Fluence (UV Dose) Required to Achieve Incremental Log Inactivation of Bacteria, Protozoa, Viruses and Algae
Revised, updated and expanded by
Adel Haji Malayeri1, Madjid Mohseni1, Bill Cairns2 and James R. Bolton3
With earlier contributions by
Gabriel Chevrefils (2006)4 and Eric Caron (2006)4
With peer review by
Benoit Barbeau4, Harold Wright (1999)5 and Karl G. Linden6
1. Department of Chemical and Biological Engineering, University of British Columbia, Vancouver, BC, Canada
2. Trojan Technologies, London, ON, Canada
3. Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB, Canada
4. Chaire Industrielle-CRSNG en Eau Potable, Polytechnique Montreal, Montreal, QC, Canada
5. Carollo Engineers, Boise, ID
6. Department of Civil, Environmental and Architectural Engineering, University of Colorado-Boulder
Where's the rest of this article?
This paper represents the second revision of a compilation that goes back to 1999. The original compilation (Wright and Sakamoto 1999) was an internal document of Trojan Technologies. The first revision was published in 2006 (Chevrefils et al. 2006). Data from the previous reviews have been included here. In addition, data from the past 10 years have been added and a new table for algae has been added. Two other reviews of the UV sensitivity of microorganisms have been published (Hijnen et al. 2006; Coohill and Sagripanti 2008).
Brief description and selection criteria for content of the tables
Tables 1-5 (only available in the full review) present a summary of published data on the ultraviolet (UV) fluence-response data for various microorganisms that are pathogens, indicators or organisms encountered in the application, testing of performance, and validation of UV disinfection technologies. The tables reflect the state of knowledge but include the variation in technique and biological response that currently exists in the absence of standardized protocols. Users of the data for their own purposes are advised to exercise critical judgment in how they use the data.
In most cases, the data are generated from low-pressure (LP) monochromatic mercury arc lamp sources for which the lamp fluence rate (irradiance) can be measured empirically and multiplied by exposure time (in seconds) to obtain an incident fluence onto the sample being irradiated; however, earlier data do not always contain the correction factors that are now considered standard practice (Bolton and Linden 2003; Bolton et al. 2015a) in order to determine the average fluence delivered to the microorganisms within the irradiated sample. Such uncorrected data are marked and should be considered as upper limits, since the necessary corrections have not been made. Some data are from polychromatic medium pressure (MP) mercury arc lamps, and in some cases both lamp types are used. In a few cases, filtered polychromatic UV light is used to achieve a narrow band of irradiation around 254 nm. These studies are also designated as LP.
None of the data incorporate any impact of photorepair processes. Only the response to the inactivating fluence is documented. The references from which the data are abstracted must be carefully read to understand how the reported fluences are calculated and what the assumptions and procedures are in the calculations.
It is the intention of the authors and sponsors to keep this table dynamic, with periodic updates. Recommendations for inclusion in the tables, along with the reference source, should be sent to:
Dr. Bill Cairns, chief scientist
Trojan Technologies Inc.
3020 Gore Road
London, ON, Canada N5V 4T7
Prof. James R. Bolton
Department of Civil and Environmental Engineering
Edmonton, AB, Canada T6G 2W2
The selection criteria for inclusion are recommended as follows:
- Data must already be published in a peer-reviewed journal or other peer-reviewed publication media; some exceptions have been allowed where data are only available in non-peer-reviewed papers;
- For the publications where an LP or MP UV lamp was used as the UV source, the calculated fluence should usually be determined by using a collimated beam apparatus; however, for other UV sources, this criterion was not strictly followed and such cases are noted;
- Ideally, the fluence rate (irradiance) should be measured with a recently calibrated radiometer, and when this has not been done, a well-characterized organism should be run as a reference to provide a comparison with the literature values to substantiate that the radiometer is within calibration;
- The publication from which the data are abstracted should describe the experimental procedures including collimated beam procedures, fluence calculation procedures along with any assumptions made, organism culturing procedures, enumeration and preparation for experiments;
- Ideally, as noted above, the protocol published by Bolton and Linden (2003) or the recently published IUVA Protocol (Bolton et al. 2015a) should be followed. In cases where this protocol has not been followed, notes to that effect have been provided. Such data should be considered as an upper limit for the fluence since the normal correction factors have not been applied. In some cases only the water factor has been applied; these are deemed to have met the protocol criterion, since the water factor is the most important correction.
- Responses should be determined over a range of fluences; that is, a complete fluence-response curve is preferred to a single fluence-response measurement.
These criteria will be applied strictly for future editions of these tables.
For the users of these tables, the following points can be helpful in understanding the information provided:
- In some papers, the authors used different methods for enumeration of their selected microorganism and based on that, they reported different fluence-responses in their work compared with the work of others. Where this has happened for a specific paper, a brief description of the implemented method is provided within the box containing the name of the tested microorganism.
- For the studies with UV sources other than an LP lamp (e.g., filtered MP lamps, UV-LEDs, excimer lamps, etc.) the full width at half maximum (FWHM) of wavelength distribution around the peak wavelength is usually about 10-12 nm, except for the tunable laser where the bandwidth is < 1 nm.
- Where the authors have reported kinetic models based on their experimental data, these models were used in fluence calculations for these tables. Where model fits were not provided, the fluence reported for each specific log reduction number was extracted by graphic linearization (Web Plot Digitizer software) between two adjacent experimental data points in the fluence range.
- In some cases, fluence-response curves have been determined at several wavelengths, so that an action spectrum can be determined. These cases are noted as “action spectrum;” however, only data for wavelengths near 254 nm are included in the tables. Data for other wavelengths can be obtained from the cited reference.
- The reader should be aware that for a given microorganism there is a data spread even after the selection criteria have been applied. Some studies have applied a Bayesian statistical analysis (e.g., see Qian et al. 2004, 2005) to obtain an average fluence-response curve and 95 percentile limits. Some of the factors that could affect the reported data are: the medium (e.g., drinking water or wastewater), differences in the nutritional state of the cells being assayed, the presence of particles because of a failure to fully disperse cells following pre-concentration for the collimated beam assay, etc.
- For a given microorganism, the fluence-response curve can depend markedly on the strain examined. This is why studies of a given strain have been grouped together.
- Note that the data in the tables below originate from highly controlled protocols usually using defined media and culture methods, irradiation methods, etc. These data are useful when validating UV technologies and envisioning regulations; however, as water quality, nutritional state, particle content and a number of other factors can impact on microbe responses to disinfection in real environmental samples or processed water, such real waters should be used for site specific assessments of UV, and design specification should benefit from the results of assays using these site-specific waters.
- In some cases, the quality of the data was questionable and did not meet some of the selection criteria listed above. In these cases, the data entries are in italics.
These tables can be used as a helpful document for understanding the fluence-responses for different organisms at different wavelengths, with different UV sources; however, if more details are important for the users of these data, they must read the reference provided for each study.
Units and nomenclature
Throughout this review, fluence rate and irradiance (units mW/cm2) are used interchangeably since they are virtually identical in a collimated beam apparatus. The term fluence (units mJ/cm2) is used, which is the proper term [see Bolton et al. (2015b) for a recommended set of terms and definitions] rather than UV dose, which was used in earlier revisions of this document; however, it should be noted that the term UV dose is still widely used. Finally, it is noted that in Europe and other parts of the world, the units W/m2 for irradiance or fluence rate and J/m2 for fluence (UV dose) are more commonly used. One mW/cm2 = 10 W/m2 and 1 mJ/cm2 = 10 J/m2.
Five tables have been prepared covering spores, bacteria, viruses, algae and other microorganisms. These tables – as well as a reference list – are too large for print, but the full review can be downloaded from the Member Zone on the IUVA website at www.iuva.org.