Control of Substances Hazardous to Health Regulations

Confined Spaces at Work Regulations.

It is up to you to decide what control measures you require, for the hazards & risks that you identify.

Hi Steve

Yes, this is for dosing at the mains for the purpose of dosing for chronic legionella. We are stuck with this system as it is the wishes of the client however, we are providing the chemicals and filling the containers when necessary. There are two containers hydrochloric acid and sodium chlorite, the unit mix the two chemicals.

The concern I have is if the unit fails then there could be a sudden release of chlorine dioxide. I have been told that the unit is standard practice in the water treatment/hygiene industry, but surely wither an alarm or, better more, a safer chemical used to dose the incoming water supply.

Sorry jarsmith - I am still a little confused... I have come across emergency dosing of mains to combat potential contamination following (or during) any intrusive works or a break that required pipe replacement. Chlorine (or chlorine dioxide) injection might then be considered to disinfect the supply from all sorts of nasty bugs. But that would also be associated with a major effort to flush through the mains any contaminated water by opening hydrants and etc...

If the system has been installed in a water treatment works (as may be the case) then again - it is for disinfection from a whole lot of nasties not just (or even majorly) legionella... The amount of dosing is carefully calculated to match the contaminant loading at any particular time.

I have never come across a mains service requirement for consumers which would use this treatment...

Using Hydrochloric and hypo seems to be a recipe for problems in small systems. Even very small process variations (which always occur) will result in massively changing levels of ClO2 in the treated water - it will be alternately noticeable by taste and / or smell - and then effectively absent... Control systems for this very vigorous process must be controlled accurately to within ppm variation... On large public supplies, this is balanced by large intermediate storage tanks - which would be impractical in any consumer situation that I can imagine... Use of weaker acid (CO2 injection) is inherently more controllable.

Is this onshore UK or are you looking at tropical or offshore installations where supply purity may be in question?

If you want to PM more info I will try to understand the issues you're facing, and help if I can. I would repeat my earlier comment though that there is no explicit statutory duty (which is what the original question asked)

I think this has gone a little off track, probably for good reasons. It is quite within normal scope to introduce chlorine dioxide to premises hot and cold water systems to deal with persistent microbiology. In fact in many older buildings, particularly healthcare, its getting to be quite common. There are some downsides but constant low level dosing helps defeat legionella, pseudomonas and number of other nasties where thermal and chemical disinfection have failed and redesign is essentially impossible.
I think it's right that alternatives to ClO2 it be considered but the actual generation unit sounds fairly standard to me. It's use in this regard is discussed at length in the new HSG274 part 2 (released only within the last week).

As for ventilation...in essence. Yes. There should be some if there is a foreseeable risk of accidental release into an area where people might be or might go e.g. to repair the fault! This would depend upon the volumes that could be released, the size of the room, speed of release etc. so would need to include knowledge of the plant but I would have thought that the tendency should be to provide the control measure rather than look for reasons not to - all coming back to need to control potential exposure under COSHH.
Look for safety data sheets for the final product (I imagine you won't have any for the mixed one) and they'll all talk about protective measures. I know that MSDS can often overstate and over control but it does have a short term WEL of 0.3ppm over 15mins.

Thus, subject to the need to risk assess, I would be suggesting that control measures of the kind you have talked about are in order.