Determining if solvent layer has water.

[~christina~]

I want to determine whether a solvent layer has water or not. (have: solvent layer + unknown compound(s) dissolved in layer)Usually there would be 2 layers if there is an organic solvent + water, however if the solvent is polar, there would be one layer.

situation:

I tried using coloured drierite, to see if there was water in solvent layer. (any a color change) I pipetted some of the solvent layer over to the drierite and the drierite stayed blue and the solvent evaporated quickly. no water?

I ran solvent layer through GC on polar column and saw sharp solvent peaks and then an unidentifiable hill shaped extended tailing peak after the sharp peaks. Could this be water ?I don't think it can be the unknown compound because it wouldn't be volatile, right? (dissolved solid compound in solvent would not be volatile...I think) How does water look like on GC? (using flame ionization detector)


Question is: If there wasn't a large percentage of water in a solvent layer, would the drierite still have the same observable colour change, or would the solvent applied, evaporate off quickly enough that it could look like there wasn't any water in sample?
IS there any other way to determine if there is water in solvent layer for sure?

=> confused as to what that observable very long tailing "hill peak" is on the GC chromatogram was and if it can be water, then I'm scheptical and question the accuracy of the looking at drierite to observe color change. :uhh:

Please help

 

[chemisttree]

Good question, Christina. Very polar compounds generally tail on the column and water is certainly polar. Tests for water include IR and the Karl Fisher test. Of course the KF test is usually performed only for solvents that don't have anything in them that might react with the KF reagent and the presence of the OH stretch might be complicated by the presence of your compound. Best thing is to remove the water before analysis. Before you begin, do a TLC and see if there is a spot at the origin that tails significantly. You won't see water on the TLC. If you see a spot, you could be dealing with a compound that has a very polar group present.

It is usual practice to change the solvent into one that can be easily dried (such as CHCl2 vs diethyl ether) and then use a drying agent like powdered anhydrous MgSO4 (finely ground Drierite without the indicator) or granular anhydrous sodium sulfate prior to your analysis.

 

[~christina~]

Good question, Christina. Very polar compounds generally tail on the column and water is certainly polar. Tests for water include IR and the Karl Fisher test. Of course the KF test is usually performed only for solvents that don't have anything in them that might react with the KF reagent and the presence of the OH stretch might be complicated by the presence of your compound. Best thing is to remove the water before analysis. Before you begin, do a TLC and see if there is a spot at the origin that tails significantly. You won't see water on the TLC. If you see a spot, you could be dealing with a compound that has a very polar group present.

It is usual practice to change the solvent into one that can be easily dried (such as CHCl2 vs diethyl ether) and then use a drying agent like powdered anhydrous MgSO4 (finely ground Drierite without the indicator) or granular anhydrous sodium sulfate prior to your analysis.

Thanks Chemisttree

I saw 2 spots on the TLC. (developing solvent: ethyl acetate[spots did not get past baseline] MeOH: 2 spots with a solvent front residue) I didn't use a different solvent, because I was just testing the unknown solvent and compound, to see what I would see with the original solvent used for the unknown.
I do have a question about this.

1. I was told that solvents can form spots on the TLC so it would mean that I don't necessarily have 2 compounds like I think I do.
Why is this?

2. What solvents will form spots? (if I use a solvent to dissolve the dry compound and if forms a spot, that will defeat the purpose of using another solvent)

 

[chemisttree]

What did you use to visualize the 2 spots on your TLC and what type of TLC media are you using? Does the TLC stationary phase have a fluorescent compound in it? Did you shine a UV lamp to visualize the spots?

Edit: AND... I should add that if ethyl acetate didn't move your compound(s) and you were forced to use methanol, you are guaranteed to have a very polar compound(s) and you should expect extreme tailing in the GC.

Looks like it is deriviatve time for you.

 

[~christina~]

What did you use to visualize the 2 spots on your TLC and what type of TLC media are you using? Does the TLC stationary phase have a fluorescent compound in it? Did you shine a UV lamp to visualize the spots?

Hm...I'm using a glass backed TLC plate but I'm not sure as to the specific type of plate coating is being used.
The stationary phase has to have a fluorescent compound in it because I visualized the spots under a UV lamp.

Edit: AND... I should add that if ethyl acetate didn't move your compound(s) and you were forced to use methanol, you are guaranteed to have a very polar compound(s) and you should expect extreme tailing in the GC.

Yes, I saw the heavy tailing of the compound in the GC.

Looks like it is deriviatve time for you.

But how can I do that if I don't know what I have?

I'm going to dry out some compound and then take that to dissolve it in some solvent (not sure which still) and then run the TLC plate with the methanol again.

 

[gravenewworld]

Sounds like you have an amine. Run a TLC plate with DCM and Methanol spiked with ammonium hydroxide solution (10%) or with 5% TEA.

 

[~christina~]

Sounds like you have an amine. Run a TLC plate with DCM and Methanol spiked with 10% ammonium hydroxide solution or with 5% TEA.

Okay, I'll try that. Thanks gravenwworld

 

[chemisttree]

That will help with the TLC but not with the GC.

 

[chemisttree]

Hm...I'm using a glass backed TLC plate but I'm not sure as to the specific type of plate coating is being used.
The stationary phase has to have a fluorescent compound in it because I visualized the spots under a UV lamp.

You should spot pure samples of the solvents you are using on the TLC after you dry them as you would do after spotting your unknown solution. Can you visualize the spot (without developing the spot in the TLC tank)? You won't see water using a UV lamp on a fluorescent TLC plate.

I'm going to dry out some compound and then take that to dissolve it in some solvent (not sure which still) and then run the TLC plate with the methanol again.

Are you using only GC to identify your unkown? How will a peak with a given retention time tell you anything? Do you have a mass spec hooked up to the system or is it just the FI detector?

If it were me, I would do the TLC on a prep-scale and scrape off the spot. Now with pure compound (1 spot!) leached from the scraping, you can determine the identity of the spot.

 

[~christina~]

You should spot pure samples of the solvents you are using on the TLC after you dry them as you would do after spotting your unknown solution. Can you visualize the spot (without developing the spot in the TLC tank)? You won't see water using a UV lamp on a fluorescent TLC plate.

I'll try that when I go to lab, but I do think I have 2 compounds after performing extraction.

Are you using only GC to identify your unkown? How will a peak with a given retention time tell you anything? Do you have a mass spec hooked up to the system or is it just the FI detector?

we are only supposed to determine the solvents using GC. We aren't allowed to use mass spec ...just IR.

If it were me, I would do the TLC on a prep-scale and scrape off the spot. Now with pure compound (1 spot!) leached from the scraping, you can determine the identity of the spot.

I think there are 2 compounds and there was some tailing visible on the plate. How would I analyze the scraping from a TLC plate? (I've heard about this, but don't know exactly what is done, procedure wise)

 

[chemisttree]

http://books.google.com/books?id=BEV_6SGtZJIC&pg=PA112&lpg=PA112&dq=%22preparative+scale+TLC%22&source=bl&ots=pgE83fkohM&sig=XDHp17VN8_ZOkN0P4T6UJqydvgM&hl=en&ei=6Av6SbuaKJGCmQeggOXDBA&sa=X&oi=book_result&ct=result&resnum=1" [Broken].