The art of separation

Chromatography in the laboratory

To those looking for help and guidance with navigating the often-tricky process of divorce or the end of a romantic entanglement, my apologies. This article deals not with maintaining a healthy and cordial relationship for the sake of the kids, but the role of chromatography in the commercial analytical laboratory.

If we try and break down the majority of chemical tests routinely performed in an environmental lab (note the specific reference to chemistry – microbiology and geological tests follow a different road) we can split the process into four parts:

  1. Extraction
  2. Separation
  3. Detection
  4. Calculation


Extraction deals with the process of taking the things you are interested in from the matrix in which they are submitted. Environmental samples cover soils, waters, effluents, sludges, filter cake, wood, incinerator bottom ash, made ground, building materials, landfill waste, waste materials intended for disposal or re-use, vegetation and in some cases biota (animal tissue).

Contrary to the indications of American TV forensics shows in the 2000s it is not possible to simply sample these materials (or matrices as they are referred to within the laboratory) and feed directly into an analytical instrument. In the main, they require a very specific type of prepared subsample, often dissolved into a very low volume of organic solvent, to avoid damage to instruments and ensure viable data can be obtained. A large part of all laboratories is made up with the process of turning that tub of soil, bottle of water or bag of waste that comes into the front end of the process into a very small amount of suitable solvent containing all the chemicals you are interested in testing for, and hopefully none of the ones you don’t!

“a large part of laboratories is the process of turning that tub of soil into a solvent you are interested in testing”


Detection we will touch on later in the article, but is basically ensuring that you are using the most appropriate method of ascertaining the presence and concentration of the chemicals you are interested in.

Getting everything right up to this point then selecting a detection system that physically can’t “see” the chemicals you are looking for won’t help – at the risk of stretching an analogy, you can’t measure the intensity of light with a decibel meter! The choice of detector is down to the chemistry of the compounds you are interested in, and often have multiple options at which points considerations of cost and fit for purpose design enter the fray… It is very possible to spend hundreds of thousands of pounds on the latest piece of analytical equipment, obtain results to nano or pico gram levels with incredibly high accuracy, but if the critical levels of interest or the specified accuracy can be achieved by a £10,000 solution you will struggle to justify to the Finance Director, or the client, that option A is the way to go.


Calculation rounds off the process and ensures that the customer receives data that relates back to the originally submitted material and provides relevant real-world information rather than the purely empirical data generated during the process. If we return back to our TV forensics shows, it’s very rare (in my experience at least) that the output from a GC-MS, for example, will include not only the correct concentration of your contaminant, but also any inferred or associated information without the input of a skilled analyst to turn the responses from the Mass Spectrometer into a fully calibrated, quality assured mg/kg result that can then be used as the basis of a decision making process.


Which eventually brings us back to point 2) and the reason we are here: Separation. Or to give the scientific term, Chromatography. Every sample received is potentially packed with all sorts of chemical contaminants and the role of the lab is to provide detail on the presence and concentration of specific ones. If we are doing our extraction process right, by the time we get to instrumental analysis we should have an extract that contains (should they be present, we’ll assume they are) everything we want. But if the intention is to measure each component individually, how do we do that?

Time to take you back to school…

The ink test

You may recall being introduced to chromatography by what’s normally referred to as the “ink test”. Someone has left a tell-tale trace on ink on the evidence, how do we know whose pen was used...

The Art of Separation

The above shows three ‘samples’ of ink, all of which look the same.

Through chromatographic separation a solvent passes up through the ink dots and ‘dissolves’ the component colours which are then split out as the solvent passes up the test, highlighting just how each ‘identical’ ink dot is actually made up. We then see that Sample 2 matches the test and our culprit is identified.

In our (hopefully) more advanced analytical lab, the basic principle of separating the component parts remains the same but the means by which we do it can vary.

During a chromatographic process there will be:

  • A Mobile Phase: this carries the sample through the process
  • A Static Phase: through which the mobile phase passes and by its interaction causes the separation

Before we delve into some examples of how we do it in the lab, I’ll try to give a visual example of the process.

Imagine, if you will, a long thin corridor with a door at each end.

Should two people (one adult, one child) enter at the same time and pass down the corridor they would likely get to the end at roughly the same time, separated slightly by variances in walking pace.

If, however, we fill the corridor with other people first:

The adult would potentially be impeded while the child could pass more easily and get to the end much quicker. This is the basis of Size Exclusion Chromatography (SEC), a key technique used in both preparation and analysis where we are separating chemicals based on molecular size. In real SEC the ‘filter’ will be a column filled with a gel consisting of beads containing pores of a specific size distribution, hence the alternate name of Gel Permeation Chromatography (GPC).

To stretch the analogy further (to look at other forms of separation) imagine two adults but one knows every other person in the room; they would likely slow down or stop to interact with people thereby lengthening their journey while the second person moves quickly through. Or three people, where Person A knows everyone, Person B knows half the people and Person C knows no one…

To put this tortured metaphor out of its misery and bring this back to actual chemistry for a while, what we actually do is create a static phase (the people in our example) that chemically interacts with the analytes of interest as they are pushed through creating the impediment to passage and therefore separation.

Real-life applications

So, what are the real life uses and applications in the lab as they process environmental samples?

The two main applications are GC (Gas Chromatography) and LC (Liquid Chromatography), the gas and liquid indicating the mobile phase used in each instance.

GC: used in conjunction with MS (Mass Spectrometer) and FID (Flame Ionisation) detectors, covers a wide range of compounds but most commonly in the environmental testing industry used in TPH (Total Petroleum Hydrocarbons), PAH (Polycyclic Aromatic Hydrocarbons), SVOC/VOC (Semi Volatile and Volatile Organic Compounds), PCBs (Poly Chlorinated Biphenyls), and certain pesticides.

The extracted sample, typically a small volume (µl levels) of organic solvent, is volatalised at high temperature and the compounds ‘pushed’ into the chromatographic column under a constant flow of gas (an inert gas, usually nitrogen, hydrogen or helium that will not interact with the process). The columns used, once glass filled with a sorbent material by hand, are now capillary columns – very thin copper with sub-micron internal diameter up to 60m in length.

The columns are usually packed with a polymeric material to interact with the compounds and generate the separation, but the analyst can also use gas flow rates and temperature variations to increase separation. The specific nature of the column packing varies by application.

LC: used in conjunction with MS, UV (ultraviolet) Absorbance or Fluorescence, ECD (Electrochemical Detector) and covers a similarly wide range of compounds including Phenoxy Acid Herbicides, Phenyl Urea’s, PFAS, Phenolics and Fluorinated surfactants amongst many others.

Often larger injection values than GC, with aqueous mobile phase as well as organic solvents. The columns vary in length and width, with various options for packing material.

The polarity of the static phase and mobile phase help create the separation, with the analyst also able to directly control pressure, flow rates, relative concentrations of multiple solvents within the mobile phase to vary parameters and further aid the chromatography.

In more recent years, the use of 2-Dimensional chromatography (GCxGC and LCxLC) has become more prevalent, the most common application being the use of GCxGC for investigatory analysis and now routine quantitative analysis of Hydrocarbons.

The principle is that two different columns of different chemistries’ (different static phases) are connected prior to the detector with the flow of mobile phase/carrier gas modulated following the first column then into the second. In the case of TPH and GCxGC, the second column is much shorter in length and separates based on polarity rather than just the boiling points as done through column one.

The Art of Separation

Adding this second dimension gives rise to chromatograms like the above where we can see horizontally on the x-axis the separation from column one based on size/boiling point and now also a z-axis where the second column further separates the peaks based on the polarity.

In this instance the chromatogram shows the separation of aromatic from aliphatic elements of TPH, something that previously would be done by a secondary separation process during the extraction (Solid Phase Extraction – SPE) where the sample is loaded onto a static phase, usually silica or florisil and the components selectively eluted by modifying the elution solvent which would then generate two separate samples for instrumental analysis. Allowing the instrument to perform the live separation gives immediate operational efficiencies as also allows for more detailed levels of data to support an investigatory process.

The development of newer detection systems (Triple Quadrupole MS systems for example) allows for more selectivity and sensitivity at the back end of the process, pushing detection limits and allowing analysis of a broader range of matrices. But these and many other developing techniques will still rely on the appropriate application of good chromatographic principles to succeed.