Chromatography — TLC, column chromatography, GC and GC-MS.

HookThe fastest man in the world lost his gold medal to a chromatogram

ModelOne idea, three techniques — the tug-of-war that separates mixtures

All chromatography stages a tug-of-war for each component of a mixture. On one side, the stationary phase holds the component back — either by adsorption onto the surface of a solid, or by dissolving it into a liquid held on an inert solid support. On the other, the mobile phase sweeps it forward, carrying whatever is dissolved in it. A component that binds strongly to the stationary phase and dissolves poorly in the mobile phase crawls; one with the opposite preferences races. Because no two compounds strike exactly the same balance, they travel at different speeds and the mixture separates. That one sentence — separation depends on each component's balance between retention by the stationary phase and solubility in the mobile phase — is the explanation AQA wants, whatever the technique.

The three techniques on the specification differ only in what plays each role. In thin-layer chromatography (TLC) the stationary phase is a thin solid layer of silica gel or alumina coated on a plate, and the mobile phase is a liquid solvent creeping up the plate by capillary action. In column chromatography the same sort of solid — or a liquid coated on an inert solid support — is packed into a vertical glass tube, and fresh solvent washes the components down through it. In gas chromatography (GC) the stationary phase is a liquid or solid coating the inside of a long, coiled column sitting in an oven, and the mobile phase is an unreactive carrier gas such as nitrogen or helium, chosen inert so it cannot react with the sample at oven temperature.

Polarity usually decides who wins the tug-of-war. A silica surface is studded with polar Si–O–H groups, so polar components adsorb strongly onto it and move slowly, while non-polar components stay dissolved in a non-polar solvent and ride near the front. Reverse the polarity of either phase and the running order reverses too — which is why quoting the phases involved is compulsory in any explain-the-separation answer.

DataTLC and the Rf value — turning a spot into a number

A developed TLC plate is read by measuring, from the pencil baseline, how far each spot travelled compared with how far the solvent itself travelled:

\[R_f = \frac{\text{distance moved by the spot (baseline to spot centre)}}{\text{distance moved by the solvent front}}\]

Both distances are measured from the baseline, the spot distance is taken to the centre of the spot, and the answer is a ratio — no units, always between 0 and 1. A compound that never leaves the baseline scores 0; one that rides the solvent front scores 1. Small Rf means the compound spent most of the run adsorbed to the silica (typically the more polar component); large Rf means it spent most of the run dissolved in the mobile phase.

The number is only as meaningful as the conditions it was measured under. Rf depends on the stationary phase, the solvent and the temperature, so a value from a different plate, a different solvent or a different day proves nothing. Identification therefore means spotting pure reference standards alongside the mixture on the same plate and comparing positions directly: same Rf, same conditions, same compound — provisionally. TLC's honest job is to count the components of a mixture, check the purity of a product (one spot = one substance), and shortlist identities for a technique with real authority, such as mass spectrometry, to confirm.

MechanismColumn chromatography — separation you can collect

TLC tells you what a mixture contains; it cannot hand the components back. Column chromatography can, which is why it is the preparative workhorse of every synthesis lab. The stationary phase — silica or alumina, or a liquid supported on an inert solid — is packed as a slurry into a vertical tube and kept wet. The mixture is loaded in a thin band at the top, and fresh solvent (the eluent) is run continuously through the column. Each component moves down at its own speed, set by the same tug-of-war as ever: strong adsorption to the packing means a slow descent, high solubility in the eluent means a fast one.

The bands separate as they travel, and each is collected as it drips from the bottom — a process called elution, with the components collected as separate fractions. The component with the greatest affinity for the mobile phase elutes first; the one gripping the stationary phase hardest elutes last, and the time each takes is its retention time on that column. Evaporate the solvent from a fraction and you hold the pure compound — this is how a pharmaceutical chemist pulls a synthesised drug away from its by-products before testing it.

Column chromatography is TLC scaled up and turned on its head, and examiners like the comparison: same phases, same principle, but TLC is fast, cheap and analytical while the column is slower, larger and preparative. A common trick is to run a quick TLC first to choose a solvent that spreads the components well, then use that solvent as the eluent on the column.

DataGas chromatography — retention times, peak areas and their limits

For volatile mixtures the mobile phase can be a gas. The sample is injected, vaporised instantly, and swept by the inert carrier gas through metres of coiled column held in a precisely controlled oven. Components that dissolve readily in the stationary-phase coating linger; those that prefer the gas stream hurry through. A detector at the far end records each arrival as a peak on a chromatogram, and the time from injection to detection is that component's retention time.

A gas chromatogram answers two questions. What might this be? — compare each retention time with pure standards run under identical conditions: same column, same oven temperature, same gas flow. How much is there? — the area under each peak is proportional to the amount of that component, so relative percentages fall straight out of the areas. Note it is the area, not the height: a tall narrow peak and a short broad one can hold the same amount. The sensitivity is remarkable — evidential blood-alcohol analysis in the UK is done by gas chromatography, resolving ethanol at the 80 mg per 100 ml limit that applies in England and Wales (Scotland's is 50) from everything else in blood.

But GC alone has honest limits, and AQA asks for them by name. Two different compounds can share a retention time and emerge as one peak (co-elution), so a matching time is evidence, not proof. Retention times drift with oven temperature and flow rate, so comparisons are only valid under identical conditions. And a compound the laboratory has no standard for gives a peak that identifies nothing at all — a time on its own carries no molecular information. Chromatography separates superbly and identifies weakly; that gap is what the mass spectrometer fills.

CaseGC-MS — separate first, identify beyond doubt

Couple the outlet of a gas chromatograph to a mass spectrometer and the two instruments cancel each other's weaknesses. On its own, MS applied to a mixture superimposes every component's fragments into an unreadable pile; on its own, GC produces clean peaks it cannot name. In GC-MS, the column delivers the components one at a time, and the spectrometer records a complete mass spectrum for each individual peak — molecular ion, fragmentation pattern, isotope signature, the full 3.3.6 toolkit — which a computer matches against a database of known spectra in seconds. The retention time shortlists; the fragmentation pattern convicts.

That is why GC-MS is the confirmatory instrument in exactly the arenas AQA cites: anti-doping laboratories (the Seoul finding against Ben Johnson stood because the metabolite's mass spectrum matched stanozolol's, not merely its retention time), forensic science (accelerant residues in fire debris, drug seizures), environmental monitoring of pollutants at trace concentrations — and space science, where NASA's Curiosity rover carries a GC-MS in its SAM instrument package to analyse Martian rock and atmosphere, and the Huygens probe carried one through the atmosphere of Titan. The exam question is nearly always the same four-marker: why combine them? GC separates the components of complex mixtures and quantifies them by peak area, but cannot identify positively; MS identifies each separated component from its unique fragmentation pattern via database comparison. Two marks for each half.

CaseRequired practical 12 — TLC done properly, and why each step exists

RP12 asks you to separate a mixture — typically amino acids — by TLC, and every step of the method carries a because that examiners test directly. Draw the baseline in pencil, because ink is itself a dye mixture that would dissolve and chromatograph alongside your sample. Spot the mixture with a fine capillary tube, keeping the spot small and concentrated (spot, dry, re-spot), because a fat spot smears into its neighbours as it runs. Stand the plate in solvent whose level sits below the baseline, because solvent above the line would simply dissolve the spots off the plate into the reservoir. Lid the beaker or tank, so the atmosphere saturates with solvent vapour and the solvent does not evaporate from the plate mid-run. Remove the plate before the front reaches the top and mark the solvent front immediately, in pencil — the solvent evaporates within seconds, and without that line there is no denominator and no Rf.

Amino acids are colourless, so the developed plate must be visualised with a locating agent: spray with ninhydrin (in a fume cupboard) and warm, and the amino acids appear as purple spots; alternatively use a plate impregnated with a fluorescent dye under UV light, where the spots show dark against the glowing background. Wear gloves throughout — not just for the chemicals, but because fingerprints deposit their own amino acids, which ninhydrin will cheerfully develop as ghost spots. Safety in the write-up: the solvent is volatile, flammable and harmful, so use a fume cupboard or good ventilation and no naked flames; avoid skin contact with ninhydrin; do not look directly at the UV source.

Evaluation marks hide in error directions. Measure to the centre of each spot, not its leading edge. If the front is marked late, the measured front distance is too short and every Rf comes out too large. If spots are overloaded they tail upwards and their centres — and Rf values — blur. And identification is only valid against standards on the same plate, because tomorrow's plate, at a different temperature with slightly different solvent, gives different numbers.

VocabularyKey terms the mark scheme pays for

TrapsMisconceptions that cost marks

ExamWhat examiners want