Advances in ultraviolet treatment of wastewater
Jon Leech of Wedeco gives the complete guide to the principles behind UV disinfection and the factors to be taken into account to ensure an effective, efficient treatment system.
One of the most common methods and technologies to achieve disinfection is the use of UV rays. The city of Marseille, France, used UV for drinking water disinfection more than 100 years ago. In the last 20 years, however, UV has increasingly been used for wastewater disinfection.
This development has been driven by several issues, one of these being data which showed UV produced no harmful by by-products during the disinfection process. Other techniques, such as chlorine treatment, produced potentially harmful by-products such as trihalomethanes (THMs).
Other benefits of UV disinfection systems include:
- Small footprint.
- Fast reaction times.
- Operational safety and easy of maintenance.
Additionally, there has been a lot of research and development undertaken to improve the technology. This addressed the principals of UV, microbiological levels of disinfection and cost-effectiveness.
Most UV wastewater disinfection systems follow the same principles. Lamps are assembled in modules, the modules are formed into banks which sit in open channels. Effluent may be either pumped or introduced by gravity flow through the channel. Weirs are used to control the level of effluent flow.
Depending on regulations, requirements, water quality, flow and lamp type, the channel construction and bank arrangement may vary in design. The most common general arrangement is based on horizontal orientation of the lamps in relation to the flows . Vertical systems are also available.
The most important part of every UV system is the type of UV lamp used. Thus choice will determine the capital, civil and maintenance costs. All lamps are based on the same general physical phenomenon, which is the emission of electromagnetic rays due to mercury plasma discharge in fluorescent lamps. In the case of disinfection, the only important discharges are between the frequencies of 200-280nm.
Lamps are made out of quartz, which does not absorb UV light in the critical 200-280nm range (absorption factor ~0.9). UV output from the lamps will fall over time. This may be due to solarisation of the quartz or diffusion of the lamp gas through the quartz. To maintain correct levels of disinfection, lamps need to be replaced after a defined period. This is known as the lamp life and will be several thousand hours in total.
If lamps are to be compared to each other, the most important factors are UVC emission per cm discharge length, and their power efficiency [%]. Efficiency is defined as the relationship between electrical power consumption [W] and UVC emission [W]. The ideal lamp should have high UVC emissions with low power consumptions.
Every UV lamp has to be operated with limited current, if not, the lamp would destroy itself due to constant current increase. Different types of ballasts are used to limit current. UV lamps and ballasts have to be defined as a unit. The operation of the same lamp with different ballasts will result in different lamp efficiencies, UV outputs and power consumption. Ballast power consumption is dependent on ballast type. Power loss is usually between 70% and 93%, in general electronic ballasts have lower losses than magnetic ones.
Types of Ballast
Magnetic ballasts are for current limitation only. They limit the maximum power input into the lamp and therefore have an effect on the operation parameters of the lamp itself. The magnetic ballast could be defined as a passive choke.
Electronic ballasts are more technical devices. They limit the current and maintain it, independently of the situation of the power supply. Electronic ballasts are able to change the net frequency into higher frequencies, which further protects the lamps. Electronic ballasts can prolong the lifetime of the lamp. Additionally, electronic ballast's increase the efficiency of the lamp due to better operational parameters.
Advanced electronic ballast's incorporate additional features such as lamp monitoring and malfunction alarms. The advanced ballast could be compared to a plug-and-play card for a personal computer. Each of these ballast's can operate two lamps.
The emission of each lamp can be expressed as UV intensity [W/cm_]. The intensity at a measure point is the summation of all intensities of the lamp expressed as the discrete emission of separate point sources. The intensity itself decreases with the distance from the lamp and the transmittance ability of the irradiated medium.
The complete intensity field within a bank of lamps will depend on the lamp arrangement. It can be visualised with numeric simulations. The calculation of the intensity at certain points is described in the US Environment Protection Agency (EPA) UV DIS program, sub-routine (Tulip).
The flow rate to be disinfected determines the velocity of the water. This is related directly to the exposure time of each infinite element of the water to the UV light. Often, the average exposure time is used to determine if these elements passed through the irradiation field with a sufficient contact time or not. Under the perspective of fluid dynamics this model does not represent the true situation if:
- Short-circuiting occurs.
- The minimum contact time is not sufficient for reliable disinfection.
There are two methods of considering the fluid dynamic situation in a more accurate manner:
- The flow may be calculated using sophisticated numeric simulation, which require high-performance computers.
- Tracer tests.
The US EPA defines limits which allow for tracer validation measurements in relation to sufficient retention times. The goal is to achieve plug flow with a maximum level of horizontal mixing. This should allow any element to find the shortest way through the channel within the lowest intensity fields.
The required hydraulic situation is also strictly linked to the type of lamp. Low-pressure, low-intensity lamps require longer retention times than medium-pressure lamps. Therefore, the range of permissible velocities for a specific flow is subject to a wide range of variation.
One of the more technical issues related to hydraulics is the need for keeping the water level constant for different flow rates. This is essential in open-channel UV systems with lamps installed horizontally. It will ensure no UV lamps become exposed or the level above the lamps becomes too high.
The UV dose is the essential parameter in designing a UV system, various terminology has been used to define the UV dose:
- Applied dose; the dose which is calculated, EPA UV DIS for example, and theoretically applied to the water. The applied dose is often expressed as the theoretical dose or calculated dose.
- Received dose; the UV dose which is equivalent to the reduction of microorganisms, i.e. the effective part of the applied dose. The term is also expressed as empirical dose or observed dose.
The difference between applied and received dose is easily understandable under the conditions of tailing effects within the inactivation curves of microorganisms. The tailing demonstrates that although the applied dose can be increased, the inactivation results are not improved further. In other words, the received dose is still constant.
Disinfection doses are often described as CT products, a name derived from concentration [mg/l] vs. time [s]. For non-chemical disinfectants like UV rays, the concentration term has to be transformed into intensity [W/m_], the equivalent to the concentration of a chemical disinfectant. This very simplistic model, describes the disinfection process as the number of photons which are effective in a certain time, based on the assumption that the total number of photons which have to be applied is constant regardless of any other factors.
The CT concept only represents the applied dose. Therefore CT concepts are under-estimating the situations and conditions under which the disinfection has to take place. This leads to the conclusion that CT is not suitable to design and evaluate UV systems. Nevertheless it is a useful tool to compare the UV susceptibilities of different microorganisms and plays an important role in dose assessments.
As mentioned in the above, it is well known that water quality parameters affect the UV dose in terms of the applied/received dose relationship. However, it is not yet clear how particles are related to the disinfection effect. To avoid misunderstanding it has to be quite clear, there is no question that particles are the most important limiting factor. But to make accurate forecasts there has to be more information about the mechanisms apart from empirical data. Possible points of interests are:
- What shape, size and surface structures are influencing the disinfection? Theoretically it is possible UV-reflecting surfaces are increasing the UV dose. It is important to study the effect of shadowing and the possibility of any build-up of material on the bulbs.
- What number of particles are tolerable for a certain disinfection goal?
- What is more relevant in terms of disinfection, higher numbers of smaller particles or lower numbers of larger particles?
- Many microorganisms are related to particles depending on the particle size.
- What measurements should be undertaken in order to control particles with respect to disinfection?
Not much information is available on these topics, however, research is being undertaken. One such project is by the German Ministry of Research and Technology at the University of Stuttgart.
There are a lot of ways to calculate UV dose under flow conditions. Generally, the basic equation of intensity x time is transformed into numeric models. One of the most common is based on the EPA model UV DIS. However, all these models are based on ideal assumptions, this will not be the case for a live plant. This leads to the question 'What is a good UV system?' In terms of dose the answer is simple. The model has to describe reality, in other words; applied dose and received dose would ideally be 1:1. Because this is not the case, the situation becomes more difficult. This poses such questions as 'How much variance may be allowed between applied and received dose for the different systems?' This is not easily addressed.
In a similar manner to the EPA, local water authorities produce official certification processes. Under these conditions the answer to the tolerable dose ratio (applied vs. received) should be none. Thus, after the certification process, the different dose calculations could be easily corrected to fit.
To design a UV disinfection plant, an assessment of the various parameters has to be made in order to evaluate the UV dose required for disinfection. These include, but are not limited to:
- What is the target microorganism?
- What is the inlet concentration of that micro-organism before disinfection?
- What is the final disinfection goal, and in consequence how many orders of magnitudee shall be reduced to achieve such a goal?
- What should the allowable tolerance be in terms of percentiles ?
After defining the dose there has to be an additional assessment:
- How can this dose be achieved in a particular flow?
- What should the lowest achievable transmittance be?
- What should the highest tolerable particle count be?
- What are the flow conditions?
In terms of received dose, the dose assessment is subject to feasibility studies. For this purpose the most valuable tool is the collimated beam device for biodosimetric studies. The collimated beam allows the generation of dose- relation curves for any water and micro-organism without disturbance of hydraulics. The disinfection goal can be compared with the collimated beam result and it is easy to decide if the disinfection goal is feasible or not. Afterwards a received dose can be specified.
In comparison to the applied-dose assessment, the received-dose method is a more accurate, safer method of making dose assessments. The next step for the collimated beam devices will be their calibration and certification.
Disinfection processes need to be controlled regardless of the accuracy of the dose calculation or design. If the process itself is not controlled the technology is worthless.
There are different methods to control the UV dose. The most simple is to connect a flow and transmittance signal with a dose calculation. The dose is controlled by the variation of flow, keeping the dose constant for different water- quality conditions. Another possibility is to pace the dose depending on flow and transmittance, which allows the system to operate more accurately. In order to allow dose pacing the UV system has to be flexible in terms of intensity variation.
To control the UV dose every parameter which has an influence on the dose should be controlled :
- Ageing of the UV lamps.
- Covering of quartz sleeves.
- Flow rate.
Because transmittance and flow are not the only relevant parameters, the intensity of the lamps also has to be monitored using UV sensors. The sensors may be based on a lot of different technologies including fluorescent materials and semiconductor materials (SiC). As there are various forms of measurement, there should be some criteria for sensors to ensure the quality and reliability of the measurement. These criteria should include:
- Sensitivity in the UVC range (this will be different for low and medium-pressure lamps).
- Temperature shift.
In terms of certification processes there should be some criteria for the calibration of UV sensors.
Besides all the technical issues, there is always the need for more detailed information on hygienic and microbiological issues. The past decade has shown that 'new' micro-organisms, like Cryptosporidium or Giardia, will not allow the pathogens book to be closed. In addition to the research on the 'silver bullet', for the pathogen of the decade, there are some general points which need to be clarified:
- The classic indicator system based on E. coli and coliforms needs to be improved and made more flexible.
- Risk assessments need to be adapted to new organisms.
- Drinking and wastewater standards have to be established with respect to treatments and catchment areas.
- Detection methods have to be developed with respect to health-related matters such as infectivity or immune status.
Some of the most interesting recent developments are the results achieved on Cryptosporidum by UV. Earlier tests showed a very low sensitivity of microorganisms to UV (2-log inactivation in the kJ range). Recent studies have shown different results when the detection methods are changed.
Previously, the common detection method was based on the count of viable cysts, unrelated to infectivity, after UV irradiation. New results were achieved with respect of the infectivity of the cysts in neonatal mouse. Although the results have to be verified, there is evidence of low UV doses (~ 200 J/m?) resulting in highly efficient reduction of the cyst reproduction (4-5 log). Research in this field is aimed at generating more information regarding action spectra, strains, viable but culturable effects, etc.
The recent available UV technology is reliable and efficient. It is a fully acceptable alternative to other (chemical) disinfectants. Due to the continuously increasing use of UV technology there is the urgent need for standardisation and evaluation on the official level such as health authorities.