Ultraviolet irradiation

UV disinfection case study

Until recently, municipal wastewater disinfection meant chlorination. Approximately 60 percent of the 20,000 municipal wastewater treatment plants in North America still disinfect water with chlorination, and another 25 percent with chlorination/dechlorination.

Today, increased awareness and new advances in ultraviolet (UV) irradiation are changing the way water treatment professionals think about disinfection. Energy-efficient design, increased lamp power and heightened attention to safety factors are all working to give municipalities a new choice for treating wastewater.

Why not chlorination?
The advantages of chlorination are well known. It's an established, proven technology with a reputation for meeting disinfection standards at a comparatively low cost. Because chlorination is so common, designing a plant for dose and contact time has become a simple matter.

There are, however, many drawbacks to chlorination to which those experienced in using the technology can attest. The toxicity of residual chlorine to other organisms in the environment can be a problem. The chlorination process can also result in the creation of disinfection byproducts such as trihalomethanes (THMs) - halogenated chloroorganic compounds suspected of being carcinogenic - that may require additional treatment. The use of high levels of chlorination for the disinfection of raw water and the treatment of municipal wastewater and the consequent formation of THMs has prompted extensive studies concerning THMs' possible toxicity and carcinogenicity. The U.S. Environmental Protection Agency (EPA) has lowered the maximum contaminant level for THMs from 0.1 milligrams per liter (mg/L) to 0.08 mg/L.

Some municipalities may also report difficulty in maintaining low total residual chlorine (TRC). But overall, the most compelling drawback to using chlorine for wastewater disinfection is safety.

The threat of chlorine gas leaks is a concern for both plant operators and the local community. Compliance with the EPA Risk Management Plan under Section 112 (r) of the 1990 Clean Air Act Amendments requires municipalities to have evacuation plans for the local community in the event of a catastrophic leak, and to share these plans with the public. This new regulation, which became effective June 21, 1999, is fueling widespread public outcry against the use of chlorine and prompting many plants to investigate safer alternative technologies.

A safer choice
UV irradiation is rapidly becoming the alternative-of-choice to chlorination, since it is an effective bacteriocide and virucide that does not contribute to the formation of toxic compounds. Today, approximately 2,000 plants in North America use UV to disinfect their wastewater, and that number is growing quickly.

The mechanism behind UV disinfection is simple and effective. UV light is a physical rather than chemical disinfecting agent. Radiation penetrates the cell wall of the microorganism and is absorbed by cellular materials, including DNA and RNA, which either prevents replication or causes the death of the cell to occur. DNA strongly absorbs UV light in the spectrum of 200 to 300 nanometers (nm); therefore, most UV systems are designed to emit light in this spectrum.

The switch from chlorination to UV disinfection is typically a painless one. Virtually all UV disinfection systems are designed for retrofit into existing chlorine contact chambers. Systems easily fit as modules into the existing channels, with almost no labor required.

Germicidal effectiveness
The effectiveness of UV disinfection will vary with each treatment plant depending on the number of microorganisms that are either clumped or particle-associated, and the distribution of particle sizes in the effluent. Because the absorptive properties of wastewater particles are composed of biological material, increasing the UV intensity will not cause a significant improvement in the performance of the UV system. Thus, it is not possible to predict a priori whether a specific degree of inactivation can be achieved with secondary effluent-containing particles.

To overcome the effects of particle shielding, either the particle size distribution must be modified by making appropriate modifications in the design and operation of the upstream treatment processes at the clarifier stage to produce an effluent with particles that can be disinfected more effectively, or the particles must be removed by some means - granular medium or membrane filtration.

To meet the stringent coliform requirements for unrestricted wastewater reuse applications, coliform count equal to less than 2.2 most probable number (MPN)/100 milliliters (mL), effluent filtration will be required.

Understanding UV disinfection systems
The common goal of any UV disinfection system is to provide UV light in a manner in which it can be readily absorbed by the DNA and RNA of the microorganisms. Most of today's systems fit into one of three categories, as defined by the type of UV lamp they employ, 1) low-pressure/low-intensity, 2) low-pressure/medium-intensity and 3) medium-pressure/high-intensity. The term "pressure" refers to the pressure of mercury in the lamp. "Intensity" refers to the lamp power in the system. Within each category, there are also several different equipment designs available.

Low-pressure/low-intensity UV systems
More than 90 percent of the UV disinfection installations in North America, many of which date back to the 1970s, are low-pressure/low-intensity systems. The standard low-pressure/low-intensity lamp has a power of 65 watts and is generally efficient in producing germicidal wavelengths necessary for damaging microbial DNA and RNA. A low-pressure/low-intensity lamp typically produces 40 percent of its output at 253.7 nm. This output matches well with DNA and RNA absorption.

Sizing and designing low-pressure/low-intensity UV systems is very straightforward. As a rule of thumb, a secondary effluent with 65 percent UV transmittance at 254 nm and with water quality of 30 parts per million (ppm) total suspended solids (TSS) and 30 ppm biochemical oxygen demand (BOD) would require 40 lamps per million gallons per day (mgd) of peak flow. A 10-mgd system, for example, would need 400 lamps. These lamps are typically configured in two banks, with each bank containing several racks, or modules, that contain approximately eight lamps per rack.

The most common lamp configuration for low-pressure/low-intensity systems is horizontal and parallel to flow. These types of systems can be designed in a range of flow capacities, from small flows with few lamps up to very large flow plants with thousands of lamps. The largest low-pressure/low-intensity plant in the world is in Calgary, Alberta, Canada, and has 11,520 UV lamps treating a peak flow of 265 mgd.

Alternatively, lamps can be positioned vertically, perpendicular to the flow of wastewater. In this configuration, lamps can slide straight up out of the channel, making them more easily accessible to the operators. This type of system does require a 4-foot minimum depth of water in the channel. Lower-flow plants with a shallower channel depth cannot employ a vertical-lamp orientation.

Low-pressure/low-intensity systems offer the best electrical efficiency of any UV system in a simple, straightforward design. For higher flow rates, these systems require a very large number of lamps that must be manually cleaned.

More lamps mean the footprint of a low-pressure/low intensity system can be very large for higher flow systems.

Low-pressure/medium-intensity UV systems
The next generation of low-pressure lamp systems is modified for higher output. Low-pressure/medium-intensity systems are relatively new and, to date, include just a few installations in North America. There are several types of equipment that fall into this category.

The first uses higher-output UV lamps, with 270 watts per lamp. These systems can use fewer lamps - about half as many as low-pressure/low-intensity systems. A 10-mgd low-pressure/medium-intensity system, for example, would require just 200 to 240 UV lamps. The power usage, however, is slightly higher than low pressure/low intensity systems, because the UVC (the region of the UV spectrum that damages microbial DNA and RNA) germicidal efficiency of the lamp drops from 40 percent to approximately 35 percent with the higher lamp power.

Another variation of low-pressure/medium-intensity equipment uses amalgam lamps that increase output by adding metal elements that actually modify the chemical content of the lamp. This means the system can be operated at higher power, typically 270 watts per lamp, thus requiring fewer lamps. Amalgam lamps are usually longer than the ones found in standard low-pressure/low-intensity systems.

Another type of system within this group boosts the power to as high as 1,600 watts per lamp. The length of these lamps is typically more than 10 feet, but it can be folded into fourths to fit into a compact reactor. With the higher power comes decreased efficiency, however - about 30 percent of the lamp's output is at 254 nm. Another drawback of these systems lies in the cost of replacement lamps, which run up to 100 times as much as their low-pressure/low-intensity counterparts.

Medium-pressure, high-intensity UV systems
For peak flows greater than 10 mgd, most engineering firms specify systems with medium-pressure/high-intensity lamps. The pressure of mercury in medium-pressure lamps is typically 100 times greater than in low-intensity ones. This makes it possible to operate at a much higher power in a compact lamp.

Medium-pressure systems substantially reduce operating and maintenance requirements in two major ways.

First, medium-pressure systems require fewer lamps. A system using 5,000-watt lamps can have less than 1/20th the amount of lamps of a low-pressure/low-intensity system. The previous example of a 10-mgd system would use 16 low-intensity UV medium-pressure lamps to meet the same treatment requirements as a low-pressure system that would need 200 to 400 lamps. Using fewer lamps makes the system smaller, resulting in lower installation expenses. Labor costs for maintenance are also cut, as there are fewer lamps to change. These cost advantages mean that a medium-pressure system will have a lower overall cost on a net present-value basis.

The second major advantage of medium-pressure systems is automatic cleaning. Because the number of lamps is more manageable, the system can incorporate the capability for automatic cleaning of the quartz sleeve surrounding the UV lamp. Cleaning can be either mechanical, using a stainless steel brush driven by a pneumatic cylinder, or mechanical/chemical, where a chamber of mild acid is moved across the quartz using a hydraulic cylinder.

Another key difference between low- and medium-pressure lamps is the spectral output. Unlike low-pressure lamps, which have all of the UVC at 253.7 nm, the medium-pressure lamp will have a broad polychromatic output across the UVC spectrum. A medium-pressure UV lamp will typically have 25 percent of its output in a region from 200 to 300 nm. Importantly, it has been shown that, in terms of germicidal efficiency, all light within this range has the same germicidal killing power as light at 253.7 nm.

Medium-pressure lamp-based systems were first used in water treatment in the 1980s for groundwater remediation and industrial wastewater. Today, there are more than 200 medium-pressure systems operating in these applications throughout North America. Most use powers from 5,000 to 30,000 watts per lamp. Municipalities began disinfecting wastewater with medium-pressure systems in 1994, when the first gravity-flow medium-pressure system was commissioned. There are now approximately 100 medium-pressure municipal disinfection systems operating in North America.

There are some things to consider when implementing a medium-pressure system. The higher number of lamps and longer contact times required by a low-pressure system also means that mixing will typically be better than in a medium-pressure system - that is, the light will be well utilized. In a medium-pressure system, retention times are much shorter, and there is a broader distribution of UV dose. This effect can be mitigated somewhat by implementing a system that incorporates turbulent mixing in the reactor.

Medium-pressure systems draw more electrical power, typically up to twice as much as low-pressure equipment. To some extent, this difference can be compensated for by the fact that medium-pressure systems can be more finely tuned to process conditions. The output-per-lamp of a medium-pressure system can be varied over a wide range, a variance not possible with low-pressure lamps. The output of the lamp, therefore, can be tied directly to the flow rate of the plant, as well as the percent transmission of the wastewater.

Low- vs. medium-pressure - What is the best fit?
When deciding between low- and medium-pressure UV systems, it's important to keep in mind that one size does not fit all. Many factors, most predominantly flow rate, must be considered before making any choice.

Table 1 provides a profile of the UV disinfection choices different types of municipalities are making. It outlines the results of a market study of 108 UV disinfection projects that were advertised to bid within an eight-month period. Table 1 shows the split of low- vs. medium-pressure for four flow rate ranges.

Table 1

Distribution of low- and medium-pressure systems vs. flow rate

Peak flow

No. of low pressure (%)

No. of medium pressure (%)

< 1 mgd

44 (97%)

1 (3%)

1-3 mgd

16 (70%)

7 (30%)

3-10 mgd

18 (67%)

9 (33%)

> 10 mgd

3 (23%)

10 (77%)

The vast majority of systems below 1 mgd are low-pressure. This is because of the capital cost advantage at the low flow rates and the ability to easily manage a system with a small number of lamps. For example, a 1-mgd system would have fewer than 100 low-pressure lamps, so the impact of fewer lamps is not substantial.

In the range of 1 to 10 mgd, the split is more even, as site-specific criteria may sway the decision one way or the other. For flow rates above 10 mgd, the choice in the market right now is clearly medium-pressure, a trend that most industry observers expect to continue into the future.


UV disinfection case study

"We wanted a solution that was safer and more environmentally friendly than chlorine," explained Robert Norby, operations supervisor of the Hinesville, Ga., wastewater treatment plant. Norby's plant switched from using chlorine to disinfect its community's wastewater effluent, to implementing an Aurora UVTM system from Calgon Carbon Corp.

"We knew that soon our chlorine permit requirements were going to be changed to no residual. We wanted to avoid dechlorination in the worst way. UV was the logical choice for us," he said.

Hinesville installed its UV system 14 months ago. The plant contracted the conversion to an outside company that spent about one month making the change.

Hinesville uses a 40 x 5 kW lamp system with a design transmission of 50 percent UVT at 254 nanometers (nm). Peak flow at the plant is 18 mgd. Upstream treatment includes a trickling filter and secondary clarifiers, but no filters. The UV dosage is 20 mWs/cm2, and the power output is paced to flow to keep UV dose constant, while minimizing electrical usage.

"We'd definitely recommend UV as an alternative to chlorine disinfection," said Norby. "It's good to know we're getting the job done without putting our workers or community at risk."

Click
here to post comments about this topic, and read what others have to say.

This article originally appeared in the 10/01/1999 issue of Environmental Protection.

About the Author

C. David Livengood, PhD, is a project manager at Argonne National Laboratory, Argonne, Ill.

Featured Webinar