Membrane filtration
The sample is passed through a fine filter that traps any microorganisms, and the filter is then placed in growth media.
(“stuh-RIL-ih-tee TEST-ing”)
Why “passed sterility” is the last line of evidence, not the first.
The short answer
A sterility test checks whether samples from a lot contain any living bacteria or fungi by giving them ideal conditions to grow and watching for growth.
Samples are transferred into two growth media, one suited to bacteria that grow with little oxygen and one to fungi and aerobic bacteria, and incubated for 14 days. No growth means the lot passes. But only a small number of units can be tested, so the test reliably catches only fairly high contamination rates. Sterility is really assured by validated manufacturing processes; the test is a final check.
Reviewed by the Codex Research quality teamUpdated Nov 8, 2026
Word study
Say it
ste·ril·i·ty test·ing
(“stuh-RIL-ih-tee TEST-ing”)
Roots and origin
RootLatin
OriginLatin
Sterile first described barren land that produced no crops. A sterility test asks the same question of a sample: given every chance to grow, does anything living appear?
Around the world
Tap a language to hear it spokenDefinition
A sterility test asks whether a sample contains any living microorganisms. Rather than looking for them directly, it gives them ideal conditions to multiply. Samples are placed into nutrient media, incubated, and watched. If anything living was present, the media turn cloudy with growth; if nothing grows, the sample passes.
The compendial test uses two media in parallel, because no single medium grows everything. One, fluid thioglycollate medium, supports bacteria that need little or no oxygen; the other, soybean–casein digest medium, supports fungi and aerobic bacteria. Incubation runs for 14 days at two temperature ranges. The sample is either filtered through a membrane that is then placed in the media, or added directly to them.
The test has a statistical limit that cannot be engineered away. Only a small number of units, commonly 20, can be tested from a lot that may hold thousands. If one unit in a thousand were contaminated, a 20-unit test would miss it about 98 times out of 100. Regulators and pharmacopeias say this plainly: the test can detect only fairly high contamination rates.
That is why sterility is assured by process, not by testing. Products that can be sterilized in their final container are processed to a validated level of sterility assurance. Products that cannot, including most freeze-dried peptides, are made by aseptic processing, where FDA guidance focuses on facility design, environmental monitoring, personnel and process simulations. The sterility test is the final confirmation on top of that, not the main evidence.
Sterility and endotoxin are different questions. A sterility test looks for living organisms; an endotoxin test looks for fragments of bacterial cell walls that remain even after bacteria are dead. A sample can be sterile and still carry endotoxin, so the two results answer different questions and both can appear on a certificate.
The test
The sample is passed through a fine filter that traps any microorganisms, and the filter is then placed in growth media.
A portion of the sample is added straight into the growth media. Used when filtration is not practical.
Fluid thioglycollate medium for bacteria that need little oxygen; soybean–casein digest medium for fungi and aerobic bacteria.
Media are incubated and inspected for turbidity or other signs of growth over two weeks.
Validated alternatives that detect microbial growth or activity faster than the classic culture method.
Interactive
Set how many units in a lot are contaminated and how many are tested. The odds that a contaminated lot is caught are lower than most people expect.
100 contaminated lots like this one
Probability of including at least one contaminated unit = 1 − (1 − rate)units, assuming contamination is spread randomly. This is why regulators treat a passing sterility test as the last check, not the main evidence: sterility is assured by validated processes, and the test can only catch relatively high contamination rates.
For buyers
A pass shows no growth in the units tested. It cannot rule out low-level contamination in the rest of the lot.
Dead bacteria can leave endotoxin behind. Sterility and endotoxin answer separate questions.
Validated sterilization or aseptic processing is what keeps contamination rates low enough to trust.
At Codex Research
Each Codex Research product page carries a lot-matched certificate of analysis for the lot in stock. For microbial testing on a specific lot, contact us and we will tell you what was performed.
See it live: the endotoxin guide in the glossary
Myth vs fact
Tap a card to flip it.
Watch for
Quick check
Membrane filtration or direct inoculation, with a lab named.
A result dated after a full incubation period.
A separate endotoxin result, not assumed from sterility.
Full walkthrough · JournalWhat endotoxin testing meansQuestions
A test that incubates samples in growth media for 14 days to see whether any living bacteria or fungi are present.
To give slow-growing organisms time to multiply to a visible level in the media.
No. It tests a small sample, so it can only reliably catch fairly high contamination rates.
Sterility looks for living organisms. Endotoxin testing looks for bacterial cell-wall fragments that remain even after bacteria die.
Fluid thioglycollate medium and soybean–casein digest medium, which together support a wide range of bacteria and fungi.
Manufacturing a sterile product from sterile components in a controlled environment, for products that cannot be sterilized in their final container.
Commonly 20 per lot, depending on lot size and container volume.
Sources and standards
Written and reviewed by the Codex Research quality team · Last reviewed November 8, 2026 · For laboratory research use only.
On real products
The current certificates for these products report a sterility testing result for the lot in stock. See it on each product page, then read the full guide.
BlendsResearch peptide
Repair Blend
Wolverine is a two-peptide research blend with equal masses of BPC-157 and TB-500, sold as 10 + 10 mg and 15 + 15 mg vials, each lot assayed by HPLC and identity-tested by LC-MS.
CellularResearch peptide
Cellular Research
SS-31 (elamipretide) is a synthetic tetrapeptide that binds cardiolipin at the inner mitochondrial membrane, studied in bioenergetics and oxidative-stress research.
RepairResearch peptide
Immune Research
Thymosin alpha-1 is a 28-residue acetylated peptide isolated from calf thymus in 1977 and generated from prothymosin alpha, studied in T-cell biology.
Journal guideWhat endotoxin testing meansRead the guide
From lab to your door
U.S.-made research vials, shipped with lot-matched, lab-tested COAs you can scan straight off the label.
21+ Only
Codex Research provides research-grade compounds for laboratory and research use only. By entering, you confirm you are at least 21 and agree to our Terms & Conditions.
By proceeding you affirm the statements above are true. Products are supplied for laboratory research use only — not for human or veterinary use, not for use in diagnostic procedures, and not evaluated by the U.S. Food and Drug Administration. Full disclaimer.