Published: June 21st, 2017 Last Modified: January 24th, 2019
Some time ago I wrote a post for Bitesizebio, a fellow biology themed blog, regarding the home brew version of phase lock gel. Surprisingly quite a few people found the recipe useful, and the interest has been pretty steady since then. One limitation of using vacuum grease (DOW Corning High Vacuum Grease) as phase lock is that it is equivalent to the “light” density phase lock. Light phase lock works for many day to day separations like phenol and phenol/chloroform extractions of minipreped DNA. However, for applications involving phenol + a high density salt solution, light phase lock does not work. Essentially, the aqueous phase which is supposed to be on top of your phase lock is now below it, meaning you have to poke through a layer of grease to get at your aqueous sample. The commercial manual explicitly states that light phase lock is not compatible with salty, dense solutions, and consequently they also sell a heavy version. Can we recreate the heavy version by manipulating simple vacuum grease? I think we can! Get your gloves on though, this is going to get messy.
UPDATE: Testing of 15% SiO2 grease here.
Straight from the horses mouth, let’s see what the recommended applications for phase lock gel is:
“PLG Heavy
Aqueous phase: High density sample (for example, high salt content).
Organic phase: Standard mixtures of phenol, chloroform, isoamyl alcohol with a chloroform fraction of at least 60% (Phenol as the sole organic solvent is not compatible with PLG Heavy).
PLG Light
Aqueous phase: Low-density sample
Organic phase: Standard mixtures of phenol, chloroform, isoamyl alcohol.”
Pretty vague, however they include a little more detail in a table, which gives some information on what constitutes “low” density solutions (<0.5M NaCl, <1mg/mL BSA).
So, how do we get ourselves some heavy phase lock gel without selling our first-born child? The vacuum grease we use needs to be made denser. To recap the previous article, phase lock is nothing more than polydimethylsiloxane (PDMS, gooey stuff) and silicon dioxide (ground glass). We can adjust the density of the vacuum grease by playing with the ratio of PDMS to SiO2. Since PDMS is less dense than SiO2, to increase the density we must ADD SiO2 to the vacuum grease.
First, I needed to get some SiO2. Luckily, I had plenty from doing the homebrew silica DNA extractions. Here’s a link to purchase some of your own (particle size 0.5-10 μm, approx. 80% between 1-5 μm), a 100g bottle is inexpensive and will last some time. Finer would be ideal, as larger particles will have a tendency to migrate during the centrifugation. I resuspended 10 grams of the Sigma SiO2 in 50 mL of water and let it settle for 2 hours. I transferred the supernatent to a fresh tube and centrifuged it, effectively size fractionating my SiO2. After drying, the next question was how much SiO2 to add to the vacuum grease.
Considering SiO2 is ~2.5 times denser than PDMS, I figured a little dab will do ya. The two variables I ended up testing were SiO2 concentration and salt concentration. I tested 0%, 2%, 5% and 10% SiO2 added to the grease by weight against 0M, 0.5M, 1M, 2M, 3M and 4M NaCl. In each tube I had 200 uL of salt solution, 200 uL of buffer saturated phenol and ~100 uL of vacuum grease. I took these tubes and centrifuged them at 16000xg for 15 minutes. The results are quite interesting, please excuse the poorly lit photos, I will update these as I explore this further.
As far as quick and dirty mock ups go, this one went surprisingly well. The yellowish solution is the phenol, the clear, barely visible liquid is the salt solution and the greasy stuff is the modified grease. A successful result we can define as aqueous on top, followed by grease, and phenol at the bottom. Unsurprisingly, the unmodified (0% SiO2) grease isn’t even able to handle a moderately dense solution of 0.5M NaCl, which explains the problems that people have when using the grease with trizol/genomic DNA preps.
With 2% SiO2, we start seeing an improvement as the grease can handle up to 0.5M NaCl. 5% is even better, it can handle up to 1M NaCl. Slight flaw in methodology here, at 2M NaCl the aqeuous is actually denser than the phenol, making most of this gradient less useful, whoops!
10% SiO2 was too dense for 100% phenol and sank straight to the bottom, which is supported by the commercial supplier who states that the heavy version is not compatible with straight phenol, has to be phenol/chloroform or a denser organic phase. One useful bit here is that at 10% SiO2 the grease is sinking below 2M NaCl. One reader mentioned that their CTAB genomic DNA extraction buffer had 1.6M NaCl, so this would probably be the extreme limit for the SiO2.
So there you have it, you can home brew your own heavy phase lock gel. If you want to separate dense solutions, it looks like the sweet spot is somewhere between 5-10% SiO2 added per weight of grease (e.g. 5g grease + 0.5g SiO2 = 10%).
Let me say plainly though, this is a messy, messy process. The SiO2 gets everywhere, and the grease gets everywhere. I mixed the grease in 50 mL tubes with a spatula as best I could and then shoveled it into syringes for use. The better you mix it, the better your phase lock will separate. Ideally you would have some sort of stainless rollers that dye makers use. I will investigate this recipe further and put any updates on this page as I get them.
– Don’t try to warm the grease up to make it liquid in hopes of mixing it better, it’s a solid gel until it starts to burn at 200 degrees C.
– The least messy way of handling phase lock gel is to squirt straight from the dow corning packaging into a large syringe and dispense from there.
Reader Submitted Tests:
LETs Genomic DNA extraction buffer:
20mM EDTA
0.5% SDS
10mM Tris-HCl
0.1M LiCl
Buffer above tested with unmodified DOW vacuum grease, left tube contains LETs buffer, grease and phenol + xylene cyanol, right tube contains LETs buffer, grease and PCI (25,24,1) + xylene cyanol. Perfect separation with buffer alone, possible modification of grease needed with high gDNA content.
E. coli gDNA Buffer:
2 mg/mL Proteinase K
1% SDS
Unmodified vacuum grease vs E.coli gDNA buffer. Left is Buffer + PCI + vacuum grease + xylene cyanol, right is water + PCI + grease + xylene cynol. Perfect separation with buffer alone, possible modification of grease needed with high gDNA content. Nice demonstration of phase locks ability to physically separate proteins from aqueous phase (light blue goo).
It would be great if people could send you their lysis buffers and you could try out different combos of buffer/phenol:phenol chloroform! I used your previous guide to start using the dow-corning grease for post-precipitation DNA cleanups, but would love to have a solid recipe down for the heavy version.
I would love that, actually! Helping readers gives me actual incentive to expand and verify these homebrew recipes. The heavy phase lock guide was motivated by a nice dude from France messaging me 🙂
Next RNA extraction I do I will figure out what SiO2/Grease ratio is best for RNAzol and Trizol, that will cover most of my personal use cases. I know that buffer/phenol is also used for genomic DNA extractions, however I do that pretty infrequently so I’ve never considered making a dedicated phase lock gel for that purpose.
But yeah, if you have a buffer you’d like me to try I’d be happy to give it a go.
Dear PipetteJockey,
We use a so called “LETS buffer” for genomic DNA extraction and I’d like to find out if it’s compatible with dow corning high vacuum grease (AKA PLG light version?)
The buffer consists of:
20mM EDTA
0.5% SDS
10mM Tris-HCl
0.1M LiCl
Thanks a lot!!
Hey Buddy,
Just tested out your buffer with straight DOW vacuum grease and both phenol or PCI, works just fine, separates like you’d expect. I’ll post the picture in the body of the post so you can have some proof.
Thanks for the update on this. Our lab has been wondering how to get around this issue. Do you have any advice on the particle size of SiO2 to purchase though? Looks like there is quite a range in particle size and a strong relationship between particle size and cost.
Hey buddy,
For making up the dense version I used this silicon dioxide, I’ll add it to the post: http://www.sigmaaldrich.com/catalog/product/sigald/s5631?lang=en®ion=CA, which is “0.5-10 μm (approx. 80% between 1-5 μm)”. The 100g portion is more than enough, and you can use it for minipreps too if you wish.
This stuff is tempting too: http://www.sigmaaldrich.com/catalog/product/aldrich/637238?lang=en®ion=CA, bit more expensive but it would make buckets of phase lock gel, much finer too (10-20nm).
Dear PipetteJockey,
What would you think about about this rather standard lysis buffer, in our case for E.coli gDNA, with 2 mg/ml proteinase K and 1% SDS? The table would suggest that we probably
need the heavy gel, would you agree or have you tested a similar buffer already?
Hey Buddy,
I’ll give your buffer a go, perhaps it will be okay with just regular phase lock.
Hello again, tried your buffer (pictures in post) and it separates fine with regular DOW vacuum grease. However it made me realize I’m not accurately mimicking your conditions, as I’m not actually lysing E.coli, and I’m thinking the gDNA may impact the density of the aqueous phase. Worst case, you’d have to make a mix of 5g grease + 0.5g SiO2.
Very cool, thanks! I guess will go for the mixture. Unfortunately, Sigma doesn’t sell the SiO2 you used, and even refuses to show me the exact product specifications. (“not available in your region”). Have you tested the nanopowder?
That sucks. I haven’t tried the nanopowder since the coarser stuff works reasonably well. I would guess that the nano stuff would actually be better, as the finer particles would disperse better in the grease and wouldn’t migrate during long centrifugations. On the downside, it’s possible that nanoparticle SiO2 isn’t optimal for plasmid DNA isolation, so it wouldn’t be dual purpose stuff.
Could you paste some of the technical specs from sigma at your location? I can try to find the most similar available here.
Here, I uploaded the spec sheet: http://pipettejockey.com/wp-content/uploads/2018/05/S5631-BULK________SIGALD_____.pdf
Hi Pipette Jockey!
Thanks for your article. Just a quick question: Did you ever manage to optimise the recipe for Qiazol/Trizol extractions? I’d love to be able to use this to get both RNA and gDNA from Qiazol/Trizol preps!
Thanks in advance!
Hello! Not a problem, happy to help 🙂
Within a week or two I’ll be doing RNA extractions on plants and will be optimizing the gel (15% added SiO2) for that purpose, as soon as I find out whether it works I’ll update the post. The most recent update is that the best way to make up “dense” phase lock is to squirt the required amount into a heavy duty ziplock bag, add the SiO2 powder, close it up, double bag it and use your hands to smoosh the ingredients together. Then, snip a corner and dispense the goo into a big syringe for further portioning.
Hi Pippette Jockey,
I want to do this and go one step further. I want to extract RNA with trireagent, then add Back extraction buffer and extract DNA using the phase lock tubes. I’m about to try it in a sample of tubes I got, but would love to know if you got the trireagent to work?
Thanks,
Tash
Actually, yes! I did get it to work, check out the follow up article at https://pipettejockey.com/2018/11/13/dense-phase-separating-gel-homemade-trizol-combo
Good luck, lemme know how it goes.
Hi! I just discovered your blog, and oh my, was I missing out! Thank you for the brilliant and illuminating articles you write to make our life so much better. There’s definitely less R rated cursing in labs, thanks to you!
Question: When you are preparing your SiO2, after you collect the 2 hour supernatant and centrifuge it to pellet the finer particles, you said to dry the it. How dry does it have to be? Do you just let it air dry?
So excited to try the homemade heavy PLG!
I’m happy you find some things useful here! It should be fairly dry before adding it to the grease, you can let it air dry or put it in a small dish and put it on low on a hot plate. In a pinch straight SiO2 from the bottle will work without getting super fine particles, you’ll just find that the larger particles will separate out after prolonged centrifugation. Doesn’t affect the separation, just looks funny.
Hi! Thank you so much for this recipe! We were using 15ml tubes for large-scale genomic DNA extractions from animal tissues. Did you try your mix in 15 ml tubes to see if this will hold?
When we used Eppendorfs 50 ml PLG tubes the gel would often slide on the side of the tube, especially when the centrifuge would warm up. So we switched to 15 ml, but now there is no place to buy these!
Hey! Glad you liked it! I haven’t tried it specifically in 15 mL tubes, put the 15% SiO2 PSG works well in 50 mL tubes, you can see the results in the following post: https://pipettejockey.com/2018/11/13/dense-phase-separating-gel-homemade-trizol-combo/
The caveat here is that you want to test how much of the grease you need for a 50 mL tube, too little and you can get an unstable layer. Considering you can make it nice and cheaply (albeit messily) there’s room to play a bit.
So for the smaller diameter 15 mL tube I believe it will work even better.
The cheapest source of SiO2, by far, is a chemistry lab. I am an organic chemist (here only because I work with biomolecules) and we use tons of silica gel for column chromatography of small molecules. The usual mesh is 60 µm but you can easily find smaller. Ours is from Macherey Nagel but many suppliers exist of course (just try “silica gel” on fisher). Alternatively, if you know a synthetic chemist (lucky you), just take 1kg out of the lab stock, I guarantee it won’t make a visible difference!
Thank you my chemically inclined brethren for the tip! Making friends with chemists is always a good idea, especially when you need something a little spicier like a liter of carbon tetrachloride in the dead of night.
Also, you know you’re in a cool field when taking a kilo of miscellaneous white powder from your neighbor doesn’t raise eyebrows 🙂
Most biochemists who come to our lab are looking for either solvents (we have thousands of liters of these) or or a rotavap to make lipid films (we have, well, not thousands but dozens). The good thing is that in exchange for a few mL of solvent or half an hour of rotary evaporation, I get to use cool stuff like microscopes or good quality centrifuges. For carbon tetrachloride it is disappearing even from synthesis labs because of environmental concerns, health issues, and mainly because it is useless except for IR (who does IR of small molecules anymore?) and – 20°c baths with acetone. So get as many liters as you can while we still have 20 years old bottles that no one uses 😉
have you tried it for phase separation chloroform versus buffer?
Don’t often do a chloroform extraction by itself, but you should get really nice separation with unmodified PSG, chloroform is pretty dense compared to buffer.
This is so awesome! I’m going to try this. You should publish this information.
It didn’t work for me 🙁
The phase lock gel (both dense and normal version) is too stiff. Our centrifuges are very tilted and after centrifuging the phase lock gel was almost at a 45° angle inside of the tube.
Why are all the pictures I can find on the internet showing that the phase lock gel is more or less horizontal? We are using regular (old) Eppendorf centrifuges.
We went back to the old way without it, because it is much easier to pipette (you can pipette from any position without having to be afraid you poke in the gel).
Hi! Super cool blog. Do you have any sample concentration results you can share after using the homemade PSG? Maybe it’s here on the site and I just can’t locate it. Especially interested in your 260/280 and 260/230 ratios. Also, have you tried Thermo Fisher’s PureLink Plant RNA Reagent with your homemade PSG? I work with succulents and that buffer works best on my plants and so curious how the PSG would hold up with that.