Practical application of Phenol/Chloroform extraction

While there are many more methods to choose from for cleaning up your RNA or DNA than there used to be, sometimes Phenol/Chloroform extraction is still the best way to go. Here I’ll discuss some of the practical aspects of using this technique.

Nick introduced the topic of Phenol/Chloroform extraction in a previous article, touching on some of the ideas about how organic extraction will remove proteins from an aqueous solution. In brief, proteins consist of both hydrophobic and hydrophilic residues, and through protein folding, achieve a compromise with water to remain soluble. However, when they are given the opportunity to transition to an environment that can accommodate both polar and non-polar residues (ie – phenol or phenol/chloroform) with no compromise required (ie – folding), they happily move over to that phase. The more highly polar molecules, like carbohydrates and nucleic acids, are “happier” in the aqueous phase (with some exceptions noted below) and remain there. Now for the nitty-gritty.

Phenol versus Phenol/Chloroform versus Chloroform
One of the most frequent questions I’m asked when training someone is what the differences are between the different organic phases used in extraction. Here is the breakdown, as best as I understand them.

Phenol – What we are actually talking about here is buffer-saturated phenol, which consists of a solution that is actually about 72% phenol, 28% water. Since phenol is a weak acid, the solutions that we use have been equilibrated with buffer to bring the pH to a particular target – either acidic for RNA purification or slightly alkaline for DNA purification. In addition to a certain amount of water dissolving into the Phenol, there is a certain amount of phenol that dissolves into water – at equilibration the aqueous phase will contain about 7% phenol. This is thought to aid in the extraction, as this dissolved phenol helps denature proteins while they are still in the aqueous solution. Buffer saturated phenol has a density that is only slightly higher than that of water.

Phenol/Chloroform – This is a mixture of buffer-saturated phenol and chloroform, usually close to 1:1 for DNA purification with other ratios sometimes used for RNA purification. Isoamyl alcohol is sometimes included as an anti-foaming agent, but is generally thought to be an inert and optional addition. This solution is commonly used in lieu of buffer-saturated phenol for a couple of reasons. As I mentioned above, the density of buffer saturated phenol is only a little higher than water. So if your aqueous phase contains enough salt or any other solutes that would increase its density, then you could end up with phase inversion during extraction, where your aqueous phase is under the phenol, rather than on top of it. Chloroform is significantly denser than water, so adding it to the organic phase increases the overall density of that phase, helping to prevent phase inversion.

In addition, the chloroform (and some say isoamyl alcohol) help reduce the interphase – the fuzzy border between the two phases populated by molecules that can’t decide where they want to go. These can be partially denatured proteins, DNA (depending on the pH), and/or partially denatured DNA binding proteins that are still clinging to DNA, and it is a real pain in the butt. If you pipette off some of this material when removing the aqueous phase, then you decrease the purity of your sample. If you are too timid while pipetting, then you hurt your yield. If you have my luck, then whatever it was you wanted to keep most is sitting in it. Adding chloroform to the mix helps reduce this. (But I have an even better solution to this problem that I’ll tell you about below.)

Chloroform – This is normally used after phenol or phenol/chloroform extractions. While pure chloroform doesn’t work as well as the organic solutions mentioned above for protein extraction, it works well for extracting phenol from aqueous solutions. Remember when I said that the aqueous phase contained ~7% phenol after equilibration? Do you also remember when I said phenol likes to denature proteins? If you don’t get rid of (or at least severely reduce) the phenol in your now protein-free nucleic acid solution, it could come back to haunt you by partially or completely inhibiting enzymes that you treat the DNA or RNA with down the line. Presented with a nice chloroform home, however, the phenol will partition away from your nucleic acids. Chloroform itself is about 10X less soluble in water than phenol (~0.8%) and is less denaturing to proteins. I was also told long ago that it is less likely than phenol to pellet with DNA during ethanol precipitation once upon a time, but I cannot find a reference to back that point up.

Ether – This can also be used to extract phenol back out of the aqueous phase. However, because of the explosive potential of ether and the tendency of biology-types to have Bunsen burners and strikers in their labs, it has been largely replaced by chloroform.

Not so Pretty in Pink
A note of caution: don’t use your phenol or phenol/chloroform if the solution is turning pink. Oxidation of the phenol produces a pink/brown compound, and this compound will cause nicking of your DNA and degradation of your RNA. Most commercial phenol solutions contain an antioxidant to inhibit this oxidation, and phenol buffered at an acidic pH seems to be resistant to oxidation, but it isn’t a bad idea to move a portion of the buffer saturated phenol (from the brown bottle that it likely came in) to a clear bottle or tube to inspect it before you start your extraction.

pH matters – a lot
Occasionally somebody does a phenol extraction and doesn’t recover any of the DNA in the sample. If this happens to you, or somebody in your lab, your first question should be “Which phenol did you use?” Labs that do both DNA and RNA work will likely have both acidic and basic buffered phenol solutions, or somebody will buy a new bottle of phenol without paying attention to the pH. Extraction of DNA containing samples with acidic phenol results in the denaturation of the DNA, and once denatured, the DNA partitions to the organic phase. This is a key feature of many RNA purification protocols, which is one of the reasons acidic buffer-saturated phenol is used.

Now, sometimes a lab’s DNA phenol extractions start failing (no recovery of the DNA afterwards) and the pH of the phenol is called into question. If you find yourself in this spot, you can’t simply dip your pH meter into it, and you cannot use pH paper, since the pH indicator on the paper was characterized in aqueous solutions. The method that I’ve used is to dilute 1 ml of the buffer saturated phenol with 9 mls of 45% methanol, mix, and then measure the pH with a standard pH meter. The safest way of adjusting the pH is by replacing the aqueous phase on top of the phenol solution with a fresh aliquot of ~100 mM buffered water (usually Tris pH 7.9 for DNA work), mix the phases well, and then let the bottle settle until the phases are well separated again. Then pH it again.

Mixing your phases
Phenol/chloroform extractions are amazingly efficient – less than 1% of the average protein remains in the aqueous phase after the first extraction has come to equilibrium. The trick is to get the extraction to equilibrium, of course. The more surface area there is between the two phases, the faster this happens, and that surface area is greater the finer emulsion you have created. This can be achieved by vortexing the phases for a couple of minutes, as many protocols call for, but not all samples can be vortexed. If you are purifying very large DNA, like genomic DNA, then you may have to mix your sample much more gently, and therefore perform each extraction for much longer. So on this point, follow your protocol and be very cautious about trying to shave time off this step.

Effects of denaturation and digestion
Some protocols call for protein denaturation and possibly digestion with Protinase K before extraction. Both of these steps are attempts to reduce the amount of material that is trapped in the interphase and therefore improve the yield of DNA or RNA recovered. I have never seen any negative effect of denaturing the proteins with SDS before extraction. On the other hand, digestion of the protein could reduce the purity of the nucleic acid that you recover. While whole proteins are almost guaranteed to partition to the organic phase, once the protein is digested into small peptides, not all of those peptides will have the same chemical ‘character’ of the whole protein, and each will have its own partition number. It may not matter a lot if you have some peptides in your nucleic acid, depending on your downstream application, but it’s formally possible that these contaminants could effect your future quantitation of the sample. However, I discovered a better way of eliminating the dreaded interphase…

Phase Lock gel®
This is one of those things that seems to be in 50% of labs, yet less than 10% of the people I’ve talked to know what it is or how it works. I discovered this at a critical point in my research, and it saved my thesis. In short, Phase-lock gel is a gooey, Vasoline-like gel that has a density that is slightly greater than water. If you add your extraction on top of it in a centrifuge tube, and then centrifuge it, the Phase Lock gel collects between the aqueous and organic phases, separating the two and preventing the formation of the DNA/RNA hungry interphase.

A search of the internet didn’t turn up any pictures of this process that I thought were good enough, so I took some of my own. In this little demonstration, red dye is taking the place of our precious nucleic acid, and blue dye is substituting for protein.

A) The Phase Lock gel® pelleted into the bottom of a 1.5 ml Eppendorf tube.
B) After adding phenol/chloroform and the aqueous phase, complete with faux DNA (red) and faux protein (blue) in the aqueous phase.
C) After gentle shaking for 5 minutes.
D) After centrifugation. Note the gel now separates the organic phase from the aqueous phase.
E) After a second addition of phenol/chloroform and gentle shaking for 5 minutes.
F) After the second centrifugation. The faux DNA could now be extracted with chloroform (in the same tube, if space allows) to remove the residual phenol.

As you can see, the gel forms a stable partition between the two phases, and if you want to extract the sample a second time and there‘s still room in the tube, then you can do it, using the same tube two or more times with no compromise of the sample purity. You cannot vortex mix the two phases in a tube containing this reagent, but you can votex mix in a separate tube, then add the sample to the tube with the gel and centrifuge. They have this gel in two different flavors – one for regular samples (light) and another for high density samples, like solutions with a high salt or protein concentrations (heavy).

Using SDS to denature the proteins in my sample prior to extraction and then employing Phase Lock gel® to separate the phases has consistently given me DNA samples with 260/280 ratios of 1.8 and greater than 98% recovery. Seriously great stuff.

In preparing this article I came across this website, which has a lot of useful information. Visit it if you would like to learn more about phenol.

Now to hear from you: what are your tricks and tips for perfect protein extractions?


  1. Sheikh Gulfam on March 10, 2017 at 5:39 pm

    Very clear and informative article

  2. Christina Lebonville on July 8, 2015 at 6:59 pm

    Where has this site been all my life!? This was so super helpful and clear. I just picked up the Phase Lock gel tubes and can’t wait to undo decades of crappy extraction protocols for my lab. Now I understand what’s happening and can fix it!

  3. alexkle on November 8, 2013 at 7:12 pm

    Did a little bit of work into reverse engineering this phase lock stuff, found an MSDS which describes it as a “Proprietary Mixture” of CAS#: 63148-62-9 and 112945-52-5. This works out to a mixture of Polydimethylsiloxane and Silicone Dioxide.

    Sourcing both ingredients is fairly cheap and easy, and could be good fun playing around with the proportions. However I read the paper that someone else posted about using vacuum grease, so I tried a phenol extraction with Beckman high vacuum grease, and it worked great! No RNA degradation either. Took a while to find the components of this brand but as expected…Polydimethylsiloxane and Silicone Dioxide. Just remember to autoclave it first!

  4. shamshu27 on July 2, 2013 at 9:13 pm

    Hi Jode,

    I was wondering if I could get some help regarding the the genomic DNA extraction protocol from bacteria using phenol-chloroform. I simply took cells and dissolved it in SDS, TE and proteinase K. Kept it at 37 for 1 hour and then added phenol chlotoform. Interestingly, the aqeuous phase came out to be very soapy and beneath it was a white insoluble precipitate and when I tried to collect aqeuous phase, the precipitate came along with it. I still went ahead and added sodium acetate and isopropanol but couldn’t see any DNA precipitate ( as expected). I could not figure out as to what have gone wrong as I followed the protocol word by word. I would appreciate if I get some help in troubleshooting it.


  5. Roberto Rosati on May 20, 2010 at 3:43 pm

    Thanks 🙂

    I’ve just noticed that NCBI holds a full copy of the paper by Mukhopadhyay and Roth:

    Also, the authors wrote a chapter in the book “RNA isolation and characterization protocols”, where they suggest the use of vacuum grease on RNA extraction:
    (see chapter 10, page 55)

  6. Suzanne Kennedy on May 20, 2010 at 4:09 am

    Wow- good find Roberto! Also, great picture 🙂

  7. Jode Plank on May 20, 2010 at 1:05 am

    I think you probably got it, Roberto. Now for the testing phase…

  8. Roberto Rosati on May 19, 2010 at 7:56 pm

    Now the search for the composition of this gel tickled my curiosity too, and I just found this paper:
    I can only see the first page, but it does help 🙂 I wonder if somebody could test it? Sadly I’m not in the USA, so if I want to, I’ll have to look for some substitute anyways.

  9. DrGenius on May 5, 2010 at 5:15 pm

    And those who don’t want to spend $122+ per 200 tubes to extend the length and cost of a protocol when accurate pipetting would generally suffice.

  10. Jode on May 4, 2010 at 11:00 pm

    Thanks for the information – that’s really helpful. I find that people break down into two categories with Phase Lock gel: those that love it and those that haven’t used it yet.

  11. Doug on May 4, 2010 at 10:43 pm

    Apparently I needed an afternoon coffee after all. Doesn’t come up in searches since it is a scanned pdf, just had to check the references in the 5 Prime handbook for the stuff. The abstract has the phrase: “A key ingredient is an easlily dispensed silicone gel barrier (Phase Lock Gel)…” I never pursued it much further. I was trying to get my hands on the stuff in bulk (like Sigma used to sell it when they held the patent) but priorities shifted.

    And DG, try vortexing in the presence of the gel with your samples. It worked for me for some applications but not others. If the density of your aqueous phase is close to that of the gel, it will be a disaster. If they are different enough it should work.

  12. Doug on May 4, 2010 at 10:07 pm

    I spent some time looking for the composition a year or so ago. Can’t find the paper where I found it but IIRC, it was a silicone gel. Petroleum jelly is soluable in chloroform FWIW.

    DG, factor in time saved into your cost equations. These can make extractions fly.

  13. Ryan on May 4, 2010 at 5:59 pm

    Very informative article. Now I can finally look undergrads in the eye and tell them why Choloroform after Phenol/Chloroform and what is Isoamyl alcohol for? I also wasn’t aware of the pH issue for RNA versus DNA extraction. As a safety issue, I also wanted to mention working with Phenol/Chloroform is a good time to don your lab coat, safety glasses, gloves, and pants (no sandals is a good idea as well). I know of two people who splashed Phenol on their legs while wearing shorts and were burned pretty much instantly. Reminiscing back to my undergrad lab class, I remember the day I was “sniffing” my tubes to see if their was any residual chloroform in my aqueous phase and had a nice temporary loss of vision and dizziness when putting my nose up to an eppie filled with chloroform. I highly recommend the wafting with your hand technique
    if you’re a “sniffer”. Again, great job on this article Jode!

  14. Jode on May 4, 2010 at 5:23 pm

    There’s a lot to be said for “If it ain’t broke, don’t fix it.” I was also doing a fair number of extractions (~50 a week), but found that the cost of having to repeat the particular experiments that I was doing was more than the cost of the product, but I was doing experiments with tiny amounts of a very precious substrate that I couldn’t spec after the extraction.

  15. DrGenius on May 4, 2010 at 2:30 pm

    Interesting & informative article. The phase lock gel seemed interesting at first until you mentioned a) it can’t be vortexed with the two phases, and b) the price. These two factors make it impractical and expensive. I routinely do ~50 RNA preps; that’s 50 trizol tubes which are chloroform extracted into 50 isopropanol precipitation tubes. This gel lock stuff would require an intermediate step, meaning labelling another 50 tubes, at 50c each that’s also $25.

  16. Jode on May 4, 2010 at 4:44 am

    I played around with a variety, but red and blue food coloring ended up working the best. (McCormick, diluted 1:300) Here they were simply used to make the extraction more photogenic, but this could be useful in a teaching situation.

    The Phase Lock gel really isn’t that expensive – ~50¢ per tube. They call it a ‘gel’, but it isn’t anything like agarose or polyacrylamide. I suspect that it is a type of petroleum jelly, but I don’t know if I would want to spend the time testing substitutes to make sure they didn’t leach anything into the aqueous phase. Vaseline, for example, doesn’t just contain petroleum jelly, but also has a variety of other compounds added to it. Vaseline has supposedly been used successfully for hot-start PCR, though, so maybe…

    • Aneesh on March 27, 2012 at 6:28 pm

      Hi Jode,
      Thanks for the article. Have you ever tried using the Phase Lock gel for recovering ligated plasmid DNA from ligation mix? I am trying to make large libraries and it requires high efficiency elctroporation. Currently I clean up my DNA after liagtion using Phenol Choloroform extraction followed by ethanol precipitation. I would like to try Phase lock gel for phenol-chloroform extraction, but I am not so sure there will be a carry over from the Phase lock gel or not? If yes will it affect the trasforamtion/electroporation efficiency?
      If you have any thoughts reagarding this issue let me know

  17. John on May 4, 2010 at 3:20 am

    Jode- What dyes did you use to highlight the two phases in your Phase Lock test run? It’d be interesting to see if we could fiddle with some gels in lab that might do the same thing more cheaply.

  18. Marco on May 3, 2010 at 1:41 pm

    Anyone can guess the gel composition?

Leave a Comment