First published 20th February 2017. Last updated 9th May 2017.
We learn at school that water is H2O. However, as with so much of the science we are taught at school, that is much too simple. The water we drink is full of anions, cations, organics, dissolved gases, etc. It has been cleaned to make it suitable for us to drink but, from a laboratory scientist point of view, it is dirty. In a laboratory we need get our water closer to the school ideal of H2O (pure water); water that is so pure it is dangerous to drink . Traditionally, this was made by distillation but, in recent years, more energy efficient and effective processes have been developed. Understanding these processes is key to choosing the type of water we need to use.
DISCLAIMER: I am not an expert on lab water. The content of this blog is what I have discovered through my efforts to understand the subject. I have done my best to make the information here in as accurate as possible. If you spot any errors or admissions, or have any comments, please let me know.
Different types of water
There are various grades of water available to most lab users, the properties of these are shown in Table 1. Some further details are discussed below. Water quality is generally defined by its electrical conductivity. Water that is pure H2O (consisting of only H+ and OH− ions) gives a resistivity of 18.2-18.3 megaohms.cm (@25ºC) any contamination will reduce this figure. Total Organic Carbon (TOC) is also measured to give an indication of the organic contaminants found in water (measured in ppm or ppb).
Tap water has limited use in the laboratory. Tap water contains many substances that, if left untreated, may react or catalyze reactions in undesired ways or just contaminate analysis. In some laboratories it has to be actively kept away; for example, where samples with low levels of chlorine are being studied. In most physical science laboratories I think the following would be a fair assessment of its use:
- Tap water generally used for: cleaning dirty glassware before further washing, cleaning the laboratory and washing excess chemicals/samples down the sink. Obviously, not drinking!
- Tap water not used for: making up solutions, final cleaning of glassware, or any other application that will directly impact the work.
Deionised (DI) water
Deionised water is the standard laboratory water. In many laboratories, it is sufficiently clean for the purposes of the laboratory. In most physical science laboratories I think the following would be a fair assessment of its use:
- DI water is generally used for: cleaning reusable equipment and other general lab based activities (e.g. making solutions, rinsing filter papers, etc.) where you are worried about impurities at the roughly ppm or higher level.
- DI water is not used for: general lab cleaning or for analytical chemistry procedures where you are worried about roughly ppb levels of impurities.
Ultra clean water is vital for work in some laboratories and in such laboratories it is guarded carefully to prevent even the smallest amount of contamination. As soon as this water leaves the production system it will start to pick up contamination. Standard glass or plastic vessels will contaminate the water and so other vessels have to be used (such as silica or ultrapure tin).
- Ultra-clean water is generally used for: preparation of solutions or other lab based activities which could be affected by water impurities at the ppb level. e.g. analytical chemistry procedures.
- Ultra-clean water is not used for: other application as it uses significant amount of energy, resources, and water to produce this product.
Production of DI water
There are 5 main classes of impurities that can be present in water and which may affect your experiments:
- Suspended particles: sand, silt, etc.
- Dissolved inorganic cations: calcium, magnesium, sodium, etc.
- Dissolved inorganic anions: chloride, nitrate, sulphate, etc.
- Dissolved organic: decaying vegetable matter, fats, oils, etc.
- Microorganisms: bacteria, fungi, etc.
- Dissolved gasses: N2, O2, CO2
A lot of impurities are removed from water during drinking (tap) water treatment. However, as discussed above tap water is not clean enough for most laboratory work and so has to be further cleaned. Also, chlorine is added to drinking water which can cause its own issues.
Stages of filtration
Figure 1 shows the main stages of filtration for laboratory water. After each stage the water can be taken and used at that quality (Type I, II, etc.) [see  for more details]:
- Pre-treatment: this stage normally includes a physical barrier to remove suspended particles (10-50µm) and a system to remove chlorine (e.g. activated charcoal)
- Reverse osmosis (RO) membrane: Removes dissolved inorganics, organics, particles, and bacteria. Only a percentage of water (around 50%) gets through, the rest is rejected to waste. RO membranes are able to reject bacteria, pyrogens, inorganic and some organic solids but dissolved gases are not as effectively removed.
- UV lamps: short wavelength UV breaks down organic molecules to they can be removed (usually 184nm). Long wavelength kills bacteria (254nm) and prevents their growth and potential contamination of the water. These properties make UV a necessity to reach low Total Organic Carbon (TOC) levels (<5ppb).
- Deionisation cartridge: usually anion and cation resins in the form of spherical beads. These cartridges can purify water to the ultra-clean level but get exhausted very quickly if water is too contaminated. To save the cartridges more than one cartridge is often used. Deionisation cartridges do not remove particles, pyrogens or bacteria; and have very limited effectiveness with many organics
- Electro DeionIzation (EDI) module: Water is passed through an ion exchange resins and an electrical field. The potential difference removes ions more efficiently than resins alone. This system can only get to 5-15 MΩ not to ultra-pure water . EDI does not remove organics, particles, pyrogens or bacteria.
- Ultrafiltration: is required to remove particles such as protein macromolecules, dead bacteria, etc. which deionisation/EDI fail to remove.
- Storage tank: a storage tank and distribution system is often used due to the relatively slow processes involved in deionisation (particularly the RO membrane) and to reduce the number of systems required in a set of laboratories. It is vital to include a recirculation system to avoid water stagnation. A UV lamp (254nm) and submicron filtration are used for microbial control.
The purer the water the more energy and water is required to produce it. This means it is really important to keep the use of DI and ultra clean water to a minimum. It should only be used when the application requires it (see Table 1 and related text).
The filters and RO membranes reject a large percentage of the water that goes into them; the rejected water is generally sent to drain. There is also a large waste of materials as filters, membranes, and UV lamps have to be replaced regularly and generally they are not recycled; although this will vary between manufacturers. Table 2 gives an idea of how often filters etc. have to be replaced.
One item that could be recycle are the deionisation cartridge. The cation and anion resins from these cartridges could be regenerated; however, in practice this process is rarely carried out. This is mostly due to difficulties in avoiding contamination and the high cost involved. This makes it easier and cheaper just to replace the whole unit.
DI water is vital in the laboratory. However, it has a significant energy, material, and financial cost. The technologies that have been developed to takes us away from the inefficient distillation are are impressive. However, please remember that every drop of DI water costs the environment so only use it where it is required.
 Laboratory water: its important and application, National Institutes of Health, March 2013.