Do cosmetic powders need sterilisation, and what does each method cost?
Start with the conclusion: not every powder needs sterilisation. Powders are low in moisture and low in water activity, and the microbial load usually comes from packaging, the environment and storage. Controlling water activity, packaging cleanliness and warehousing costs less than treating at the end.
Where treatment is needed, the three routes have different costs. Gamma irradiation from cobalt-60 penetrates well and runs at ambient temperature; 5-10 kGy typically drops the total plate count by three to four orders of magnitude. The cost is that organic treatment layers — silane, stearic acid — can degrade, and colour powders can fade or yellow, so colour difference and oil absorption must be re-measured before and after. Some markets also require an irradiation mark on the pack. Ethylene oxide penetrates poorly and needs vacuum cycles; a loose powder bed does not help gas diffusion, residual EtO and 2-chloroethanol have to be tested, and in practice the method suits packaged finished goods rather than bulk powder. Dry heat at 160 °C for two hours, or 180 °C for half an hour, leaves no residue and is simple; the cost is caking in powders that carry moisture, and oxidation of organic treatment layers.
Work in this order: measure the microbial limits first — total plate count, moulds and yeasts, thermotolerant coliforms, Staphylococcus aureus — and only choose a method if they fail. Then verify the chosen method on a small batch for treatment integrity and colour shift before scaling.
One point is often missed: the effect on colour powders is not linear. Dark shades such as carbon black and black iron oxide usually come through unchanged, while pale and red shades show the shift first, so a single pale sample is not a fair trial.