UV disinfection of municipal water and waste water is becoming increasing popular for a variety of reasons. So that a UV disinfection system can be correctly selected, sized and implemented for a given application, a clear understanding of how UV systems perform their function is required. It is also very important that the necessary process information is gathered and communicated to the UV system supplier so that a suitable UV system can be designed to deliver the required performance. Finally, it is important to understand the process by which UV systems are designed and verified. This verification is sometimes referred to as “validated” and is vitally important as it determines whether or not the UV system is suitable to perform the intended duty.
What is UV light anyway?
Ultraviolet (UV) light is a naturally occurring component of sunlight. It falls in the region between visible light and X-Rays in the electromagnetic spectrum (Figure 1). Generally, UV light is considered as falling between 100nm & 400nm in wave-length, however UV light in itself can be categorized even further into separate regions. Although scientists hold varying opinions as to the exact boundaries of these regions, they are generally considered to be approximately as follows: Far UV (or “vacuum”) 100nm – 220nm, UVC 220nm – 290nm, UVB 290nm – 320nm and UVA 320nm – 400nm. Of these UV regions, UVC is recognized as having significant germicidal properties. UVC light is however, almost entirely filtered out by the Earth’s atmosphere. As such, if we are to utilize the germicidal properties of UVC light, we have to artificially generate it here on Earth using commercially produced UV lamps. Figure 1 – Electromagnetic Spectrum
How does a UV lamp work?
UV lamps contain a small amount of mercury, either in a free state within the lamp tube, or imbedded within the lamp tube’s surface. When electricity is applied to the lamp electrode, electrons flow between them, these vaporize the Mercury, which when bombarded with electrons emit UV light. The exact wavelengths emitted depend on the vacuum pressure within the lamp tube itself. Low Pressure (LP) UV lamps are evacuated to relatively “low” pressures (between 1-10 Pa) and emit germicidal (I.E. UVC) light at a single UVC wavelength of approximately 254nm. Medium Pressure (MP) lamps are evacuated to what is termed “medium” pressure and emit a broader spectrum of UV light with higher intensities between around 254 – 265nm. Figure 2 represents this diagrammatically. Low pressure and so called “Amalgam” lamps are about twice as efficient at converting electrical energy into UVC light as compared to medium pressure lamps. However, medium pressure lamps emit far more UVC energy per unit length than a low pressure or amalgam lamp. There are a variety of considerations to be taken into account when choosing which of these lamps should be used for a given application. The nature of this decision is quite detailed and can be left to a future discussion. Suffice to say that both low pressure (including amalgam) and medium pressure lamps are germicidally effective. But why exactly is UVC light germicidal?
UV Lamp Output and it’s relationship to the UV Absorbance of DNA
Figure 2 – Low Pressure Vs Medium Pressure UV lamp emission spectra
How does UVC light inactivate pathogens?
Deoxyribose Nucleic Acid (DNA) is often referred to as “the building blocks of life”. Each and every cell of a living organism contains DNA, which acts as the “blueprint” by which the cell is able to function and reproduce. UVC light is able to penetrate the cells of microorganisms and disrupt the structure of their DNA molecules (Figure 3). In doing so, the microorganism is prohibited from surviving and/or reproducing, thereby rendering it inactive and no longer pathogenic. This may at first seem to be a very simple principle, however many barriers exist that can potentially stop the UVC light effectively penetrating the target organism’s cells. For this reason, it is important that a UV system is carefully selected to ensure these barriers no not affect performance.
Figure 3 – UVC Radiation disrupts DNA
How is a UV system selected?
There are three key parameters that need to be considered when selecting a UV system to perform a disinfection duty. They are:
1. Water quality
2. Water flow rate
3. Pathogen(s) to be inactivated
To understand the principles of UV system selection fully, each of these parameters need be examined in more detail
1. Water Quality
The nature and quality of the water to be disinfected is critical, not only in selecting an appropriate UV system, but also in deciding if UV disinfection is even possible or suitable at all. Of all water quality parameters, Ultraviolet Transmissivity (UVT) is the most important. This is because the UVT of the water will determine how well the UVC light will penetrate the water so the pathogens in the water are exposed to sufficient UVC light to be inactivated. The UVT of the water is determined by taking a water sample in a quartz cuvette and passing UV light at 254nm through the sample. The percentage of UV light that penetrates this sample is referred to as the sample’s “UVT”. Most typically, the cuvette used has a path length of 10mm, in which case the UVT reading is referred to as being “T10”. The UVT of the sample is critical for a very simple reason; although parameters such as BOD, COD, turbidity and TSS all influence the extent to which UVC light penetrates the water, and although correlation to UVT is difficult if not impossible, they are all effectively accounted for by the single UVT reading. Of all other water quality parameters apart from UVT, Total Suspended Solids (TSS) and Total Dissolved Solids (TDS)/Salinity are also important. TSS is important because anything in excess of 20mg/l TSS can result in a phenomenon known as “shielding” whereby the pathogens can be “shielded” from the UVC light by the particles suspended within the water. TDS/Salinity is important because at very high levels, attention must be paid to the materials of construction of the UV system itself so as to avoid the risk of corrosion.
2. Water Flow Rate
A key factor in determining how effective UVC light will be in deactivating a given pathogen, is the length of exposure time that pathogen has to the UVC light for a given UV intensity. The longer the exposure time, the more UVC radiation will penetrate the pathogen’s cells and therefore the more effective the inactivation process will be. The slower the flow rate of the water through the UV system, the longer the UV exposure time and vice versa, and so the maximum and minimum flow rate of the water should be considered. This is because many UV systems have the ability to adjust the power output of the lamps in relation to changes in water flow rate. By doing so, energy may be conserved when water flow rates are lower than peak flows. When determining maximum and minimum flow rates, it is important to establish the instantaneous flow rates as it is this that that will determine the instantaneous minimum and maximum UV exposure times. Daily and hourly flow rates are usually misleading in this respect, as they can mask important “peaks and troughs” in the instantaneous flow rate, thereby resulting in spurious calculations of the true UV exposure time during these peaks and troughs.
3. Pathogen(s) to be inactivated
Different pathogens have differing resistance to UV; some are more susceptible than others and so require different amounts of UVC exposure in order that they are inactivated. In order to correctly size and select a UV system, it must be established which pathogen(s) are to be inactivated. But what does inactivation truly mean? Does it mean that every single pathogen that ever passes through the UV system will be inactivated? In reality, this is impossible. Indeed, this is impossible regardless of what disinfection method is used, whether it be UV, chlorine or anything else. What is possible is that the pathogen is reduced by a predictable amount. This predictable amount is referred to as a “log” reduction (as in “Logarithmic” reduction). A “one log” (most commonly referred to as 1 log) reduction will see the pathogen of interest reduced by 90% from the influent level. A 2 log reduction will see a 99% reduction, 3 log by 99.9%, and so on. Scientists have calculated the amount of UV exposure required to inactivate a whole range of different pathogens by various log reductions. Examples appear in Figure 4.
Figure 4 – Excerpt from page 1-7 from the USEPA UV Disinfection Guidance Manual showing log reductions applicable to typical water borne pathogens.
Up to this point, the amount of UVC energy delivered in order that a pathogen be inactivated has been referred to as “UV exposure”. In fact the correct term for this exposure is “UV dose” or more correctly “UV fluence”. Further, the relationship of UV fluence to log reduction as illustrated in Figure 4, is described as a pathogen’s “Dose Response Curve”. As “UV dose” remains the most common expression for UV exposure, this is what will be used for the remainder of this discussion.
It is important to note from Figure 4 that the UV dose required to inactivate a given pathogen to a given log reduction level is rarely linear. A common mistake often made is to take the UV dose required to achieve a 1 log inactivation and simply multiply it in order to calculate a higher log reduction. Although one very common pathogen, E. coli, has a dose response curve that is almost linear, most are not, and so this means of calculating log reduction versus UV dose is not correct.
What is UV Dose?
UV dose is measured in millijoules seconds per cm2 (mJ/cm2) and is calculated using the following parameters
- UV Intensity (I) measured in milliwatts per cm2 (mW/cm2)
- Exposure time (t) (seconds)
In addition, the UV intensity at any point in the reactor isinfluenced by the UVT.
Fig 4 – Dose or Fluence
It is important to understand that actual equations used by UV systems are more complex than this and vary from UV system to UV system to account for UV reactor design differences. The relationship between these parameters can be described in general by the following equation.
(I/UVT) x t = UV dose or UV Fluence
The important thing to understand from this relationship is that UV Intensity (UVI) and UV dose/Fluence are two different things. These two parameters are often (incorrectly) used interchangeably, or one is confused with the other. UVI (Intensity) measures the “amount” of UV energy in the water and varies throughout the reactor . UV dose/Fluence is the amount of UV energy penetrating the water, multiplied by the amount of time the water is exposed to this energy, and it is this that determines the log reduction of the pathogen. UV Intensity is measured by a UV intensity monitor mounted in the reactor. Both of these should not be confused with UVT (Transmissivity) which is the amount of UVI that is adsorbed by the water when UV light travels from the Lamp to the end point (wall) in the reactor.
UV dose is usually quoted as either the “average” dose, computational fluid dynamic dose (CFD) or Reduction Equivalent Dose (RED). Further discussion on the true meaning of RED will follow later.
With all reactors, the delivered dose will cover a range of doses (the Dose Distribution). The narrower the dose distribution, the more efficient the reactor. For any stated dose, there is always some water that will receive less dose and some more. For log kill, the amount of water receiving less dose than the target is critical, just 1% of the water not receiving the target dose for inactivation will limit the performance to 1 log reduction! Average dose, as the name implies, is simply the average throughout the reactor. It takes no account of dose distribution, and so can give a false view of reactor performance. The average dose value will always be higher than an equivalent CFD or RED dose, often by as much as 70%.
CFD can be used to predict an RED dose, which takes into account the dose distribution and the dose response of the target microbe. It uses a computational model to predict particle velocity and path through the reactor. Over many years, Berson have refined their CFD modelling using test data and are able to accurately predict the real log kill performance – however it is still a theoretical calculation.
A Biodosimetric tested RED dose results from the real life tests with actual microbes to measure the reactor performance. It it this technique that is used for all validated UV systems.
How are UV systems validated?
Validating a UV system proves the performance of a given UV system. Individually validating each and every UV system in situ is completely impractical, if not impossible, and so, water treatment authorities world-wide began their search for a means of validating a system design. Over the years, there have been many systems developed aimed at validating UV system designs. The system that is now accepted by the international community as the most appropriate, verifies system performance by “first principles” – the so called “biodosimetric” approach. This approached is based on the use of actual microbes to test the log reduction achieved by a given UV system. Figure 5 provides a representation of the Biodosimetric approach to UV system validation.
Figure 5 – UV Disinfection System Validation using Biodosimetric Principles
In reference to the numbering in Figure 5, system validation using biodosimetric principles proceeds as follows:
1. The microbe of interest is cultured under controlled and reproducible laboratory conditions.
2. This culture is then prepared for testing, again under controlled and reproducible conditions such that:
a. There is a known concentration and/or distribution of the cultured pathogen.
b. The medium in which the culture is to be presented is of a known UVT.
3. The culture is then exposed to a known UV intensity, of known wavelength, for a fixed period of time, delivering a known UV dose to the known area of the presentation plate (hence dose and intensity are measured per cm2 (area) rather than cm3 (volume). The apparatus used to perform this test is called a Collimated Beam apparatus.
4. The exposed content, then is re-cultured to see how many of the microbes survived.
5. This procedure is replicated many times at systematically increasing doses to build a Dose Response Curve. This curve enables the log survival (and by inference, log reduction) for the microbe of interest to be determined for any given UV dose.
· Step 2
6. After the various Dose Response Curves have been constructed in the laboratory, these then need to be applied to test an actual UV system in order that it might be validated. The conditions under which the UV system is tested depends on which international testing protocol is used. More on these protocols later. The important thing is that the microbes used to test the UV system is cultured (albeit in much higher volumes) under exactly the same conditions as was used in the laboratory (I.E. Step 1). This microbe culture is presented to the UV system mixed with the water of known UVT and tested over the range of UVT’s and flow rates of interest.
7. A sample of the water is taken at the entry and exit of the UV system and re-cultured to quantify the reduction of the microbe.
8 Using the observed log reduction of the microbe and comparing it to the microbe’s Dose Response Curve, the UV dose delivered can be determined. This UV dose is termed the Reduction Equivalent Dose – RED.
In the test procedures a challenge microbe such as for instance MS2 is used for health and safety reasons and repeatability.
Fig 6 - Biodosimetry Dose Determination “RED”
Who determines the rules for UV system validation?
Internationally, UV systems are validated against either Drinking Water or Water Reuse protocols. Interestingly, there is as yet no internationally recognized validation protocol for Wastewater.
The internationally recognised validation protocols for Drinking Water are:
- O-Norm (Austrian)
- DVGW (German)
- USEPA (USA – as per the UV Disinfection Guidance Manual – UVDGM)
Of these USEPA & DVGW are the clear leaders. In general, with some exceptions, most countries regulation authorities will accept UV systems validated against an internationally accepted validation protocol, which includes either of these two protocols. Table 1 compares these two validation protocols. It is important to recognize that this table is not intended to be exhaustive, but rather it is intended to compare some of the most fundamental aspects of the two protocols.
Who is permitted to carry out the UV system validation?
Anyone who can prove that the validation protocol outlined in the UVDGM has been followed.
Only a DVGW certified facility.
What is the result of the validation procedure?
A detailed report proving the UVDGM protocol has been followed.
Certification of the validated UV system.
What UV dose is required to achieve validation?
As much as is required to inactivate a given pathogen by a specified log reduction.
40mj/cm2 RED. This is based on the principal that almost all common water borne pathogens will experience at least a 4 log reduction at this dose.
Table 1 – Comparison of key aspects of the USEPA UVDGM & German DVGW validation protocols
Currently, the only internationally recognized Water Reuse validation protocol is:
- USEPA – National Water Reuse Institute (NWRI)
This validation protocol is administered in a similar way to the UVDGM protocol for Drinking Water. The main difference is that the dose required to meet the validation standard is affected by the nature of the pretreatment of the water upstream of the UV system. This introduces the concept of “log credits”. This concept is best illustrated by way of the following example.
Let’s assume that a pathogen requires a 7 log reduction on its passage through a water reuse disinfection system. The filter system in use upstream of the UV system has been validated to provide a 3 log reduction in the pathogen (I.E. it provides a “3 log credit”), therefore the UV system is required to provide only a 4 log reduction to achieve the 7 log target.
Various filter media perform better than others when it comes to providing log credits. In general media filters are less efficient than membrane filters, which are in turn less efficient than, say Reverse Osmosis (RO).
This discussion concerning the principles of UV disinfection and validation was not intended to be exhaustive in its coverage of all aspects of the topic. Rather, it was intended to provide an educational over view of the subject. It is hoped that this review will provide an understanding of what needs to be considered when specifying or selecting a UV system for a given duty. Most importantly, those selecting a UV system must have a clear understanding of:
1. The quality of the water in question, in particular the water’s minimum UVT.
2. The peak, instantaneous minimum and maximum flow rate of the water passing through the UV system.
3. Either the required dose (otherwise known as fluence) OR the log reduction requirement with respect to the pathogen(s) of interest.
What, if any validation is required of the UV system.