Research Blog, Reference Library, Data Repository

Why Is There a Fluoride Shortage? Inside America’s Water-Fluoridation Supply Chain

American cities reduced fluoride in drinking water during 2026, even though the United States still manufactures fluoridation chemicals. A review of federal data, company filings, and municipal contracts reveals supplier exits, rising replacement bids, misleading import statistics, and a major plant closure planned for 2027.
Sherafy editorial card about 2026 fluoride-additive shortages, showing an abstract phosphate source, industrial process building, chemical tanker and municipal water pipe joined by a dotted route with a small gap. It depicts procurement vulnerability, not unsafe drinking water.
Contents

Updated November 15, 2026

The short answer: American water systems faced shortages of the chemicals used to add fluoride to drinking water in 2026 because of disruptions in a specialized manufacturing and distribution market. Suppliers reported producer withdrawals, some overseas deliveries became unreliable, fertilizer manufacturers faced their own operating pressures, and municipal buyers sometimes could not obtain acceptable bids. The problem was not uniform: some utilities reduced fluoridation, others secured new shipments, and some restored normal levels.

The most recent evidence is concrete. On November 5, South Milwaukee announced it would reduce added fluoride to about 0.3 milligrams per liter (mg/L), roughly half its previous level, to stretch supplies until an expected shipment in early 2027. Milwaukee Water Works had already reduced its target in July after its former supplier declined to renew a contract and a replacement solicitation received no bids. Meanwhile, WSSC Water in Maryland restored its normal 0.7 mg/L target in September after an earlier interruption.

Those reductions do not mean the affected water became unsafe to drink. Fluoridation is a dental-health measure, not the treatment that disinfects the water or makes it potable. Removing or reducing added fluoride may change expected cavity-prevention benefits; it does not, by itself, indicate contamination or failure of other drinking-water standards.

The deeper story concerns something most residents have never needed to investigate: Can a city reliably trace, purchase, test, and replace the industrial chemical it adds to the public water supply? And why is the national supply chain so difficult to reconstruct from information that is already public?

Our review of federal reports, company disclosures, certification records, and municipal procurement files finds four particularly revealing facts:

  • A major water-treatment distributor reported that a manufacturer withdrew from the municipal market around mid-2025, before many of the international disruptions blamed for the 2026 shortage.
  • Some utilities received replacement offers dramatically higher than competing or existing prices, but the records do not establish a single national price trend or prove profiteering.
  • The EPA's frequently quoted import figures combine production of one specific chemical with trade statistics for a much broader category of acids. Those figures cannot establish what percentage of American water-fluoridation chemicals is imported.
  • On November 4, 2026, Nutrien announced plans to close a Florida phosphate facility in the second quarter of 2027. A drinking-water chemical certification directory lists that facility for fluorosilicic acid, but the volume of municipal supply potentially affected remains unknown.

What follows separates what the records establish from what manufacturers, distributors, and advocates claim, and from what the public still cannot verify.

Which Cities Reduced Fluoride in 2026, and Is the Shortage Still Happening?

Yes, documented shortages were still affecting municipal fluoridation in November 2026. But there is no evidence that every American water system stopped fluoridating or that every system faced the same suppliers, prices, or risks.

Water system What happened What the records establish as of November 15, 2026
Milwaukee, Wisconsin Reduced its fluoride target from approximately 0.6–0.7 to 0.3 mg/L beginning July 31. Its former supplier would not renew, and a new bid solicitation received no bids. The utility's published notice still describes the reduction as temporary. Milwaukee Water Works
South Milwaukee, Wisconsin Reduced its target from approximately 0.6–0.7 to 0.3 mg/L beginning November 5. The utility expected its next chemical shipment in early 2027 and stated that the water remained safe to drink. City notice
WSSC Water, Maryland Reduced its target from 0.7 to 0.4 mg/L in April. Restored 0.7 mg/L on September 16, after securing supply. It serves customers in Montgomery and Prince George's counties. Restoration announcement
Battle Creek, Michigan Announced a reduction to approximately 0.5 mg/L in August. Its October notice continued to describe reduced fluoridation during supply constraints; a subsequent restoration was not confirmed in the records reviewed. City water notice
Central Arkansas Water Reportedly exhausted its additive supply at the end of August, then received a partial shipment in September. Conway Corp's September account described Central Arkansas Water resuming at a reduced rate. Conway Corp also warned about its own possible future interruption; the two systems should not be confused.
Davie County, North Carolina Local officials directed the water system to stop adding fluoride effective September 1. This was a county policy decision during a period of supplier concerns, not simply proof that chemical inventory ran out. Reporting and county resolution

This table is a dated set of documented examples, not a live nationwide outage tracker. The absence of a city does not establish that its supplies were unaffected; a temporary notice does not establish that a reduction remains in effect indefinitely.

The distinction is visible in Maryland. WSSC Water's April announcement linked its shortage partly to the Middle East conflict. Its September restoration showed that additional supplies could be obtained even while other cities remained constrained.

There is no single national on/off switch for fluoridation. Thousands of water systems make purchasing decisions under different contracts, laws, storage conditions, and delivery schedules.

How many Americans are potentially affected?

The CDC's published 2022 national fluoridation statistics counted approximately 209.1 million people receiving fluoridated community water and 5,561 water systems that actively adjusted fluoride. About 72.3% of people served by community water systems received fluoridated water in that reporting year.

Those are important scale indicators, but they are not a 2026 count of people facing shortages. The 209.1 million figure also includes people served by systems with sufficient naturally occurring fluoride or by systems purchasing treated water from another utility. It cannot be treated as the number of Americans whose water depends directly on a purchased additive.

To measure that exposure properly would require matching each actively dosing system to the population it serves, its current chemical supplier, and its operating status.

What Chemical Is Actually Added to Drinking Water?

This is the first correction necessary to understand the supply chain. Sodium fluoride is not the only fluoridation chemical, and it is not the principal liquid chemical described in the shortage records.

According to the EPA's fluorosilicic-acid supply-chain profile and NSF's treatment-chemical fact sheet, three products are commonly certified for this purpose:

Additive Chemical formula What it is Why the distinction matters
Fluorosilicic acid (also called hydrofluosilicic or fluosilicic acid) H₂SiF₆ A concentrated liquid acid, often recovered during phosphate-fertilizer production. Central to many 2026 supplier and municipal purchasing records.
Sodium fluorosilicate (sodium silicofluoride) Na₂SiF₆ A solid fluoride compound used in some fluoridation systems. Not interchangeable with sodium fluoride when comparing trade statistics.
Sodium fluoride NaF A distinct fluoride salt, supplied as a solid and used in some smaller or other treatment applications. Its import or production ranking cannot simply be assigned to fluorosilicic acid.

In properly operated water-treatment systems, these compounds dissolve and contribute fluoride ions to the finished water. That is the ion being measured when a utility reports a fluoride concentration in mg/L. The dose and chemical specifications differ, however, and the products should not be treated as identical industrial commodities.

A fluoride concentration of 0.7 mg/L is approximately 0.7 parts per million in drinking water. It describes the amount of fluoride in the finished water, not the concentration of the liquid acid that arrived at the treatment plant.

Why can't a utility simply switch to another fluoride compound?

The three products can supply fluoride ions, but they come in different physical forms and concentrations. A plant designed to meter a corrosive liquid acid may need different storage, feed equipment, safety procedures, operator controls, delivery arrangements, and approvals to use a dry salt instead. A substitute also needs to be available from a qualifying supplier on acceptable contract terms. Chemical interchangeability at the level of fluoride ions does not mean logistical interchangeability at a working treatment plant. EPA chemical supply-chain profile; NSF fluoridation product guidance.

How does fertilizer production create a drinking-water additive?

The industrial process works roughly as follows:

Phosphate rock → phosphoric-acid fertilizer production → fluoride-containing process gases → gas capture and scrubbing → fluorosilicic-acid solution → product specification, certification, transport, and utility dosing.

Phosphate ore naturally contains fluorine. When ore is processed with sulfuric acid to make phosphoric acid for fertilizer, fluoride-bearing compounds enter the process gas stream. Equipment can capture those gases in water and recover fluorosilicic acid. The result can be treated as a waste stream or processed and sold as a commercial product.

That is why the description "fertilizer-industry byproduct" is materially correct. It is also incomplete. It says where the chemical originated, not whether a particular shipment meets water-treatment specifications, which impurities it contains, or its final concentration in drinking water. Those are separate questions answered by product standards, testing, and utility operating records.

The concentrated product is corrosive and requires industrial handling precautions. The finished drinking water contains a much smaller fluoride concentration. Confusing the hazards of a concentrated bulk chemical with the exposure from correctly diluted drinking water is a serious analytical error; assuming every certified shipment is perfect would be another.

Who Actually Makes and Sells America's Fluoridation Chemicals?

A water utility may buy an additive from a familiar distributor without buying directly from the plant that originally produced it. The chain can include a phosphate processor, a chemical manufacturer or marketer, a transport company, a distributor, and the utility.

A certificate listing a supplier's American address does not necessarily identify the original manufacturing country. Likewise, one producer can supply multiple distributors, and a distributor may purchase from more than one manufacturing site.

The EPA's 2022 profile illustrated the distinction with 103 NSF-certified suppliers in a 2021 listing, compared with five U.S. manufacturing locations in a separate 2017 count. The figures come from different years and should not be equated, but they make the basic point: a long supplier directory does not mean there are 103 independent factories.

A source-to-supplier evidence ledger

Company or facility Documented role What is established What remains unknown
ICL Rotem, Israel Phosphate producer and fluorosilicic-acid marketer ICL's own reporting identifies fluorosilicic-acid exports, including to the United States, for multiple industrial applications; a related Israeli facility appears in NSF certification listings. Exact U.S. municipal tonnage, water-grade share, and the origin of any specific city's shipment.
Itafos Conda, Idaho Domestic phosphate and fluorosilicic-acid producer Reports both rated chemical production capacity and actual production, sales, and realized prices. How much output was municipal grade, contracted to particular distributors, or available for new utility customers.
Nutrien White Springs, Florida Phosphate operation associated with an NSF-listed fluorosilicic-acid source Nutrien plans to shut the site in Q2 2027. Current potable-grade output and whether utility customers will lose supply or be transferred elsewhere.
Nutrien Aurora, North Carolina Continuing phosphate operation; also NSF-listed Nutrien intends to retain and optimize Aurora. Whether it could offset any relevant lost White Springs fluorosilicic-acid sales.
Mosaic Major phosphate producer and historical supply-chain participant Company filings describe phosphate curtailments and rising sulfur costs. A late-2019 agreement transferred certain acid-marketing and logistics assets to Univar. How much municipal-grade additive output, if any, was lost because of specific 2026 curtailments.
Univar Solutions Chemical marketer/distributor Acquired certain Mosaic hydrofluorosilicic-acid contracts and related supply-chain assets. Whether any particular 2026 utility shipment originated from a Mosaic manufacturing facility.
Pencco Water-treatment chemical supplier/distributor Reported a manufacturer's withdrawal from the municipal market; withdrew an Albany bid and declined renewal in Milwaukee. The identity and precise scope of the manufacturer that allegedly withdrew in 2025.
George S. Coyne Chemical Distributor Its documented account links a supplier disruption to ICL. How much of ICL's total U.S. business was affected versus Coyne's specific supply channel.

Source basis: ICL's 2021 manufacturing disclosure; NSF certified chemical listings; Itafos operations; Nutrien's November announcement; Mosaic's SEC filing; Univar's 2020 transaction announcement; Albany's procurement record; and Coyne's supplier-disruption letter.

The table deliberately avoids drawing unsupported connections. "Nutrien produces a certified additive" is not the same claim as "Nutrien supplied Milwaukee." Establishing that last step requires dated invoices, bills of lading, manufacturer certifications, or comparable shipment records.

The Federal Import Numbers Do Not Answer the Question People Think They Answer

One of the most consequential findings appears in the EPA's own published supply-chain analysis.

Its December 2022 fluorosilicic-acid profile includes an estimate of 58 million kilograms of apparent U.S. consumption in 2019. The component figures are:

2019 quantity shown by EPA Amount What the underlying data actually measures
Domestic production 29 million kg Fluorosilicic acid specifically
Imports 39 million kg A broader category of inorganic acids, not just fluorosilicic acid
Exports 10 million kg The same broader category of inorganic acids
Apparent consumption 58 million kg Calculated as 29 + 39 − 10, combining those unlike data categories

The EPA explicitly says the imports and exports come from tariff classification 2811.19.61, while domestic production measures fluorosilicic acid specifically. The U.S. International Trade Commission's Harmonized Tariff Schedule independently shows that the relevant classification is not a clean, standalone record of drinking-water-grade fluorosilicic acid.

The calculation is arithmetically correct but chemically mismatched. You cannot reliably estimate how much fluorosilicic acid America imports by adding production of that single chemical to imports and exports of several different acids.

For example, 39 ÷ 58 ≈ 67%. It would be tempting to write that the United States imports roughly two-thirds of its water-fluoridation acid. But the numerator is not exclusively fluorosilicic acid, and the denominator mixes data categories. That percentage is not supported by the underlying figures.

This is not a discovered secret that an agency concealed. The caveat is in the public EPA report. The accountability issue is that a widely useful report can still be misread or repeated without its methodological warning, while the more precise question remains unanswered.

Did an EPA table rank Israel second in the world?

Yes, but not for the chemical or quantity commonly alleged online.

The EPA reproduces a 2021 international trade table for Harmonized System category 2811.19: inorganic acids other than hydrogen fluoride. Its top exporters in that broader category include:

Country Reported 2021 exports in the broad HS 2811.19 category
China 92 million kg
Israel 40 million kg
Poland 27 million kg

The agency expressly warns that fluorosilicic-acid-specific trade data was unavailable from the cited data source. This table does not establish that Israel was the world's second-largest manufacturer of sodium fluoride, the second-largest producer of fluorosilicic acid, or the second-largest supplier of additives to U.S. water systems. Those are three different claims. EPA, worldwide trade table and methodology.

Nor do trade totals for sodium fluoride automatically substitute for fluorosilicic-acid imports. They describe different chemical products and may include non-water-treatment uses.

What would it take to measure actual import dependence?

A defensible estimate would require chemical-specific import records or shipment-level classifications identifying fluorosilicic acid, actual manufacturers, quantities, concentrations, countries of origin, and intended uses. Those records would need to be reconciled with domestic manufacturing and municipal purchases, not merely combined with broad tariff data.

Even an accurate total for fluorosilicic-acid imports would not, by itself, establish how much entered public drinking water rather than other industrial markets.

The practical question is not simply where a chemical crosses a border. It is which manufacturers supplied certified material that American utilities could buy and use during a shortage.

The Shortage Was Developing Before the Most Widely Cited International Disruption

The chronology matters because there are competing explanations for the same crisis.

Date Documented development Why it matters
2012 An American Water Works Association warning, later cited by the EPA, discussed seasonal shortages of fluorosilicic acid. Supply fragility was a known industry problem long before 2026.
December 2022 EPA published its chemical supply-chain assessment, highlighting production concentration, transportation issues, and periodic interruptions. The risk was not newly discovered during the 2026 crisis.
Approximately mid-2025 Pencco later said a major manufacturer had withdrawn from the municipal market about six months before its January 2026 letter. If accurate, a significant market exit predated the year's better-publicized international events.
January 22, 2026 The Great Lakes Water Authority documented a supplier alert and the need to consider higher-priced alternatives. A utility-scale shortage was being reported early in the year.
January 27, 2026 Pencco sought a substantial contract price increase in Kingsport, Tennessee, citing supply-market changes. Supplier withdrawal translated into a municipal purchasing dispute.
February 13, 2026 Pencco withdrew an Albany, New York, bid, saying it could no longer guarantee consistent supply. A specific purchase failed; Albany subsequently paid a higher competing bid.
February 16, 2026 According to a later distributor notice, Israeli producer ICL informed George S. Coyne Chemical of a planned disruption or withdrawal of supply to that distributor. International instability affected at least one reported U.S. supply relationship.
April–September 2026 Multiple utilities announced reductions; some later restored target levels. The impact was real but geographically uneven and changeable.
November 4, 2026 Nutrien announced its planned White Springs closure in Q2 2027. A future domestic production question emerged while shortages were unresolved elsewhere.
November 5, 2026 South Milwaukee reduced added fluoride to extend available inventory until an expected early-2027 shipment. The problem had not ended nationwide by November.

Documents: EPA supply-chain profile; Kingsport municipal packet containing the Pencco letter; Great Lakes Water Authority presentation; Albany award and withdrawn-bid correspondence; Coyne supplier letter; Nutrien announcement; South Milwaukee notice.

The timeline does not demonstrate that the alleged mid-2025 producer exit caused every later shortage. It demonstrates that an explanation based entirely on events beginning in 2026 cannot account for all the reported warning signs.

The 2025 manufacturer exit remains unidentified

In a letter included in Kingsport's municipal meeting materials, supplier Pencco described the withdrawal of a major manufacturer from the municipal fluoridation market around six months earlier.

In its February 13 email to Albany, Pencco again identified the loss of a major source of production as a reason it could not promise reliable delivery.

Those are documented supplier statements, not independently verified evidence identifying the factory or confirming why it exited. Pencco did not name the manufacturer in the cited correspondence. The disappearance of a producer, the end of a marketing arrangement, and a distributor's decision not to serve municipalities are different events. Establishing which one occurred would materially improve the public understanding of the shortage.

What Happened to Prices? The Contracts Tell a More Complicated Story

A shortage becomes tangible when a water utility has to replace a failed supplier. Several municipal records show how disruptive that process can be.

Albany: A low bidder withdrew, leaving a much higher offer

In a January 23, 2026, Albany bid opening, Pencco offered $2,948 per dry ton for hydrofluosilicic acid. Slack Chemical Company offered $5,907 per dry ton. Univar did not submit a bid.

Then Pencco sought to withdraw, citing the changed supply situation and uncertainty about fulfilling the contract. The March 6 municipal record shows Albany accepting Slack's bid at $5,907 per dry ton.

The accepted bid was approximately 100.4% higher than the withdrawn bid: a difference of $2,959 per dry ton. This is a particularly strong comparison because both offers appear in the same solicitation with the same stated pricing unit. It is not, however, evidence that the price paid under a previous Albany contract rose by 100% or that every municipality faced the same increase.

Kingsport and Great Lakes: More evidence, but not directly comparable prices

Pencco's January letter in the Kingsport packet requested an adjustment from $0.249 to $0.332 per wet pound. That is an approximately 33.3% increase in the specified unit.

The Great Lakes Water Authority recorded an existing figure of $320 per ton and a backup quote of $809 per ton, illustrating a potentially substantial replacement-cost shock.

Record Figure A Figure B What can fairly be concluded
Albany, NY, Jan.–Mar. 2026 Pencco $2,948/dry ton Slack $5,907/dry ton The accepted bid was about twice the withdrawn bid in the same procurement.
Kingsport, TN, Jan. 2026 $0.249/wet lb $0.332/wet lb Pencco requested a roughly 33% contract-price increase.
Great Lakes Water Authority, Jan. 2026 $320/ton Backup quote $809/ton The authority faced a much more expensive reported alternative. The contract and product basis must be checked before comparison with other cities.

Do not combine these prices into a national average. Some are quotes and one is an accepted bid. They may involve different acid concentrations, freight arrangements, package types, contract periods, locations, or definitions of a ton. A wet pound includes the water in the supplied solution. A dry ton is a different purchasing basis. Comparing the raw dollar amounts as though they describe one standardized product would mislead readers.

Why might a city pay more while a producer reports lower prices?

That apparent contradiction appears in the latest corporate data.

The phosphate producer Itafos reported, for the first nine months of 2026, that its fluorosilicic-acid sales rose to 4,662 metric tons, from 4,467 in the same period of 2025. Yet its average realized selling price fell from $765 to $553 per metric ton, according to its November 2026 management report.

A manufacturer's realized average and a city's delivered price are not interchangeable. The customer mix, chemical specification, contract obligations, freight, timing, and geography may differ. Still, the coexistence of rising local bids and falling realized prices at one producer is evidence against assuming that every part of the industry faced the same price pressure.

A useful investigation must distinguish factory output, commercial availability, and last-mile delivery. A chemical may exist in a manufacturer's inventory while being unavailable to a city under the city's purchasing conditions. Conversely, a local shortage does not prove national production has collapsed.

Why the Israel Explanation Is Both Real and Incomplete

Israel is relevant to this story because an Israeli producer has documented fluorosilicic-acid exports and an American distributor attributed a major supply disruption to that producer. Neither point should be dismissed. Neither is enough to establish that Israel controls the American fluoridation market.

What ICL's own disclosure establishes

In a detailed 2021 sustainability disclosure, Israeli company ICL Rotem said it sold roughly 5,000 metric tons of fluorosilicic acid per year for reuse, shipped fluorosilicic acid to several countries including the United States, and produced material meeting water-fluoridation standards.

It also identified multiple end uses, including metal treatment, aluminum, silicate-related industries, and water fluoridation. The company's listing of American exports does not break out how many tons were used in U.S. drinking water.

A separate ICL report described approximately 70,000 tons annually generated at the Rotem site in a later reporting period. That is not evidence that 70,000 tons were sold, exported, or used for U.S. fluoridation. Generated byproduct, commercially sold product, exports, and municipal consumption are four different measurements. ICL 2023 corporate responsibility report.

The NSF drinking-water chemical certification listings for Israel also identify an Israeli facility associated with a certified fluorosilicic-acid product. That supports the existence of a qualified production source, not a particular U.S. shipment or market share.

A U.S. distributor says the Israeli supply relationship was interrupted

A 2026 force-majeure notice from George S. Coyne Chemical Company, discussed in North Carolina Health News' examination of utility records, describes ICL informing Coyne in February that supply to that U.S. relationship would be discontinued or disrupted. The distributor attributed the change to wartime workforce and operating conditions.

The account is relevant because municipalities purchase from distributors, not necessarily from original manufacturers. The loss of one qualifying supply relationship can leave a distributor unable to fulfill commitments even if a producer continues operating elsewhere.

ICL's 2026 SEC reporting acknowledges multiple rounds of Israeli military reserve call-ups involving approximately 15% of its Israel-based employees over the preceding two years, along with operational and logistics pressures. At the same reporting point, the company did not characterize those problems as materially damaging its overall consolidated operations.

The statements can coexist. An event too small to materially affect a multinational company's consolidated accounts can still materially affect one product, delivery route, or municipal buyer. On the other hand, workforce disruption does not prove the producer stopped manufacturing the chemical or that all U.S. supply depended on that producer.

The industry's alternative explanation

A September 2026 briefing by the American Dental Association and American Fluoridation Society argued that domestic production constraints, supplier exits, and industrial-market dynamics explained much of the U.S. shortage. It disputed the idea that Israeli production problems were the principal nationwide cause.

These groups advocate community water fluoridation and have an institutional interest in maintaining programs. That does not make their factual claims wrong; it means their account should be evaluated alongside independent company and municipal records rather than treated as a final market audit. Likewise, the distributor's account should not automatically become the explanation for every affected city.

A producer can have material available in Israel while a particular U.S. distributor lacks an acceptable supply contract or reliable delivery route. A domestic supplier can withdraw from municipal sales while another producer increases output. Several causes can operate simultaneously.

What remains unverified is the exact amount of Israeli-produced, municipal-grade fluorosilicic acid delivered to the United States in 2024, 2025, and 2026, through which intermediaries, and to which water systems. Neither the broad EPA international trade table nor the available company statements provide that complete answer.

Why Did Israel Stop Fluoridating Its Own Water in 2014?

Israel's water-fluoridation policy did change. But the full chronology is more complicated than the claim that it banned a dangerous substance while exporting that same substance abroad.

The Israeli Ministry of Health's current public guidance confirms that proactive public-water fluoridation stopped in 2014 and has not resumed as a routine national program.

The historical record also documents an attempted reversal. An Israeli government environmental-health report describes a 2016 decision intended to restore fluoridation at 0.7 mg/L, with implementation contemplated for 2017. That restoration did not become the country's ongoing practice.

This means that a policy decision to stop fluoridation, a later governmental effort to resume it, and an industrial company exporting a fluoride-containing chemical all occurred within the same country's history.

Those facts warrant careful discussion. They do not establish that Israel's government secretly determined water fluoridation was poisonous, that the policy was designed to promote exports, or that Israelis stopped using the product because its manufacturer knew something U.S. regulators did not.

Indeed, the Israeli health ministry's own public material continues to discuss the dental-health benefits of fluoride. The stronger investigative question is why an attempted restoration was not implemented, rather than treating the 2014 change as conclusive proof of a health finding that the cited records do not establish.

The industrial and health-policy issues should both be investigated, but a commercial export relationship is not proof of a governmental motive.

Domestic Production Did Not Simply Disappear

The U.S. Geological Survey's 2026 Mineral Commodity Summaries reports the following historical U.S. recovery of fluorosilicic acid from phosphate processing:

Year Fluorosilicic acid recovered in the United States
2021 40,000 metric tons
2022 43,000 metric tons
2023 43,000 metric tons
2024 45,000 metric tons
2025 (estimate) 45,000 metric tons

The 2025 estimate refers to acid recovered from three phosphate-processing plants. It does not tell us the amount sold to water utilities, the concentration of every sale, available inventory, certified product output, or imports. A nationwide 2026 production total was not yet available in this report.

Nevertheless, the historical series is meaningful. It does not show a nationwide collapse in recovered fluorosilicic-acid production through 2025. That means any account of the 2026 shortage must explain why market-access and distribution problems could occur despite substantial domestic recovery.

There is another common source of confusion. The USGS also describes the United States as heavily or entirely import-dependent for fluorspar, a mineral feedstock known chemically as calcium fluoride (CaF₂), under its reported import-reliance definition. That statement is not a claim that the United States imports all the fluorosilicic acid used to fluoridate drinking water. One is a mineral commodity; the other can be recovered domestically during phosphate processing.

The Fertilizer Connection: Sulfur Prices, Curtailments, and Hidden Dependence

Water-fluoridation chemical supply depends partly on an industry whose main economic purpose is not municipal water treatment.

The EPA explains that fluorosilicic acid is commonly recovered as part of wet-process phosphate fertilizer production. If fertilizer plants reduce throughput, change their operating configuration, or decide that recovery and sale are uneconomic, the amount of commercial acid available may change.

In its September 2026 Form 10-Q, The Mosaic Company reported that its average cost of sulfur consumed in North American production increased from $220 per long ton in the first nine months of 2025 to $524 during the same period of 2026.

That is approximately 138% higher. The filing also describes phosphate-production constraints and output reductions.

Mosaic's North American sulfur cost First nine months of 2025 First nine months of 2026
Average cost consumed $220/long ton $524/long ton
Change – +138%

A long ton here is a 2,240-pound weight measure, not the metric-ton measure used in other company reports.

The connection is economically plausible: lower phosphate processing throughput can reduce opportunities to recover fluorosilicic acid, while higher input costs can change whether producers market the recovered product. But the available filing does not quantify how much certified drinking-water-grade fluorosilicic acid Mosaic lost because of the 2026 sulfur-cost shock.

That missing quantity matters. We can report the upstream pressure and its technical connection without converting it into an unsupported estimate of municipal additive losses.

Why distributors matter as much as plants

A March 2020 announcement by Univar Solutions says the distributor had acquired certain Mosaic hydrofluorosilicic-acid business assets in late 2019, including customer contracts and rail and supply-chain agreements.

This was not an announced acquisition of every Mosaic manufacturing site. It demonstrates that commercial relationships, sales arrangements, and freight logistics can be transferred separately from a chemical plant.

For a utility, the immediate problem may not be that the original factory vanished. It may be that the company holding the contract no longer has access to a qualifying producer, shipment route, or economically acceptable quantity.

The Itafos Puzzle: Why Is Output Far Below Stated Capacity?

One of the clearest opportunities for further industrial reporting comes from Itafos, which operates the Conda phosphate facility in Idaho.

On its corporate operations page, Itafos lists approximately 27,000 metric tons per year of hydrofluorosilicic-acid production capacity. In its 2025 management discussion and analysis, the company reported 5,648 metric tons of actual annual production.

That works out to about 20.9% of the stated annual capacity. The gap is approximately 21,352 metric tons.

This is a question, not proof of 21,352 tons of unused, immediately marketable supply. Nameplate capacity may assume operating conditions, upstream phosphate throughput, technical recovery arrangements, product mixes, and customers that differ from reality. It also does not automatically establish available certified potable-water-grade output.

For the first nine months of 2026, the company's latest management report provides another revealing comparison:

Itafos fluorosilicic-acid business Jan.–Sep. 2025 Jan.–Sep. 2026
Production 4,109 metric tons 4,392 metric tons
Sales 4,467 metric tons 4,662 metric tons
Average realized price $765/metric ton $553/metric ton

Production and sales increased, while the average realized price declined by about 28%. Those results coexist with municipalities describing shortages and expensive alternative bids.

The appropriate questions are specific: What limits commercially viable production at Conda? How much is potable-water grade? How much is available under new contracts? What transportation costs and specifications separate the factory from cities struggling to buy the product?

The records do not yet answer those questions. But the apparent capacity-to-output discrepancy is exactly the kind of concrete issue that can be examined with plant data, contracts, and expert technical evidence rather than speculation.

A New 2027 Risk: Nutrien Plans to Close White Springs, Florida

On November 4, 2026, Canadian fertilizer company Nutrien announced plans to shut down its White Springs phosphate operation in Florida in the second quarter of 2027.

The company cited the approach of the end of the site's economically recoverable reserves, increasing operating costs, and the desire to improve its broader phosphate business. It said it would retain and optimize its Aurora, North Carolina, operations.

The NSF certification directory for fluorosilicic acid lists facilities associated with Nutrien at White Springs and Aurora. That makes the closure directly relevant to the supply-chain investigation.

However, three quantities remain unestablished from the records reviewed:

  1. How many metric tons of municipal-grade fluorosilicic acid White Springs is actually producing or delivering in 2026.
  2. Which utilities or distributors, if any, currently rely on that plant's output.
  3. Whether production, contracts, or deliveries could be transferred to Aurora or another source before the shutdown.

Nutrien said White Springs represents approximately 20% of Nutrien's expected 2026 phosphate sales. That is a share of the company's phosphate business, not 20% of America's fluoridation chemicals, and it must never be presented as the latter.

The closure establishes a potential supply-chain exposure, not a proven nationwide 2027 fluoride shortage. If it removes water-treatment-grade output that cannot be replaced, affected buyers could face more pressure. If its municipal deliveries are limited or transferable, the impact might be minor. The necessary facility-level evidence has not been published in the records examined here.

One other government signal shows why 2027 cannot be ignored. Connecticut's Department of Public Health extended temporary enforcement discretion through January 31, 2027, for certain documented fluoridation-supply problems. That is a response to procurement difficulty, not a wholesale repeal of state water-fluoridation requirements.

Are Fluoridation Chemicals Tested for Arsenic, Lead, and Other Contaminants?

Yes. Products used to fluoridate drinking water are covered by treatment-chemical standards, and the fact that a chemical originates as an industrial byproduct is not, by itself, evidence that a certified product makes finished drinking water unsafe.

But that is not the same as claiming there are no impurities, no risks, or no need for independent oversight.

NSF/ANSI/CAN 60, the drinking-water-treatment-chemical health-effects standard, addresses fluorosilicic acid, sodium fluorosilicate, and sodium fluoride. NSF's published explanation describes product formulation disclosure, contaminant testing, annual inspections of production and distribution sites, and evaluation of expected contaminant contributions after the product is added at its certified dosage.

A useful point of transparency is that the certification process does not merely pretend contamination is impossible. It measures contaminants.

In NSF's historical summary of 328 fluoridation-product samples tested from 2012 through 2017, arsenic was detected in approximately 52% of samples under laboratory testing conditions. After adjusting to the expected product dosage in water, the estimated mean arsenic contribution was 0.16 micrograms per liter (µg/L) and the highest reported contribution was 0.9 µg/L in the table. The U.S. EPA's primary drinking-water standard for arsenic is 10 µg/L. The sampled products met NSF's applicable health-effects criteria. NSF fact sheet and testing table.

Those figures mean neither "arsenic never appears" nor "arsenic is being added at the levels found in a drum of concentrated acid." Laboratory measurements of concentrated material and their calculated contribution after dilution must be distinguished.

They also do not substitute for current, producer-specific records. A historical certification study cannot tell a resident the exact content of the shipment delivered to their own utility in September 2026.

What residents should be able to verify

The most revealing records would identify the product manufacturer and plant, lot or batch, chemical concentration, NSF/ANSI/CAN 60 certification status, the supplier's certificate of analysis, the receiving utility's acceptance testing, and the fluoride concentration measured after dosing.

A certificate-of-analysis value for a concentrated solution is not a tap-water value. A product specification is not a measured finished-water sample. A supplier's certificate is not proof that every shipment was independently tested. A rigorous source-to-tap investigation needs to distinguish all three.

Also, a maximum-use figure displayed next to a product in a certification directory refers to that treatment chemical's approved use concentration or conditions, not an instruction that finished drinking water should contain that many milligrams of fluoride ion. NSF certification directory.

Is Fluoride in Drinking Water Safe? What the Evidence Actually Shows

This is a legitimate public-health question, distinct from the question of who supplies the chemicals. Supply shortages should not be used to manufacture certainty about health effects in either direction.

The three concentrations people often confuse

Fluoride concentration What it means
0.7 mg/L The U.S. Public Health Service's recommended level for community-water fluoridation intended to balance dental benefits and risks such as dental fluorosis. It is not a federal law ordering every utility to fluoridate.
Around or above 1.5 mg/L A higher-exposure range particularly relevant to the body of research assessed by the National Toxicology Program concerning children's neurodevelopment. It is not an established universal dividing line between harmless and harmful exposure.
4.0 mg/L The U.S. EPA's federal maximum contaminant level for fluoride in drinking water, a separate legally enforceable standard for covered public systems. It is not the recommended fluoridation target.

Sources: U.S. Public Health Service fluoridation recommendation, EPA National Primary Drinking Water Regulations, and the National Toxicology Program's 2024 monograph.

Dental benefits: Evidence exists, but magnitude and setting matter

Fluoride has well-established effects on tooth enamel and dental-caries prevention, and community water fluoridation has a long history as a population-wide intervention. The CDC's oral-health guidance describes its role in reducing cavities.

The 2024 Cochrane systematic review is particularly important for interpreting that history in modern settings. It found that contemporary studies offer a less certain, and potentially smaller, incremental benefit than older studies suggested, partly because widespread use of fluoride toothpaste and other preventive measures has changed the comparison. The evidence base is not equally strong for every population, setting, or outcome. Cochrane review and supporting full-text link.

This does not make water fluoridation pointless. It means a serious assessment should ask how much additional benefit exists today, for whom, and compared with which alternatives, rather than relying on one historical slogan.

Neurodevelopmental concerns: Higher exposure is associated with lower IQ, but the U.S. target remains a distinct question

The National Toxicology Program's August 2024 systematic review concluded with moderate confidence that higher fluoride exposures are associated with lower IQ in children. Much of the evidence involved drinking-water concentrations above approximately 1.5 mg/L or other measures of higher exposure.

The NTP also explicitly stated that its review does not determine whether fluoride added to drinking water at 0.7 mg/L, by itself, is associated with a measurable IQ effect. Its work assessed exposure from multiple sources and was not designed as a full dental-benefit-versus-neurodevelopmental-risk analysis.

A later 2025 systematic review and meta-analysis published in JAMA Pediatrics reported inverse associations between fluoride exposure and children's IQ across several analyses, including some studies using urinary fluoride measurements. But urinary fluoride concentration and the fluoride concentration in a municipal tap-water supply are not interchangeable: urine reflects exposure from water, food, dental products, and other sources, while study design and potential confounding affect causal interpretation.

These findings justify serious attention to exposure, cumulative sources, lower-dose uncertainty, and the quality of the underlying studies. They do not justify saying either "all fluoridation at 0.7 mg/L has been proved dangerous" or "there is no remaining scientific question."

What about the federal fluoride court ruling?

A 2024 federal district court decision under the Toxic Substances Control Act concluded that fluoride in drinking water presented an unreasonable risk. But on May 21, 2026, the U.S. Court of Appeals for the Ninth Circuit vacated the decision and remanded the case, principally because of how the district court handled the evidentiary presentation and related procedural issues. The appellate court did not issue a definitive scientific ruling resolving fluoride's neurodevelopmental effects. Read the Ninth Circuit opinion.

The legal status of that decision, the scientific evidence on childhood exposures, and the industrial purity of certified additives are three separate matters. Conflating them produces a misleading answer no matter which position the writer favors.

Is Drinking Water Safe if a City Stops Adding Fluoride?

Usually, yes, when the only change is a reduced fluoridation dose and the utility confirms all other drinking-water standards are met. The reduction does not, by itself, suggest a failure of filtration, disinfection, contaminant monitoring, or drinking-water potability.

For example, South Milwaukee's November 2026 notice specifically says its water remains safe to drink despite the reduction. Milwaukee Water Works similarly states that residents do not need to switch to bottled water because of the fluoride-supply issue.

There may still be implications for dental health over time, particularly for people at greater risk of tooth decay or with less access to preventive care. Families should continue ordinary oral hygiene, including age-appropriate fluoride toothpaste as advised by their dental professionals. People with individualized medical or dental concerns should consult their clinicians rather than interpreting a utility's emergency purchasing notice as personal medical guidance.

A shortage announcement is also not the same as a boil-water advisory. If a utility issues an actual contamination or disinfection warning, follow that separate notice. Boiling water does not solve a shortage of a dental-health additive.

How to Find Out What Is in Your Own Drinking Water

A citywide or national headline may be outdated by the time a resident reads it. The most accurate current answer comes from the utility that supplies their particular address.

  1. Identify your water system. Your water bill or local utility website should name the provider. If you live in an apartment or receive water from a neighboring city, the billing entity and treatment facility may differ.
  2. Look for a dated operational notice. Search your utility's website for fluoride, fluoridation, chemical supply, or water quality. A city may have reduced its target in April and restored it in September.
  3. Check the annual Consumer Confidence Report. It gives regulated water-quality information, but it is generally retrospective, not a real-time record of today's fluoride dose.
  4. Ask for the latest measured fluoride concentration, not just the target. Naturally occurring fluoride remains in water even when a utility stops adding it. The target dose and the measured finished-water concentration are different figures.
  5. If chemical origin matters to you, request procurement records. Ask for the current supplier contract, manufacturer and facility of origin where recorded, product name and concentration, NSF/ANSI/CAN 60 certification, recent shipment documents, and lot-specific certificates of analysis. State or local public-records laws may apply.

The CDC's My Water's Fluoride database can help identify fluoridation information for participating systems, but it should not be treated as a live dashboard confirming that a temporary 2026 shortage has ended.

The most important distinction for readers is simple: a supplier's country, a product's certification, and the concentration measured at a tap are three different facts. Each requires its own supporting record.

The Questions the Public Records Still Cannot Answer

After comparing the documents, the most important mystery is not whether fluoride is manufactured in Israel or whether fertilizer production creates a marketable byproduct. Both are documented.

It is why the public cannot easily move from a city water-treatment contract back through the distributor to the actual producing plant and forward again to a verified chemical batch, while also measuring how much of the national market depends on each source.

The following records would materially advance that investigation:

Unanswered question Evidence required Why it matters
Which manufacturer left the municipal market around mid-2025? Pencco's upstream supply agreement, a named producer statement, notices to purchasers. Would clarify a potential prewar root cause.
How much of the U.S. municipal-grade market did ICL supply? Producer shipment totals, U.S. customer breakdowns, distributor invoices, origin records. Would separate verified Israeli involvement from unsupported claims of market control.
Why is Itafos's reported output far below its rated capacity? Plant utilization, phosphate throughput, recovery yields, product-quality and market constraints. Could explain whether the bottleneck is production, economics, specifications, or delivery.
What fluoridation output will Nutrien's White Springs closure remove? Facility-specific production and contract figures, transition plans, confirmation from customers. Would turn a plausible 2027 risk into a measurable forecast, or rule it out.
How much do cities pay for comparable delivered chemical? Standardized bids and invoices with concentration, ton basis, freight, dates, and volume. Would distinguish actual price increases from accounting and contract differences.
Can a resident verify the exact lot used by their utility? Manufacturer-origin records, certificates of analysis, receiving tests, final-water measurements. Would establish traceability and product-quality accountability.
Did phosphate-production curtailments reduce water-grade output? Recovery-volume records by facility and customer segment. Would test the proposed causal link to fertilizer-market pressures.

There is no responsible basis in the records reviewed to conclude that any named company deliberately engineered a shortage, that a foreign government secretly controls American water fluoridation, or that search platforms intentionally hid these relationships.

There is a responsible basis to question supplier concentration, procurement resilience, incomplete trade reporting, the accountability of distributors and manufacturers, and the quality of the public data used to explain a nationwide public-health supply chain.

Those are not lesser questions. They are the ones the actual contracts, financial reports, and chemical records can answer.

Frequently Asked Questions

Why is there a fluoride shortage in the United States in 2026?

Documented problems include manufacturers or distributors leaving parts of the municipal market, constrained deliveries, international supplier disruption reported by some utilities, and operating pressure in upstream phosphate production. There is no verified single cause for every city. EPA supply-chain profile; Albany procurement record.

Is the U.S. running out of fluoride itself?

Not necessarily. Fluoride is an ion found in many minerals and compounds. The reported municipal shortage concerns specific commercially available and qualified water-treatment additives, delivered under usable purchasing and transport conditions. Domestic fluorosilicic-acid recovery was substantial through 2025; that does not guarantee every city could buy what it needed in 2026. USGS mineral commodity data.

Is drinking-water fluoride made from fertilizer waste?

Fluorosilicic acid is commonly recovered from the gases generated during phosphate-fertilizer production. Calling it a byproduct is accurate. Whether a shipment is suitable for drinking-water treatment depends on its specifications, certification, contaminants, and use, not the word byproduct alone. EPA production explanation.

Does Israel supply chemicals used for American water fluoridation?

An Israeli manufacturer, ICL Rotem, has publicly identified exports of fluorosilicic acid to the United States, among several destinations and industrial uses. A U.S. distributor also attributed a 2026 disruption to its ICL supply relationship. What fraction of American municipal fluoridation came from that source is not established by the broad international trade totals. ICL disclosure; Coyne letter.

Is Israel the world's second-largest sodium fluoride producer?

The evidence reviewed does not establish that claim. An EPA table ranks Israel second in 2021 exports of a broader category of inorganic acids. That is not a sodium-fluoride production ranking or a measure of water-treatment chemical exports to the United States. EPA trade table.

Why doesn't Israel fluoridate its own water?

Israel stopped proactive water fluoridation in 2014. A later effort to restore it did not take effect. Its health ministry still discusses dental benefits of fluoridation. The record does not establish that the policy was designed to facilitate chemical exports or amounted to a finding that water fluoridation at recommended doses was inherently poisonous. Israeli Ministry of Health; 2017 government report.

Will the fluoride shortage continue in 2027?

Some utilities anticipated continued disruption into early 2027. Separately, Nutrien plans to close White Springs in the second quarter of 2027. The effect on municipal fluoridation cannot be responsibly forecast without knowing the facility's current water-grade output and customer relationships. South Milwaukee notice; Nutrien announcement.

Will I notice anything if my city reduces fluoride?

A utility's reduction of a dental-health additive should not normally produce a noticeable change in water's appearance, taste, or odor, and it does not by itself mean the water is unsafe. The practical implication concerns the intended long-term dental-health benefit. Follow your utility's latest water-quality notice rather than assuming every city has the same status. Milwaukee Water Works FAQ.

What the Evidence Shows, and What It Does Not

Verified: American utilities reduced fluoridation in 2026 because of additive supply problems; some restored it. Fluorosilicic acid is recovered during phosphate processing. Domestic production exists. ICL has documented exports of that acid, including to the United States. Pencco's procurement records document a failed municipal bid and a claimed manufacturer withdrawal. Federal import statistics frequently used for fluorosilicic acid are chemically broader than their label suggests. Nutrien has announced a White Springs shutdown planned for 2027.

Claims that require qualification: The identity and timing of the manufacturer allegedly leaving the municipal market, the extent to which ICL's wartime operating constraints caused shortages in particular U.S. cities, and industry assertions about which cause mattered most nationally. These statements are attributable to specific businesses, utilities, or advocacy organizations, but not independently quantified in a national source-to-tap accounting.

Reasonable inference: A specialized supply chain with few producers, qualified product requirements, distributor relationships, and geographically constrained delivery can suffer local interruptions even if total annual output remains substantial. The upstream economics of phosphate fertilizer can contribute to the vulnerability. Neither inference establishes the numerical responsibility of a specific company.

Unknown: The precise U.S. municipal market share of each original producer and country, the source of every affected city's shipments, the full consequences of White Springs closing, and the amount of immediately available potable-grade capacity at facilities such as Conda.

A federal production statistic cannot answer those questions. A distributor's letter cannot answer them. And a viral claim cannot answer them.

The fundamental issue is traceability: When a public-health program depends on a specialized industrial commodity, the public should be able to establish who produced it, where it came from, what was delivered, what it cost, whether it met specifications, and how quickly a substitute can be secured. The 2026 shortage demonstrated why those answers matter.

Reporting Methodology and Research Origin

This article is a document-based research synthesis, not an account of confidential interviews or unpublished shipment records. We examined published U.S. federal supply-chain and mineral-commodity data, NSF certification materials, SEC and corporate disclosures, municipal purchasing and operational records, foreign government policy documents, and original scientific reviews.

We calculated the Albany bid difference, Kingsport's requested percentage increase, Mosaic's reported sulfur-cost change, and Itafos's production-to-nameplate-capacity ratio from the stated figures. Those calculations do not convert a product-grade or unit mismatch into a like-for-like comparison. We treated supplier allegations as allegations, company financial disclosures as statements about their expressly defined scope, and dated utility notices as evidence of conditions on the dates published.

Research origin: Questions about Israeli production, U.S. imports, and water-fluoridation policy were initially brought to our attention by an October 2026 Instagram video from Kali The Kid (@kali_the_kid805). The article's documentary findings and larger supply-chain framing come from independent examination of primary and authoritative sources. Crediting the original research lead does not imply confirmation of every claim in that video, including its allegations about artificial intelligence or intentional concealment.

References and Further Reading

Federal Data, Chemistry, and Trade Classification

Municipal Procurement and Supply Disruption Records

Manufacturers, Distributors, and Corporate Financial Disclosures

Health Science, Regulation, and International Policy

Editorial currency note: Supply-chain conditions, facility operations, utility dosing targets, certifications, court proceedings, and government policies can change. The findings above reflect documents available and verified through November 15, 2026. Historical figures retain their original reporting years; the municipality table is not a real-time status service. Corrections or new primary records should be incorporated with their dates and sources preserved.

Cite this article

Published November 15, 2026

Think something here is wrong, incomplete, outdated, or insufficiently supported? You can challenge a factual claim, source, interpretation, missing context, or privacy issue.

Learn How the challenge process works


More to think on...