Chlorine (Cl2) has a strong warning property: most people can detect Cl2 at approximately 0.002 ppm in air (the official odor threshold per the Agency for Toxic Substances and Disease Registry) and 0.31 ppm in water. However, most data indicate an odor threshold of 0.2-0.4 ppm. In other words, even the most minute leak will be detected by most humans, provided that your process is not allowed to run with known leaks. However, some chlorine rooms and processes are large and complex, and identifying the exact leak point can be challenging, especially when time is critical. Remember, Cl2 has an IDLH of 10 ppm and TLV of 1.0 ppm, so looking for a leak point REQUIRES adequate PPE – in my book, that entails a Supplied Air Respirator at a minimum! But this article is about the “leak detection” methods – how do we actually find the leak?
Most facilities will use the “Ammonia Gas” method. Small leaks of Cl2 (less than 25-50 ppm) are invisible to the eye; a larger leak will have a greenish-yellowish color, so when we are searching for the leak, we can smell – we may never see the actual gas being released. But when we “puff” ammonia gas into the air around our containment systems, the ammonia gas (also a colorless gas) will make a white smoke-like cloud, actually ammonia chloride, when it comes into contact with the Cl2 gas.
WARNING! We are using NH3 GAS to find the leak – NOT NH3LIQUID and even then, we MINIMIZE the Cl2 piping to NH3 gas exposure.
Some folks prefer a lab-style squirt bottle of household ammonia for their test method; others claim that a commercial 26º Be ammonia (28-30%) must be used.
NEVER EVER, NEVER EVER, NEVER EVER use Anhydrous Ammonia for Cl2 leak detection as we will make EXPLOSIVE NITROGEN TRICHLORIDE!
A commercial-grade (not home-use grade, which is typically only 16º Be) of Aqua Ammonia will work just fine – I do not recommend using home-use grade ammonia.
Some folks will use the modified squirt bottle and “puff” the ammonia gas from the headspace into the air in the proximity of the Cl2-containing equipment as they walk down the process searching for the leak point. Just be sure the liquid level in the squirt bottle is well below the lid to ensure we are not squirting liquid aqua ammonia onto the Cl2 process equipment. I was always instructed to take a cotton ball (or two) and soak in the 26º Be ammonia liquid and drop them into the squirt bottle so that we have little to no liquid ammonia in the bottle. And by using the 26º Be ammonia (vs. home-use concentrations) the concentration of ammonia is strong enough that the gas coming off the 1-2 cotton balls in the squirt bottle will work just fine. Some practices will use a cellulose acetate cleaning sponge – which is a fancy name for a sponge we would use at our kitchen sink! The idea is that the cotton balls or sponge will hold the liquid, which will release ammonia gas (which is what we want) and nearly ensure we do not squirt liquid ammonia into our process.
So what does it mean when I say “modified squirt bottle”? It’s a simple modification, but it is CRITICAL to ensure we are not squirting liquid ammonia onto our Cl2 process. Here is what the “Chlorine Manual” says:
4.4.2 Detection of Minor Releases and Leaks
A plastic squeeze bottle containing 26º Baumé aqua ammonia can be used to detect a minor release or leak. If ammonia vapor is directed at a leak, a white cloud will form, indicating the source of the leak. If a wash bottle is used, the dip tube should be cut off so that squeezing the bottle directs vapor, not liquid, out of the nozzle. Avoid contact of aqua ammonia with brass or copper.
Others used the “soaked rag” method – usually because they have corroded their Cl2 piping from squirting aqua ammonia liquid onto it by mistake and now they want to be certain there is no liquid contact with the Cl2 process. But this “soaked rag” method is also very handy and is used to find leaks too high to find with the “puff method”; although the ammonia gas will rise, it becomes very diluted with the ventilation air and is less effective at finding leaks high up in the area.
The “soaked rag” method will use the same concentration of liquid ammonia, but instead of “puffing” the gas, we soak a rag – a 1-time use rag!!!! – with the liquid ammonia and then attach it to a stick – much like a broom/mop handle. PLEASE make sure this “stick” is just an ordinary WOODEN stick that is not painted, stained, treated, etc.
Remember – Cl2 is a VERY STRONG OXIDIZER.
REMEMBER: if we can see Cl2, it is NOT a minor release and a complete emergency response is required.
| A PSM/RMP note: Using this 26º Be ammonia within your process, especially since its use is “critical” as it is being used to locate leaks, PLEASE be certain to include this chemical in your PSI and HAZCOM/PSM training. Using the 26º Be ammonia MUST be done via an annually certified SOP as well as a certified PPE Hazards Assessment. This activity will also fall under 1910.119(n) Emergency planning and response…
The employer shall establish and implement an emergency action plan for the entire plant in accordance with the provisions of 29 CFR 1910.38. In addition, the emergency action plan shall include procedures for handling small releases. Employers covered under this standard may also be subject to the hazardous waste and emergency response provisions contained in 29 CFR 1910.120 (a), (p) and (q). |
If you like a nice neat little package deal, our friends at Indian Springs sell a kit that contains:
- 1pc. 16oz Wash Bottle
- 5pcs. Cotton Swabs Instructions
- 1 pc. 16oz Aqua Ammonia (shipped separately-HAZMAT charges apply)
- MSDS