“It is good to expose water to sunlight.” This citation from the Indian Sanskrit Ousruta Sanghita, dating from 2000BC, extracted by Francis Evelyn Place in 1905, describes the basic principle of ultraviolet (UV) irradiation of water – but only 4,000 years later was it developed into a large-scale technology, and today is one of the most popular means for water disinfection.

While wastewater disinfection is common practice in the US, the main driver for disinfection in Europe is the revised EU Directive on Quality of Bathing Water, proposed in 2002 to replace the Council Directive 1976/160/EEC. The new proposal contained two parameters to define bathing water as “good quality” or “excellent quality”, depending on the Intestinal Enterococci and Escherichia coli concentration as set out in Table 1 (below right) at 95% compliance.

The proposed good-quality bathing water standard is roughly equal to the guideline standard in the existing directive. In 2002 (Blue Flag) guideline standards were achieved at 68% of coastal bathing waters in the UK, while more than half of the 550 bathing areas would meet the excellent standard. Nevertheless, 25-30% of bathing waters in the UK in 2002 would fail the proposed “good” standard.

UV disinfection is regarded as the most appropriate technology to help achieve improvements in microbial contamination of wastewaters. The electromagnetic energy emitted at wavelengths of 235-270nm is biocidal to waterborne bacteria and viruses. The pathogens’ deoxyribonucleic acids (DNA) readily absorb UV light, which alters their molecular structure so they cannot replicate. The result is disinfected wastewater.

UV disinfection systems have gained popularity over the past 20 years, and have become more sophisticated, reliable, and cost-effective. Today, UV disinfection systems can be divided into three major classifications: low-pressure; medium-pressure; and low-pressure, high-intensity.

Low-pressure lamps

Available for more than 20 years, low-pressure (LP) lamps are available in horizontal or vertical configurations submerged in relatively shallow flow channels. Enclosed and Teflon-tube systems are also available. Systems consist of ballasts, reactor, cleaning equipment, power supply, and controls and instrumentation.

Low-pressure lamps are typically used at facilities where the design flow is less than 20Ml/d, as they consist of a large number of low-power (60-80W) lamps that emit a near monochromatic radiation of 253.4nm. Control of these types of lamps is limited to either “on” or “off”. The disadvantage is that the number of lamps increases as flow increases, and with a typical life span of 8,000 to 14,000 hours. Because of the large number of lamps involved, cleaning of a low-pressure system usually entails removal of the bank of lamps and dipping it into a chemical solution, which increases maintenance costs at large facilities over other types of UV systems.

The lamps operate at a temperature around 40-50ºC, but low temperature effluents can lower the temperature of the lamps, and thus reduce their efficiency. Conversion of applied power into UV radiation is about 35-40%.

Medium-pressure lamps

During the last decade, medium-pressure (MP) lamps became available in open-channel and closed-pipe configurations. Because they have higher UV output, medium-pressure systems use about one-tenth of the number of lamps that a low-pressure system requires. However, they use more power and need an automatic cleaning system to periodically remove the solids that coat the quartz sleeves.

Low-pressure, high-intensity

Introduced within the last five years, low-pressure, high-intensity (LP-HI) lamp systems use about one-third the lamps of low-pressure systems, but three times more than medium-pressure systems. Early installations were deliberately over-designed, involving multiple banks of lamps and cumbersome hydraulic diversion controls designed to turn lamp banks on and off as operating conditions dictated. Originally, all lamps in the bank or channel operated at full intensity, but improvements allow for the wattage output of these lamps to be varied, so to optimise dose delivery. In addition these systems are equipped with automatic cleaning.

The survey said…

As part of a recent study funded by the American Water Works Association Research Foundation (AwwaRF), 19 large WwTWs, with more than a third located outside the US, were assessed. Installations with design flows of more than 350Ml/d, different water quality requirements and discharge limits were surveyed, to assess the efficiency and reliability of the technology on large scale. The purpose of the survey was to examine issues such as level control, the relationships between low flow and treatment effectiveness, and the use of back-up power.

The challenge in operating a UV light disinfection system is ensuring the water receives the correct UV dose: too much, and you are wasting electricity; too little, and the water is not thoroughly disinfected. That is why the proper control system is critical. UV systems can be controlled manually or automatically, depending on system components. A successful control system will deliver the correct dose at the correct time to ensure compliance with discharge permit limits. If automated, it will also interface seamlessly with the treatment plant’s overall supervisory control and data acquisition (SCADA) system.

Level control was established using a control device and then adjusted to reflect the flow discharged from the plant. The most frequently used flow control devices were found to be gates, weirs, and penstocks.

Performance impacts attributable to low flow were difficult to quantify because, in many cases, plant operators indicated that they did not review data unless the system was in non-compliance.

Average daily flows at the 19 plants were found to be one-third to one-half the design flow, which provides sufficient capacity at low flow conditions. The survey found that large UV systems can be successfully used at the design flows greater than 350Ml/d. Of the 19 WwTWs surveyed, only one was more than ten years old. Twelve plants had peak design flows between 350 and 500Ml/d, and five between 750 and 950Ml/d, of which one was under design, and one under construction. Of the two facilities with peak flows larger than 1,300Ml/d, one was operating and one was under construction.

Any difficulties with scale-up have been successfully addressed by the UV system manufacturers. The systems appear to be meeting their discharge limits regardless of the discharge, influent water quality, or geographic location of the facility. One of the facilities surveyed had installed an LP system in the 1990s and is now converting to an LP-HI system, indicating the confidence to continue to use UV, and the tendency that its use at large wastewater facilities will increase.

UV systems used to disinfect WwTW effluent are designed to achieve compliance with a microbiological limit that is based on the designated use of the receiving stream or of the effluent. These limits can vary widely. For instance, a plant discharging to the ocean may be subject to less restrictive limits than one that discharges to an effluent-dominated stream. The most restrictive limits apply to systems whose effluent is used as reclaimed water.

Upstream processes were found to affect UV disinfection performance: the type of

biological process (fixed film/suspended growth) and filtration. Eighteen facilities reported using some form of the activated sludge process and one of the systems

surveyed uses a fixed film process. Studies conducted by WERF (1999) indicated that 13-14% of the coliforms in the effluent from an activated sludge WwTW are associated with particles. If the plant uses a nutrient removal process, the amount of coliforms associated with particles decreases to 3-4%.

The second upstream process that was found to impact performance was filtration. Nine of the systems surveyed use filtration upstream from the UV system. The survey did not attempt to determine how the type of filtration system affected the effectiveness of the UV disinfection system. One facility blends its chemically enhanced primary treatment effluent with the effluent of the activated sludge process ahead of the UV system.

The types of UV lamps used at the facilities surveyed, according to the three types mentioned above, are listed in Table 2. Results of the study indicate a general trend toward the use of LP-HI and medium pressure systems at larger WwTWs, as mechanical cleaning makes LP-HI and MP systems more attractive. Key to process control is the ability to deliver the correct dose of UV light at the correct time to ensure compliance with the discharge limits. Flow, intensity, and transmittance are used to control the dose emitted from the UV system as indicated in Table 3. With flow control, the number of lamps energised increases or decreases in response to the flow being discharged from the facility. This type of process control does not take into account the decrease in lamp output as a result of aging or changes in water quality. Flow control was generally used on the LP-LI systems.

Dose pacing using intensity is generally used on the LP-HI systems. In this method, the intensity is used with flow (changed to retention time) to calculate dose. This system allows the dose to adjust based on lamp output and changing water conditions.

Dose pacing using transmittance uses measurements from an online transmittance device with flow to calculate dose. This type of control also allows adjustments to be made based on lamp output and water quality, but maintenance/calibration of intensity sensors and online transmittance needs to be performed frequently to ensure compliance with discharge limits.

One of the key parameters in the design of UV disinfection systems is UV transmittance (UVT) at a wavelength of 254nm. Literature indicates that the effluent from a WwTW using the activated sludge process should have a UVT of 60-65%, which is considerably lower that the 85-95% measured at WTWs, and leads to a much larger number of lamps.

The design transmittances for the facilities that reported their UVT are listed in Table 4, showing that around half of the facilities responding had a design transmittance lower than 60%. A lower transmittance value could be a result of a number of factors including high effluent solids concentration, a larger industrial waste component, or trace amounts of ligins and tannins in the domestic water stream.

More details on energy efficient disinfection will be presented at the CIWEM Energy Event 24-25 April, Biocity – Nottingham: [email protected].

Action inspires action. Stay ahead of the curve with sustainability and energy newsletters from edie