Org Prep Daily

January 1, 2008

Notes on extraction

Filed under: procedures — milkshake @ 7:55 pm

ftaci.jpg

The emulsions

Emulsions are a nasty. When the phases won’t separate add some brine and gently swirl around (no shaking). If this does not help try to add some ethanol and swirl again. If possible, acidify with diluted HCl or sulfuric acid. If nothing else helps, filter the whole damned emulsion through, using a medium-to-fine-porosity large sintered glass Buchner funnel. This takes some time and effort but it usually does the trick of breaking the milkshake. You can also do the filtration through a plug of charcoal for the extra emulsion-busting effect but you may end up losing some material on the charcoal. (Be sure to use fine porosity funnel as to prevent the charcoal from getting through).

In extreme desperation, chemists were using centrifuge to break emulsions.

You have to expect emulsions with greasy acids in basic conditions and greasy amines in acidic condition. Some compounds are natural detergents – phospholipids for example. One can pair up quartenary ammoniums with greasy anions like perfluorobutyrate or tetraphenyl borate to obtain salts that will extract into organic  – and even separate reasonably on silica column – but these classes of compounds are best not purified by aqueous extraction.

What therefore God hath joined together, let not man put asunder.

Highly soluble compounds

Another class of compounds that should not be purified by aqueous workup are the highly water soluble compounds like polyamines, polyols, unprotected peptides etc. If you have to do aqueous workup, use a minimal volume of water and best saturate it with some salt. Ethyl acetate is the usual choice for the polar compound extraction but in desperate cases one can do multiple re-extractions with a solvent like methyl ethyl ketone, n-butanol or chloroform+methanol 2:1(v/v) mixture. One shoud expect to extract large quantities of water and salts (sodium acetate for example) into the organic phase when using these alcohol-containig solvents.

In the most unfortunate cases people set up continuous extraction apparatus and let it run overnight – though it is not something commonly seen in the organic lab nowadays. There is always some more convenient non-aqueous workup/isolation alternative.

Please be sure to collect all your aqueous washing and save them until the moment when you hold a flask with your crude product that you can weight.  If you are not sure that you extracted all your product from aqueous phases, run HPLC on them (you can take a sample, blow off organics from it by a stream of nitrogen and apply undiluted onto the column; strongly basic solutions must be acidified to protect the column. It is very pleasing to compare an HPLC of your crude product extract – with the HPLC from the combined aqueous washings that contain all the baseline grass of the impurities you just removed).

Small-scale extractions

If you don’t have a miniature sep funnels the polypropylene 20 mL syringes will work fine for the purpose (the all-plastic kind, without a rubber plunger; the slip-tip syringes are preferable over the luer-lock). I like a needle with sawed-off tip that can be stopped by finger – but other people like to use a sharp needle with a rubber or silicone piece as a stopper.  When you are unstopping the needle do it always with the needle pointed up and with slight negative pressure produced by a pull on the plunger. Squeeze out all air void before inverting the syringe needle down as to prevent your mix from squirting out due to the changing vapor pressure

Please note that polypropylene syringes swell in organic solvents: toluene and dichloromethane swell it quite fast but even ethyl acetate will mis-shape the syringes after while. This will cause syringes to leak over time so it is best to discard the syringes after the extraction.

For small volume extraction done in syringe I like chloroform inplace of dichloromethane – it separates a bit faster (its heavier) and it has a lower vapor pressure.

Small extract volumes can be conveniently dried by direct filtration trough a plug of MgSO4 in a small Buchner funnel (14/20 joint).

Order of washing

The recommended order of washing is diluted acid followed by water (or brine) followed by bicarbonate. One can dry extracts after bicarbonate directly, no additional washing with brine is needed  – drying with MgSO4 takes care of the mild residual basicity – but one should re-wash the extract after an acid wash. (For example citric acid is pretty soluble in ethyl acetate.)  Also washing the extract after acid wash directly with bicarbonate will produce effervescence and pressure buildup in the sep funnel so it is best to add one neutral wash inbetween.

Drying agents

MgSO4 is much more powerful drying agent than Na2SO4 and it comes in different grades – use the the powdered grade, the one that has the consistency of confectioner sugar; the lumpy grade is useless. When MgSO4 fails (typically the mixtures that form stable emulsions also provide milky extracts that are difficult to dry) the comercial 4A powdered activated sieves will work; use Celite plug for filtration from sieves since the powdered 4A sieves have rather fine particles.

Another good drying agent is Na3PO4 and Na2CO3; it works better than K2CO3. (I am not a great fan of KOH or NaOH as drying agents). Since Na-triphosphate and  carbonate are strongly alkaline, they are not compatible with ethyl acetate.

You may want to give the drying agent some time to do its work, like 20 min stirring in a round flask. If you are concerned about dropplets of moisture stuck above the level + on the neck of the flask, you can sprinkle some drying agent onto a funnel that was wetted with the solvent and then wash down the drying agent stuck on the funnel with some additional solvent; that takes care of the problem.

Washing away DMF, DMAc

Both ether and ethyl acetate are the recommended solvets for washing away small volumes of DMF, DMAc, DMSO and tetramethyl urea with water and bicarbonate solution. I don’t know the exact reason why they work better than chlorinated solvents (which tend to extract DMF and others into organic phase) but I suspect the relatively high water solubility of ether and ethyl acetate in water helps to improve the partitioning of DMF, DMAc, etc. into the aqueous phase.

Ether and ethyl acetate can cause some precipitation of salts from saturated salt wash solution (sat NaCl, NaHCO3, NH4Cl) – in that case one should want to dilute the mixture with additional water, to bring the inorganics back into the aqueous phase.

More on extractions here

43 Comments »

  1. For emulsions, using a warm to hot aqueous layer can dramatically reduce their occurrence. For acid washes, dilute phosphoric is preferred since it is non-oxidizing and does not react readily with olefins. In addition, diluting o-phosphoric acid is much easier and safer (less exothermic) than sulfuric of HCl. Acetic (or citric) acid may be used, but a stronger base (K2CO3) may be needed to fully remove it from the organic layer.

    Comment by milo — January 1, 2008 @ 10:25 pm

  2. excellent advice!

    THF/H20 emulsions: add et2o equal volume to the THF in question (sometimes it takes 2 volumes)

    another trick for getting rid of the last traces of DMF is washing the organic layer w/ 5% LiCl a few times

    Comment by cookingwithsolvents — January 1, 2008 @ 10:40 pm

  3. THF separates quite well with sat NaCl but of course adding extra ether helps.

    I did not know the LiCl trick and it makes sense – I could try it for large scale reactions where wash with huge volumes would be impractical.
    The other alternative one can use is an oil pump hooked up to a cold finger rotavap – I have one and it pulls between 0.5 and 1 Torr on a spinning rotavap so I could pull off the majority of DMF, DMAc from a 30-40C bath before the workup.

    Comment by milkshake — January 2, 2008 @ 5:34 am

  4. I am a fan of Celite. It’s great for filtering an emulsion and you don’t risk losing any of your product, like you can with charcoal. It’s also useful because you can often simply rotovap your reaction onto Celite and then pour the Celite onto the top of your column. This works with or without prior addition of the aqueous phase. Now, obviously you have to be sensible. Don’t concentrate an H2SO4 aqueous layer onto Celite but most neutral salts are fine.

    Comment by Bunsen Honeydew — January 2, 2008 @ 10:58 am

  5. Re: getting rid of residual dipolar aprotic solvents – a while ago there was an OPRD paper on how best to do this: http://dx.doi.org/10.1021/op060154k
    The data suggest that toluene and isopropyl acetate are superior to ethyl acetate as extraction solvents.

    Comment by dtb — January 2, 2008 @ 3:32 pm

  6. Thank you dtb – this is a pretty useful reference!

    Comment by milkshake — January 2, 2008 @ 5:59 pm

  7. Regarding post #5, this looks like a useful paper. I wonder if anyone maintains a list of papers like this, about general, in-the-lab type stuff? Papers like this one, and the JOC NMR solvent impurities paper, etc. Practical references that don’t really get cited very often in literature on, say, organic synthesis, but which are really quite useful to the practitioners of such arts.

    Comment by zt — January 2, 2008 @ 10:57 pm

  8. ZT- you can try looking at the NOT VOODOO website. There is a lot of useful stuff there. I don’t think there are many papers per se but there is a lot of good advice.

    http://chem.chem.rochester.edu/~nvd/

    Comment by Bunsen Honeydew — January 3, 2008 @ 9:23 am

  9. Vogels guide to practical organic chemistry and Advanced Practical Organic chemistry were quite useful to me.

    The Not Voodoo site was also used, along with this one… Organic Syntheses also helped me learn a lot of synthetic procedures(like how to purify solvent, reagents and so on). I just searched for a reagent on Org. Synth. and looked at the notes… often times they say how they were purified.

    Comment by Tom — January 3, 2008 @ 10:54 pm

  10. During my Ph. D., I was lucky enough to deal with relatively non-polar compounds, so whenever I needed to wash away DMF, DMSO or HMPA, I extracted my product into roughly 15% EtOAc/hexanes. Then, wash 3x with dilute NaCl, and I guarantee you won’t see any polar solvent residue in the crude NMR.

    Comment by HPCC — January 4, 2008 @ 1:22 pm

  11. M.S., this is great. I’ve made it suggested reading for my advanced organic students as a reference for practical extraction advise.

    Comment by Chemgeek — January 5, 2008 @ 12:57 pm

  12. When working up THF or DMF solutions of your product I recommend to remove the solvent firts and sissolve the residue in dichloromethane. DMC also forms a nice azeotrope with water so no drying with MgSO4 is necessary.
    Breaking up the small layer of emmulsion between yor aquoes phase and the organic: I prefer to collect it with the organic phase and filter it through a small plug of Na2SO4 or MgSO4. For small volumes of emulsions in the organic phase it is a viable way (and you get rid of some undissolved stuff (dust, teflon grease, salts, other CRaP)).

    Comment by Jan — January 5, 2008 @ 9:48 pm

  13. Solvent impurities paper: Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512. doi:10.1021/jo971176v

    I’ve found the two tables from this paper published on the back covers of a few textbooks.

    In a previous project we used DMSO as a solvent. Distilled off under vacuum (60-70 C). When almost dry and pasty, washed with water, with sonication to remove the entrained DMSO. Worked pretty well.

    Comment by andrew — January 5, 2008 @ 9:50 pm

  14. Jan,

    Do you mean DMF or DCM form the azeotrope. According to the literature, dichloromethane forms a 98:2 (DCM:water) azetotrope (bp = 39C), I cannot find any data on DMF or THF with water.

    Comment by milo — January 10, 2008 @ 1:35 pm

  15. DMF forms upper azeotrope with water,like DMSO or TFA – it means that wet DMF boils higher than dry DMF.

    The other thing is that DMF and DMSO should not be heated to their boiling point at atmospheric pressure because they decompose easily.

    Also, one may want to dry and filter the organic phase after the extraction because even if the last wash was done with pure water there is no guarantee the dropplets remaining in the organic phase do not contain some inorganics or some insoluble stuff or from the reaction mixture. And water dropplets cause “bumping” on rotavap.

    Comment by milkshake — January 10, 2008 @ 3:09 pm

  16. […] noted by Org Prep, extractions are a handy (and often powerful) way to separate  compounds. Separating organics […]

    Pingback by Extractions: The Math — January 17, 2008 @ 11:40 pm

  17. For polar compounds, both amines and carboxylic acids, a 4:1 mixture of chloroform : isopropanol has been working wonders for me. Just make sure you dry organic layers with magnesium sulfate, NOT sodium sulfate (it dissolves in iPrOH). I have shared this trick with a number of collegues and saw a couple of bottles labelled as Eve’s mix around the labs… I got this trick form my master’s advisor…

    Comment by Moebius — January 25, 2008 @ 2:59 pm

  18. Another note: when I work up reactions with a high salt content (Grignard etc.) then I prefer using diethylether. Working up reactions involving THF, EtOH, MeOH or any other watersoluble solvent/byproduct I use dichloromethane. The ratio behind it is to maximize the difference in density of the two phases. This facilitates the phase separation.

    Milo: DCM forms an azeotrope with water. When you rotovap the organic phase the residual water will evaporate easily (when you have lots of water you will see a phase separation in the collecting flask, water appears oily in DCM). During my whole thesis I used NaSO4 only to break up emulsions…

    Comment by Jan — January 26, 2008 @ 6:18 pm

  19. For really nonpolar compounds I like to use an organic/organic bilayer (for example pentane/DMF). This is a great first step to remove polar organics. I used it a few times for sensitive silyl-enol ether synthesis. The polar solvent removed any unreacted SM, amine salts…etc. The only thing in the pentane is enol ether and a little DMF or DMSO.

    Comment by Dennis — April 7, 2008 @ 9:11 am

  20. acetonitrile also does not mix freely with aliphatic hydrocarbons, you can do hexane or cyclohexane/acetonitrile partitioning.

    Comment by milkshake — April 7, 2008 @ 4:22 pm

  21. wht abt large scale extraction ??:S:S?

    Comment by Hamid — April 20, 2008 @ 9:11 am

  22. Hello milkshake,

    I sincerely need your advice on work – up of my compound. I did reaction of chloro diamino pyrimidine with cyclohexyl methyl alcohol (cyclohexyl methyl alcohol was used as a solvent – 50 ml, its very high boiling (bp : 180 oC), so its difficult to distill off, syrupy, very viscous,. I tried few times but was unsuccessful…column proved to be futile..The reaction mixturte contains 6-alkylether diamino pyrimidine and large excess of cycolhexyl methanol. Could you suggest any method so that I can get rid of alcohol..

    Cheers
    Marto

    Comment by Marto — August 12, 2008 @ 2:21 pm

  23. I would put the crude mix into a shortpath vacuum distillation apparatus (or a Kugelrohr apparatus) and pull off the alcohol excess on highvac – with a decent oil pump you can reach easily 0.1Torr and your alcohol should distill of, at least the majority of it. Then I would take the residue (which will probably still contain some alcohol impurity) and try to precipitate it by adding benzene or cyclohexane or ether.

    if the precipitation does not work I would then try to precipitate it as a HCl salt from ether, by bubbling HCl gas into the solution in dry ether (filter the formed precipitate quickly and preferably under positive pressure of nitrogen – as the anhydrous HCl solution is extremely hygroscopic). It is easy to make HCl gas by adding conc HCl into conc H2SO4 or you can use the commercial anh HCl solution in ether (or anh HCl in dioxane).

    If this hydrochloride precipitation trick fails to give a filtrable precipitate, you could portion the mixture between ether and water, your diaminopyrimidine is probably basic enough to be extracted into the aqueous phase as a HCl salt and the greasy alcohol should stay in ether. Then you alkalise the aqueos phase with ammonia and your product should precipitate out.

    Comment by milkshake — August 12, 2008 @ 3:14 pm

  24. Hi MS,

    many thanks for the suggestions..Cheers Marto

    Comment by Marto — August 13, 2008 @ 7:26 am

  25. Dear all,
    We are struggling for distillation time of DMF. To distill DMF 400L it is taking 8 days in the batch distillation.
    we are distilling DCM+water+DMF+product+ortho xylene under vacuum our objective is to remove DMF completely. can any one give the solution to do the distillation in short time.
    I appreciate for your answers.
    Regards
    M Narendra Kumar

    Comment by Narendra — January 26, 2010 @ 7:13 am

    • Your vacuum is probably too poor. At 1-2 Torr DMF should distill just slightly above the room temperature. Make sure the distillation condenser operates on properly chilled antifreeze-water below zero so that all DMF condenszs. (Dry ice condenser would be even better choice but you would use up lots of dry ice).

      Since you are doing it on process scale you will need to buy a powerful oil pump with a very high pumping rate (over 400L/minute free air displacement rate – the pump will be pretty big-sized) and have an efficient and large-sized cold trap attached in between, to protect the pump, and use a very wide tubing that is as short as possible between the distillation apparatus and the cold trap. You need to have the distillation apparatus that fits the scale. Maybe you can distill it straight from the reactor if you can seal the leak around the stirrer shaft, or if you could replace the shaft with a metal-based sparge capilary for vacuum distillation.

      It is not worth doing extraction in your case – to wash away 400L of DMF you would produce many thousands L of aqueous washings that would have to be processed and disposed as hazardous aqueous waste (not mentioning the losses of your product). It would be too expensive.

      Comment by milkshake — January 26, 2010 @ 7:51 am

      • Dear Milkshake,

        Firstly thanks to you for giving valuable suggestions
        whether we can do any Reactive distillation or Extractive distillation in this case to reduce the Energy or operating cost

        Regards
        Narendra

        Comment by Narendra — January 26, 2010 @ 2:49 pm

  26. Can anyone give the Continuous Hydrogenation process upto 30 bar references/journal/experience

    Regards
    Narendra

    Comment by Narendra — January 27, 2010 @ 5:23 am

  27. Hi I have a question. My compound is soluble in water, methanol, CHCl3, CH2Cl2, THF. I want to separate my compound from a mixture containing LiOH, 2-pyridinecarboxaldehyde. I tried to use extraction to separate them, but yield is very terrible or not successful. Could anyone give some suggestions? Thank you.

    Comment by staphlee — February 19, 2010 @ 5:59 pm

    • I don’t have a specific answer because you did not tell me what is your compound and what exactly is in your reaction mix and how you got it. Sometimes one cannot use extraction at all

      Comment by milkshake — February 20, 2010 @ 11:32 am

  28. Hi,
    I have run a reaction in DMF and the product of the reaction is highly water soluble. can anybody tell me how to separate it from the DMF?

    Comment by Ranajit — March 10, 2010 @ 4:41 am

    • You need to evaporate DMF. I would recommend to use an oil pump with high pumping rate (above 150 L/min) connected to rotovap with good seals and dry ice-cooled condenser. You may want to put a second cold trap inbetween the pump and rotovap, to protect the pump oil from contamination. When the vacuum in your rotovap is below 1 Torr DMF evaporates quite nicely.

      Comment by milkshake — March 10, 2010 @ 8:19 am

  29. After conducting a Parikh-Doering (Swern type) oxidation in DMSO I was left with a fairly water soluble product (hemiaminal) in DMSO (staring material was also quite water soluble, very soluble in acetonitrile). The extraction I tried was EtOAc washed with 2 x water, which left only about 20% of my product in the organic phase.

    Any tips on how to run the extraction in such a way as to separate my product from DMSO.

    Comment by Octo — November 29, 2011 @ 6:57 am

    • yeah, you might try to add few spoons of lithium chloride to your workup mix, to better keep DMSO in aqueous phase as a complex with Li+, and give it few more extractions with EtOAc but my guess is that you won’t be able to extract your product selectively away from DMSO (and you may get some LiCl carried into the org extract too) . I would extract everything, by using some really polar solvent mix like chloroform+isopropanol 3:1 mix, or pure isobutyl alcohol, then pump on the evaporation residue with stirring overnight from a warm bath. Change your pump oil, fill the cold trap with liquid N2 (instead of dry ice) to get the best vacuum you can from your pump. Possibly chromatography at the end because you will still have some DMSO left there.

      You may need to do real Swern next time (rather than SO3.pyr/DMSO) because with Swern you can get away with just 1.2 equivs of DMSO. Or Corey-Kim oxidation (Me2S/NCS/NEt3).

      Also, since your material is basic, you might try mildly acidic ion exchanger (with carboxylic acid moiety) to absorb your product from the reaction mix and then wash the product from the resin with diluted AcOH in water and lyophilize the solution. Your product and all other amino stuff from the reaction mix would be in the form of acetate salt. (In the old days when peptides were cleaved with anhydrous HF from Merrifield resin, this ionex/AcOH was a routine technique, to remove fluorides from the crude peptides.)

      Comment by milkshake — November 29, 2011 @ 1:16 pm

      • LiCl was a nice tip, thanks for that, seems to help remove the last dregs of DMSO. regarding your other solutions, I tried the Swern first, but the starting material was too insoluble at such low temps in DCM, and the desired hemiaminal product is not acid stable and eliminates to an undesired ene-amine.

        I settled upon precipitating all insolubles with diethyl ether (x2) and then dissolving the precipitate in a small volume of water (with LiCl, ty) and extracting the sparingly soluble organics with several portions of chloroform. Not elegant, but good enough to go ahead with column chromatography.

        thanks again.

        Comment by Octo — November 30, 2011 @ 10:53 am

  30. hi,
    i have a question about my compound. i did a methoxy deprotection reaction with AlI3 and tried extracting it with EtOAc and running a column. But it turns out that my compound is soluble in methanol. any suggestion on how to quench the reaction and do the workup?

    Comment by dripta — November 15, 2012 @ 5:54 pm

    • getting all produced aluminum salts from the deprotected product is difficult because of phenol complexing with Al, which depends on the phenol structure. If your phenol product is poorly soluble in water, quenching the reaction mixture with diluted aqueous hydrochloric acid or sulfuric acid solution could be a good alternative – you could collect the poorly soluble phenol product by simple filtration, followed by thorough washing with water.

      I cannot suggest anything more specific without knowing what are you using as a solvent for the demethylation, or what is the solubility and other properties of your phenol product in water, ethyl acetate, dichloromethane etc

      Comment by milkshake — November 15, 2012 @ 8:44 pm

      • thank you so much for your prompt reply. i am trying to deprotect two methyl groups meta to each other simultaneously. i am using benzene and Al and I2 to generate AlI3 in situ.generally i quench the rxn with 2M HCl and extract with EtOAC but i am getting horrible yields. i know for sure that my pdt is soluble in methanol and only partially soluble in dcm and ethyl acetate. so i was wondering if there is any other way to quench the rxn or use different solvent system to extract my pdt.

        Comment by dripta — November 15, 2012 @ 8:55 pm

  31. Hi, my product in the alcooxide form and I need to convert the product to alcohol and extract by EtOAc, so how to get rid of DMF and the best order of these steps.
    Thank you.

    Comment by Mohammed Isam — March 25, 2014 @ 3:55 am

    • sorry, these are freshman questions, please post it on reddit chemistry or some similar forum

      Comment by milkshake — March 25, 2014 @ 10:41 am

  32. I need to get malachite green out from nutrient broth medim ..tried ethyl acetate extraction..didi not work too well
    Suggests any alternatives.

    Thank you

    Comment by righton — March 26, 2014 @ 5:36 pm

    • there is no good way of removing this dye from a broth media – by extraction or otherwise. You can try to pair it up with a greasy anion – for example by adding sodium tetraphenylborate, and try to extract the tetraphenylborate of malachite green into ethyl acetate – but I suggest that you avoid having this dye in broth media to begin with. (The tetraphenylborate excess can be removed later by adding KCl, potassium tetraphenylborate is poorly soluble in water, but first you ned to evaporate ethyl acetate, which has solubility about 100g/L in water, for K-tetraphenylborate removal by precipitation)

      Comment by milkshake — March 26, 2014 @ 6:23 pm

  33. Thanks so much for the tips! Really appreciated the notes!

    Comment by Phyllis — August 28, 2014 @ 8:25 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

The Shocking Blue Green Theme. Blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.

Join 139 other followers

%d bloggers like this: