Water content of organic solvents Log Out | Topics | Search
Moderators | Register | Edit Profile

NIR Discussion Forum » Bruce Campbell's List » I need help » Water content of organic solvents « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

Andrew Hind (Andrew)
Posted on Friday, August 20, 2004 - 6:19 am:   

A while back there was some interesting discussion concerning analysis of water in acetonitrile. I am interested in measuring water content in organic solvents in the range 0.1 - 1%. I know it should be possible in the NIR but, before I try, wondered if anyone has any comments on likely detection limits and/or could refer me to any relevant methods or publications.
Thanks.
Andrew
Top of pagePrevious messageNext messageBottom of page Link to this message

Bruce H. Campbell (Campclan)
Posted on Friday, August 20, 2004 - 6:31 am:   

Andrew,
One example is Dupont using NIR to determine water in a solvent. The specification, as I remember, is on the order of 0.0001 or 0.0002%.
Bruce
Top of pagePrevious messageNext messageBottom of page Link to this message

Tony Davies (Td)
Posted on Friday, August 20, 2004 - 8:55 am:   

Hello Andrew,

While there are many papers concerned with the determination of moisture in agricultural and food products, I cannot recommend a single paper for water in solvents.

During IDRC-2004 (last week) there was a discussion about calibrating a method for moisture in a pharmaceutical using Karl Fischer reagent. I attempted to point out that as Karl Fischer has to be calibrated (known concentrations of water in methanol) it would be better to do the calibration by spectroscopy but this caused an argument that was not resolved!

You do need to be careful about changes in the shape of the water peak especially in non-polar solvents. At very low concentrations the water cannot form hydrogen bonds but with increasing concentration water molecules will being to form hydrogen bonds with other water molecules and the peak becomes broadened. Thus you will have a non-linear calibration if you try to calibrate across such a range. You might consider using the method of standard additions if you can maintain the peak shape over the required range.

I think you should be able to get down to about 0.01%

Does this help?

Best wishes,

Tony
Top of pagePrevious messageNext messageBottom of page Link to this message

Gabi Levin
Posted on Monday, August 23, 2004 - 8:08 am:   

Hi Everybody,

Water in organic solvens - well no complete rule, depends greatly on:
1. Absorption of solvent in 1450 and 1950 regions.
2. Interaction of the water with the solvent - polar solvents interact more, lead to lower sensitivity. Non polar or weakly polar solvets (such as dischlorobenzene)have limited solubility, as low as 500ppm and sometimes less.

In solvents with limited solubility - levels of 20 ppm can be conveniently measured.

However, I am skeptical that0.0001% (1ppm?) can be safely meausred by NIR.

In methanol, 20 to 40 ppm are OK.

In solvents - yes there can be an issue of inter water molecular interactions once the concentration is high enough, however, in non polar I would expect the solubility to reach maximum before those interactions become significant, thus, the measurement becomes irrelevant past the solubility (saturation and formation of separate water pahse). In polar solvents, such as methanol, I would expect the methanol molecules to provide "sufficient shielding"of the water molecules from each other to sufficiently high concentrations - say above 10%. It is interesting though to see when we start to see "deviations"from linearity.

Thanks, Gabi Levin
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Monday, August 23, 2004 - 8:34 am:   

Good point, Gabi. As for water-alcohol mixtures, there are two important points ro remember:

1) The lower alcohols (at least through propanol) are miscible with water in all proportions.

2) The mixing is non-linear, even for the cruder characteristics such as density and volume. Alcohol is the classic example of "partial molal volumes": if you mix 1 volume (whatever the unit is: cc, gallons, liters, etc.) each of alcohol and water you get about 1.95 volume unit of mixture. I'm sure you could find tables of the measured non-linearity values for the intermediate mixtures in the literature, but you'd have to go back quite a ways to find those, I think.

So I'd expect that the spectroscopic response would exhibit at least that degree of non-linearity, as well.

Howard

\o/
/_\

Add Your Message Here
Posting is currently disabled in this topic. Contact your discussion moderator for more information.