Toothpaste Silica Abrasive: Balancing RDA and Cleaning Power

In short

A toothpaste silica abrasive has to remove stain and pellicle while keeping the finished paste inside its RDA window. Cleaning power and RDA both rise with abrasive load and particle hardness, but particle size distribution, the coarse tail and paste structure decide how far apart they sit. Set the RDA band first, then screen abrasive grade and level in your own paste.

Few numbers on a toothpaste brief cause as much rework as RDA. A development team picks a cleaning silica, builds a paste that looks and extrudes well, sends it for testing, and the result lands outside the agreed window. The usual reaction is to change the abrasive level and send it again. That works eventually, but it treats RDA as a property of the raw material when it is really a property of the whole paste. This article sets out what a silica abrasive actually controls, what it does not, and a screening order that gets cleaning and RDA right in fewer rounds.

What RDA measures, and what it does not

Relative Dentin Abrasivity is a laboratory comparison of how much a finished dentifrice wears dentin against a reference abrasive, run under ISO 11609. It describes the finished paste: the same silica can return quite different RDA values in two formulations because the slurry that reaches the surface depends on abrasive level, humectant and water balance, binder and the other solids present. No combination of raw-material figures on a TDS predicts it.

RDA also says nothing directly about cleaning. Stain and pellicle removal is measured separately, typically with a pellicle cleaning test on stained substrates. The goal of abrasive selection is not the lowest RDA or the highest cleaning score on its own, but the best cleaning result available inside the RDA band your product positioning, your market and the applicable standard allow. Check the current edition of ISO 11609 and your market's requirements for the limits that apply to your SKU.

Why cleaning power and RDA do not move in lockstep

If cleaning and RDA rose together in a fixed ratio, the only decision would be how much abrasive to add. In practice the ratio between them depends on how the abrasive is built and how it sits in the paste:

This is why two silicas with similar surface area can give a different cleaning-to-RDA balance, and why the comparison has to be made in your own recipe.

Properties to compare when choosing a toothpaste silica abrasive

The controlled TDS for each grade is the reference for these properties. Compare candidates on the same set of methods so differences are real, not artefacts of different test conditions.

PropertyMethodWhy it matters for abrasive duty
Particle size D50ISO 13320Sets the scale of cleaning action
Sieve residue (45 µm)ISO 2591-1Controls the coarse tail behind RDA outliers and grit
BET specific surface areaISO 9277Separates cleaning-route from structuring-route silica
Oil absorption (DBP)ISO 4652Indicates how much liquid the silica ties up and how it thickens the paste
Tapped densityISO 697Compact, dense grades lean toward cleaning; open grades toward structure
pH (5% suspension)ISO 6588Compatibility with actives and the paste's own pH target

Censil does not publish typical-value tables for these properties on the site. The figures for each grade come on the controlled TDS and lot COA sent with the sample, so your evaluation runs on the documents that will govern supply. The grade specifications page shows how each grade is positioned.

Where the Censil oral-care grades sit

Three grades are put forward on the oral care silica track, each in a different role in the screening:

The point of running them together is that the abrasive cannot be judged in isolation. Raising the cleaning level adds solids and ties up free water, so the paste thickens even with the structuring grade untouched; changing the structuring level changes how the abrasive is carried to the surface. Confirm first whether cleaning power, paste structure or storage stability is the priority, then arrange the comparison around it.

A screening plan that balances RDA and cleaning

  1. Fix the RDA band and the cleaning claim

    Write down the RDA window and the cleaning positioning before choosing a grade. Together they decide how much room the abrasive has and how much is left for structure.

  2. Check the cleaning grade's distribution

    Compare D50 and 45 µm sieve residue on the TDS and COA of each candidate. A tight distribution with a controlled tail gives the most cleaning per unit of RDA.

  3. Run abrasive level against structure level

    Build a small matrix — at least two cleaning levels against two structuring levels — rather than changing one factor at a time. It usually reaches the target formulation in fewer development cycles.

  4. Hold the paste constant for RDA and cleaning tests

    Send the same batches for RDA and cleaning so the two results describe the same paste. Record viscosity and density alongside, so a shifted result can be traced.

  5. Age the leading candidates

    Check extrusion, viscosity, sedimentation and any syneresis on centrifuged and temperature-cycled samples. A cleaning fraction that settles during storage changes both performance and appearance over shelf life.

Troubleshooting common RDA and cleaning problems

RDA high, cleaning adequate. Look at the coarse tail before reducing the abrasive level. If sieve residue or the upper end of the size distribution differs between lots or grades, that is the first suspect. Reducing level alone often costs cleaning as well.

RDA in range, cleaning weak. The abrasive may be under-represented at the surface rather than too soft — check whether the paste is over-structured. If structure is reasonable, a higher cleaning level with a matching cut in structuring silica is the next trial.

Results drift between batches. Compare lot COAs for particle size and sieve residue, then check process variables such as mixing order, vacuum and water addition. When bench and plant results diverge, the cause can sit in the material, the dosage or the process step; separating them is part of Censil's trial support.

Gritty feel reported in use. Treat it as a coarse-tail or dispersion problem first. Fineness, uniformity and mouthfeel are part of the validation set on the oral-care track, alongside RDA.

What to send when you request samples

A useful sample request covers the product type and positioning, the RDA window and cleaning claim, the current abrasive and structuring system, the actives and pH target, the mixing and filling equipment, and the problem you want to solve. With that, Censil can propose a lead grade plus contrast samples, each with its COA, so your trial is a real A/B rather than a single data point. Registered users can pull the TDS, COA and SDS set for the oral-care grades from the Download Center, or you can talk to an expert about the screening plan.

Screening a cleaning silica for a new or reformulated paste?Tell us your RDA window, cleaning claim and current abrasive, and we will propose a CENSIL 310, 320 and 150 comparison with the TDS, COA and SDS for each sample.
Talk to an expert

Related: Oral Care Silica  ·  Request a sample  ·  All news