We are developing a UV sterilization device for in-line and bottled water.
Any idea what standards do we have to conform with and registration that we need to do? FDA? EPA? or other?

Thanks,
Yoram

We are developing a UV sterilization device for in-line and bottled water.
Any idea what standards do we have to conform with and registration that we need to do? FDA? EPA? or other?

Thanks,
YoramTake a look at NSF for some relevant standards. (http://www.nsf.org/business/drinking_water_treatment/standards.asp?program=DrinkingWatTre)

Yoram;
Several items require clarification for developing an answer:
When you say "sterile" - for what purpose; e.g., drinking, injectable, process, other?
Where do you want to sell your device - US, EU, Israel, other?
Who are your target customers - drug companies, food processors, home use?

Most of the companies use UV sterilizer at 254 nm. There are no USP limits for purified water or water for injection on as to what should be the efficiency of the sterilizer.

But I think there are limits for purified water for total plate count (100 CFU/ml) and its less for water for injection. I dont know what it is but refer to USP 22 and 23 for these limits.

Also gewater.com has good information.

Chirag

Check out this quote from CFR21 part 129:

Subpart E—Production and Process Controls
§129.80 Processes and controls.

(a) Treatment of product water. All treatment of product water by distillation, ion-exchanging, filtration, ultraviolet treatment, reverse osmosis, carbonation, mineral addition, or any other process shall be done in a manner so as to be effective in accomplishing its intended purpose and in accordance with section 409 of the Federal Food, Drug, and Cosmetic Act.

You can take this to mean that so long as you're water is in spec, then your UV is great.

For the full monty, go to:

www.washingtonwatchdog.org/documents/cfr/title21/part129.html

The EPA guideline for drinking water bacterial contamination is 500 CFU/mL using the pour-plate method.
The current USP (28) does not set limits for bacterial contamination for Purified Water or Water for Injection. General Information chapter <1231> Water for Pharmaceutical Uses, under the heading Alert and Action Levels suggests that "generally considered appropriate Action Levels are 500 cfu per mL for Drinking Water, 100 cfu per mL for Purified Water and 10 cfu per 100 mL for Water for Injection."
Ultimately, it is up to you to set your own limits based on the functioning of your system. We have set an Alert LEVEL of 25 CFU/mL and Action LEVEL of 50 CFU/mL for our Purified Water, USP systems. Notice that we call them levels rather than limits. A subtle distinction, but one that can save you tons of heartache down the road.

If you are processing 'raw water' (potable water) into a grade of water for pharmaceutical use (or if you are claiming Purified Water or 'Sterilised') then you have to meet the relevant pharmacopoeial criteria for that grade of water.

For drinking water, the typical acceptable TVC (total viable count) is given as 500cfu, with a total absence of Coliforms (including E.coli), Cyrptosporidium, Legionella, Viruses and Giardia.

For Purified Water (not sterile) the TVC is lowered to 100cfu per ml and it is usually good practice to set an action limit of 100cfu and an alert limit of 50cfu. The same situation applies with respect to the other named species (pathogens)

For WFI (water for injection), the TVC limit is again lowered to less than 10cfu per ml.

What MUST be emphasised is that NONE of the above grades of water are 'Sterile'. Sterile materials must be demonstrated by verification measures (Sterility Assurance), as defined in the USP/Ph.Eur. Sterilisation of 'water' can be conducted by terminal sterilisation or aseptic processing using ultra-filtration. It is usually practiced to produce WFI and to then 'sterilise' this material, in order to create a 'sterile' product. Suitable information on this is probably best found from a business manufacturing opthalmic preparations (e.g. contact lens solution manufacturers).

It isn't simply acceptable to 'set your own limits' and then make claims of 'sterile', without proving sterility by analytical means.