Sciencemadness Discussion Board

Removing non water contaminants from alcohols

alking - 4-1-2017 at 10:25

I have some anyhdrous ethanol with a bp around 74-76.8C, and some anhydrous methanol with a bp around 62.5C. I assume the bp supression is likely from aldehydes and ketones. How would I go about removing them? The only thing I can think is reducing them to alcohols, but I'm wondering if there's a more efficient/cheaper way. The alchol was dstilled and dried from vodka and the methanol is tech grade. They seem to be azeotropic as is unless my column is just not efficient enough to separate them further. The thermometer measures boiling water at 100.1C so I assume it's reasonably accurate.

[Edited on 4-1-2017 by alking]

alking - 4-1-2017 at 14:31

On second thought I'm actually wondering if my thermometer is reading accurately, it's a probe style so perhaps its size results in a slightly reduced reading due to a larger heat capacity. The head of the still for instance is not 62.5C when distilling methanol so I would not be surprised if there is some heat loss on the probe. I've only tested boiling water in the liquid phase, not the vapor phase, I will do that shortly to verify. If that reads a little low then I probably do have relatively pure alcohols here.

JJay - 4-1-2017 at 20:28

You can test your thermometer with boiling water and ice to see if it is reading correctly.

DutchChemistryBox - 4-1-2017 at 22:13

Quote: Originally posted by JJay  
You can test your thermometer with boiling water and ice to see if it is reading correctly.


You can indeed use purified water ice. I don't like the boilling water idea, think about the atmospheric pressure.

unionised - 5-1-2017 at 01:43

If the variation in atmospheric pressure is big enough that you can't rely on water boiling at 100C then you also can't rely on the boiling point of the alcohol as a measure of purity.

However, if the boiling point remains constant throughout the distillation it's a good indicator that the material is pure (or an azeotrope, if you are unlucky).

JJay - 5-1-2017 at 02:01

Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
You can test your thermometer with boiling water and ice to see if it is reading correctly.


You can indeed use purified water ice. I don't like the boilling water idea, think about the atmospheric pressure.


You can get local barometric pressure in realtime off of the Internet in most places, which may allow you to use this handy calculator: http://www.csgnetwork.com/prescorh2oboilcalc.html

You should be able to use a triple-point diagram or suitable table (or software) to find the boiling point of other substances at a given pressure.

DutchChemistryBox - 5-1-2017 at 08:57

Quote: Originally posted by JJay  
Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
You can test your thermometer with boiling water and ice to see if it is reading correctly.


You can indeed use purified water ice. I don't like the boilling water idea, think about the atmospheric pressure.


You can get local barometric pressure in realtime off of the Internet in most places, which may allow you to use this handy calculator: http://www.csgnetwork.com/prescorh2oboilcalc.html

You should be able to use a triple-point diagram or suitable table (or software) to find the boiling point of other substances at a given pressure.


Until you turn on your fume hood and the pressure in the room drops.

alking - 5-1-2017 at 09:09

Is a fumehood really going to significantly affect the pressure though? I wouldn't think it would. Anyway I have tested it against boilong water as I said, the thermometer is accurate for sure, but I question if the distillation rate is slow that despite the distillate coming over at, say 64.5 for MeOH that the thermometer loses a bit of heat to the surrounding air and thus reads low. I'm going to distill some dH2O and test the vapor temp of that to verify. I believe that the solvents are impure still but I need to rule this out first to be sure.

JJay - 5-1-2017 at 09:57

Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
You can test your thermometer with boiling water and ice to see if it is reading correctly.


You can indeed use purified water ice. I don't like the boilling water idea, think about the atmospheric pressure.


You can get local barometric pressure in realtime off of the Internet in most places, which may allow you to use this handy calculator: http://www.csgnetwork.com/prescorh2oboilcalc.html

You should be able to use a triple-point diagram or suitable table (or software) to find the boiling point of other substances at a given pressure.


Until you turn on your fume hood and the pressure in the room drops.


You need a fume hood to boil water?

wg48 - 5-1-2017 at 10:04

Yes if you have a small lab/workshop and you don't want your tools to rust.

unionised - 5-1-2017 at 10:06

Quote: Originally posted by JJay  
Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
Quote: Originally posted by DutchChemistryBox  
Quote: Originally posted by JJay  
You can test your thermometer with boiling water and ice to see if it is reading correctly.


You can indeed use purified water ice. I don't like the boilling water idea, think about the atmospheric pressure.


You can get local barometric pressure in realtime off of the Internet in most places, which may allow you to use this handy calculator: http://www.csgnetwork.com/prescorh2oboilcalc.html

You should be able to use a triple-point diagram or suitable table (or software) to find the boiling point of other substances at a given pressure.


Until you turn on your fume hood and the pressure in the room drops.


You need a fume hood to boil water?


The thread isn't about boiling water.

JJay - 5-1-2017 at 10:13

I don't feel you can gauge a thermometer's accuracy across a range with one data point.

unionised - 5-1-2017 at 10:30

Quote: Originally posted by JJay  
I don't feel you can gauge a thermometer's accuracy across a range with one data point.

How fortunate we are that nobody said you should.

JJay - 5-1-2017 at 10:38

Let's not make this personal, unionized.

In an ideal world, we'd gauge our thermometers by how well they measure the freezing point of water and absolute zero. But taking readings at absolute zero is not really practical... if someone has a better suggestion than boiling water and ice, I'd like to know what it is.

unionised - 5-1-2017 at 11:19

OK: how fortunate we are that nobody said anyone should.

The freezing point is good enough for most of us; the purists would go for the triple point of water.

There's no real need to clean up the ice before you use it for a melting point. Put crushed ice in a funnel and stick the thermometer in it.
Impurities that don't dissolve don't matter. Impurities that do dissolve get washed away as the ice melts- so when the temperature settles to a constant value, the ice is clean enough.
Getting a reliable boiling point is tricky unless you know the atmospheric pressure. ( Also, beware of weather sites that tell you the atmospheric pressure "corrected" to sea level).

The traditional approach is to use the melting points of chemicals that are well documented and easy to get in a pure form. This sort of thing.
http://www.thinksrs.com/downloads/PDFs/ApplicationNotes/MP_P...

One day I hope to sort out a few melting point cells with gallium, hydrated sodium sulphate and maybe mercury and tin.

aga - 5-1-2017 at 12:46

Quote: Originally posted by wg48  
Yes if you have a small lab/workshop and you don't want your tools to rust.

That tends to happen only with tools that are Unused and uncared for.
Quote: Originally posted by unionised  
The freezing point is good enough for most of us; the purists would go for the triple point of water.

Agreed.

This an Amateur Chemistry forum, so ice-cubes and boiling (distilled) water are certainly close enough.

In any event, the formula $$y=mx + c$$ will work at all earthly altitudes.

y is the actual temperature. x is the reading on the thermometer.

m and c are calculated by measuring boiling water (x1, y1=100 C) then ice water (x2, y2=0).

m is given by $$\frac {y_1-y_2}{x_1-x_2} = \frac {100}{x_1-x_2}$$ and c is given by $$100 - mx_1$$

Calibrate, get the m and c numbers, then press buttons on a calculator to get the Real temperature.

If the thermometers are never ever used, just stare at them and Wonder, saving all that tedious work and effort.

unionised - 5-1-2017 at 13:00

To a reasonable approximation (i.e. lots of people are too lazy to do any better) the effect of pressure and/ or altitude on boiling point will be comparable for all liquids so, if you "define" your own local temperature scale where water boils at 100 and freezes at 0 then acetone will boil pretty near 56 degrees - even if you are on top of Everest.

The assumption that the effect of pressure is the same for all liquids is, of course, laughable.
However it's common practice
I remember when this was a paper chart
http://www.sigmaaldrich.com/chemistry/solvents/learning-cent...

Realistically, boiling point isn't a very good way to determine purity.
Density is pretty good.