# Choise of solvent for extraction

In the synthesis of (S)-Diphenyl(pyrrolidine-2yl)methanol, I need to extract the product using an organic solvent and water. Which organic solvent is the most suitable in this case?

• Ethyl acetate: I think that it is miscible with water and it can form an emulsion, for this reason I would not choose it.
• Hexane: it is an apolar solvent so maybe the product is not very soluble in it.
• Ethanol: I think I cannot use it because it is miscible with water.
• Diethylether: it is an apolar solvent so maybe the product is not very soluble in it.
• Dichloromethane: I think that this is the best choice because it is not miscible with water and the product should be soluble in it.

Is my reasoning correct?

• Is it diphenyl- or biphenyl-? – Mathew Mahindaratne Aug 5 '20 at 15:54
• It's very likely to be soluble in ether. In my experience, ether is preferred when it works. It's easy to remove, and not a huge ecological problem. DCM is OK, but chlorinated solvents aren't the best for the environment. – Zhe Aug 5 '20 at 16:53
• @MathewMahindaratne I confirm that it's diphenyl- – Violet Aug 5 '20 at 19:09
• @Zhe Could you please detail how can you tell it is likely to be soluble in ether? – Violet Aug 5 '20 at 19:12
• Ether is more polar than you think. I have no rationale that would hold water, but I've done this long enough that I have a feeling. – Zhe Aug 5 '20 at 20:15

The perspective about ethyl acetate should be revisited. EtOAc is slightly soluble in water; not infinite (as e.g. ethanol), nor this little as hexane(s). It is greener than halogenated solvents such as dichloromethane, and the boiling point higher than the for dichloromethane or diethyl ether*) is an advantage during summer. Thus, EtOAc is an often used polar solvent of extraction.

Equally recall the partition coefficient -- multiple extractions with a small volume of organic are more efficient than one with a large volume as long as after shaking the organic and the aqueous layer separate well from each other. Typically, for the isolation of your raw-product later to be purified by crystallization or chromatography, three extractions just «good enough» may suffice for you. Except you intend a Craig-Extraction to separate similar compounds from each other (as here), a precursor of the countercurrent chromatography.

The same logic of using multiple small volumes instead of one large is applied when washing the organic phase, e.g. by brine, and back extraction. Indeed, the presence of salts in the aqueous phase may break an emulsion during an extraction; prior to application, it should be tested on small (e.g., test tube) scale ahead though.

*) Compared to diethyl ether, consider MTBE ether as alternative: it is much less prone to the formation of peroxydes, has a higher boiling point, and is cheaper (per L) than the former, too.

• What do you think about using hexane? – Violet Aug 5 '20 at 19:15
• @Violet I have no hands-on experience with this compound. The pyrrolidine and the methanol moiety add polarity to the molecule which is still relatively small thus speculate your product dissolves better («more g product per L solvent») in general purpose solvent EtOAc, than in hexane(s). Speculating you used MeOH as solvent which anyway has to be distilled off at the rotary evaporator, prior to an extraction, check at test tube scale how well a tip of the spatula of the dry «cake» of raw product dissipates in a) 1 mL water + 1 mL hexanes, compared to b) 1 mL water and 1 mL EtOAc. – Buttonwood Aug 5 '20 at 20:38
• @Violet Equally I would prefer EtOAc over hexanes because the identification and integration of its signals in 1H NMR (and comparison [e.g., estimate of purity] with your product) is much easier, than with hexane(s). Check e.g., sdbs.db.aist.go.jp/sdbs/cgi-bin/cre_index.cgi for their spectra. In addition, labs purchase hexane by the 1 L bottle or buy as a blend of isomeric hexanes, the later may be even more difficult to sort out in the 1H-NMR and recycle within the lab, than «just» EtOAc. Possibly your group already has some experience and guide you along MTBE, EtOAc, Et2O, too – Buttonwood Aug 5 '20 at 20:48

(S)-(-)-$$\alpha,\alpha$$-Diphenyl-2-pyrrolidinemethanol is a chiral ligand, which is commercially available (e.g., for $$\pu{53.00 USD}$$/$$\pu{5 g}$$ from Spectrum; Manufacturer: TCI). That is much cheaper price compared to other vendors and the trouble preparing it (assuming this is not the project you are assigned for). To be honest, I didn't have experience with this compound ever, but read a lot of articles about its stereoselectivity. Here what I found online search:

• (S)-(-)-$$\alpha,\alpha$$-Diphenyl-2-pyrrolidinemethanol is a solid with melting point of around $$\pu{77-80 ^\circ C}$$ (you can find vendors and corresponding melting points at ChemSpider).

• (S)-(-)-$$\alpha,\alpha$$-Diphenyl-2-pyrrolidinemethanol is soluble in chloroform ($$\ce{CHCl3}$$). Therefore, I expect it to be soluble in dichloromethane ($$\ce{CH2Cl2}$$). Therefore, your choice is justified. However, keep in mind that some compounds make emulsion with $$\ce{CH2Cl2/H2O}$$ systems more than they do in $$\ce{EtOAc/H2O}$$ systems (based on my 30+-year experience in organic labs).

Based on your scale of synthesis, I'd say you may able to purify this compound by dissolving it in $$\pu{2 M}$$ $$\ce{HCl}$$ solution (making its $$\ce{HCl}$$ salt), then neutralized it to precipitate it as free base again, and recrystallize the solid using methanol or ethanol.

If you are afraid of possible racimization (and synthesis is on a small scale), you can extract the final reaction mixture with $$\ce{CH2Cl2/H2O}$$.