Skip to content

Established in the mid 1970's, New England Biolabs, Inc. (NEB) is the industry leader in the discovery and production of enzymes for molecular biology applications and now offers the largest selection of recombinant and native enzymes for genomic research. NEB continues to expand its product offerings into areas related to PCR, gene expression, sample preparation for next generation sequencing, synthetic biology, glycobiology, epigenetics and RNA analysis. Additionally, NEB is focused on strengthening alliances that enable new technologies to reach key market sectors, including molecular diagnostics development. New England Biolabs is a privately held company, headquartered in Ipswich, MA, and has extensive worldwide distribution through a network of exclusive distributors, agents and seven subsidiaries located in Canada, China, France, Germany, Japan, Singapore and the UK. For more information about New England Biolabs visit neb.

Faster, Cooler DNA gels

Content brought to you by New England Biolabs

fast DNA gelsAll over the world, molecular biologists are tragically wasting hours of their life running DNA gels using tris-based conduction buffers like TBE or TAE.

These buffers are known to overheat at high voltages, causing problems with gel integrity, sample denaturation and more. Because of this, molecular biologists are forced to keep the voltage of their gels to a maximum of 5-10 volts/cm (e.g 100 volts for a 10 cm gel) and extend the running time, sometimes to hours.

Although long gel runs, like long restriction digests, are often used as a convenient coffee break opportunity they can also eat into the molecular biologist’s precious time, leading to longer and less efficient working days.

But, in 2004, a team of scientists from Johns Hopkins came up with solutions (pardon the pun) to this problem. They have developed and verified three conductive buffers that stay cool during electrophoresis, allowing the voltage to be racked up to a massive 35 volts/cm without any problem, reducing the time taken to run gels by up to 7 times.

Between them, the three buffers cover all of the molecular biologist’s DNA gel needs. The buffers are:

  • 10mM sodium boric acid (Na2B4O7/Borax)
    For standard applications (separation of DNA fragments from 100bp-5kbp).
  • Our lab’s standard DNA gel buffer has become 20 g of Borax into 1 l of water which gives a 20x stock solution. Small gels run in 10-15 minutes at 200V.

  • 5mM lithium acetate (LiOOCCH3, CAS:546-89-4)
    For separation of fragments longer than 3 kbp.
  • 1mM lithium boric acid (Li2B4O7, CAS:12007-60-2)
    For separating small DNA fragments and ssDNA

The sodium borate and lithium acetate buffers can also be used for RNA gels in place of MOPS buffer.

Simply make up the required buffer, use the same buffer in the gel and the tank, turn up the voltage to 10-35 volts/cm and watch that DNA go.

Despite this excellent work, there are still thousands of molecular biologists who have not yet been shown that there is an alternative to wasting their precious hours using tris buffered gels.

Help us to eliminate this tragedy by spreading the word about this method for faster, cooler DNA gels. You can do this by clicking the link below this article to e-mail it to your molecular biologist friends. We, and they, thank you for your help.

Further reading:
1. Brody and Kern(2004) Biotechniques 36 p214
2. Brody et al (2004) Biotechniques 37 p598 (Free registration required)
3. Hudson, Biocompare protocols (a protocol giving a quick overview)

Photo: Klis


  1. Carlo on February 3, 2012 at 8:26 pm

    Is making up a solution of 5mM lithium acetate all I have to do to run my gel? How about the pH?
    Some say to add EDTA to a final concentration of 1mM, does anyone know whether it make any difference?

  2. uhale on February 1, 2012 at 10:15 pm

    I’ve been using boric acid gels for several years, and love them, but just found out boric acid is now a suspected teratogen: may harm the unborn child. Also other potential toxicity concerns. I’m probably going back to TAE.

    • user-48565 on November 14, 2013 at 10:31 pm

      On the boric acid package there are similar warning.

  3. Evi S on June 29, 2011 at 2:04 pm

    Most applications in our group require the addition of MgCl2 in order to stabilize the DNA in specially folded structures.

    Does anyone of you think there might be disturbing effect by mixing Sodium Borate with Magnesiumchloride?

    Thanks in advance for your opinion!

  4. 741216 on March 13, 2011 at 2:35 am

    It is my time to try SB. My gel is 13 cm long. I tried 1.5% gel and 3% gel at 150V for 2 hours and 2% gel at 200V for 3 hours. My samples’ performance were even worse than on TBE and TAE gel. But it might be caused by samples. Let’s just say the ladders. The ladders I used were Bioline Hyperladder I and IV.

    1.5% gel: 100bp was at ~8cm; 1000 bp was at ~3 cm; >1500 bp were stuck together
    3% gel: 100 bp was at ~4cm; >500bp were stuck together
    2% gel: 100bp was ~11 cm; 1000 bp was at ~4 cm; >1500 bp were stuck together

    at 150V, I did not feel hot.
    at 200V, the gel was hot.

    My conclusion:
    SB is good for small size DNA, epically for DNA <500 bp, for two reasons: 1) time saving, 2) good resolution.
    for big size DNA: 1) time saving is not significant, 2) resolution is bad.

    • user-48565 on November 14, 2013 at 10:39 pm

      I’ve similar experiences. Nevertheless it’s works fine for PCR product control. 3% agarose gel electrophoresis with SB now takes about 27 min. to compltete at 300V :). The only problem is that orange G migrates very quick now – far faster than in TAE/TBE buffer – so when it comes to the end of ca. 10 cm gel 50bp band is barely 5 cm from the start. Bromophenol blue is a bit better.
      Any of you have any recommendations for replacement?

  5. rnaplus on February 10, 2011 at 4:29 am

    Can anyone comment on their experiences using lithium boric acid buffer to resolve short (<40 nts) DNA and RNA?  I routinely use TBE in a 20% PAGE gel and good results, but in hours.  I’m hoping to speed up the analysis by using the new buffer system.

  6. Misha Golynskiy on January 18, 2011 at 5:47 pm

    I guess this a bit of a follow up to Brian Cady’s question – I’ve used some old Na2B4O7 x10H2O that we had around the lab and that gives pH around 9.4… I’m assuming 20g borax/1L H2O gives pH that’s still a bit off. Do you just adjust pH with boric acid? the elevated anion concentration does not affect the gel properties?

  7. Matt G on July 15, 2010 at 3:44 pm

    My lab only runs the sodium borate gels now. We used to run them at 350V but burned up a power supply so now we only run them at 250V.

    I use the gels from 0.7% to 4.0% depending on the purpose. The 4% gels end up running about 20 minutes for good resolution and the buffer gets hot but I have never melted a gel.

    The best ladder we have found is the Fast DNA Ladder from NEB. All the others we tried would smear very badly but this one runs beautifully.

  8. James Taylor on July 14, 2010 at 8:21 am

    I’ve also used the Lithium acetate buffer to resolve topoisomerse of intact plasmids. You need to add 1 mM EDTA and pH it with LiOH. Unfortunatly high voltages seem to make the plasmids run strangely (maybe there adopting a new conformation?). The resolution at lower voltages is great though.

  9. Brian Cady on April 21, 2010 at 5:30 pm

    When adjusting the 20X SB stock solution pH, does one use solid boric acid? I tried preparing 0.5M boric acid solution, but its pH is only around 4, and the pH of the SB buffer has barely budged with 15 ml of 0.5M boric acid solution added to 500ml of 0.1M disodium tetraborate solution. This has a pH of about 9.4 originally, similar to what was noted above.

  10. Nick on July 6, 2009 at 2:53 pm

    I really like this buffer system very much, except for the following issues:

    It is REALLY sensitive to salts and glycerol, so if you’re doing the classic alkaline lysis miniprep and restriction digests, you will get substantial gel retardation problems unless you either dilute your sample or clean up a column, which in my opinion defeats the purpose of cost and time savings, not to mention that the ladders become somewhat useless. That being said, it is GREAT when analyzing PCR fragments as the resolution is very good and migrates well with the ladders. I use 5mM final concentration of Borax (20X stock is 100mM which is close to the limit of solubility). Lastly, because of the time savings, I’ve started to tinker with the boiling miniprep protocols as they have substantially less salt in them and it looks fairly promising to combine with the Borax buffers.

    I have not tried yet the LiOAc but will be doing that soon as most of what I have been analyzing is lower than the 3kb limit.

  11. max on December 11, 2008 at 1:58 pm

    Ivan Ivanov added the following by mail, when I asked him:

    I will try to explain in more details:
    The optimal final concentration of Na+ in the buffer should be around 10mM and lower, according to Brody et al. 2004.

    There are several different forms of borax on the market. The one that you use, seems to be the anhydrous (waterless) form with molecular weight 201.2 (and molar mass 201g/mol). That’s why if you dissolve 20.1g in 1L you will obtain 100mM solution of borax but 200mM of Na+, since the formula is Na2B2O7. From here I am sure you can do the math.

    There is one more thing… my experience is that borax is very difficult to dissolve (use boiling water) and gives pH around 9.5. That’s why I always add boric acid to pH 8.3.


  12. Nick on December 10, 2008 at 3:17 pm


    Glad you have had success with this! Thanks for pointing out the typo – I have fixed it now 🙂

    • McFuch on April 2, 2014 at 9:59 pm

      Isn’t 40 g/L correct? I believe borax is is the decahydrate version (Na2B4O7*10H2O) which would correctly make it approximately 10 mM Na+.

  13. Ivan Ivanov on December 10, 2008 at 3:08 pm

    Hi, guys
    I just wanted to express my excellent experience with these media. I have obtained very good resolution on 3% standard agarose gels (2 bp for microsatellites).
    There is a mistake on the above recipe for the SB !!! If prepared according to the above (40g/l borax for 20x) you will end up with 20mM Na+ final conc. (instead of 10mM) as 1 mol of borax (Na2B4O7) provide 2 mol of Na+.

    • McFuch on April 2, 2014 at 9:58 pm

      Isn’t 40 g/L correct? I believe borax is is the decahydrate version (Na2B4O7*10H2O) which would correctly make it approximately 10 mM Na+.

  14. max on May 14, 2008 at 8:42 am

    Thanks for the link Caroline! The FAQ says that if you use HIGH salt restriction digest buffers, you’d better dilute them a little before putting them onto a gel. As SBA/LB are lower salt than TBE/TAE, the salt difference samples/gel starts to play a role.

    It seems that I never used high salt buffers until now with the gels so I cannot say how strong the effect is.

    I guess it is a good idea in general, also for TBE, to dilute the ladder with the buffer of the samples you’re running and, in the case of LB/SBA, to dilute the samples with the running buffer a bit.

    Has anyone experienced this? Has anyone calculated which of the standard buffers (e.g. NEB 1-4) has a salt content > 50 mM ?

  15. Caroline on May 14, 2008 at 4:07 am

    I’ve thought about trying these buffers. Isn’t it the basis of the flash gel and invitrogen precast (EZgel?) systems? I read that it doesn’t work well with restriction digests, something about the salt concentration of the digests not being compatible. Has anyone tried this?

    Here’s a link to a FAQ from a company selling the buffers…


  16. maximilian on May 4, 2008 at 10:54 pm

    i’ve never had any problems with extracting a band from my gels and I don’t see why the buffer would make any difference for the gel extraction.

  17. Brandon on May 2, 2008 at 4:25 pm

    Does anyone know how well these gels work with the typical gel extraction protocols?

  18. Liam on April 15, 2008 at 7:35 am

    Ok, perhaps I should have suggested this first. We normally depending on the amount of DNA loading dye added to a well, start the gel at 100V, then proceed to higher voltages, this normally works well.

    Of course if you’re filling your gel wells to the top, things don’t run very well, but if you have bigger wells, and are able to add more loading dye to make it sink better, or can load a smaller volume, it generally gives better resolution. If nothing else, run this buffer at your normal voltages, it is cheaper than the normal buffers.

  19. Max on April 14, 2008 at 8:11 pm

    CK: Yes, a gel at 300V is worse than a 80V-TBE-gel, I agree. For nice gels, I’d slow them down, to 150V, for instance. SBA is still the better, cheaper generic buffer in my opinion, if run at the same voltage as TBE.

  20. CK on April 14, 2008 at 3:22 pm

    Tried this ~ 2 years ago and, whilst succesful at running gels without overheating in < 15 min, the bands were definitely more smeared and less sharp than a typical ‘slow’ TAE/TBE gel. In the original paper the bands look fine but I was unable to reproduce that at the higher voltages. In my experience: useful to check quickly if a PCR has worked but not something I’d put in a thesis/paper.

  21. bala on April 11, 2008 at 2:21 pm

    I tried running my gel with lithium acetate buffer at 300v but i observed the heating effect and it did not turn well!I’ve attached the picture below.Any suggestions

    20x Lithium Acetate: 6.6g lithium acetate, Made up to 1 L with water.


  22. Liam on April 11, 2008 at 8:19 am

    You can also prepare this buffer by making up 10mM NaOH and adjusting the pH to 8.5 with boric acid powder. We routinely run our gels at 150v, and sometimes at 300v although the gel chambers don’t like it very much. The drop well systems seem to handle the higher voltages much better. Perhaps even run it in the cold room. I know in one part of the papers above, they resolved nucleic acid at 1000v in 5 minutes.

  23. Max on April 9, 2008 at 8:53 pm

    I want to add that the new buffers (which really are old, but just haven’t been used for agarose gels before) also give you cleaner bands. In addition, SBA is cheaper than TBE/TAE.

  24. Kyle on March 3, 2009 at 11:34 am

    Some of the links say this buffer should be used for 100bp – 5kbp resolution. I will experiment with this, but I wanted to know if anyone regularly uses this buffer for larger bands? It seems like running the gel longer and a little cooler (more like tradtional TBE gels) should allow for this, but I’m pretty ignorant about electrophoresis buffers.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll To Top
Share via
Copy link