[Doc Oc]

Hold on, did I read that correctly?  You're trying to separate 20 g of  androstene and 2 g of estradiol?  If that is the case I don't think you have enough solvent the separate the compounds, 100 mL of hexane certainly isn't enough.  This would also require quite a large glass column and a large amount of silica.  All are things that complicate the separation by column chromatography (not to mention that if you're doing gravity drip rather than flash chromatography you could very well be running this column for days, and that's not an exaggeration).

[littlebug]

There is 20,000mg of androstenone and 2,000mg of estrodial benzonoate in solution.

Hold on, did I read that correctly?  You're trying to separate 20 g of  androstene and 2 g of estradiol?

My best guess is that he's using a comma instead of the decimal, as is done in some European countries. 

[Mr.rose]

Yes you did read that correctly.

That is all the hexane i have on me at the moment, i will be ordering more. How many L or hexane and Ethyl Acetate will it take?

My setup involves a 100mL column, with a length of 1m , so it is a small column. I have ordered some Alumina as i think it would be a better and more 'stable' choice over silica gel. I just don't know which type of Alumina is best, so i ordered a few activation levels.

My TLC plate are in route to me, I am just waiting for them to arrive. I have ordered silica gel backed plates, would this be a problem since i am switching to Alumina, I am assuming the Rf will change?

Days, hmm, that is quite a bit of collection.

[OC pro]

Alumina is quite different to silica gel. You can not combine silica gel TLC plates with an alumina column. Rf values can differ dramatically.
Gravity column is not by far the best but it will not take days. This is nonsense. But at least 2 to 4 hours you can expect. Fractions should be around 500 ml.
I assume ca. 5 to 10 l of hexane. 1l of ethyl acetate should be enough.

[Mr.rose]

Ok, 100mL column, silica gel 60-200mesh 60A, 4L Hexane and 1L Ethyl Acetate. I think i am all set.

Hopefully i have enough solvent. Am i suppose to run the Ethyl acetate first as it is more polar, then after i use a liter, i run the hexane until i run out.

So that will be 10 fractions, at 500mL fractions. Is that good? or should i do 250mL fractions. I would hate to get one fraction with both compounds in it.

What should the flow rate be? 2-3 drops per second?

Thankyou again for all your help guys.

[helenn]

You need to run the least polar hexane first, and I would run a gradient increasing the amount of ethyl acetate. But you should find the best ratio of sovents to give the best separation by running the TLCs first.

[democanarchis]

There are a lot of problems here. You say that your materials to be separated are dissolved in MeOH. How do you plan to load them onto the column? It doesn't sound like you have a lot of equipment, which would make it difficult to dry load it (need rotary evaporator, silica, silica trap), and they probably aren't soluble enough in the mobile phase to use that.

Do i read you correctly in that you have a 100mL column? IE the total internal volume is 100mL? I'm sorry, but thats completely ridiculous. I think silica gel is something like 0.7g/cm³ wet. That means even without allowing for headspace for solvent or sample, you're talking about 70g of silica. To separate 22g of material. Usually if you have good separation (>0.2 rf), you want something like 25g silica/g material to be separated. Thats about 550g, or nearly 8x the capacity of your column.

I really really disadvise this. If you go ahead with the column, run a small column with like 0.5g of material and verify that you get separation with your mobile phase. Have you tried recrystallising a small portion of your material? With a 10:1 ratio I would be very surprised if you could not find a suitable recrystallisation technique.

[Doc Oc]

Helenn is absolutely right, if you don't run TLC's first you're doing this purification blindly.

1) To clarify my earlier question, do you have 20 g of the testosterone or 20 mg?  That's a HUGE difference that will considerably change the way you need to purify the compounds (you won't be able to purify 20 g on the 100 mL column you are using, you will need a MUCH larger column).

2) You are thinking of running the solvents backwards again, it should always be least polar to more polar.  However, you don't run the solvents one after the other, you mix them.  So that's why you need to do TLC, to find out which combination works best.  It may be 9:1 hexane/EtOAc or it could be as high as 1:1 (although this is probably too polar for your compounds).  What you're looking for is the solvent combination that gives the best separation between your compounds.

3) Fraction size will depend on how well your compounds separate on the TLC.  If the two spots are far apart you can collect larger fractions.  But if they're close (which I'm guessing they will be) then you'll want to make smaller fraction collections to minimize the mixed fractions.

[Mr.rose]

Thanks for the criticism, that is why I am here .

Ok new equipment. 600mm column with a 1L reservoir, 4L of n-hexane, 1L of Ethyl Acetate, Silica TLC plates are on there way, then i will start with a 9:1 mix of hex and ethy, and then 8:2, 7:3, etc etc until i find the best separation.

Yes its 20g of testosterone acetate and 2g of estrodial benzonoate.

I can easily recrystallize the testosterone and estrodial by dripping some de-ionised water into the container. But i may have to do this several times, and i double i will get 100% yield from this method. What would be the best method for recrystallization?

[democanarchis]

I'd try a fractional recrystallisation - Recrystallise, separate crystals and mother liquor, evap mother liquor to dryness, recrystallise etc. I wouldn't use deionised water, you ideally want an organic solvent that is reasonably volatile, you'll need to try a few small recrystallisations, ethanol and cyclohexane are the two solvents I go to first.

[Mr.rose]

Sorry I'm not to familiar with many chemistry methods, I am an engineer, my specialty is with machines not with chemistry.

So what your saying is that with my solution of 10:1 test/estro dissolved in methanol, I have to add either ethanol or cyclohexane which will recrystallize my products? I though that test and estro are both soluble in alcohols?

Will this recrystallize both the test and the estro?

Out of curiosity, to save a bit of money, can a column be made from clear PVC pipe?

[democanarchis]

You'll need to evaporate the methanol, you can generally only recrystallise a solid. Basically you are looking for a solvent that the compound isn't soluble in/is sparingly soluble in at room temperature, that when heated fully dissolves the compound. If your compounds are soluble in ethanol at room temperature then that isn't a good choice. I would try the hexane.

Basically what you do: Take a small sample of the methanol solution and evaporate it. Take a small amount of that solid in a test tube. Add a few drops of hexane and see if it dissolves. If not, then take a separate testtube and place the remaining solid in it. Heat a small amount of hexane to below boiling point. Add it dropwise to dissolve the solid in a minimum amount. Allow it cool to room temperature on its own then place in an ice bath. The major constituent of the solution should crystallise leaving the minor compound in solution.

[Mr.rose]

I see how that works. So what would the yield be of the major constituent? Would greater than 90% crystallize. I'm sure a small amount of estrodial will crystallize as well, correct? So if i run a TLC on it, wouldn't it be like running a TLC on the original product?

[Mr.rose]

I dug up a bit of information

In the hexane, the ratio of solubility of TP to EB was about 100 to 1.

Estrogen has a solubility of 2.5mg/L in ethanol and 20mg/ML in DMSO

Testosterone proprionate is practically insoluble in water; freely soluble in ethanol (~750 g/l) TS and ether R; soluble in vegetable oils.

and also from another source, TP in ethanol: soluble10 mg/mL, 45% (w/v) aq 2-hydroxypropyl-β-cyclodextrin: soluble4.4 mg/mL, H2O: insoluble

Maybe we can use ethanol instead of hexane. With only 2.5mg/L solubility of EB in ethanol, and virtually 100% solubility of TP.

Also with the mobile phase, after a bit of reading these two seem to be popualr choices for these chemicals.

1) 1:1:2 ethyl acetate to tetrahydrofuran to hexane
2) 92 volumes of dichloroethane R, 8 volumes of methanol R, and 0.5 volumes of water
3)9 volumes of chloroform R and 1 volume of methanol R

[democanarchis]

I didn't realise the two products had such drastically different solubilities. In this case, the answer is easy:

20g of testosterone and 2g estrogen. If as you say testosterone has a solubility of 750g/L in EtOH, 20g should dissolve in 20(g)/750(g/L)= 0.02666L or 26.66mL EtOH. If estrogen only has a solubility of 2.5mg/L, only 0.0665mg will dissolve which is negligible.

[EDIT] Just read your other source gives a solubility of 10mg/mL for testosterone. In that case, you'd need 2L of EtOH, but that would still only dissolve 5mg of estrogen which is still going to give you 99.975% pure. What you could do is add 100mL, filter and weigh the solid. If its greater than 2g, put it back in solution and add more EtOH accordingly until you only have 2g of solid filtered material.

So: evaporate your MeOH soln to dryness. Add the solid to a flask with say 100mL EtOH and and add a magnetic stirrer. Put on a hotplate and stir at RT. Filter the solution and weigh the solid. If >2g return to solution and add more EtOH, if not then filter the solution and remove the EtOH from the filtrate in vacuo. Your filtered material is pure estrogen, your dried filtrate is pure testosterone.