit as long ago as 1990 and by 1996 already had around 10 sites with high quality chalk groundwater where medium pressure UV and marginal chlorination was the only treatment. Depending on the source, the company used super and dechlorination, UV and marginal chlorination, or for some high quality sources, just marginal chlorination. With the recognition of aquifer changes due to extreme weather and the Cryptosporidium risk, the company realised that disinfection had to be an absolute barrier. Early advice was that only physical removal of Cryptosporidium was effective and the company installed a number of ultrafiltration plants. In 2007 the regulations changed to allow Cryptosporidium to be ‘rendered harmless’ following recognition that it was sensitive to UV. Since UV offers economic benefits compared to ultrafiltration, the company is moving to replacing ultrafiltration by UV plants and has adopted a standard design for such plants.

Robert Pitchers (WRc) referred to the latest WHO Guidelines which promote the application of health targets, these being applied as DALYs (Disability Adjusted Life Years) (Australia, Canada) or Infection Risk levels (USA, Netherlands). The UK requires water suppliers to have Water Safety Plans which necessitate having site-specific pathogen-focussed disinfection policies. To assist water companies WRc has developed a toolbox which relies on published data and, where necessary, extrapolation and allows companies to optimise chlorination.

Roger Marlow (the Pump Centre) described the Water Industry Mechanical & Electrical Specifications (WIMES) project. One of its key objectives is the procurement of equipment on the basis of lowest life cycle costs (LCC). The WIMES specification for UV equipment (8.01(B)) was issued in July 2014 and is due for review in 2019. The project has also produced LCC model 05 to assist in the comparison of LCC as this is more appropriate than simple capital cost.

After lunch, Andrew Elphinston (Black and Veatch) talked about experiences in the design, installation, and operation of UV plants in Wales. Following experiences with Cryptosporidium at a hitherto good quality upland source in 2005, Welsh Water reviewed all its sources and concluded that over 25 were at risk and UV would be the preferred solution. Comparing low pressure (lower OPEX and power) with high pressure (smaller footprint) UV plants led the company to adopt both types depending on location and power availability. He discussed the practicalities of operating with duty and standby units. He also described the routine maintenance requirements of UV plants and problems with, and control of, fouling.

Jolyon La Trobe-Bateman (MWH) discussed experiences with fail-safe disinfection in 90 works including 30 UV plants. For chlorination the residual should be continuously monitored before and after contact. While in some situations it might be possible to rely on alarms and manual intervention it is much safer to have automatic triggering of shutdown and in order to prevent shutdown on monitor malfunction, all monitoring should have triple validation. Fail-safe monitoring equipment ideally should be completely separate from operational monitors. It is also normal to set a duration before which shutdown is initiated. There are many different configurations for a fail-safe installation but in all cases shutting down should leave the works in a condition from which it can recover. He then described some instances where fail-safe had not worked.

Ludwig Dinkloh (XYLEM) spoke about the application of Advanced Oxidation Processes (AOP) to seasonally occurring micropollutants. He explained that AOP rely on the production of highly reactive OH radicals which can be generated by ozone + hydrogen peroxide, UV plus hydrogen peroxide, or UV plus chlorine. The main seasonal micropollutants are geosmin and methyl iso-borneol (MIB) and they tend to only occur for 2–3 months per year. Studies showed that low pressure (LP) offered significant savings over medium pressure (MP) for removal of MIB. AOP can also be used for removal of some pesticides in particular metaldehyde which can be a problem for some water companies. Designing a plant to tackle metaldehyde investigations showed that while ozone + hydrogen peroxide can be effective, it leads to production of bromate although this can be controlled to a degree by a series of sequential mixings. UV alone would require doses much above the validated range for the equipment. The most cost effective solution over 10 years of seasonal use which was adopted was a combination of ozone + hydrogen peroxide followed by UV. This, despite its higher CAPEX, offered the lower life cycle cost.

Bruno Pollet (Power and Water) gave a detailed account of the mode of action and many applications of ultrasound and sonoelectrochemistry. He also presented data showing how an ultrasound unit could reduce the microbial content of wastewaters as well as effecting changes in the water's chemistry. Ultrasound can remove chlorine, reduce particle size and also reduce chlorine need.

Richard Joshi (atg UV Technology) finished the meeting with an interesting forward looking presentation on the future of UV. He talked about the way in which new ideas progress through inflated optimism via disappointment before stabilising in their application. Any new product must meet a market need, and technically possible does not necessarily mean practically viable. He compared LP, MP and LP High Output (Amalgam) and suggested in the short to medium term that these technologies were unlikely to be overtaken, any developments being in increased efficiency. One interesting development could be UV LEDs. These would have many advantages, being small, wavelength-tunable, solid state, instantly switching, mercury-free and low voltage. Unfortunately at present their efficiency is low, costing per watt of UV 1000 times more than LP and 100 times more than MP. He surmised that UV LEDs are at least 10 years away for large volume water treatment but some small volume applications could happen sooner. UV could have a role in water re-use where resources are scarce and could be used in small package plants for small communities. He suggested that AOPs were likely to be a key element in developments of advanced water treatment.