By Rick Andrew
Distillation as a water treatment technique has been used for hundreds of years. It can be effective in the reduction of total dissolved solids (TDS) in general, as well as that of specific minerals and heavy metals. Distillation can also be an effective method for microbiological purification of water.
NSF/ANSI Standard 62 contains test methods for assessing the effectiveness of residential distillers in reducing these contaminants. As with all of the NSF/ASNI Drinking Water Treatment Unit (DWTU) Standards, the NSF DWTU Joint Committee has structured these test methods to challenge distillation technology specifically. This approach of technology-specific test methods results in different test conditions for different types of water treatment. These methods ensure that products conforming to the requirements have been evaluated in a challenging, conservative manner.
Because distillation as a technology and process has been widely known and studied, the contaminant reduction test methods in Standard 62 are straightforward and easy to understand. This article will briefly describe these methods.
A surrogate approach
NSF conducted a study in 1991 to assess the potential to use TDS as a surrogate for a variety of inorganic ionic contaminants, shown in Table 1. 
These specific contaminants were chosen because their mechanism of reduction by distillation is the same as the mechanism for reduction of sodium chloride, chosen as the surrogate for TDS.
The study involved conducting contaminant reduction tests with two different distiller systems. One of the test systems was a 1.5-gallon batch system capable of producing three batches per day. The other was an automatic unit capable of producing 12 gallons per day in a 12-gallon reservoir. Both systems had all carbon filters removed so that only distillation performance was evaluated.
Batches of water containing all of the contaminants in Table 1, along with sodium chloride, were processed through each distiller system at pH 5 and pH 8. Deionized water was used to prevent precipitation of certain ions with certain anions found in Ann Arbor, Mich., tap water. The deionized water was chlorinated to allow potentially volatile chlorides of any of the ions to form and possibly carry over into the product water.
The results of this study indicated that TDS was an effective surrogate for all of these inorganic ionic contaminants except fluoride and mercury 
(see Table 2), which had considerably poorer percentages of reduction under certain conditions than did TDS, causing them to be eliminated from inclusion in the TDS surrogate list.
Claims for reduction of fluoride and mercury must be specifically tested on the system in question, regardless of its performance in the TDS reduction test.
Conducting the TDS reduction test
NSF/ANSI 62 requires that the system, with all carbon filters removed, reduce a challenge of 1,000 mg/L sodium chloride in deionized water by at least 99 percent at all sample points throughout the course of the test. For batch systems, the operational cycle of the test is per Table 3. 
For systems that generate a maximum of one batch per day, the test protocol is lengthened to 12 days.
Automatic systems are operated similarly to batch systems, with the two-day stagnation period on days five and six, except that one-gallon samples are collected every six hours of operation. If the production rate is such that more than one gallon is collected every six hours, any product water over and above the first gallon is discarded.
Microbiological reduction claim
The organism Bacillus subtillus is used to establish conformance for claims of microbiological reduction of residential distillers. Chlorine free water with 200-500 mg/L TDS and 106 spores of Bacillus subtillus is used as the challenge water. Fewer than 10 colony-forming units of the organism are permitted in each of four product water samples collected throughout the course of one operational cycle of the test unit.
Performance indication
Manufacturers who wish to make claims on distillers for the reduction of health contaminants (i.e., anything above and beyond TDS reduction) must provide a method for consumers to determine that the system is functioning properly. Manufacturers have the following options, per Standard 62:
- Providing a sampling kit for analysis of TDS or other appropriate contaminants
- Providing a TDS monitor
- Having the system terminate discharge of treated water
- Sounding an alarm or flashing a light
- Providing the operator with an obvious, readily interpretable indication that the system is not functioning properly
- Providing a sampling service by the manufacturer, either directly or through an authorized dealer, at least once every six months
Manufacturers wishing to make claims of nitrate reduction or microbiological purification must provide for an automatic shutdown or diversion of distillate to waste if a system failure occurs.
Systems that use a TDS monitor to trigger a performance indication warning, or to provide for automatic shutdown or diversion of distillate to waste, must be tested for the sensitivity of the monitor. A 500 mL solution of 10 mg/L TDS is injected into the condenser coil and the performance indication warning must be triggered, or automatic shut down or diversion of distillate to waste must occur.
This TDS monitor test has proven to be very difficult to pass. Ten mg/L TDS is a very low threshold for these monitors to differentiate from the usual level of performance of the system. This difficulty has limited the ability of manufacturers to make claims of nitrate reduction and microbiological performance. As a result, a task group is working to revise the TDS monitor test requirements.
Conclusion
Compared to some of the other NSF/ANSI DWTU Standards, the test methods for residential water distillation systems are relatively simple to understand and to implement. This is reflective of the fact that the technology is well established and its mechanisms are well understood. An NSF study has proven valuable in simplifying and reducing the Standard 62 test requirements to make a number of claims for reduction of inorganic contaminants. Despite these straightforward aspects of the Standard, it still needs work to address the difficulties with the current TDS monitor test. A dedicated task group of volunteers is working to resolve that issue so the full potential of distillers can be recognized in conformance with the Standard.
About the author
Rick Andrew has been with NSF International for over six years, working with certification of residential drinking water products. He has been the Technical Manager of the Drinking Water Treatment Units Program for over three years. His previous experience was in the area of analytical and environmental chemistry consulting. Andrew has a bachelor’s degree in chemistry and an MBA from the University of Michigan. He can be reached at 1-800-NSF-MARK or email: [email protected]