Manual Liquid-Liquid Extraction – Time for a Change
Posted on February 2nd, 2018
If you still carry out liquid-liquid extraction manually, and are fed-up with the associated costs, mistakes and hazards, then this will be the most useful thing you read today. I am going to show how liquid-liquid extraction can be scaled-down, sped-up and give better quality data than you ever thought possible, by fully automating the process.
21st Century Labs, 19th Century Technology
The glass separation funnel emerged in the 1800s and it passed into common use when labour was cheap, people were less concerned with risk, and the number of samples analysed was much lower than it is today.
In analytical chemistry, the separation funnel is the principle means of performing liquid-liquid extractions (LLE) and is mostly commonly used to extract lipophilic compounds from water samples using hydrophobic solvents. LLE remains overwhelmingly, a manual process.
In the context of LLE, chlorohydrocarbons are useful solvents, but the potential for chemists to be poisoned by vapour inhalation, has meant that nowadays, the least toxic solvent, dichloromethane (DCM), is the one most commonly used. That said, DCM does present some degree of hazard and with manual LLE, a large amount used for every extraction, so the manual procedure should be carried out in a fume cupboard.
For a typical 500mL water sample, around 100mL of extracting solvent can be used, with still more solvent used for washings and rinsing glassware.
Some labs employ mechanical shakers, but many rely upon hand shaking to some extent. Extracting 20-30 samples per day by hand it physical work and, if done routinely, poses a risk of repetitive strain injury.
As with any manual process, there is scope for samples to be processed in an inconsistent fashion, to the detriment of data quality.
In the past, the relatively large scale of LLE has made it prohibitively expensive to automate, (although some have tried), so laboratories have learned, grudgingly, to accept the costs, risks and dodgy data due that comes when LLE is performed by hand.
Bringing LLE into the Modern World
Mass Spectrometers are now more sensitive than ever. From a theoretical perspective, the volume of samples that need to be extracted, is simply a function of sensitivity of the mass spectrometers used for the measurement. Every time the sensitivity of the mass spec doubles, the amount of sample you need to extract halves. If you re-examine the design of your extraction, taking into account the sensitivity that modern instruments offer, you will find that you can achieve the required LODs with much smaller volumes of both sample and extracting solvent. Which brings us to the important bit:
It is now perfectly feasible to make measurements at the 1ng/L level or lower, by extracting water samples of less than 10mL.
Automated Dispersive Liquid-Liquid Micro Extraction
We have been working on the development of a fully automated solution for dispersive liquid-liquid extraction (DLLME), in collaboration with several customers.
DLLME is similar in principle to traditional LLE, but differs, in that a dispersant is added to the sample, that enables it to form a milky emulsion with the extracting solvent. The surface area in contact with the two phases is far greater and this greatly improves extraction efficiency as a result. This gives two major advantages:
- The extraction happens quickly, with just 15-30 seconds of shaking needed,
- The extraction efficiency is very high, giving excellent analyte recovery.
Using our automated LLE solution, you can have it both ways; you can automate conventional liquid-liquid extraction; however, it is also possible to take advantage of the gains in performance that come from DLLME.
DLLME is nothing if not simple. Firstly, sufficient isopropyl alcohol (IPA) is added to each water sample to create a 10% solution. Then when the extracting solvent (DCM) is introduced into the sample a fine colloidal suspension forms, as the IPA acts as a surfactant.
With a huge surface area between the two phases, it needs just a few seconds of vigorous mixing, for the analytes to achieve a distributive equilibrium between the two phases.
When spun in a centrifuge, the emulsion breaks quickly, and the DCM forms a layer at the base of a high recovery vial, from which a microliter syringe can retrieve a 1uL portion and inject it into the analyser.
The entire process is automated, and each sample is processed in just a few minutes. This means that, in most cases, samples can be prepared faster than a GC-MS can analyse them, opening the way for a single robotic system to serve several analysers. Alternatively, the robot can be installed directly on the GC or LC-MS. Then samples can be prepared and analysed on seamless, a just-in-time basis.
The system uses cheap disposable glassware; so there is nothing to clean and the risk of cross-contamination is eliminated.
- Samples can be processed quickly with no operator intervention. This means not only is the data quality excellent, compared to the old way of doing things, the costs of preparing samples is much lower. The same number of people can get much more done.
- The volume of the samples to be extracted is cut 10-50 times. This means much less sample to transport, store and handle.
- You also use much less solvent and what is used, is mostly handled by the robot. Your entire health, safety and environmental footprint is slashed – and you save money, through buying and disposing of less solvent.
Try Out Automated LLE – Risk-Free
Anatune’s experienced application chemists will help so that you can try-out automated LLE and DLLME yourself; so, you can be sure it will work for you. We then help you generate the information you need to make a case for purchase.
Will this work for you? There is one way to find out – contact us now! Email us, or call the office on +44 (0)1223 279210.