[jclarholm]

Hi

First of all I want to point out that I am not a chemist, I am an electronics engineer. I have no idea if what I want to do is even possible to do DIY fashion and without to much fancy equipment and that is why I turn to you guys who are the expert on these things.

As a hobby project I am building a breath-acetone meter. Then sensor I use is an electro chemical tin dioxide (SnO2) semiconductor sensor from Figaro sensors called TGS822. It is sensitive both to alcohol and acetone and I have it working so that I can take readings from it. However to be able to link the sensor readings to what gas concentration it corresponds to I need to get a know fix point. In the sensor manufacturer used ethanol vapor at 300ppm as calibration level so that would be optimal but not necessary, any known concentration between 100-1000 ppm should work.

I guess that you have to start with a know concentration of alcohol say 40% (vodka?) and then dilute that further with water until the alcohol concentration is lower. Then take a small enough amount of that and dissolved in a known amount of air and place the sensor in that mix. However I have now gotten lost so many times in my calculations with molecular mass and unit conversion between ug/l and ppms etc that I have almost given up, but before I give up I thought that I would ask you guys for help first.

Is what I want to do possible without to much special equipment?
If it is possible can someone tell me what proportions to use?
Any word of advice is highly appreciated.



If anyone is interested to read more about what I have done so far please take a look at
http://jenslabs.com/category/electronics/ketosis-detector/

Thank you in advance.

Jens

[Arkcon]

Hmmm ... tricky.  It seems you need a concentration of 100 - 1000 ppm of ethanol, in air, to calibrate your instrument.  That is really more complicated than trying to make a solution in water.  There are many variables to consider, for example: air saturated with moisture, which is what humans exhale, can likely hold more ethanol (some of it in the droplets of water) than dry air can.  Or can your sensor even "see" what's in water.  Or will it be damaged by water.  At any rate, making an air mixture to your specs doesn't seem trivial to me.  Maybe someone on this sub-forum has more experience, so I've moved your post.  Maybe some sort of calibrated mixer is sold.  Or some sort of engineering trick with an air tank and an alcohol mister.  Or something. 

[jclarholm]

Thank you Arkcon both for your reply and for moving the thread to where it best belongs.

You are absolutely correct that I want to have the ethanol in air as you said.

I had not thought about that the moisture in the breath it self could hold acetone as well, but with a small amount of liquid and with a decent amount of air it might be possible to vaporize it so that we don't have any large amounts of liquid ethanol-water in the measuring chamber.

I don't think that the sensor can detect the alcohol unless it is in gas form, the sensor has a small heater built into it so I guess that will help to vaporize some ethanol-water but as little excess moisture as possible is of course better. Also just to be clear it is a gas sensor that can handle some moisture without breaking but it can't be dipped in water.

The sensors characteristics do change slightly depending both on temperature and humidity but the sensor behavior under different temperatures and humidity is known so I should be able to account for those factors.

If I am not able to produce an ethanol gas with a know concentration another method, which is a bit of a cheat, to get the calibration done would be to find an already calibrated alcho-meter and then produce my own ethanol-air mix by drinking and blowing. Perhaps not so scientific but it could be fun.

Perhaps someone else here has any good ideas.

[Borek]

Composition of gas mixture that is in equilibrium with solution is - for a given temperature - constant. As you don't need a lot of gas, perhaps just equilibrating small volume of solution of known concentration (relatively easy to prepare) in a large bottle and then squeezing it will produce short, but reliable stream of air of known ethanol concentration.

[jclarholm]

Hello Borek, thanks for your input. Now I'm starting to feel a bit more hopeful again.

I have also been thinking similar thoughts like you with a bottle and then a drop or two with a know concentration. However my problem is that I don't know how to calculate how large the "small volume of solution of known concentration" is supposed to be relative to the size of the bottle. All my attempts at calculating it my self has ended up with me having ridiculous values where I have to either have something like 100000 liters of air or 1 ul of alcohol. It is obvious that I don't have my units in order somewhere or there is some other beginners mistake, thats why I am asking for help here.

[Borek]

I have also been thinking similar thoughts like you with a bottle and then a drop or two with a know concentration. However my problem is that I don't know how to calculate how large the "small volume of solution of known concentration" is supposed to be relative to the size of the bottle.

I think you are missing the point. It is not about using a volume that will be completely evaporated (although it could be probably done as well), it is about using excess volume of the solution. Even if there is way too much solution, partial pressures of the vapors in equilibrium with the solution won't be higher than those described by Dalton's and Raoult's law. That in turn means that getting right concentrations can be relatively easy. I can be wrong, as I have not checked the numbers (especially the mass balance), but that would be my line of tackling the problem.

[Stepan]

It is a little bit more complicated than it sounds. At ppm level, solvents get adsorbed to the vessel walls and the actual concentration will be way lower than expected. On the other hand the household jars and plastic bags emit enough vapours to give you  positive reading without any ethanol.  In lab environment, people usually create steady air flow of purified air, and place so called "permeation tube" (which can a vial with a pinhole in the lid) with ethanol (pure) weighed before and after experiment. Weight loss in mg/min gives you evaporation rate which is a constant for this tube at given temperature. Then you divide the evaporation rate (mg/min) on air flow rate in m3/min, you get ethanol concentration in mg/m3.

[curiouscat]

It is a little bit more complicated than it sounds. At ppm level, solvents get adsorbed to the vessel walls and the actual concentration will be way lower than expected.

Not to contradict you but how does adsorption change anything if you equilibriate the system? Won't those surfaces eventually saturate?

[Borek]

I know what he means - while the surfaces will saturate, absorption can substantially change concentration of the trace substance, so we can't be sure what its "active" concentration is. That in turn means we don't know what is concentration in the gaseous phase.

[jclarholm]

Thank you both Borek and Stepan.

I think you are missing the point. It is not about using a volume that will be completely evaporated (although it could be probably done as well), it is about using excess volume of the solution. Even if there is way too much solution, partial pressures of the vapors in equilibrium with the solution won't be higher than those described by Dalton's and Raoult's law.

You were absolutely right when you said that I had missed the point so thanks for clearing that up for me.

Stephan, I appreciate that you took the time to tell me how it is done professionally and giving me the keyword "permeation tube" which proved a good starting point for further google'ing. Perhaps I have to let this one go and just find an already calibrated sensor that I can use as a baseline.

I do have a question though, when you say that:

On the other hand the household jars and plastic bags emit enough vapours to give you positive reading without any ethanol.

I take it that it applies to plastic jars/bags but is it also true for glass jars?

However I am still very much open to sugesstions if anyone has an ideas.

[JGK]

As someone who used to work with volatile materials on a regular basis. You can make standard gas mixtures in inert sampling bags relatively easily

http://www.sigmaaldrich.com/analytical-chromatography/air-monitoring/gas-sampling-bags.html.

Know volumes of air and your sample can be added by Syringe (or a flow rated pump) and you can use a heat source (hair dryer) to get your sample into the vapour state.

I've made test samples in bag sizes up to 240L using this type of equiment.

[jclarholm]

So if I can get my hands on an "inert sampling bag" we could be back in business. Then I'm back at the original problem of determine how large my sample should be and what the ethanol concentration of it should be. I will try my best with the calculations again tomorrow, bed time here in Sweden now. Thanks for all the replies.

[curiouscat]

I know what he means - while the surfaces will saturate, absorption can substantially change concentration of the trace substance, so we can't be sure what its "active" concentration is. That in turn means we don't know what is concentration in the gaseous phase.

Ah! Got it. Thanks.