Reviewing trace element carryover from polyphosphate glass raw materials is really three questions, not one. What trace elements are present in the raw material? How much of the glass actually dissolves into the water under real operating conditions? And what concentration does that dissolved fraction produce at the point of use? Total impurity content answers none of these on its own, and a supplier certificate that reports only “heavy metals, max 10 ppm” will not tell you whether a given polyphosphate glass grade is acceptable for your cartridge, your source water, or your destination market.

This article is written for the people who have to make that call: filter and cartridge manufacturers specifying a polyphosphate or silicate-polyphosphate medium, water-treatment equipment builders qualifying a new raw material source, media blenders, importers doing supplier due diligence, and procurement teams comparing quotations from two phosphate glass suppliers. The goal is a defensible review procedure, not a slogan.

Start By Separating Three Different Numbers

Most disagreements about “carryover” come from comparing measurements that are not equivalent. Keep these three values distinct throughout the review.

A review that stops at the first column is a raw-material screen. A review that reaches the third column is an application assessment. Both have a place, but they are not interchangeable.

Define the Allowable Contribution Before You Test Anything

Testing without a target produces an interesting number and no decision. Set the target first.

1. Identify the applicable limit for each element of concern in the destination market and for the intended use. Drinking-water limits, action levels, health advisories, and industrial-water specifications are all different, and they change over time.
2. Decide what share of that limit the medium is allowed to consume. A common internal rule is 10 percent or less, leaving the rest of the budget for the source water and the distribution system. The number is a judgement call, but make it explicit and record the reasoning.
3. Account for the source water. If the incoming water already carries 2 µg/L of an element and the limit is 10 µg/L, the medium’s allowance is smaller than it would be on a cleaner supply. Carryover is additive, not absolute.

As illustrative reference points only — and these must be confirmed against the current regulation for your market and application — the WHO guideline value for arsenic in drinking water is 10 µg/L, the US EPA maximum contaminant level for cadmium is 5 µg/L, and the US EPA action level for lead is 15 µg/L. Do not treat those figures as a specification for any product; treat them as the kind of number you must look up yourself.

Work From Composition to a Worst-Case Water Concentration

Once you have a target, you can screen the raw material arithmetically before spending money on leach testing. The conversion is simple because the units align:

Cwater (µg/L) = Cimpurity (mg/kg in the solid) × Dglass (mg of glass dissolved per litre of water)

The reason is that 1 mg/kg equals 1 µg/mg, so multiplying by a dissolution expressed in mg/L gives µg/L directly.

Worked example with invented numbers: a raw material contains 5 mg/kg of an element of concern, and under the worst credible operating conditions you estimate that 1 mg of glass dissolves per litre of water. The worst-case contribution is 5 µg/L. Compare that with your allowance from the previous step and the decision is immediate — accept, reject, or investigate further.

Two practical notes on the dissolution term Dglass:

• If your supplier quotes a dissolution rate as phosphate (mg/L PO4), you can convert it back to glass mass using the phosphate content declared for that specific grade. Confirm the declared figure rather than assuming one, because grades differ.
• If no dissolution data exists, do not substitute a number from a different product or a different manufacturer’s datasheet. Treat the dissolution term as unknown and measure it.

Where the Trace Elements Actually Come From

It helps to know which part of the supply chain you are auditing, because the control measures differ.

• The phosphate source. Phosphates derived from mineral rock can carry cadmium, arsenic, lead, uranium, and fluoride at levels that depend on the ore body and on how much purification the phosphate received. Purified or food/pharmaceutical-grade phosphate feedstocks are generally held to tighter heavy-metal specifications than technical grades, but “generally” is not a specification — ask for the number.
• The co-reactants. Sodium, calcium, and aluminium salts used to build the glass each carry their own impurity profile.
• Processing and handling. Milling media, drying air, furnace refractories, and packaging can all introduce contamination. A result that looks impossible from the formulation alone often traces back to the mill.

Do Not Confuse Intentional Constituents With Contamination

Some elements in a polyphosphate glass are there on purpose. Aluminium, for example, is a deliberate component of several sodium-calcium-aluminium polyphosphate glass formulations, and calcium is a structural component rather than an impurity. Reporting aluminium as “carryover contamination” in such a product is a category error — the correct question is whether the intentional release profile is acceptable for the application, which is a formulation and dosing question, not a raw-material purity question.

The same logic applies to phosphate itself. Phosphate release is the function of the product, not a defect. Trace element carryover is about the minor and unintended fraction.

Running the Leach Test Properly

If the arithmetic screen leaves you close to the target, test. Keep these points in mind so the result means something.

1. Make the test conditions represent the application. Temperature, pH, contact time, liquid-to-solid ratio, and water chemistry all change dissolution. A test in ultrapure water at 20 °C may understate or overstate carryover relative to warm, low-alkalinity water in a real system.
2. Test the fines separately. Fines dissolve much faster than whole beads or coarse crystal because of their greater surface area. If a cartridge can release fines at start-up, the fines fraction deserves its own test.
3. Use an analytical method with a reporting limit well below your target. A method whose detection limit sits at or above the concentration you care about produces a “not detected” that means nothing. For µg/L work in water, ICP-MS is the usual choice; ICP-OES suits higher concentrations.
4. Control contamination. Digest solids in a closed microwave system with trace-metal-grade acid. Avoid grinding samples in steel or tungsten carbide mills, which add iron, chromium, nickel, cobalt, and tungsten. Run method blanks and a certified reference material with every batch.
5. Do not substitute TCLP or a waste-classification leach test. Those tests exist to classify waste; they are not contact-material assessments and their conditions rarely resemble a drinking-water or process-water application.
6. Sanity-check with a mass balance. Over the expected service life, the total mass of an element released cannot exceed the total mass present in the medium. If the numbers imply otherwise, the measurement or the dissolution estimate is wrong.

Factors That Change Carryover in Real Equipment

• Temperature. Dissolution of phosphate glass generally accelerates as water warms.
• pH. Acidic or poorly buffered water tends to attack the glass more aggressively than hard, alkaline water.
• Contact time and stagnation. Long residence times — overnight stagnation in a cartridge housing, for instance — produce higher concentrations than continuous flow.
• Surface area. Particle size, fines content, and how the medium is packed all matter.
• Dose and dissolution rate. The more glass that dissolves, the more of everything it contains is released. Carryover scales with the function.
• Flush protocol. First-flush volumes after installation or cartridge replacement are typically the highest-carryover period. Define the flush and sample after it, or sample the flush itself deliberately — just be clear which you are reporting.

What a Certificate of Analysis Can and Cannot Do

A supplier CoA is useful for lot-to-lot consistency and for screening. It is not proof of application suitability. Check whether the CoA states the test method, the digestion procedure, and the reporting limit for each element. A result reported as “<10 mg/kg” with no method information cannot be compared with a result reported as “2 mg/kg by ICP-MS after microwave digestion,” even if they look like the same measurement.

Also verify the scope of any certification. A material described as food grade has met a specification for food contact or food additive use; that does not by itself establish suitability for drinking water in any particular market. Drinking-water suitability is normally established through the applicable national approval scheme — for example, NSF/ANSI/CAN 60 in North America or the relevant European standard and national approval route — and it applies to a specific product, a specific use, and a specific maximum dose. Confirm the scope of the approval rather than assuming it transfers.

One more naming caution: “Siliphos,” “siliphos crystal,” and generic “polyphosphate media” are not guaranteed to be the same material. Branded products, generic equivalents, and silicate-containing variants can differ in composition, dissolution behaviour, and approvals. Review the actual grade in front of you.

Frequently Asked Questions

Does a low total heavy-metal result guarantee low carryover?

No. It limits the maximum possible carryover but says nothing about how fast, or how much, the glass dissolves in your application. A cleaner material that dissolves quickly can contribute more than a slightly less clean material that dissolves slowly.

Is carryover the same as water softening?

No. Polyphosphate glasses are used for sequestration and threshold scale inhibition, and historically for corrosion control in distribution systems. They do not remove calcium and magnesium the way an ion-exchange softener does, and they do not filter trace elements out of water.

Can a polyphosphate medium reduce metal levels in water?

Polyphosphates have a long history in corrosion-control programmes, where they can help limit the release of pipe-derived metals such as lead and copper. That is a distribution-system effect and is entirely separate from the trace elements the medium itself may contribute. Both effects can be present at once, which is why source-water and outlet sampling are both needed.

How often should carryover be re-verified?

At minimum when the raw material source, supplier, or grade changes; when the equipment design or operating conditions change; and when the applicable regulatory limit changes. Ongoing verification then rests on incoming-lot CoAs against a specification you have already validated.

What if the data simply does not exist?

Then the honest conclusion is that carryover is uncharacterised. State which elements, which conditions, and which reporting limits are missing, and either commission the testing or restrict the application until the gap is closed. Filling the gap with a plausible-looking number from a different product is the most common and most expensive mistake in this kind of review.

Building the Review Into Your Specification

The end product of this exercise is not a report — it is a short incoming specification. It should list the elements of concern, the maximum total content for each, the test method and reporting limit, the sampling plan, and the leach or in-service data that supports the limit. Once that document exists, subsequent lots can be accepted or rejected in minutes rather than weeks.

If you are evaluating a polyphosphate glass raw material for a specific programme and want to compare it against your acceptance criteria, send us the details that determine the answer: intended application and equipment type, expected contact time and operating temperature, source-water pH, alkalinity, hardness and background metal levels, the elements and limits you are working to, the grade and quantity you are considering, and the destination market. With those inputs, ENVO CHEMICAL can discuss the relevant material data and what still needs to be confirmed for your case.