Body-safe silicone: what to require

Is "body-safe silicone" a material grade I can put in a contract?

No — it is an adjective, and an unmeasurable one. According to the Wikipedia entry Silicone, silicone is a family of polymers whose properties depend on how they are compounded and cured, so the phrase says nothing a laboratory could test for. Write the cross-linking chemistry, filler content, Shore hardness and the substances that must be absent into the specification instead, and ask for a signed material declaration for every skin-contact component. The declaration formats are listed in the certification pack.

Is body-safe silicone a grade, and which test should you name instead?

No standards body defines body-safe as a material grade. It is an adjective doing the work of a specification, and it is effective precisely because nobody can be proven wrong. According to the Wikipedia entry Silicone, the material is a synthetic polymer family whose properties depend entirely on how it is compounded and cured — which is why the adjective alone tells a buyer nothing. What actually protects a buyer is specifying the things that are measurable: the curing chemistry, the filler loading, the hardness, and the substances that must be absent.

Our own range uses medical-grade silicone over a food-grade ABS chassis, and every skin-contact component carries a material declaration. That is stated as a specific claim rather than a general one because the general claim is unusable in a compliance file.

Which chemistry question sorts silicone suppliers fastest?

Silicone is cross-linked from a polymer into an elastomer, and there are two common routes. Platinum addition curing uses a platinum catalyst, produces no reaction by-products, and needs no extended baking step to remove them. Peroxide curing is cheaper and tolerates a wider range of compounds, but it produces volatile by-products that require a post-production baking step; shorten that step and residues remain.

Every row above is answered on paper or not at all. A supplier who explains curing chemistry in a call and cannot put it in a material declaration has told you something about their document control — which is why these answers belong inside the shortlisting method, at the evidence stage, rather than in a conversation.

Platinum addition curing leaves no peroxide by-products to extract, which is why it is the route specified for skin-contact parts. [Noll, 1968] covers both chemistries and their residues in the detail a material engineer would want, and the practical difference shows up in the extraction test rather than on the datasheet.

Why do filled silicones explain most quality complaints?

Industrial compounds are extended with fillers to reduce raw material cost and improve processing. Heavily filled compounds tear more readily at the surface, leave a chalky residue when handled, and vary in colour between batches because the filler distribution varies. They also complicate testing, because the filler chemistry has to be assessed alongside the base polymer rather than assumed inert.

You can detect heavy filling without a laboratory. Stretch a sample slowly and release it: nothing should whiten, tear, or stay deformed. Press a fingernail gently into the surface — it should recover cleanly. These are field tests, not certification, but they catch the majority of substitutions before you pay for anything.

Which silicone certificate do you actually need?

• FDA 21 CFR 177.2600 sets composition requirements and extractables limits for rubber articles intended for repeated use. It is a material benchmark in this category, not an approval of a finished device — see the chemical standards guide.
• REACH restricts substances of very high concern in articles sold in the EU. Evidence is a third-party report against the current candidate list.
• RoHS restricts lead, mercury, cadmium, hexavalent chromium and certain flame retardants, and covers the electronic subassemblies rather than the silicone.
• MSDS gives the exact chemistry of the material supplied. It is the document most often promised and least often delivered.
• ISO 3533:2021 is the only international standard written for this category rather than for a compound. Published in September 2021, it covers design, materials, risk management and the information that must reach the user — not extractables limits. It is voluntary, and it is the nearest thing this trade has to a specification for the product itself.

The set we issue per skin-contact component, with the issuing laboratory named, is listed in the document pack for every shipment. The measurement conditions behind every published figure are on the test methods page, and the material question is answered model by model in the wholesale FAQ.

What does ISO 3533 cover, and why does it matter to a buyer?

Every document above is a rule about a material or an electrical property. None of them says anything about whether a product is shaped in a way that will not have to be removed in an operating theatre — and that gap is what ISO 3533:2021 was written to close. According to the Wikipedia entry ISO 3533, the standard began as a Swedish proposal to the ISO after hospital data showed a rise in retained products, was opened to stakeholder comment through ANSI, and was published as a first edition in 2021.

In practice it asks for four things a material certificate cannot give you. A documented risk analysis across the product life, not a one-off test. Design that resists retention and avoids burrs, sharp edges and parts that can come loose — with specific provisions for anal-use products, where the failure mode is the expensive one. Coverage of vibration and electrical safety, and of what happens when a wireless remote loses its link. And user information in the box: cleaning, storage, warnings, and the material facts a buyer would otherwise have to guess at.

It is a voluntary standard, and no customs officer will ask for it. That is exactly why it is diagnostic. A supplier who has read it can tell you how their anal-use models are shaped to be extracted, what their restricted-substance list contains, and what the insert says about cleaning a motorised unit. A supplier who has not will answer the same question with the word "silicone" — which is a material, not a design, and answers none of it.

The scope is narrower and more useful than the summaries suggest. ISO states it in one sentence: "This document specifies safety and user information requirements relating to the materials and design for manufactured products intended for sexual use" (ISO 3533:2021). Read that twice. It names three subjects — materials, design, and the information that reaches the user — and it limits itself to products in direct contact with genitalia and/or the anus, while stating that it is not primarily intended for products classified as medical devices, cosmetics or assistive products, lubricants and massage oil among them. A supplier who claims the standard covers their entire catalogue is describing something broader than the document does.

What the free record does not give you is the clause text. The standard runs to 16 pages, is sold by ISO for CHF 100, and its clause numbering and annexes sit behind that paywall. That is worth saying plainly, because this is a category where pages recite "ISO 3533 clause 5.2" as though it were public knowledge: unless the writer has bought the document, they are reciting another blog post. We do not reproduce clause numbers here, and a supplier who quotes them fluently should be asked to show you the page they came from — it is a two-second question that separates a technical file from a sales deck.

Nor does it replace the numeric gates above. 21 CFR 177.2600 still sets the extractables limits a rubber article has to meet, REACH still governs which substances may be present at all, and RoHS still covers the electronics. ISO 3533 asks a different question — whether anyone assessed the design against foreseeable failure, and whether the person opening the box was told what they need to know. A file that clears every number but has no design review is complete on paper and incomplete in the standard's own terms; a design review with no extractables report is the mirror image. Buyers need both, and the second one is cheaper to obtain than most expect: it is a table of contents and a signature, not a laboratory invoice.

What durometer should a vibrator be, and what does the number mean?

Hardness is the only material property a buyer can hold in their hand and still misread. Shore A is measured by pressing a calibrated indentor into the surface under a fixed spring load and reading how far it sinks — a number from 0 to 100, taken at a stated thickness after a stated delay, because the reading drifts as the material relaxes. Two labs can disagree by several points on the same part if the specimen is thin or the reading is taken late.

Hardness and filler loading interact, which is why the two questions belong together. A compound can be made softer by adding plasticiser or by choosing a softer base polymer, and only one of those is stable: a plasticised compound loses its softness as the plasticiser migrates, which is what a buyer experiences as a part that "went tacky" after a year in a warehouse. [Brydson, 1999] sets out the compounding relationships; the practical takeaway is that softness achieved by formulation holds, and softness achieved by additive does not.

Why is post-curing the step most easily skipped?

Both curing routes leave something behind. Peroxide curing produces volatile by-products that have to be driven out in a bake after cross-linking; platinum addition curing produces none in the reaction, but the compound still contains low-molecular-weight fractions that migrate to the surface over time unless the part is finished properly. The bake is a time-and-oven cost with no visible output, which is exactly the kind of step that gets shortened when a line is busy.

The symptoms are consistent and a buyer can test for them without a laboratory. Seal a sample in a clean glass jar for a day at room temperature and open it: a sharp chemical smell, or a film on the glass, means something is still coming out of the material. Wipe the surface with a white cloth — a grey or chalky transfer is filler or mould release, not silicone. Neither test certifies anything, and both catch a shortened bake before you place an order.

What will the customer put on the surface, and what does that rule out?

A skin-contact elastomer is never used alone. It meets lubricants, cleaning agents and the inside of a storage pouch, and each of those is a chemical exposure that a material certificate does not cover. Silicone oil is the obvious case: it is chemically close to the elastomer, so a silicone-based lubricant can swell and soften a silicone part over repeated contact, while a water-based one generally does not. That is not a reason to ban a product category — it is a reason to test the combination you intend to sell.

1. Name the lubricant base in the brief — water, silicone or oil — rather than "compatible with standard lubricants".
2. Ask for an immersion test on the actual compound: a stated duration at a stated temperature, with swell and hardness change measured before and after.
3. Check the pouch too. A soft PVC pouch with plasticiser can mark a silicone surface in transit, and the mark arrives as a customer complaint about the product.
4. Write the cleaning instruction you can defend. Alcohol wipes are common and generally tolerated; solvents are not, and a claim that survives one will not survive the other.

None of this appears on a data sheet, because a data sheet describes a material rather than a product. It is the sort of question a supplier either has an answer to or has never been asked — and the difference tells you how the material was chosen. The material declaration pack covers what ships with an order; the published specification sheets give the hardness, dimensions and battery figures model by model, so the material question and the measurable question can be asked against the same document; and the the bench, the meter and the four QC gates covers how each claim is measured.

References

1. Brydson, J. A. (1999). Plastics Materials, 7th ed. Oxford: Butterworth-Heinemann.
2. Noll, W. (1968). Chemistry and Technology of Silicones. New York: Academic Press.
3. U.S. Food and Drug Administration (2024). 21 CFR 177.2600. Code of Federal Regulations, Title 21.