Sciencemadness Discussion Board - Drying Ethanol with Portland Cement

Drying Ethanol with Portland Cement

aga - 3-2-2016 at 13:10

This experiment is to quantitatively determine the drying effect of Portland Cement on azeotropic ethanol over a period of 1 hour.

Portand cement composition :-
(source: http://iti.northwestern.edu/cement/monograph/Monograph3_6.ht...)

CaO 63.8% insoluble in methanol/Et2OH
=> Ca(OH)2 insoluble in EtOH
SiO2 21.5% ?? insoluble in water
Al2O3 4.5% insol. in water. practically insol. in EtOH
Fe2O3 3.0% insol in water. insol in EtOH
MgO 2.4% insol in both
P2O5 0.1% hydrolises. wiki says it is a powerful dessicant
TiO2 0.2% insol in water
Na2O 0.2% =>NaOH Yes, Yes both forms
K2O 0.8% =>KOH Yes, Yes, both
SO3 2.5% => reacts to H2SO4

Many of these components may have already reacted to become different species, given the high temperatures employed in manufacture, however these are the only accurate data available relating to the composition, and will be used as-is for the time being.

Calculation of the components reactive to water and the stoichimetry, it was determined that 1g of a cement powder will react with a maximum of 0.213g of water.

With this figure, the minimum weight of cement required to dry x grammes of ethanol with a concentration of c w% was calculated to be approximately :-

4.7x - 0.047cx

Therefore, to dry 100g of OTC 37.5w% vodka requires at least 293g of cement !

Fractional distillation will generally bring the ethanol to near the 96 w% azeotrope, so the requirement falls to 0.188g of cement per gramme of azeotrope, giving a more manageable requirement of ~19g cement to dry 100ml.

Objective:-

Plot the concentration of azeotropic EtOH treated with cement powder over a period of one hour or more.

Experiment:-

1. Materials used.

Fractionally distilled ethanol.
Portland Cement powder
Spatula
0.01g scale
Filter papers
80ml disposable plastic beakers
Disposable plastic stirrers
Funnel
Ethanol refractometer
Distilled water
Egg timer
Tin foil, cling film, playing cards, cd-rom cases - anything to cap off the beakers.

2. Make & standardise the stock solution.

Distill a quantity of ethanol using a Vigreux column, yielding at least 100g of azeotropic ethanol.

Dilute 5g of the ethanol with 15g distilled water, then measure with an ethanol refractometer.
Multiply the reading by 4 to arrive at the actual reading.

Repeat 3 times with a clean beaker each time, then average the results.

3. Calibrate the instrument

Calibrate the refractometer to Zero according to the manufacturer's instructions.

4. Sample Preparation

Measure approximately 5g (representing a 33% excess) of cement powder into 7 beakers, marked 0 through 7.
Record the exact weight added to each beaker.
Add a stirring rod to each beaker.

5. Add Ethanol

Fill a clean beaker with approximately 20g of the stock ethanol solution.
Record the weight then add to each beaker, starting with #7.
Mix with the stirring rod.
Discard the rod and immediately cap the beaker and set it aside.
Repeat for all beakers.

When done, start the timer, set for 10 minutes.

6. Prepare to Filter

Add a clean filter paper to the funnel (fluting is not necessary).
Place a clean beaker under the funnel.

7. Filtering

Beginning immediately with beaker 0 (not waiting 10 minutes) tip the entire contents into the filter funnel.
Discard the beaker.

If Not processing beaker 0, reset the timer now.

8. Prepare a sample to measure

Place a clean beaker on the scales, and Tare the scales.
Measure out approximately 2.5g ethanol filtrate and record the weight.
Without re-taring the scales, add enough distilled water to reach approximately 4x the ethanol weight.
Record the total weight of ethanol and water.

9. Measurement

Place a few drops of the ethanol/water mixture onto the aperture of the refractometer.
Wait 30 seconds for temperatures to equalise then record the reading.
Discard the beaker.

Repeat three times from step 8 for each sample.

Working at a resonable rate, there is plenty of time between readings for a quick sip of beer.

10. Process all samples

When the timer expires, repeat from step 6 until all samples are processed.

11. Recalibrate the refractometer

Re-test the calibration of the refractometer as per manufacturer's instructions.

If the result is not Zero, discard the entire experiment and re-do from start.

12. Process results

Adjust each reading to account for the actual weights recorded, then take an average of the 3 adjusted readings for each sample.

Plot on a graph.

13. Work out what the hell it all means.

Results:

The calculated w% values ranged from 107.84 w% for the stock solution through to 117.29 w% for the maximal Outlier 20 minute sample.

Estimating the stock solution to be around 96 w% a correction factor of 0.89 was applied to the other results, giving a range of 96.0 w% to 100.3 w%

An estimated 'best fit' of the plotted data gives an increasing range from 96.0 w% (stock solution) to 99.5 w% (70 minutes treated sample).

Observations / Concerns:

Given the diversity of the components in cement powder, tests should be done for contaminants introduced, such as Iron and Sulphate ions.

The cement-treated then filtered ethanol was water-clear with no traces of any haze or other colouration, indicating that any contaminants introduced must be solvated, if present.

Conclusions:

1. Portland cement powder can remove approximately 3 w% water from an ethanol/water azeotropic mix in one hour, representing approximately 75% efficacy in that time.

2. Despite the (assumed yet scientifically unproven) superior efficacy of K₃PO₄, CaO, K₂CO₃ et al, the low cost and universal availability of Portland Cement renders (sic) it a viable option for amateur chemists to achieve near-anhydrous ethanol, starting from an azeotropic (96%) mixture.

3. This refractometer and experimenter might not be the sharpest tools in the box for this experiment.

Raw data :
Attachment: Calcs.xlsx (20kB)
This file has been downloaded 411 times

Please deride, slate and comment.

[Edited on 3-2-2016 by aga]

blogfast25 - 3-2-2016 at 14:45

Did you check these cement-dried samples with KMnO₄?

It's faster than I anticipated.

Plus side: you may now have enough cement to finally build that explosion-proof concrete outhouse?

[Edited on 3-2-2016 by blogfast25]

aga - 3-2-2016 at 14:50

Not yet.

Darkness approached towards the end, and the usual night-time preparations had to be made against the Night Evils.

Tomorrow shall it be so-tested (a new 1 hour treated sample from the same feedstock).

The pot permanganate test is very handy.

blogfast25 - 3-2-2016 at 14:55

Quote: Originally posted by aga

Not yet.

Darkness approached towards the end, and the usual night-time preparations had to be made against the Night Evils.

Tomorrow shall it be so-tested (a new 1 hour treated sample from the same feedstock).

The pot permanganate test is very handy.

A photo, dried v. azeotropic with t'KMnO₄, would be of high appreciability, Squire.

aga - 3-2-2016 at 15:01

OK.

chemrox - 3-2-2016 at 20:47

99.5 is pretty good but try it with everclear or 95% EtOH. I'd be impressed if you could get to 99.5.

Fulmen - 3-2-2016 at 23:37

The listed composition of portland is a bit misleading, in reality the main components are di- and tricalcium silicate, tricalcium aluminate, tetracalcium aluminaferrite and hydrated calcium sulphate (gypsum). Some free Ca(OH)2 will also be present, about 1%.

aga - 4-2-2016 at 00:30

Quote: Originally posted by Fulmen

The listed composition of portland is a bit misleading

Yeah, however that was the best i could find in terms of percentages to work with.

Would you happen to have a more detailed breakdown of cement ?

Fulmen - 4-2-2016 at 00:52

The wikipedia page has some data: https://en.wikipedia.org/wiki/Portland_cement
This has some data on the hydrated species found in cured cement (and CCN abbreviations): https://en.wikipedia.org/wiki/Cement_chemist_notation

It's been a while since I worked with cements (and then mainly physical testing) but I do remember some. I have to add that I listed the clinker composition, most cements also contain blast furnace slag and/or fly ash. Both of these are rich in calcium and aluminum silicates.

Sulaiman - 4-2-2016 at 01:45

from vague memory, Portland cement kilns are sometimes used for toxic waste disposal,
what isn't destroyed by high temperature will be trapped in the concrete.
So chemical purity of the alcohol as a reagent may be variable,
and human consumption would be unwise.

Could you use Plaster of Paris for a similar effect?

EDIT: I just looked at the Wikipedia entry for Portland cement pointed to in the post above, it mentions waste disposal there.

[Edited on 4-2-2016 by Sulaiman]

Fulmen - 4-2-2016 at 01:52

I'm pretty rusty on the hydration mechanisms, but I believe the free CaO is the key. It dissolves, lowering the pH so the other ingredients can dissolve and form gels that eventually hardens as the water is consumed. So other than trace amounts I don't think you need to worry about solubility in alcohol.

IIRC the actual hydration process chemically consumes about 25% of the cement weight in water, but water is also bound physically in the gel phase. For concretes a water/cement-ratio of 0.4-0.6 is usually employed, but this is determined by workability / flow properties and not chemical consumption. But even then the concrete is kept wet after the initial setting to ensure enough water for complete curing. I would try an experiment with alcohol in excess to determine how much water is consumed under these conditions.

Most cement kilns are indeed used to incinerate wastes and hazardous materials, but the temperatures involved virtually guarantees that no organics are left in the product. Heavy metals will of course remain, but I doubt there will be more than trace amounts due to stringent environmental and occupational hazard requirements.

Hexavalent chromium (usually from trace amounts in the ore or metal wear in the mills) can represent a real hazard to cement workers (contact dermatitis), at least in northern Europe this is reduced by adding ferrous sulfate to the milled cement.

Consumption is of course out of the question. You do not add industrial grade chemicals to anything meant for human consumption.

[Edited on 4-2-16 by Fulmen]

aga - 4-2-2016 at 02:00

Quote: Originally posted by Fulmen

I would try an experiment with alcohol in excess to determine how much water is consumed under these conditions.

That'll be for someone else to explore.

I just needed to make dry ethanol for the TCA synth.

Fulmen - 4-2-2016 at 02:36

Fair enough. Nevertheless an interesting approach, and like many good ideas it's pretty obvious once you start to think about it. I know refluxing with CaO can be used to get to 99.5%, but cement should do much of the same and it's far more available.

diggafromdover - 4-2-2016 at 02:44

Did you evaporate any of the product to determine the weight of dissolved material? If so, how much was there, and can you speculate on it's composition?

hissingnoise - 4-2-2016 at 06:15

More ways to skin a . . .

blogfast25 - 4-2-2016 at 06:25

Quote: Originally posted by diggafromdover

Did you evaporate any of the product to determine the weight of dissolved material? If so, how much was there, and can you speculate on it's composition?

That's quite a good idea, actually. Just evaporating a teaspoon and looking for solid residue could potentially be revealing.

aga - 4-2-2016 at 07:38

Nice find hissingnoise.

Today the ethanol got dunked in cement again, for 3 hours this time.

Didn't dry it completely, as the test with pot permanganate shows

Left is the treated ethanol, right the untreated. Barely any percieveable difference.

The still-not-dry ethanol was tested (left to right) with silver nitrate, potassium hexaferrocyanate and barium chloride.

Not sure what the yellow stuff means, yet there seem to be no chloride or sulphate ions present.

arkoma - 4-2-2016 at 07:47

excellent work. I've been roasting small amounts of oyster shell for CaO. Portland MUCH easier.

aga - 4-2-2016 at 08:30

Quote: Originally posted by arkoma

roasting small amounts of oyster shell for CaO

Oysters are a bit too expensive for my taste - i'm more of a cement-and-chips man

Edit:

(the CaCO₃ => CaO + CO₂ thing requires 800+ C)

[Edited on 4-2-2016 by aga]

arkoma - 4-2-2016 at 08:35

Quote: Originally posted by aga

Quote: Originally posted by arkoma

roasting small amounts of oyster shell for CaO

Oysters are a bit too expensive for my taste - i'm more of a cement-and-chips man

I have the ground shell fer my chikins

*edit* Can't stand the thought of EATING an oyster

[Edited on 2-4-2016 by arkoma]

blogfast25 - 4-2-2016 at 09:52

Quote: Originally posted by aga

Not sure what the yellow stuff means, yet there seem to be no chloride or sulphate ions present.

Almost certainly that the hexaferrocyanate is insoluble in EtOH!

You need longer drying times (but NEVER, EVAH listen to me!

), as hissing's page suggests.

blogfast25 - 4-2-2016 at 09:54

Quote: Originally posted by arkoma

Can't stand the thought of EATING an oyster

[Edited on 2-4-2016 by arkoma]

Be the First Billy in the History of Your People to eat one! Lead Them to the Promised Land! Yee-Haw!

Fulmen - 4-2-2016 at 10:08

I agree, more time is needed. Vogel's procedure with CaO is 6hours of refluxing and standing over night.

aga - 4-2-2016 at 11:07

The surprising thing is that nobody has jumped in to say 'I always dry my ethanol like this ...' and detailed their usual process.

It is hard to imagine that i'm the first amateur chemist ever to physically dry ethanol (as distinct from 'knowing what to do')

Magpie - 4-2-2016 at 11:50

Quote: Originally posted by aga

The surprising thing is that nobody has jumped in to say 'I always dry my ethanol like this ...' and detailed their usual process.

It is hard to imagine that i'm the first amateur chemist ever to physically dry ethanol (as distinct from 'knowing what to do')

I use the CaO method in Vogel to get to 99.5%, then 3A mole sieves to get very near 100%.

I make CaO by calcining slaked lime, Ca(OH)2, available dirt cheap by the lb at my local weed & feed store. Bring your own container. This is also available as pickling lime at my grocery store.

I calcine the Ca(OH)2 in my muffle furnace at ~700°C, IIRC, until constant weight is achieved. Likely 1-2hrs does it.

aga - 4-2-2016 at 12:35

Excellent info Magpie.

I shall do the same !

Purifying the CaO made from limestone took a while, although now there is a pot of pure white Ca(OH)₂ to stick in the furnace.

I guess buying some tripotassium phosphate, or clean KOH and H₃PO₄ to make tripot are options, but where's the Fun in that ?

Let's see what the KMnO₄ test says when your suggested process is followed.

careysub - 4-2-2016 at 12:42

I have been using potassium carbonate with 75% Everclear (highest proof sold in California), which gets you to 94%, and 3A sieve pellets from there, though that is expensive (if you do not do vacuum/head desiccation to regenerate).

I'll have to try the Portand cement technique with the 94% ethanol before applying sieves. Thanks!

deltaH - 4-2-2016 at 12:45

I have a bit of a bromance with lime, it's dirt cheap and readily available, but I would like to speculate about another alternative... using a cheap reactive metal like scrap aluminium.

Saturate your crude alcohol with caustic soda (sodium hydroxide) by stirring it for some time with excess prills and then filter the liquid from the excess undissolved prills (or simply decant), then add aluminium and fit a good water cooled condenser in the reflux position (but without heating... the reaction is itself very exothermic once it get's going). When the reaction is done (no more gas evolved), distill to isolate the pure ethanol.

/end speculation

[Edited on 4-2-2016 by deltaH]

Magpie - 4-2-2016 at 13:16

lit. Ca(OH)2 decomposition temp. = 580°C

When I have needed absolute ethanol I found out the hard way (poor yields) that I have to be very careful and conservative in my preparation.

Looking in my lab notebook for that last time I did this (2014) I find:

"Weighed out 107g of slaked lime into large crucible.
3:20PM set muffle furnace for 800°C. Set crucible in furnace.
4:20PM Nearly reached 800°.
6:20PM Turned off muffle furnace.
6:43PM T=600°C, transferred crucible to dessicator to cool."

When distilling the 99.5% ethanol off the CaO I distill it onto mole sieves in an RBF using tapered glass joints to exclude atmospheric moisture. Also I place a CaCl2 moisture guard tube on the vacuum adapter.

"3A mole sieves are rated at 22wt% H2O absorption. Assuming a very conservative 1% water in the CaO treated ethanol, use the required amount based on 15% absorption." Again I'm being very conservative.

I have used CaC2 and anhydrous CuSO4 for moisture detection per Brewster. (I know, they are not perfect.)

What's this test for water using KMnO4?

blogfast25 - 4-2-2016 at 13:39

Quote: Originally posted by aga

The surprising thing is that nobody has jumped in to say 'I always dry my ethanol like this ...' and detailed their usual process.

Most of us have been drying their EtOH over cement for years, mate. You've just exposed our dirty little secret, that's all...

blogfast25 - 4-2-2016 at 13:46

Quote: Originally posted by deltaH

Saturate your crude alcohol with caustic soda (sodium hydroxide) by stirring it for some time with excess prills and then filter the liquid from the excess undissolved prills (or simply decant), then add aluminium and fit a good water cooled condenser in the reflux position (but without heating... the reaction is itself very exothermic once it get's going). When the reaction is done (no more gas evolved), distill to isolate the pure ethanol.

That'll work! Does sound a bit like taking a 155 mm Howitzer to shoot a mouse!

aga - 4-2-2016 at 15:04

Funny how the slightest ego-provocation brought out the 'Here's how I do it' replies.

As an Amateur Chemist i would have offered what i Know Works much earlier, simply to help out another Amateur Chemist.

These responses raise many questions in my mind (such as it is).

Edit:

Too many questions, so i'm out.

[Edited on 4-2-2016 by aga]

blogfast25 - 4-2-2016 at 15:16

Quote: Originally posted by aga

Funny how the slightest ego-provocation brought out the 'Here's how I do it' replies.

As an Amateur Chemist i would have offered what i Know Works much earlier, simply to help out another Amateur Chemist.

These responses raise many questions in my mind (such as it is).

Huh? Ego-provocations?

So:

Quote:

The surprising thing is that nobody has jumped in to say 'I always dry my ethanol like this ...' and detailed their usual process.

... was a trap?

aga - 4-2-2016 at 15:27

In the trade we call it a Mole Hunt.

Magpie - 4-2-2016 at 15:34

Quote: Originally posted by aga

Look at the title of the original post.

blogfast25 - 4-2-2016 at 15:37

Quote: Originally posted by aga

In the trade we call it a Mole Hunt.

What trade?

deltaH - 4-2-2016 at 22:19

Quote: Originally posted by blogfast25

Quote: Originally posted by deltaH

That'll work! Does sound a bit like taking a 155 mm Howitzer to shoot a mouse!

Yeah, but this is where the boy in me fights against the engineer in me

A word of caution to those not aware of this, but it will generate A LOT of hydrogen gas, so not something to do in an enclosed shed unless you want to blow it to smithereens.

A company very close to me sells aluminium powder by the kilo, dirt cheap, as a metallic filler for plastic composites (eqv. 3.65 euros/kg incl.).

Caustic soda is sold in 25kg amounts for (eqv. ~1.35 euro/kg, ok, actually a little cheaper, but I like easy math).

1kg aluminium powder will dispose of exactly 2kg water based on stoichiometry, but you need 1.5kg caustic soda if you going the alkaline activation route.

On that theoretical basis, 1kg aluminium powder can treat 50kg 96 wt.% ethanol, at a cost of 1 euro/10kg (~12.5l) wet ethanol... not so bad.

For completeness, the reaction in question is of course:

Al(s) + 3H2O(l) + NaOH(aq) => NaAl(OH)4(s) + 1.5H2(g) + heat

One could also use a little mercury or salt thereof to activate the aluminium and so dispense with having to use caustic soda, but I'm not a fan of working with [and having to recovery] mercury. Incidentally, using mercury, one can prepare aluminium alkoxides with alcohols directly from the metal. So one should not use too much of an excess of aluminium or you will consume alcohol after first consuming all the water.

[Edited on 5-2-2016 by deltaH]