There’s something in the water, and it would love to go after your experiments.

Straight out of the tap, water contains microorganisms, endotoxins, DNase and RNase, salts and other impurities that could gobble up your experiment in one bite. Of course we avoid this drama completely by using purified water from which these nasties have been removed. But how is this purification done?

Well in practice a number of techniques are used, each of which can remove a different set of impurities.

So here are the techniques:

1. Distillation. A technology as old as the hills (or at least as the stills that were hidden in the hills). Water is heated to its boiling point then condensed back to liquid. This will remove many impurities but impurities with a boiling point equal to or less than that of water will also be carried over in to the distillate.

2. Microfiltration. In this technique, pressure is used to force the water through a filter with pore sizes of 1 to 0.1 micron in order to remove particulate matter. Filter diameters lower than 0.2 micron removes bacteria – so-called cold sterilisation.

3. Ultrafiltration uses even smaller pore sizes (down to 0.003 micron). These are essentially molecular sieves, which remove molecules with a diameter larger than the pore size. It can be used to remove viruses, endotoxins, RNase and DNase

4. Reverse osmosis. If you thought that ultrafiltration used impressively small pore sizes, you’ll be even more impressed by reverse osmosis . Reverse osmosis filters have pore sizes of less than 0.001 microns, which allows them to sieve ions depending on their diameter. This is used for desalting the water.

5. Filtration through a bed of activated carbon is useful for removing things like chloride ions and organic compounds, which are adsorbed onto the surface of the carbon.

6. UV radiation. We all know what UV radiation, at specific wavelengths, can do to DNA and microorganisms. So UV is an obvious way to remove microorganisms from the water. It can also clean up the water by breaking down certain organic compounds into less harmful products.

7. Deionization/ Ion exchange. This technique removes ions from the water by passing it through a resin bed containing a mixture of cationic and anionic resins. Positive ions in the water are attracted to the anionic resin particles and negative ions are (yes, you’ve guessed it) attracted to the cationic resins. The result is that nicely deionised water comes out of the other end of the resin bed.

Commercially available water, or water purification systems will typically use a combination of these. The higher the water purity grade, the more techniques used.

Any questions or comments? Just jump in and join the discussion in the comments section below. The water’s lovely.

More 'DNA / RNA Manipulation and Analysis' articles


  1. <>

    Just a chemist’s quibble… Cationic resins will have positively charged ions in them — so these will attract *negative* ions, not positive. This confusion is not infrequent and might come from the names of the common “exchange” resins; for example strong anion exchange resin, SAX, binds to strong anions but is therefore itself cationic.

  2. Great post Nick.

    Lab water is so often overlooked as an important factor when designing experiments and (as mentioned by John here) when looking for sources of errors. The whole area of lab water has been carefully looked at by a number of agencies/groups including the American Society of Testing and Materials (ASTM) and the Clinical and Laboratory Standards Institute-Clinical Laboratory Reagent Water (CLSI®-CLRW). These organisations have similar (but not identical) definitions for highly purified water, which are determined based on parameters such as the water’s conductance, resistance, the presence of colloids, bacterial count, organic content and pH. Using ASTM nomenclature, type IV is the lowest grade of purity, suitable for most routine lab work, while type I is the highest grade. The most commonly used standard, ASTM D1193-6 is summarized in table below. This standard can then be further sub-divided into A, B and C, where A is the most pure, as a measure of heterotrophic bacteria count (CFU/ml) and endotoxins (units per ml).

    Measurement (unit) Type I Type II Type III Type IV
    Resistivity (M?-cm) at 25 °C >18 >1 >4 >0.2
    Total organic carbon (ppb) <50 <50 <200 No limit
    Sodium (ppb) <1 <5 <10 <50
    Chloride (ppb) <1 <5 <10 <50
    Total silica (ppb) <3 <3 <500 No limit

    It is also important, I think, to note that the purification technologies are often combined in lab water generators and can be used to replace commonly used techniques: for example, the DEPC treatment of water as used for RNA work can be replaced by using a water generator using a combination of UV and UF (ultrafiltration) – e.g. Thermo Scientific Barnstead Nanopure UV/UF.

  3. Interesting that often reagents such as Taq and other enzymes, dNTPs etc are qizzed first for contamination rather than looking at the reagent that makes up probably 90% of the reaction
    Still, have been reports of DNA sequences (legionella comes to mind) and another one I cannot recall the publication, being amplified from commercial water supplies.
    Nice review Nick

Leave a Reply

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