Don’t throw leftover silica DNA purification columns/buffers away! Use them up! (DIY Silica Spin Kits)

Published: June 16th, 2017   Last Modified: November 23rd, 2019

Silica based nucleic acid purification columns and kits are a common consumable in the lab. They’re used for cleaning up PCR reactions, purifying gel slices and plasmid DNA minipreps. Inevitably, when you reach the bottom of the box you are usually left with one of two scenarios: 1) You don’t have enough of a specific buffer or 2) You don’t have enough columns. Both of these scenarios usually result in the remaining components of the kit being tossed out or forgotten in your cupboard for a decade (that somehow make their way into my hands).

Why not source your own columns, mix your own buffers to both reduce waste and save money! Grab a snack, ease into your bunny slippers, this is a long one.

There are two major technologies when it comes to purifying DNA column-style; silica based purification and anion-exchange resin purification. In this post I’ll be talking about silica kits, but the anion exchange post is coming too.

Silica Based Kits


Silica purification is by far the most common technology due to the low cost of materials, and is generally found in kits for smaller scale purifications, such as miniprep kits, PCR cleanup kits and gel purification kits. Silica will bind DNA in the presence of a high concentration of chaotropic salts such as guanadinium hydrochloride (Gu-HCl) or sodium iodide (NaI). The silica is then washed a few times with a high concentration of ethanol and finally eluted in a low volume of pH 8-8.5 Tris.

So, what if you run out of columns for a particular kit? Well, as long as it’s based on silica/chaotropic salt technology, you can use them interchangeably. One caveat here is that different manufacturers deposit differing amounts of silica into their columns, which translates to a higher or lower binding capacity of DNA. Also, silica columns can dry out, significantly impacting binding capacity. Dried up columns can be re-humidified and reused.

Don’t have old silica columns in a drawer somewhere? You can buy the silica spin columns by themselves much cheaper than you could get from Qiagen/GE/etc. Enzymax is the cheapest at 0.38$ per column, with Epoch Life Sciences at 0.64$ per piece. Other manufacturers do exist but I have had good experiences with both companies.


Now, what if you have the columns, but are missing a specific buffer? Don’t be scared! Since these silica kits all use the same underlying technology, the buffers themselves are interchangeable and vary little in their composition. Openwetware has the compositions of the majority of Qiagen buffers, taken straight from their manuals and patents. I’ve also scoured patents from other manufacturers for their recipes so that you can see how similar they are. The trick here is to use the appropriate buffer for your specific application.

For silica kits you will generally use three types of buffers: binding buffers that allow your DNA to bind to the silica, wash buffers which clean away unbound DNA/RNA/protein/salt, and elution buffers which dissolve your DNA that you can then use.

Miniprep Buffers:

Qiagen Buffer P1: 50 mM Tris-HCl pH 8.0, 10 mM EDTA, 50 ug/mL RNase A

Qiagen Buffer P2: 200 mM NaOH, 1% SDS

Binding buffer for PCR product/enzymatic reaction cleanup:

Qiagen Buffer PB :5 M Gu-HCl, 30% isopropanol, add 5 volumes to 1 volume of PCR/enzymatic reaction, load onto column.

Qiagen Buffer PB (Enzymax variant): 5 M Gu-HCl,  20 mM Tris-HCl pH 6.6, 30% ethanol

The isopropanol/ethanol is likely added here to prevent smaller DNA fragments (i.e. primers) from binding to the membrane. You could likely adjust the concentration of isopropanol to exclude binding of larger pieces. Machery Nagel does an excellent job demonstrating this in their PCR cleanup manual, although they dilute their PCR binding buffer with water.

Binding buffer for agarose gel slice purification:

Qiagen Buffer QG: (5.5 M guanidine thiocyanate, 20 mM Tris HCl pH 6.6) Add 300 uL of buffer to 100mg of gel slice, heat to solublize, load onto column. Enzymax suggests same recipe for their columns.

Promega Gel Buffer: (6M guanidine thiocyanate, 10 mM EDTA) Recipe from their patent. EDTA used to prevent discoloration of guanidine. Use ~333 uL of buffer per 100 mg of gel slice, heat to melt, load onto column.

Binding/neutralization buffers for plasmid minipreps:

Qiagen Buffer N3: 4.2 M Gu-HCl, 0.9 M potassium acetate pH 4.8. This is an interesting buffer which was gleaned from the patent, Qiagen calls it proprietary. Essentially it is Solution 3 from your classic alkaline lysis miniprep with Gu-HCl (1.73M final conc.) thrown in to get the plasmid to bind to the silica. Resuspend cells in 250 uL Buffer P1, lyse with 250 uL Buffer P2, neutralize with 350 uL of Buffer N3, load unto column.

Qiagen Buffer N3 (Enzymax Variant): 4M Gu-HCl, 0.5M Potassium Acetate pH 4.2


GE Neutralization buffer (4.4M Gu-HCl, 0.65M potassium Acetate
and 3.1M Glacial Acetic Acid). Got this from a GE healthcare patent, slightly higher Gu-HCl (2.2M final conc.) content. The acetate/acetic acid portion is very close too, you’d be adding that much acetic acid to potassium acetate to make it pH 4.8 as in the Qiagen buffer. GE uses 350 uL of this buffer for 175 uL of lysis buffer (200mM NaOH, 1%SDS) + 175 uL resuspension buffer (100 mM Tris-Cl pH7.5; 10 mM EDTA; 400ug/ml RNase A). Slightly different resuspension buffer, but identical lysis buffer.

Wash buffers for PCR/Enzymatic cleanup, gel purification/plasmid preps:

Qiagen Buffer PB: 5 M Gu-HCl, 30% isopropanol. This is an optional wash buffer used to remove residual nucleases/carbohydrates that may be stuck to the silica, recommended when isolating plasmids from endA+ strains. Wash once with 500 uL. Proceed with regular washing steps.

Qiagen Buffer PE: (10 mM Tris-HCl pH 7.5, 80% ethanol) Wash twice with 750 uL.

5X Qiagen Buffer PE (Enzymax Variant): (80 mM NaCl, 8mM Tris-HCl pH 7.5) Dilute to 1X with 100% ethanol, 1X conc: 16 mM NaCl, 1.6 mM Tris-HCl pH 7.5, 80% EtOH.

GE Wash buffer: (2mM Tris-Cl pH8; 0.2mM EDTA and 80% ethanol) Very similar to Qiagen, would work interchangeably. GE uses 400 uL for two washes, but 750 uL like Qiagen is probably better.

Promega Wash Buffer: 80% Isopropanol

Elution buffers for PCR/Enzymatic cleanup, gel purification/plasmid preps:

Qiagen Buffer EB: (10 mM Tris·Cl, pH 8.5)

Qiagen Buffer EB (Enzymax Variant): (10 mM Tris-HCl, pH 8.0)

Machery Nagel AE Buffer: (10 mM Tris·Cl, pH 8.5)

GE Elution Buffer: (10 mM Tris-HCl, pH 8.0)

Any high pH (8-8.5) buffer will work, as will water at a lower efficiency. The presence of EDTA can also lower recovery. The amount of buffer used to elute will dictate recovery, more buffer will recover more DNA but at a lower concentration.

So, those are the buffers you need to use any brand of silica spin kit. As you can see most of the magic is in the binding buffers, which when applied to your sample have a final Gu-HCl/thiocyanate concentration of ~1.5-4M. As long as your sample has that concentration, the DNA will bind to the silica. Wash buffers are 80% ethanol/isopropanol, with Tris/EDTA added for flavour. Elution buffers are even simpler, with basically all manufacturers disclosing it to be Tris pH 8-8.5.

48 thoughts on “Don’t throw leftover silica DNA purification columns/buffers away! Use them up! (DIY Silica Spin Kits)”

  1. You’re so right about these kits and solutions… It’s almost they “setup” the kit to make one component to be short of another… A gel pcr purification kit is directly related to the Plasmid prep kits… And like you pointed out…if we really spent the time studying the solutions comprising the kit…gone would be those dreaded days of running out of sometimes only “one” solution requiring us to buy a new kit. This happened often when I used a Machery Nagel Maxi prep kit… And due to the large reaction volumes always led to some overuse of some solutions… Especially the annoying resuspension buffer which is essentially TE buffer plus RNase A… Bought a cheap vial of RNase A and made some TE and saved 200 bucks lol

    1. Yeah buddy, you would regenerate them akin to anion exchange columns where you run 1M HCl through them several times, let them soak in HCl overnight, then rinse with distilled water and run some guanidinium buffer to get the silica ready to bind again.

      1. Thanks. Should have looked into your site better before asking…You had the information on another post. Great site and good detective work.

        Two more questions and excuse my ignorance. You show two binding buffers, one for plasmids and one for PCR products. The third one “minipreps” is for what? And related to this, can one of the above be used for genomic DNA from bacteria.

        1. The miniprep buffer is for extracting plasmid DNA from E.coli, agrobacterium, yeast. It essentially combines the classic Buffer 3 of a plasmid prep, which contains acetic acid to neutralize Buffer 2, as well as guanidinium to get that plasmid DNA to bind to the silica.

          Interesting question about the genomic DNA, hadn’t thought about it as I do genomic extraction pretty infrequently with the CTAB/phenol based method. Spin kits are also used for genomic DNA extraction, the magic comes in the initial lysis buffer which releases the genomic DNA, it involves lysozyme and proteinase K. After the genomic DNA has been liberated then it’s the usual fare of adding a guanidinium buffer (5-6M) to the lysate, and put it through the spin kit as usual. Just some quick reading from the Qaigen DNAeasy kits suggests that ethanol is added at this step too, putting the composition close to the PCR purification buffer listed. Give me a bit of time to dig you up a solid recipe for a genomic DNA spin kit lysis and binding buffer.

          1. I also use the CTAB/phenol and would be nice to be able to clean any residual phenol an salts this way. Thanks!

  2. Do you have any suggestions for a binding buffer for genomic DNA purification on silica spin columns, or ny binding buffers that don’t contain guanidine?

    1. Genomic DNA purification by spin kit would follow the same basic steps as the other methods, the magic being in the way that you initially lyse your cells. I’m trying to find a decent recipe for you and another poster, but from my reading of the DNeasy kit the composition is close to the PCR purification buffer but with ethanol vs iso.

      If for some reason you couldn’t use guanidine then Sodium Iodide would make a reasonable substitute for it, check out this paper for DIY NaI based genomic DNA extraction

      If you can’t use chaotropes like GuHCl or NaI, then the DEAE resin kits would be the way to go, which purify DNA using salt gradients.

      But yeah, the recipe is in the works, I’ll post it as soon as I can, been a bit busy submitting a paper, but that’s done now, yay! 🙂

        1. Ya bro, should be interchangeable. Just make sure you’re not following Qiagen’s mini tip (?) protocol, those use the anion exchange technology, but that’s the only caveat I know.

          We use the same columns from Enzymax for gel extraction, PCR cleanup and miniprep interchangeably.

    1. Unfortunately Qiagen has never released the recipe for RW1 either in their manuals or patents for whatever reason. The best I can offer you is a look at one of their patents (US 8,624,020 B2), where they looked for alternatives to ethanol in their buffers so they could avoid restrictions on shipping ethanol.

      Their RW1 replacement is: 20% Tetraethylene glycol (or 20% 1,3 butanediol), 900 mM guanidine isothiocyanate and 10 mM Tris pH 7.5, Buffer RPE: 70% Tetraethylene glycol (or 80% 1,3 butanediol), 100 mM NaCl, 10 mM Tris pH 7.5

      Allegedly this washing combo worked better than the ethanol version. I don’t have the RNAeasy kits to try the recipe out for myself, but depending on how bad you want more RW1 (vs buying a bottle), you could try the ethanol free version, or substitute 30-50% ethanol for the tetraethylene glycol in RW1

  3. Thank you for this information, I stumbled upon it a few days ago and I honestly have to say it was enlightening to the processes I have been using.
    Anyways, I had a question about the Agarose binding buffer recipes you had there. So I don’t have the guanidine thiocynate but would guanidine HCl be a resonable substitute? Also, since I already gave it a shot, I tried using 4M Guanidine HCl in 20mM Tris 6.6 pH and it took forever for the gel pieces to dissolve(almost two days at 70C). Could my concentration of Guanidine HCl be too low ( I did see it was 6M there but I remember seeing 4M somewhere else) or should there something I may need to do to differently? Just looking to troubleshoot my problem. Again thank you for your information and thank you for any help.

    1. Hey Buddy, sorry you’re having trouble.

      The Qiagen patent does mention using Guanidine-HCL, however they don’t actually give a recipe of those buffers, just the thiocyanate which I guess they settled on. It’s possible the Gu-HCL will work, but perhaps at a lower efficiency of binding to the silica. I’d bump the concentration up to 6M, but I’m not sure if that will help with your dissolution problem. You mention you apply heat, but what type of agarose do you use? We use low-melt grade, which is quite fragile but melts very easily. I don’t think I ever tried a gel extraction from regular agarose.

      1. Thanks for the reply. It is the melting temperature. I thought the agarose was low melting but I was wrong, I have to bump my temp up to 88 to break it down. I will also bump the Gu-HCl concentration up and see what happens. Thanks again.

  4. Hey there. Useful stuff you have around here, it’s great.
    I have some questions. I’ve been using the NucleoSpin by Macherey Nagel columns for RNA isolation. The kit consists in silica-membrane columns and the following steps: cell lysis, RNA binding, DNAse treatment, RNA washing and elution. Unfortunately, they were not very useful for the isolation (we obtained a ridiculously low quantity of RNA, but that’s another story), but we decided to use the columns for the DNAse treatment and washing after a phenol-cloroform extraction.
    Can these columns be reused, if we intend to use them for DNAse treatment only? How can they be washed?
    Also, you’ve inspired me to search the components of the DNAse buffer from these guys. I don’t live in the US so this stuff takes forever to arrive down here. No way I’m going to wait 2 months for a 7 mL solution.

    1. Hello, I’m glad you find some of the stuff useful! 🙂

      Yeah, RNA extraction columns are pretty notoriously low yielding, we found that too. We have regenerated Qiagen anion exchange (DEAE-silica) columns (mostly maxi preps) with the following method about 10 times per column. Basically let it soak in 1M HCl for a while, wash it plenty with H2O and re-equilibriate with buffer QBT. I haven’t applied this to RNA silica/guanidinium columns, but I would guess that the method would work the same, except you would equilibriate with whatever the binding buffer for the RNA is. Haven’t dug too deep into RNA kit patents but I believe they are out there.

  5. i ran out of AW1 buffer of qiagen ffpe tissue dna extraction kit. can AW1 buffer of other qiagen dna extraction be used?

    1. Yup, should be cross compatible, assuming the AW1 from the other kit wasn’t prepared too long ago, the ethanol can evaporate…

  6. This is a really cool site. I’m glad the mega biotech corporations don’t quite have all the power, haha.

    I have just a quick question that could use your expertise: I do quite a few PCR purifications and recently our lab transferred from using machery-nagel kits to qiagen kits. The binding buffer in macherey-nagel’s (Buffer NTI) has a feature where by diluting it with h2o, smaller fragments can be selected for based on size, and washed out.
    Qiagen’s binding buffer (PB Buffer) which I now know is using gu-hcl, dosent seem to have this feature.
    I have an inkling macherey-nagel may use guanidium thiocyanate in their NTI buffer rather than gu-hcl as well.

    I would love to see some more information on how macherey-nagel’s NTI buffer can do this. I can’t seem to find their recipe anywhere.

    1. Hey! Yeah, they don’t have all the power 🙂

      As for Buffer NTI, I can’t seem to find any patent information regarding the formulation and you’re right, it is thiocyanate based. However with Qiagen’s PB buffer the dilution trick should work in a similar way, you are essentially just adjusting the binding conditions and the more concentrated the binding buffer is the tighter the binding to the silica will be. So, a higher ratio of PB will mean smaller pieces will bind. Unlike MN, Qiagen doesn’t describe this in their manual but I’m confident it would work the same way, you’ll just have to do a gradient to figure it out, use MN’s manual as a guide.

      I’ll let you know if I find any literature as to the composition of NTI, this is the best I could find:

      Good luck! 🙂

  7. Hello, we recently moved the Lab and the movers decided to throw away all solutions for GenJet Maxiprep. Now I have ~600$ worth of columns without any solutions. Can I use the same you suggested for the mini preps, scaled up? GenJet is also silica-based.

    1. Hello! Yes, assuming that they’re using silica/guanidinium binding technology, the same buffers as the miniprep kit would work, looks like the older msds’s say that the neutralization solution do indeed contain guanidinium, which would confirm it. Try it out, if it doesn’t work then potentially it’s a DEAE column, but likely it’s a big old silica column.

    1. Mr Mao Zhang is the sales representator of Aidlab, China FYI 🙂

      I have used Aidlab columns before, 0.15 cents/column and some shipping fee, you can get a bag filled with 1000 columns (or two bags, one for columns and one for collection tubes) that work around 60-70% efficient compared to a commercial column (with homemade Qiagen Buffers – my comparison is to MN NucleoSpin). They also sell other stuff in bulk, like SYBR 😉

      Here kits cost 2.5X the price compared to US so cutting corners is a must…

      1. Oh man thank you so much for saying what the comparative efficiency is. I’ve been researching the different efficiencies of columns for weeks with almost no results.

    2. Thanks for the tip! Yeah, I’ve been looking at consumables suppliers in China for some time, mostly off alibaba/aliexpress, but never actually tried. I have had success ordering custom endmills from there though.

      I would wager most of our disposable labware are manufactored by the same companies with different advertising. Next time we need to order I’ll give them a try, I like trying out different suppliers.

  8. I need the recipe for buffer PN (to which we add isopropanol to make PNI), PE, PB and EB – how do I calculate the volumes of the components listed to make up a standard solution?

  9. I’ve got a question, do you need two washes? would it be possible to combine the PE wash and the PB wash steps together? As in: creating one super wash step with Gu-HCl, Tris-HCl and ethanol? I’ve been searching for a while and can’t find an answer.

    1. At the end of the day you need the wash out the Gu-HCl so it doesn’t inhibit downstream reactions. So, I suppose you could do one wash of the combined buffer and one wash of just PE, vs 1 wash with the combined buffer and two PE washes. However just washing with the combined buffer will likely leave quite a bit of Gu-HCl when you elute.

  10. This is good stuff. Thanks for setting up this site.
    I’ve been using the Qiagen Plasmid Plus kits for higher yields. It uses (thicker) silica columns and gives about 10-fold increase in pDNA yield for the midi kit and much higher for the Maxi and Giga. The kit contains buffers S3 and BB. Any idea what they are? The Safety Data sheet for S3 contains acetic acid, pH 5.5. The Safety Data sheet for BB lists CTAB, but not its concentration. Any thoughts?

  11. my plasmid stocks in the column. It is about 16kb and I do every step correctly. sometimes it is eluted and sometimes not. I preheat the elution buffer. what else can I do?

    1. I feel you on the large plasmid troubles, my viral plasmid is about the same. What I do is heat up my buffer, elute twice with 50 uL of elution buffer then concentrate down in a Centrivap concentrator.

  12. Cool article! I really enjoying your blog.
    Here is a relatively recent study which maybe of interest regarding this topic:
    They use both silica column regeneration and even more simplified DNA PCR/gel purification buffers.

    Their previous protocol on column regeneration using phosphoric acid instead of 1M HCl (which is more just slightly more nasty in handling and smelling =) ) was already mentioned somewhere else here in the blog:


    1. Thank you for the compliment, glad you like it cause’ I like doing it 🙂

      I read the paper, reminds me a bit of a method I used for quite a bit:

      There they made their own silica binding “milk” and Sodium Iodide as their chaotrope.

      The addition of the sulfite as a preservative is a good idea, I wondered how to stop the NaI from slowly going a pale straw yellow…

      Stopped using that when we realized we could buy silica columns cheaply from Enzymax. I’m getting in touch with some Chinese suppliers and I’m getting quotes of 15-30 cents per column at 1000 quantity.

  13. Has anyone had problems cleaning Qiagen PCR purification columns by soaking in 1M HCl? The Siddappa (2007) paper says they can be left ‘indefinitely’ but I left some columns soaking for a few months and I’m getting much lower DNA yields with them. I’ve been cleaning the same PCR rxn in parallel and I’m getting 30-50ng/ul with new columns but around 1-2ng/ul using the cleaned. I’ve been removing them from the HCl spinning them out, filling and spinning out with ddH2O 2X, and then rinsing with PB buffer.

    Also, I put an NEB Monarch plasmid prep column into 1M HCl to soak yesterday and by today most of the silica had fallen out so fyi columns without frits may not be able to handle soaking.

  14. Hi, I came across this page after sorting through boxes of old DNA and PCR cleanup kits in my lab and googling whether they’d still be usable. All are expired and some will have been sitting around for up to 10+ years so I’m interested in your section about re-humidifying old columns. Silly question, could I just spin some water through them instead of following the humidifying protocol or is not that simple?


  15. Hmm, you could likely just extend the initial incubation period of your lysate/PCR product/solublized gel with the silica from 2 minutes to whatever, have a coffee. Water could work, although I haven’t personally tried that, and I believe Sigma suggested in an old manual something like 5M Gu-HCl, which could get expensive.

    The patent mentions keeping your old columns in a space with a few 72% humidity packets, but likely a sealed box with some wet paper towel would rehydrate them in a few days if you’re not in a rush.

  16. I like your low-tech option of paper towels in a box and have set some up now. I’ll leave it a few days then test a couple of purifications with and without humidifying and see how they compare. Will report back with my findings.


  17. What about ATL and AL buffer, any guess? For total nucleic acids extraction from virus/bacteria we are using ATL plus protease K, then AL, wonder what their composition would be? thanks!

  18. Hiya,
    I’ve been doing some tests and it appears that good old solution 3 without guanidine may just work, if you dilute the whole lysate with isopropanol 1X before loading onto the column (taken from miraprep protocol). I suspect the isopropanol makes it just chaotropic enough for DNA to bind the column…
    Something rather unrelated, there are some new silica-based PCR (DNA) purification kits out there that has a size-selective binding buffer and allows you to select for higher mw DNA (e.g. Thermo Purelink or Promega ProNex). My suspicions is this is probablya play between Gu-HCl and/or isopropanol concentration, wondering whether you have any insight on this? Thanks!

Leave a Reply

Your email address will not be published. Required fields are marked *