Start with the answer: control one variable, and measure the other

When a polyphosphate media performs differently after a change, two variables are usually moving at the same time: the glass formula and the ball size. Formula determines how fast the glass dissolves per unit of exposed surface. Ball size determines how much surface area a given weight or volume of media exposes to water. Change both, and you cannot tell which one caused the result.

The practical solution is to express dissolution on a surface-area-normalized basis, and to run comparison batches where only one variable moves. Everything below explains how to do that in a development or supplier-qualification program for phosphate glass media used in scale-inhibition applications.

First, a terminology check that prevents wasted development time

Polyphosphate media is not one product. “Siliphos” historically refers to a family of silicate-phosphate glass balls used for scale and corrosion control in potable and process water systems. Generic “polyphosphate crystal,” “phosphate glass beads,” and other trade names may differ in composition, silicate content, additive ions, density, and dissolution behaviour. Treating them as interchangeable is a common source of misleading test results, and it also makes it easy to attribute a size effect to a formula effect, or the reverse.

It is also worth restating what these materials do. Polyphosphate media:

• Inhibit scale by threshold effect — a small phosphate residual interferes with calcium carbonate crystal growth rather than removing hardness.
• May contribute to corrosion inhibition depending on formulation and other water treatment chemicals present.
• Do not soften water. Hardness ions remain in the water.
• Do not filter. If particulate removal is required, that is a separate mechanical stage.
• Do not disinfect. Phosphate glass has no biocidal function.

Keeping these functions separate matters, because a “performance failure” after a media change may actually be a scale-control failure, a dosing-residual failure, or simply an expectation mismatch.

How formula and ball size each affect performance

The table below summarises what typically changes when each variable is altered on its own. It is a framework for designing experiments, not a set of guaranteed values — actual response depends on the specific glass and the water chemistry.

Property	Changing only the formula	Changing only the ball size
Dissolution rate per unit exposed surface	Changes — this is the intrinsic material property	Unchanged
Surface area per gram or per litre of bed	Broadly unchanged, except via density and porosity shifts	Changes substantially; smaller balls expose much more area
Initial phosphate release into the water	Changes with intrinsic rate	Changes with available area
Time to depletion at fixed mass	Changes with intrinsic rate	Changes inversely with total surface area
Pressure drop across the bed	Minor, unless density or shape changes	Significant; smaller balls raise head loss
Channeling and flow distribution	Minor	Can worsen with wide size distributions or fines
Fines and attrition behaviour	Depends on glass toughness	Smaller balls have more handling and attrition exposure

The key structural point: formula is a rate variable, and ball size is an area variable. Multiplying them gives the observed release into the water. Any test that reports only the product of the two cannot separate them.

The metric that separates them: surface-area-normalized release

For near-spherical media, surface area per unit mass is inversely proportional to diameter. Halving the nominal ball diameter roughly doubles the surface area available per gram. This relationship is why a size change alone can produce a large apparent performance change that has nothing to do with the glass chemistry.

To isolate the formula effect, express results as a rate per unit of exposed surface area — for example, phosphate released per unit area per unit time under defined conditions — instead of per gram, per litre of water treated, or per housing.

Two test modes are useful together:

1. Static or stirred-bottle test. A known mass of media in a known volume of water at controlled temperature and mixing. Sample over time. This gives you a release curve for a defined size fraction and lets you calculate an area-normalized rate.
2. Dynamic column test. A fixed bed with controlled flow, contact time, and inlet water chemistry. This gives you the effluent residual and the service behaviour that actually matters in equipment.

The static test isolates intrinsic dissolution. The column test shows what the customer experiences. Using only the column test is where most confusion starts, because bed geometry, flow, and water chemistry all interact with both formula and size.

A practical test plan that keeps the two variables apart

Step 1 — Fix the water chemistry

Use the same source water for every run in a comparison series, or a synthetic water prepared to a fixed specification. Temperature, pH, alkalinity, hardness, and chloride all affect phosphate glass dissolution and polyphosphate hydrolysis. If these drift between runs, no conclusion about formula or size is defensible.

Step 2 — Narrow and characterize the size fraction

Screen the media into tight size fractions before testing. Report the nominal diameter, the size range, and the method used. A batch supplied as “5–8 mm” behaves differently from a sieved 5–6 mm cut, even if the glass is identical.

Step 3 — Run three comparison arms

• Same formula, two sizes. Shows the pure geometric effect.
• Two formulas, same size. Shows the pure chemistry effect.
• Reference arm. An existing qualified product, run under identical conditions, to anchor the results.

Step 4 — Choose and state the comparison basis

Each basis answers a different question, and mixing them up causes most disagreements between development and field data:

Basis	Question it answers
Equal mass	What happens if the same weight of media is used?
Equal bed volume or depth	What happens if the same housing is filled?
Equal exposed surface area	What is the intrinsic formula effect?

Step 5 — Track cumulative release, not just peak residual

A high initial phosphate reading can come from fines, surface dust, or a small ball size with high surface area. Track cumulative phosphate released over the full depletion curve, plus any marker ions relevant to the formulation, such as silicate or zinc where present. The shape of the curve tells you more than any single sample point.

Interpreting results: what the numbers actually mean

If two media give the same area-normalized release rate but different effluent residuals, the difference is geometric or hydraulic — ball size, packing, contact time, or channeling. The formula is not the cause.

If area-normalized rates differ at the same size fraction and the same water chemistry, that is a genuine formula effect.

If a smaller ball size raises effluent residual but shortens service life, that is the expected trade-off between area and mass, not a quality problem. The correct response is usually to re-specify the size range, not to reformulate.

If a formula change was intended to slow dissolution but the field result shows a lower residual, check whether the size fraction also changed between the old and new supply. This is the single most common confounding error in polyphosphate media qualification.

Limits of the two-variable framework

Formula and size are not always fully independent. Very small beads can suffer more attrition, generating fines that dissolve quickly and distort early samples. Large beads in shallow beds may allow preferential flow paths, so a portion of the media contributes little. Wide size distributions pack differently from narrow ones, changing both bulk density and head loss. In these cases, run a third arm that varies size distribution width while holding nominal diameter and formula constant.

It is also important not to transfer dissolution data from one supplier’s product to another. Composition, thermal history during quenching, and post-treatment all influence dissolution behaviour. Any figure used for equipment sizing should come from testing of the specific material being specified.

What to record before comparing any two media

• Glass type and formulation description, with silicate or additive content if disclosed
• Nominal diameter, size range, and screening method
• Bulk density and, where relevant, apparent density
• Test water chemistry: pH, alkalinity, hardness, temperature, chloride
• Flow rate, contact time, bed depth, and column diameter
• Sampling intervals and analytical method for phosphate
• Whether results are reported per gram, per litre, per bed volume, or per unit surface area

Without this record, a comparison is anecdotal and cannot be repeated or defended.

Frequently asked questions

Does a smaller polyphosphate ball always release more phosphate?

At equal mass, yes — smaller balls expose more surface area per gram, so release is faster. At equal bed volume or equal surface area, the outcome depends on the formula and on how the bed hydraulics change.

Can I judge a formula change from a single effluent sample?

No. Polyphosphate release changes over the life of the bed, and polyphosphate chains hydrolyse to orthophosphate over time. A single sample reflects one moment in a curve.

Is food-grade polyphosphate media automatically suitable for drinking water?

No. Food-grade status describes one aspect of material quality. Suitability for drinking water depends on the exact product, the application conditions, the dosing level, and the requirements of the destination market. Those must be verified for the specific product and market before use.

Does polyphosphate media reduce water hardness?

No. It inhibits scale formation at substoichiometric doses. Hardness ions stay in the water. If hardness reduction is required, that is a different treatment step.

What if my formula and size both changed and I cannot re-run the old material?

Order the current formula in at least two size fractions, or obtain the previous size fraction from a retained sample. Without one anchor point, the two effects cannot be separated retroactively.

Planning a media development or qualification program?

If you are comparing polyphosphate or silicate-phosphate media for a specific application, ENVO CHEMICAL can review the technical picture with you. Useful information to include: the application and equipment type, source-water conditions (pH, hardness, alkalinity, temperature), the required media diameter and size range, your treatment objective, the quantity involved, and the destination market. With those details, the formula-versus-size question can be framed around your actual system rather than a generic specification.