FDA, LFGB and USP Class VI: What Each Certificate Actually Tests
Three names that get used interchangeably in quotation emails, and three completely different tests. Here is what a buyer is actually paying for.
There is no such thing as “FDA approved silicone”
The FDA does not approve materials. It publishes regulations, and 21 CFR 177.2600 sets out what a rubber article may release when it contacts food. A supplier claiming “FDA approved” is either simplifying or bluffing. What exists is a compound that passes the extraction limits.
The three standards, side by side
| Standard | Market | What it tests | Validity |
|---|---|---|---|
| FDA 21 CFR 177.2600 | United States | Extractables in distilled water and n-hexane after reflux | Compound + process specific |
| LFGB § 30/31 | Germany / EU | Overall and specific migration, plus a sensory panel | Compound + process specific |
| USP Class VI | Pharma / medical | In vivo and in vitro biocompatibility on the finished material | Material specific |
| ISO 10993-5/-10 | Medical devices | Cytotoxicity, irritation, sensitization | Device and contact-duration specific |
The part buyers miss
All of these are tied to a specific compound and a specific cure. Change the colour masterbatch, move from peroxide to platinum, or skip post-cure, and the previous report no longer describes the part in the box.
A test report is a statement about a sample at a moment in time. It is not a property of the material in general.
Choosing by application
- Food contact, US market: FDA 21 CFR 177.2600, platinum cured, post-cured.
- Food contact, EU market: LFGB plus a migration test; the sensory panel is where most compounds fail.
- Pharma, single-use: USP Class VI on the exact compound, plus the extractables package your QA asks for.
- Medical device: ISO 10993-5 at minimum, and -10 if the part is in prolonged contact.
Related technical notes
Both are silicone rubber, and on a drawing they look identical. The cure system decides what leaches out of the …
28 Oct 2026 · 7 min read
ID, wall and concentricity are three separate promises. Confusing them is the most common reason a tube that &ld…
21 Oct 2026 · 8 min read
Two tubes, same ID, same wall, 12% different flow rate. The variable was hardness — and it is usually the last t…
07 Oct 2026 · 6 min read
Questions about your own part?
Send a drawing or a sample. We will tell you which compound and hardness we would run, and what it costs at your volume.