By Douglas Frederick
Over the last few years, more and more authorities are requiring that drinking water treatment chemicals meet ANSI/NSF Standard 60: Drinking Water Additives. Not only do the manufacturers of a particular chemical need to meet the standard, but it’s often required that the re-packagers and bulk distributors also carry the certification on their own to sell their product. Often, the distributors down the supply chain don’t know how these products are evaluated. The product evaluation consists of a formulation review, testing the product in the laboratory, normalization of the results and comparison of the results to the pass/fail criteria.
The initial step in the product evaluation process is to gather the information on the product while answering a few basic questions—what are the ingredients, where do the ingredients come from, and what is the proposed field use level? This information will be used in the formulation review. A blended phosphate will be used as an example.
For most of the common drinking water additives, there’s already a prescribed use level; some of the levels are based on chemicals themselves having a component that directly limits its use in the field. Others have been established by a survey of common industrial use levels, and others are blends or chemicals that need to have a use level verified as part of the formulation review.
The proposed, maximum-use level of the blended phosphates is determined by the available phosphate with the limit being set at 10 milligrams of phosphate per liter (PO4/L) in the final drinking water. The sum of each contributor needs to be less than 10 mg PO4/L. This is calculated using the equation: (number of available PO4) × (the molecular weight, MW, of PO4) ÷ total molecular weight of PO4 = the amount of PO4 contributed. This value is used in the equation (maximum use of blend) × (amount of PO4 contributed) × (percent of contributor in blend) = mg of PO4 contributed. The blend is a mixture of 10 percent sodium tripolyphosphate (STPP), 10 percent sodium acid pyrophosphate (SAPP), 20 percent phosphoric acid, and the balance of water with a suggested maximum use level of 28 mg/L. The resulting PO4 contribution is as follows:
(Na5P3O10) MW = 368 has three available PO4 (MW = 95)
28 mg/L suggested max use level and 10 percent
(3)(95)/368 = 0.774
(28)(0.774)(0.1) = 2.17 mg PO4
(Na2 H2P2O7) MW = 222 has two available PO4 (MW = 95)
28 mg/L suggested max use level and 10 percent
(2)(95)/222 = 0.856
(28)(0.856)(0.1) = 2.39 mg PO4
(H3PO4) MW = 98 has one available PO4 (MW = 95)
28 mg/L suggested max use level and 20 percent
(1)(95)/98 = 0.969
(28)(0.969)(0.2) = 5.43 mg PO4
This blended phosphate contributes a total PO4 of 9.99 mg/L—2.17 + 2.39 + 5.43 = 9.99. Therefore, this maximum use level is acceptable for this product. There are direct water additives that don’t contain a limiting compound such as sodium hydroxide. Their use levels are based on common industrial usage. As a consequence, the use levels could be higher if the contaminants permit a higher rating.
Formulating the test
The next step is to test the product. The results of this testing aren’t assay of the material, but an exposure typically of two to 100 times the use level diluted to 1 liter. The test parameters are established for a particular chemical by the standard and the results of the formulation review. In the case of our example, regulated metals, radionuclides and fluoride are the prescribed tests for a blended phosphate. The sample would be prepared using a 28-mg/L use level × 10 times overexposure yielding a solution of 280 mg of the sample in 1 liter. This is the sample that’s analyzed for the aforementioned parameters.
Once the data are collected from the instruments, it’s normalized to the “at the tap” concentration. This is the level that’s expected to be present if the product is properly used. There are two parts to this normalization factor—the laboratory-prepared solution and the maximum use level. The laboratory prepared solution is the actual weighed sample. The (laboratory prepared solution) × (the maximum use level) yields the general normalization formula:
In the case of blended phosphate, the laboratory analyst actually weighed out 295 mg of blended phosphate in a 1-liter sample. The normalization factor would be (1)/295 × (28)(1) = 0.095. This normalization factor then would be multiplied by the contaminant concentration to obtain the normalized result.
After the sample is analyzed and the results are normalized, they are compared to the Single Product Allowable Concentration (SPAC). The SPAC is the maximum concentration of a contaminant in drinking water that a single product is allowed to contribute under Annex A of ANSI/NSF Standard 60. For regulated contaminants, the SPAC is 10 percent of the maximum contaminant level (MCL). The MCL is defined as the maximum concentration of a contaminant permitted in a public drinking water supply as defined by the U.S. Safe Drinking Water Act.
Using the blended phosphate sample, the instrument detected 1.2 parts per billion (ppb)—or micrograms per liter (µg/L)—of lead and 5.4 ppb of chromium. The normalized values are 0.11 ppb and 0.51 ppb for lead and chromium, respectively. The MCL for lead is 15 ppb and the MCL for chromium is 100 ppb; therefore, their respective SPACs are 1.5 ppb for lead and 10 ppb for chromium. In the example, there were no other detected analytes and, thus, this product meets the requirements of the standard.
To further clarify the difference between the SPAC and MCL, here is one more example. The product, sodium hypochlorite, is analyzed for regulated metals, volatile organic compounds (VOCs) and bromate (see Table 1). The normalization factor for the solution is 0.09. Two compounds were detected in the VOC scan—2,3-dichloro-1-propanol at a concentration of 15 ppb, and 1,3 di-t-butyl benzene at a concentration of 2 ppb. Neither compound is a regulated compound so no MCL exists. The normalized concentrations are 1.4 ppb for 2,3-dichloro-1-propanol and 0.2 ppb for 1,3 di-t-butyl benzene. There’s an existing SPAC for 2,3-dichloro-1-propanol at 9 ppb, but no SPAC for 1,3 di-t-butyl benzene; therefore, 1,3 di-t-butyl benzene must be sent for further toxicological review per Annex A of the standard to see if this compound can be cleared to a SPAC of 0.3 ppb. If there’s sufficient information on 1,3 di-t-butyl benzene to clear it to a SPAC of 0.3 ppb, the product would pass. If there isn’t sufficient information, the product would fail.
About the author
Douglas Frederick has been an environmental chemist with Underwriters Laboratories for seven years. He has a bachelor’s degree in chemistry from Northern Illinois University. Frederick can be contacted at (847) 515-1557, (847) 313-2231 (fax) or email: firstname.lastname@example.org