Sciencemadness Discussion Board

Building a solubility tester with arduino chip

alchemizt - 31-5-2022 at 20:00

I want to build a device which determines solubilities of samples in organic solvents. I'm contemplating how to do that. I'm thinking that with a photometer and light source, you should be able to plot a dissolution curve due to the mixture becoming gradually more translucent as there are less and less undissolved particles in there, and eventually should be able to record the point at which the solvent becomes saturated as the change in translucency tapers off. You could calibrate an arduino chip so it knows the spectral properties of a saturated solution of your sample in the given solvent and use the dissolution curve to determine how much of the sample dissolved in a given amount of solvent.

What do you think?.

Sulaiman - 31-5-2022 at 22:10

I think that ;

You will need a calibrated temperature measuring device

You will need temperature stabilisation
(or solubility vs. temp. data)

You need good stirring

So, stirring would keep particles suspended,
which I guess would reduce transmitted light

Making or buying reference solutions sounds tedious and/or expensive.

RustyShackleford - 1-6-2022 at 05:16

Sounds very doable and the ideas you have seem reasonable. As Sulaiman said i think having integrated temperature control and stirring will be necessary to get reasonable data. If you plan on using a test tube or such as the container, then for the stirring i think a vortex stirring mechanism would be perfect, should be really easy to build yourself.

Sulaiman - 1-6-2022 at 08:15

PS if you use a visible laser you can see when there are any suspended particles.
(I forget the name of this effect)

alchemizt - 1-6-2022 at 15:46

Quote: Originally posted by Sulaiman  
I think that ;

You will need a calibrated temperature measuring device

You will need temperature stabilisation
(or solubility vs. temp. data)

You need good stirring

So, stirring would keep particles suspended,
which I guess would reduce transmitted light

Making or buying reference solutions sounds tedious and/or expensive.


Yeah the temperature part is a bit tricky, I'm thinking to connect a thermocouple to a PID controller programmed to keep the temperature constant. I don't know how to do this in the case of exothermic dissolutions, not sure how to electrically cool the liquid. It can wait for the mixture to cool back down to the test temperature before recommencing recording data. Then the possibility of supersaturation needs to be factored in somehow.

Stirring at a steady RPM will definitely will be necessary to keep the suspension uniform. Over time it would be possible to add a lot of data to a database that would be useful for machine learning and making extrapolations.

alchemizt - 1-6-2022 at 15:50

Quote: Originally posted by Sulaiman  
PS if you use a visible laser you can see when there are any suspended particles.
(I forget the name of this effect)


Is it the Tyndall effect?

alchemizt - 1-6-2022 at 16:22

Quote: Originally posted by RustyShackleford  
Sounds very doable and the ideas you have seem reasonable. As Sulaiman said i think having integrated temperature control and stirring will be necessary to get reasonable data. If you plan on using a test tube or such as the container, then for the stirring i think a vortex stirring mechanism would be perfect, should be really easy to build yourself.


Is a vortex stirring mechanism the one where the machine shakes the tube in a circular motion like the ones they use in microbiology labs? I've never tried that method of stirring. I was thinking of using a magnetic stir bar because its easy to find a hotplate stirrer and set it up to hold the temperature constant. The only thing is the stir bar blocks light so the laser will need to be either above it, or else the chip needs to be calibrated to know exactly how much light it blocks during a given time interval, that should be easy since at a high RPM it will be relatively uniform. Can you control the temperature easily with a vortex stirring mechanism? Would you need to use an immersion heater? Like a cartridge heater?

I think it will be necessary to submerge the thermocouple in the solvent, because if you were to submerge it in a sand or water bath, you can only really measure the heat going into the mixture, not whats coming out of it.

Tsjerk - 2-6-2022 at 01:17

Maybe put the thing in an insulated box? That way you can control the temperature as a whole. Also light from outside would be blocked which could interfere with your readings.

JohnnyBuckminster - 2-6-2022 at 07:31

I have used an old commercial turbidity meter, I think it came from Fischer, to check the solubility of some dyes in various solvents.

My main concern was always the quantity sample required for a mesurement vs cost, it may be attractive to design something that can work with small volumes, like 100 - 500 uL.

Heating and cooling might be achived with a Peltier element.

There are some DIY projects on YouTube, see for example

https://www.youtube.com/watch?v=Bu8y-6Bpv0U


In my experience with dyes, the stirring was actually not that important. If the sample did not dissolve properly, then there was always some scattering stirred or not, and further dilution was required.

[Edited on 2022-6-2 by JohnnyBuckminster]

DraconicAcid - 2-6-2022 at 12:00

What kids of samples are you looking at? I'm just thinking of the tendency of many organic compounds to form supersaturated solutions until crystallization is induced.

DocX - 2-6-2022 at 12:51

interesting! One should probably be able to do the same with melting point determination. Laser through capillary tube filled with salts, measure the intensity, plot melting point.