[strokebow]

Hi,

I have a dissolved CO2 sensor. It  gives an output which is related to the amount of CO2 dissolved in the water.

My question/problem: How can I calibrate this/verify this?

Ideally: I would make up a solution of KNOWN dissolved CO2 concentration. Making a few different solutions up. e.g.) 5, 10, 20, 40ppm. Then I would measure the reading in each case for each standardised solution.

How would one do this in practice? My issues:
1. I guess first the water would need to be degassed. How do people do this normally? By boiling water to 100degerees c then cooling?
2. Then measure the volume of the degassed water.
3. Add CO2 gas. How do you do this? (I'm guessing straight after I have boiled the water gas will start dissolving back into the water straight away from the atmosphere?) Adding the CO2. Would you need to have pure CO2 and then bubble through a given volume of CO2 gas by measuring the flow over a given time. How do you ensure that all the gas 'bubbled' through actually gets dissolved in the water?
4. At this point a known conc of CO2 dissolved in water should have been achieved. How should this be stored in order to stop gas diffusing in/out of the water. If test is done straight away - would this be accurate?

I'm guessing people have done this sort of things many times before. Can anyone help/offer advice or answer any of my questions raised?


Thanks!!

[curiouscat]

What if you take a CO2 cylinder & bubble gas but through a Temp. controlled water bath? The saturation conc. must be available easily.

[strokebow]

Thanks for reply curious cat. Can you explain how this will work - not quite sure what you mean.

thanks

[Borek]

What kind of sensor is it?

CO₂ equilibrium in water is highly pH dependent.

[strokebow]

Thanks for your input.

its a mass spectrometer. So signal is directly proportional to concentration. Just need to create some standard concentrations solutions so I can plot signal intensity against conc and have some appreciation for how signal intensity relates to conc.

[curiouscat]

Thanks for reply curious cat. Can you explain how this will work - not quite sure what you mean.

Say you have a const. temperature 1 Litre water bath filled with distilled water at 40 C.  Now you sparge / CO2 through it for a fairly high rate for say 20 minutes. Now the water ought to be saturated with CO2 corresponding to its saturation conc. at 40 C. From a plot like the one below this seems to be about 1 gm CO2 / kg H2O. You got one callibration point.

Now change bath to 30 C water and repeat. Get another point. And so on.

Not sure if this is the best method, but just an idea.

[Borek]

its a mass spectrometer. So signal is directly proportional to concentration

What about HCO₃^-, CO₃^2-? I have no idea how they behave in such situation, but won't they interfere. decomposing to CO₂?

Generally speaking - are you determining just the dissolved CO₂, or total CO₂? What about dissolved CO₂ that turned into carbonic acid and dissociated?

[strokebow]

Thanks for your posts.
Seems like a decent approach curiouscat. Good idea and relatively straightforward BUT I can see issues in trying to get a calibration points for lower concentrations.

Thanks Borek. Good point. My main interest is in dissolved CO2.


I would really appreciate if anyone has any further advice/suggestions

[curiouscat]

 Good idea and relatively straightforward BUT I can see issues in trying to get a calibration points for lower concentrations.

How low do you have to go? Can you tell us more about your application?

[strokebow]

Measuring CO2 in fish pond. Needs to be in range 1-30ppm

[curiouscat]

Measuring CO2 in fish pond. Needs to be in range 1-30ppm

Are you sure your sensor is designed to be sensitive to a CO2 level as low as 1 ppm?

[strokebow]

Yes, absolutely. Detection limit would be in the ppb range

[Corribus]

Have you tried contacting the manufacturer? They may have a procedure worked out. Also, is there possibly an ASTM method for measuring CO₂ concentrations which might include something about method validation or device calibration? That may be a good place to look.

http://www.astm.org/