4 NSF Fluoride Fact Sheet 2019.pdf

link to page 7 link to page 9 link to page 9

Published January 14, 2019

Fact Sheet on Fluoridation Products and Fluoride

Background

NSF International (www.nsf.org) is an independent, global not-for-profit organization that facilitates
standards development, and tests and certifies products for the food, water, health sciences and
consumer goods industries to minimize adverse health effects and protect the environment. Founded in
1944, NSF is committed to protecting human health and safety worldwide. NSF International is a Pan
American Health Organization/World Health Organization (WHO) Collaborating Center on Food Safety,
Water Quality and Indoor Environment.
NSF/ANSI 60

NSF/ANSI 60: Drinking Water Treatment Chemicals - Health Effects was developed to establish minimum
requirements for the control of potential adverse human health effects from products added directly to
water during its treatment, storage and distribution. The standard requires a full formulation disclosure
of each chemical ingredient in a product to allow for a comprehensive evaluation of the products and
their ingredients. The standard requires testing of the treatment chemical products, typically by dosing
these in water at 10 times the maximum use level (MUL), so that trace levels of contaminants can be
detected. A further evaluation of test results is required to determine if the concentrations of any
detected contaminants have the potential to cause adverse human health effects. When health effects
criteria have not been established for a given product or contaminant, the standard requires that health
effects criteria be derived according to the requirements of Annex A prior to approval under this
standard. For More Information on NSF/ANSI 60…

Product Certification

NSF’s testing and certification program for drinking water treatment products was developed in the late
1980s to ensure that individual U.S. states and waterworks facilities have a mechanism to determine
which products are most suitable for use. The NSF certification program requires annual, unannounced
inspections of production and distribution facilities to ensure that the products are properly formulated,
packaged and transported with appropriate safe guards in place to protect against potential
contamination. NSF also requires annual testing and evaluation of each NSF certified product to confirm
contaminants do not exceed drinking water health effects criteria. For More Information on Product
Certification…

Fluoride

Fluoride Properties and Water Applications

Water fluoridation is the practice of adjusting the fluoride content of drinking water. Fluoride is added
to water for the public health benefit of greatly reducing the incidence of tooth decay and therefore
improving the health of the community. For more information please visit the U.S. Centers for Disease
Control and Prevention: www.cdc.gov/fluoridation/.

NSF certifies three products in the fluoridation category:
1.  Fluorosilicic acid (aka fluosilicic acid or hydrofluosilicic acid)
2.  Sodium fluorosilicate (aka sodium silicofluoride)
3.  Sodium fluoride

All  three  products  readily  dissociate  in  water  to  release  fluoride  and  related  ions.  In  the  case  of  the
fluorosilicates,  they  fully  dissociate  to  fluoride  and  silicate  ions  in  association with  either  hydrogen or
sodium ions1. In the case of sodium fluoride, it fully dissociates to form fluoride and sodium ions.

Treatment products that are used for fluoridation are specifically addressed in Section 7 of NSF/ANSI 60.
The standard requires that the treatment products added to drinking water, as well as any impurities in
the products, are supported by an evaluation of potential health effects resulting from exposure to the
products or associated contaminants.  The following text explains the allowable levels established in the
standard  for  1)  fluoride,  2)  silicate  and  3)  other  potential  contaminants  that  may  be  associated  with
fluoridation chemicals.

Fluoride Drinking Water Criteria

NSF/ANSI 60 requires, when available, that the U.S. or Canadian regulatory values be used to determine
the acceptable level for a chemical of interest.  The EPA MCL for fluoride ion in water is 4 mg/L and the
Health  Canada  MAC  for  fluoride  is  1.5  mg/L  to  protect  against  skeletal  fluorosis.    However,  in  2003,
representatives  from the EPA and Health Canada reviewed the  evaluation criteria in  NSF/ANSI  60 and
recommended  that  the  TAC  and  SPAC  for  fluoride  ion  in  drinking  water  contributed  by  fluoridation
products be lowered to 1.2 mg/L to match the upper bound of the U.S. Public Health Service/Centers for
Disease  Control  recommended  fluoride  concentration  range  to  prevent  dental  fluorosis  (an  aesthetic
effect).  The Drinking Water Additives Joint Committee for NSF/ANSI 60 supported this recommendation
and set the acceptance criteria for fluoride ion equivalent to the maximum recommended fluoride ion
dose.  For treatment chemicals other than fluoridation products, the allowable fluoride contribution is
0.12 mg/L, or a tenth the TAC.

In  2015,  the  U.S.  Department  of  Health  and  Human Services  recommended  that  the optimal  range  of
water fluoridation of 0.7 to 1.2 ppm (mg/L) be updated to an optimal concentration of 0.7 ppm (mg/L)
due to observations of increasing amounts of fluoride in food that is processed with fluoridated drinking
water. Some U.S. states have elected to adopt this optimal concentration for fluoridation of community
water supplies. Testing these chemicals at the higher use level of 1.2 ppm (as is currently done) provides
a more conservative screening for contaminants in or associated with use of these products.

More  recently,  the  U.S.  Centers  for  Disease  Control  and  Prevention  (CDC)  has  proposed  that  the
recommended  operational  range  for  community water  supplies  be  narrowed  to 0.6  mg/L  to  1.0 mg/L
fluoride  ion2.  The  CDC  advised  that  this  change  is  to  reflect  demonstrated  ability  of  water  utilities  to
maintain  control of  fluoride  concentration  around  the  0.7  mg/L  optimum  concentration.  The  Drinking

1 Finney WF, Wilson E, Callender A, Morris MD, Beck LW. Reexamination of hexafluorosilicate hydrolysis by fluoride
NMR and pH measurement. Environ Sci Technol 2006;40:8:2572)
2 Centers for Disease Control and Prevention. Proposed Guidance Regarding Operational Control Range Around
Optimal Fluoride Concentration in Community Water Systems that Adjust Fluoride. Federal Register Vol. 83, No.
135.
2

Water Additives Joint Committee for NSF/ANSI 60 is reviewing whether it should lower the fluoride ion
typical use level from 1.2 mg/L to 1.0 mg/L. If this request is approved by the Drinking Water Additives
Joint Committee, it will be incorporated into NSF/ANSI 60 in 2019.

Silicate Drinking Water Criteria

Fluorosilicates do not  require  a toxicological assessment specifically for the fluorosilicate ion, because
measurable levels of this ion do not exist in potable water at the fluoride concentrations and pH levels
typical of public drinking water.3 There is currently no U.S. EPA-derived MCL or Health Canada MAC for
silicate in drinking water.  The current sodium silicate typical use level (TUL) of 16 mg/L is listed in Table
5.1 of NSF/ANSI 60, which was based on the value for sodium silicate published in the Water Chemicals
Codex.4    A  fluorosilicate  product,  applied  at  its  maximum  use  level  (noted  below),  results  in  silicate
drinking water levels that are substantially below the 16 mg/L TUL.  For example, a sodium fluorosilicate
product dosed at a concentration into drinking water that would provide the maximum concentration of
fluoride currently permitted by NSF/ANSI 60 (1.2 mg/L) would only contribute 0.8 mg/L of silicate – or 5
percent of the TUL allowed by NSF/ANSI 60 for silicate.

Fluoridation Product Use Levels

Allowable use levels for fluoridation chemicals are limited by the NSF/ANSI 60 limits for fluoride ion.  Per
the previous section, the silicate contribution is not a limiting factor.  The allowable maximum use levels
(MUL) for NSF/ANSI 60 certified fluoridation products in 2018 are:

1.    Fluorosilicic acid: 6 mg/L
2.  Sodium fluorosilicate: 2 mg/L
3.  Sodium fluoride: 2.3 mg/L

Potential Contaminants in Fluoridation Chemicals

The product review conducted by NSF for a water treatment product considers all chemical ingredients
in  the  product,  as  well  as  the  manufacturing  process,  processing  aids  and  other  factors  that  have  an
impact  on  the  chemicals  attributable  to  the  products  present  in  the  finished  drinking  water.    The
identified chemicals of interest are subsequently evaluated during testing of the product.  For example,
fluosilicic  acid  is  produced  by  adding  sulfuric  acid  to  phosphate  ore.    This  is  typically  done  during  the
production of phosphate additives.   The manufacturing process is documented by an NSF auditor at the
initial  audit  of  the  manufacturing  site  and  during  each  subsequent  annual  unannounced  audit  of  the
facility.  The manufacturing process, ingredients and potential contaminants are reviewed annually, and
the  product  is  tested  for  any  potential  contaminants  of  interest.  A  minimum  test  battery  for  all
fluoridation  products  includes  heavy  metals  of  toxicological  concern  and  radionuclides  because  they
may be contained in phosphate ore.

Many  drinking  water  treatment  additives,  including  fluoridation  products,  are  transported  in  bulk  via
tanker  trucks  to  terminals  where  they  are  transferred  to  rail  cars,  shipped  to  distant  locations  or
transferred into tanker trucks, and then delivered to the water treatment plants. These tanker trucks,

3 Finney WF, Wilson E, Callender A, Morris MD, Beck LW. Reexamination of hexafluorosilicate hydrolysis by fluoride
NMR and pH measurement. Environ Sci Technol 2006;40:8:2572)
4 National Academy of Sciences, Water Chemicals Codex, 1982

3

transfer  terminals  and  rail  cars  are  potential  sources  of  contamination.  Therefore,  NSF  also  inspects,
samples, tests and certifies products at rail transfer and storage depots.  It is always important to verify
that the location of the product distributor (the company that delivers the product to the water utility)
matches that shown in the official NSF listing for the product (available at www.nsf.org).

NSF has compiled data on the levels of contaminants found in or through use of all fluoridation products
that have applied for, or have been approved for, certification by NSF under NSF/ANSI 60. The results in
Tables 1, 2, and 3 include those from the initial and annual monitoring tests for fluoridation products
that NSF certified to NSF/ANSI 60 from 2012 to 2017 (Table 1), 2007 to 2011 (Table 2) and 2000 to 2006
(Table 3). This summary includes 245 separate samples analyzed during the time period of 2000 to 2006,
216 samples in the period of 2007 to 2011, and 328 samples in the period of 2012 to 2017. The
concentrations reported represent contaminant levels expected when the products are dosed into
water at the manufacturer’s maximum use level (MUL). For the time periods summarized below, the
typical product certification and evaluation were based on a fluoride ion dose of 1.2 mg/L.  Lower
product use levels would produce proportionately lower contaminant concentrations (e.g. a 0.6 mg/L
fluoride dose would produce one half the contaminant concentrations listed in Table 1).

The data reported in Tables 1, 2, and 3 demonstrate that very low concentrations of contaminants are
associated with fluoridation chemicals. In fact, NSF was only able to detect the reported trace amounts
by dosing the chemicals into reagent water at 10 times the manufacturer’s maximum use level (as
required by NSF/ANSI 60) and then mathematically adjusting the laboratory results to expected field
dose. If the products had been dosed into water at the manufacturer’s maximum use level, only one
copper contaminant concentration would have been above the analytical method detection limits. This
is demonstrated by comparing the results in columns 3 and 4 with the detection limits in column 5. The
low concentrations of contaminants documented for this most current time period are explained, at
least in part, by the ongoing effectiveness of NSF/ANSI 60 and the NSF certification program for drinking
water treatment additives. The levels in Table 1 are comparable to those documented earlier for the
2000-2011 time period, which is further attested to by a 2004 article in the Journal of the American
Water Works Association entitled, “Trace Contaminants in Water Treatment Chemicals.”3

Summary
All the fluoridation products tested by NSF, when evaluated at their maximum use level in water, meet
the health effects requirements of NSF/ANSI 60.  Arsenic was periodically detected in half of all samples.
However,  the  mean  arsenic  concentration  is  1/50th  of  the  U.S.  EPA  MCL  and  none  of  the  samples
exceeded 1/10th the U.S. EPA MCL. The majority of fluoridation products certified by NSF do not contain
detectable concentrations of lead, radionuclides or other heavy metals when dosed into water at their
maximum use level.  In summary, fluoridation products certified by NSF do not contribute a significant
contaminant burden to drinking water.

4

Table 1. Fluoridation product chemical contaminant analysis from 2012-2017 (328 samples)

Percentage
Mean1
Maximum
Analytical
NSF/ANSI 60
U.S. EPA
of Samples
Contaminant
Contaminant
Method
Single Product
Maximum
With
Concentration
Concentration
Detection
Allowable
Contaminant
Detectable
in All Samples
in Detectable
Levels
Concentration
or Action
Levels
(ppb)
Samples (ppb)
(ppb)
(ppb)
Level (ppb)

Antimony
<1%
0.022
0.07
0.5
0.6
6
Arsenic
52%
0.16
0.9
1
1
10
Barium
<1%
0.045
0.5
1
200
2000
Beryllium
0%
ND
ND
0.5
0.4
4
Cadmium
0%
ND
ND
0.2
0.5
5
Chromium
<1%
0.044
0.2
1
10
100
Copper
6%
0.061
2.8
1
130
1300
Lead
5%
0.037
0.3
0.5
1.53
15
Mercury
<1%
0.0088
0.02
0.2
0.2
2
Radionuclides
0.12 pCi/L
0.5 pCi/L
3 pCi
1.5 pCi/L
15 pCi/L
<1%
– alpha
Radionuclides
0.16 pCi/L2
0.6 pCi/L2
4 pCi
0.4 mrem/yr2
4 mrem/yr2
<1%
– beta
Selenium
1%
0.096
1.1
2
5
50
Thallium
<1%
0.0088
0.02
0.2
0.2
2
1Mean values were calculated by setting the non-detectable (ND) values to half of the method detection level.
2The corrected gross beta detection in radioactivity (pCi) per volume of water (L) is compared against the U.S. EPA
limit of 4 mrems/yr for beta particle emitters in drinking water. A mrem is a dose equivalent unit from ionizing
radiation to an organ, organ system or the total body. The conversion factor from pCi/L to mrem is based on the
strength of the beta radiation from a particular beta emitting isotope. The beta detections above were determined
to be less than the 0.4mrem/yr limit of NSF/ANSI 60.
3The NSF/ANSI 60 single product allowable concentration (SPAC) for lead was lowered from 1.5 µg/L to 0.5 µg/L in
the published 2017 edition of NSF/ANSI 60. However, this standard edition was published at the end of 2017 and
all the lead results summarized in Table 1 were compared against the 1.5 µg/L SPAC.

5

Table 2. Fluoridation product chemical contaminant analysis from 2007-2011 (216 samples)

Percentage
Mean1
Maximum
Analytical
NSF/ANSI 60
U.S. EPA
of Samples
Contaminant
Contaminant
Method
Single Product
Maximum
With
Concentration
Concentration
Detection
Allowable
Contaminant
Detectable
in All Samples
in Detectable
Levels
Concentration
or Action
Levels
(ppb)
Samples (ppb)
(ppb)
(ppb)
Level (ppb)

Antimony
0%
ND
ND
0.5
0.6
6
Arsenic
50%
0.15
0.6
1
1
10
Barium
2%
0.042
0.6
1
200
2000
Beryllium
0%
ND
ND
0.5
0.4
4
Cadmium
0%
ND
ND
0.2
0.5
5
Chromium
<1%
0.039
0.3
1
10
100
Copper
<1%
0.039
0.091
1
130
1300
Lead
<1%
0.037
0.088
1
1.5
15
Mercury
0%
ND
0.04
0.2
0.2
2
Radionuclides
<1%
0.18 pCi/L
0.3 pCi/L
3 pCi/L
1.5 pCi/L
15 pCi/L
– alpha
Radionuclides
<1%
0.19 pCi/L2
0.5 pCi/L2
4 pCi/L
0.4 mrem/yr2
4 mrem/yr2
– beta
Selenium
0%
ND
ND
2
5
50
Thallium
<1%
0.0079
0.01
0.2
0.2
2
1Mean values were calculated by setting the non-detectable (ND) values to half of the method detection level.
2The corrected gross beta detection in radioactivity (pCi) per volume of water (L) is compared against the U.S. EPA
limit of 4 mrems/yr for beta particle emitters in drinking water.  A mrem is a dose equivalent unit from ionizing
radiation to an organ, organ system, or the total body. The conversion factor from pCi/L to mrem is based on the
strength of the beta radiation from a particular beta emitting isotope. The beta detections above were determined
to be less than the 0.4mrem/yr limit of NSF/ANSI 60.

6

Table 3. Fluoridation product chemical contaminant analysis from 2000-2006 (245 samples)

Percentage
Mean1
Maximum
Analytical
NSF/ANSI 60
U.S. EPA
of Samples
Contaminant
Contaminant
Method
Single Product
Maximum
With
Concentration
Concentration
Detection
Allowable
Contaminant
Detectable
in All Samples
in Detectable
Levels (ppb)
Concentration
or Action
Levels
(ppb)
Samples (ppb)
(ppb)
Level (ppb)

Antimony
0%
ND
ND
0.5
0.6
6
Arsenic
43%
0.12
0.6
1
1
10
Barium
<1%
0.001
0.3
1
200
2000
Beryllium
0%
ND
ND
0.5
0.4
4
Cadmium
1%
0.001
0.12
0.2
0.5
5
Chromium
<1%
0.001
0.2
1
10
100
Copper
3%
0.02
2.6
1
130
1300
Lead
2%
0.005
0.6
1
1.5
15
Mercury
<1%
0.0002
0.04
0.2
0.2
2
Radionuclides
0%
ND
ND
3 pCi/L
1.5 pCi/L
15 pCi/L
– alpha
Radionuclides
0%
ND
ND
4 pCi/L2
0.4mrem/yr2
4 mrem/yr2
– beta
Selenium
<1%
0.016
3.2
2
5
50
Thallium
<1%
0.0003
0.06
0.2
0.2
2
1Mean values were calculated by setting the non-detectable (ND) values to half of the method detection level.
2The corrected gross beta detection in radioactivity (pCi) per volume of water (L) is compared against the U.S. EPA
limit of 4 mrems/yr for beta particle emitters in drinking water.  A mrem is a dose equivalent unit from ionizing
radiation to an organ, organ system, or the total body. The conversion factor from pCi/L to mrem is based on the
strength of the beta radiation from a particular beta emitting isotope.

Additional information on fluoridation of drinking water can be found on the following websites:

American Dental Association (ADA)

https://www.ada.org/en/public-programs/action-for-dental-health/prevention-and-education

U.S. Centers for Disease Control and Prevention (CDC)
http://www.cdc.gov/fluoridation
7

More Information on NSF/ANSI 60

The standard requires that the treatment products added to drinking water, as well as any impurities in
the products, are supported by an evaluation of potential health effects resulting from exposure to the
products and/or associated contaminants during the anticipated use(s) of the product. It is a
requirement of NSF/ANSI 60 (Annex A) that a determination be made as to whether a published and
peer-reviewed quantitative risk assessment for the substance is available. If the substance is regulated
by the U.S. EPA or Health Canada, these regulatory limits are to be used to derive the drinking water
criteria for the regulated chemical of interest. If the identified published assessment is not the basis of a
drinking water regulation, the assessment and its corresponding reference dose must be reviewed for its
appropriateness in evaluating the human health risk of the substance.  If a published assessment is not
identified or an identified assessment is deemed unacceptable, it is required that a comprehensive risk
assessment be conducted from which drinking water criteria are derived.  The non-regulatory drinking
water criteria derived from a risk assessment are known as total allowable concentrations (TAC), as
stipulated in Annex A.

Single Product Allowable Concentrations (SPAC)
Guidance for derivation of drinking water criteria known as single product allowable concentrations
(SPAC) is provided in Annex A of NSF/ANSI 60. The SPAC is defined as the maximum concentration of a
contaminant in drinking water that a single product can contribute.  For contaminants regulated by the
U.S. EPA or Health Canada, the SPAC is set to a default level that is not to exceed 10 percent of the
regulatory level to address the possibility that multiple sources of the contaminant exist in the water
supply. For contaminants that are not regulated, the SPAC is derived from the total allowable
concentration (TAC), which is defined as the maximum concentration of a non-regulated contaminant
permitted in public drinking water supply. The default SPAC in these instances is one-tenth of the TAC
to account for the possibility that more than one product in the water and/or its distribution system
could contribute the contaminant of interest to drinking water.  Whether the chemical of interest is
regulated or not, a lower or higher number of sources of the chemical in the drinking water supply can
be specified if chemical-specific data are available to warrant deviating from the default.

How the Standard Was Developed
Development of NSF/ANSI 60 was initiated in 1985 at the request of U.S. EPA as part of a larger EPA/FDA
MOU under the Safe Drinking Water Act, to establish modern testing, evaluation and acceptance criteria
for product added directly to water during its treatment, storage and distribution. The NSF
International-led consortium, including the American Water Works Association (AWWA), the American
Water Works Association Research Foundation (AWWARF), the Association of State Drinking Water
Administrators (ASDWA) and the Conference of State Health and Environmental Managers (COSHEM),
completed development in 1988.

Who Maintains the Standard
NSF/ANSI 60 is one of nearly 90 consensus-based standards to which NSF has contributed significantly.
While NSF’s Standards department provides the support and structure for the development and
publication of product standards, it is the Joint Committee on Drinking Water Additives which continues
to review and maintain the standard annually. This committee consists of representatives from the
original stakeholder groups including the U.S. EPA as well as other regulatory, water utility and product
manufacturer representatives. Every NSF standard is reviewed by NSF’s Council of Public Health
Consultants (CPHC) whose role is to determine that the health-effects standards developed with support
8

from NSF continue to aid in the protection of public health. The American National Standards Institute
(ANSI) has an oversight role in the standards process to ensure that the documents are developed and
maintained according to their guidelines; ANSI makes all proposed revisions to ANSI standards available
for public comment on its website at www.ansi.org

More Information on Product Certification

Product Review and Testing
The product review conducted by NSF for a water treatment product considers all chemical ingredients
in the product, as well as the manufacturing process, processing aids and other factors that have an
impact on the chemicals attributable to the products present in the finished drinking water.  The
identified chemicals of interest are subsequently evaluated during testing of the product.  The
manufacturing process is documented by an NSF auditor at an initial audit of the manufacturing site and
during each annual unannounced inspection of the facility. The manufacturing process, ingredients,
potential contaminants and label information are reviewed annually, and the product is tested for any
potential contaminants of interest at NSF laboratories.

NSF’s Mark and Public Product Listings
All NSF certified products bear the NSF mark, maximum use level, lot number or date code, and
production location on the product packaging or documentation shipped with the product.
NSF maintains listings of certified products at www.nsf.org/certified-products-systems. These listings are
updated daily and list the products at their allowable maximum use levels.

History of the Product Certification Program
The NSF testing and certification program for drinking water treatment products was developed in the
1980s at the request of the U.S. EPA to provide an updated, science-based process to evaluate the use
of drinking water treatment chemicals in public water supplies, based on NSF/ANSI 60.    This
certification program was intended by U.S. EPA to eventually replace the EPA’s Advisory Program to
States, which it did in 1990. The states quickly adopted EPA’s recommendations to rely on certified
products to help ensure that individual U.S. states and waterworks facilities have a mechanism to
identify and select treatment chemicals for use in public water supplies.

Who Uses the Standard
According to the latest Association of State Drinking Water Administrators (ASDWA) Survey on State
Adoption of NSF/ANSI Standards 60 and 61, 49 U.S. states require that chemicals used in treating
potable water meet NSF/ANSI 60 requirements.  In Canada, nine provinces/territories require drinking
water treatment chemicals to comply with the requirements of NSF/ANSI 60. Certification to NSF/ANSI
60 is also recognized in several other countries, including Brazil, Israel, UAE, Saudi Arabia, Singapore and
South Africa. Manufacturers can choose to apply to NSF International or another ANSI-accredited
certifying body for certification to the standard. If you have questions on your state's requirements, or
how the NSF/ANSI 60 certified products are authorized in your state, contact your state's Drinking Water
Administrator or download a state survey summary at
http://www.nsf.org/newsroom_pdf/water_asdwa_survey.pdf.
9

Abbreviations Used in This Fact Sheet

ANSI – American National Standards Institute

AWWA – American Water Works Association

AWWARF – American Water Works Association Research Foundation

ASDWA – Association of State Drinking Water Administrators

CDC – Centers for Disease Control and Prevention

COSHEM – Conference of State Health and Environmental Managers

EPA – U.S. Environmental Protection Agency

MAC – maximum allowable level

MCL – maximum contaminant level

mrem/yr – millirems per year – measurement of radiation exposure dose

MUL – maximum use level

NSF – NSF International (formerly the National Sanitation Foundation)

ppb – parts per billion

pCi/L – pico curies per liter – concentration of radioactivity

SPAC – single product allowable concentration

TAC – total allowable concentration

10