By Rick Andrew

Ultraviolet (UV) water treatment systems use UV energy to break bonds in the DNA of microorganisms, which prevents them from reproducing and causing infection. Over the years, a number of research studies have been conducted to establish the energy, or UV dose, required for UV disinfection to be effective on many waterborne pathogens, including bacteria, viruses and cysts. A concise summary of the relevant studies considered in setting the requirements can be found in the NSF/ANSI 55 Foreword.

Based on the results obtained in these studies, it is generally agreed that almost all pathogenic microorganisms can be adequately disinfected with treated with UV dose of at least 40 mJ/cm2 at 254 nm. Likewise, it has been established that UV dose of 16 mJ/cm2 at 254 nm is sufficient for control of non-pathogenic nuisance bacteria. These UV dose requirements have been adopted into NSF/ANSI 55 for POU and POE UV systems for Class A and Class B disinfection, respectively.

UV dosage versus log reductions
Microbiological disinfection technologies are usually evaluated based on log reductions of microorganisms. As discussed in the Foreword, one of the major bases for log reductions has been the Guide Standard and Protocol for Testing Microbiological Water Purifiers, published by US EPA in 1987. This Guide Standard lays out log reductions for microbiological water purifiers utilizing various treatment technologies such as mechanical filtration and chemical disinfection, as described in Figure 1.

UV systems, however, are often specified in terms of energy output, or UV dosage. For this reason, it was decided to develop the requirements of NSF/ANSI 55 in terms of UV dosage. So, the approach was to use the log reductions to UV dosage correlations developed through the research studies cited in the Foreword to establish the UV dosage of 40 mJ/cm2 as being sufficient for disinfection and 16 mJ/cm2 as being sufficient for control of nuisance organisms.

A new protocol
The original NSF/ANSI 55 test protocol was established around low-pressure mercury UV emission, which has been the predominant technology for POU/POE UV disinfection for many years. Low-pressure mercury UV lamps emit radiation very specifically at 254 nm, which is very close to the ideal wavelength for disinfection. In recent years, however, there has been significant development of alternate technologies that can emit at different wavelengths from 254 nm. Recognizing these new technological developments, in 2019 a new protocol was added to NSF/ANSI 55 to evaluate UV systems emitting UV radiation at wavelengths other than 254 nm.

To simplify the evaluation of these systems, this new protocol departed from the UV dosage approach and instead uses a direct log reduction of the Q beta microorganism used for testing. When this alternate protocol was developed, there was limited data correlating the log reduction of Q beta to UV dosage, but it was at that time determined that the data was sufficient to establish log-reduction requirements for the standard. At the October, 2020, meeting of the NSF Joint Committee on Drinking Water Treatment Units, however, additional data was reported and shared. This data included dose-response values establishing the relationship between UV dosage and log reduction of Q beta compiled from several labs over the past year using low-pressure systems. This data showed more variance than expected.

The data also indicated that the log-reduction requirements for Class A and Class B qualification should be revisited. Accordingly, a task group working under the auspices of the Joint Committee was formed. The task at hand was to confirm the appropriate log-reduction level for the Q beta organism based on correlation to dosage of at least 40 mJ/cm2 at 254 nm for Class A systems and 16 mJ/cm2 for Class B systems.

Upon review of the extensive dose-response data from the 2015-2020 time period that was shared, the task group determined that the original 4.00 log reduction (Class A) and 2.14 log reduction (Class B) requirements were too high and therefore overly conservative. The task group concluded that 3.5 log reduction of Q beta correlates to equal or greater than 40 mJ/cm2 dose with a 95 percent statistical confidence level based on historical does-response studies. Therefore, the task group recommended that the Q beta log-reduction requirement for Class A systems be revised to 3.50. The task group also recommended that the Class B Q beta log-reduction requirement be revised from 2.14 to 2.00. These recommended changes were developed into a proposal to revise the standard. This proposal is summarized in Figure 2.

Next steps
The next steps in the process are for the Joint Committee to consider the recommendations of the task group to revise NSF/ANSI 55 according to their recommendations. This is done through a vote based on a ballot describing the proposed changes, the reasons for the changes and the technical rationale providing justification for the validity of the changes. This process can be iterative if some Joint Committee members offer suggestions to improve the recommended changes, so it can take several months to complete. Once the process is complete, provided that consensus can be reached among the Joint Committee members that the changes are beneficial, a revised NSF/ANSI 55 including the changes will be published.

Continuous improvement
The NSF/ANSI DWTU Standards are based on the best available technical data, research information and testing and evaluation practices. As newer and better information and data become available, the Joint Committee reviews and evaluates the new information to assess potential updates to the standards to assure that they continue to be state of the art. This continuous improvement process results in new versions of standards being published nearly every year, sometimes containing multiple updates to improve them in several different aspects.

Users of these standards, including manufacturers designing and producing treatment products to conform to them, regulators relying on them to help assure compliance and consumers purchasing products independently certified to them to be assured of product safety and performance, can rest assured that the best available science is being incorporated into their requirements.

About the author
Rick Andrew is NSF’s Director of Global Business Development – Water Systems. Previously, he served as General Manager of NSF’s Drinking Water Treatment Units (POU/POE), ERS (Protocols) and Biosafety Cabinetry Programs. Andrew has a Bachelor’s Degree in chemistry and an MBA from the University of Michigan. He can be reached at (800) NSF-MARK or email:


Comments are closed.