A plain-language look at HPLC

High-performance liquid chromatography, almost always shortened to HPLC, is the workhorse behind a large share of the purity and assay results on laboratory certificates. The instrument looks complicated and the output looks cryptic, but the principle is simple: push a dissolved sample through a tube packed with fine particles, and different components come out of the other end at different times.

The parts of the instrument

An HPLC system has four essential parts. A pump pushes a liquid, called the mobile phase, through the system at a steady flow and high pressure. An injector introduces a small, precisely measured volume of the dissolved sample into that stream. A column, a narrow steel tube packed with small particles coated in a chemical layer called the stationary phase, is where the separation happens. A detector at the outlet registers each component as it emerges.

The most common arrangement is reversed-phase chromatography. The stationary phase is water-repelling, often a long hydrocarbon chain bonded to silica particles, and the mobile phase is a mixture of water and an organic solvent such as acetonitrile or methanol. Components that are more water-repelling cling to the stationary phase longer and elute later. More water-loving components pass through faster.

The mobile phase composition can be held constant for the whole run, which is called isocratic elution, or changed gradually so that it becomes stronger over time, which is called gradient elution. Gradients let a single run separate components with very different behavior.

Reading a chromatogram

The detector’s output is a chromatogram: a trace of signal against time. A flat baseline means nothing is emerging. A peak means a component is passing the detector. Two properties of each peak matter.

The first is retention time, the time from injection to the top of the peak. Under fixed conditions a given compound elutes at a reproducible time, so matching a sample peak’s retention time to that of a reference standard run on the same system is one piece of evidence for identity. It is supporting evidence, not proof, because different compounds can elute at the same time.

The second is peak area. Over a validated range, the area under a peak is proportional to the amount of that component that passed the detector. That proportionality is what turns a chromatogram into a number.

From peak to result

There are two common ways to turn areas into a reported value. Area percent divides the main peak’s area by the total of all integrated peaks. It needs no reference standard and is quick, but it assumes every component responds equally at the detector, which is rarely exactly true, and it ignores anything the detector cannot see.

External standard quantitation compares the sample’s peak area with the area produced by a reference standard of known concentration, prepared and injected under the same conditions. Because the standard’s content is known, the sample’s content can be calculated directly. This is how assay results are normally produced, and it is why the reference standard’s own certificate matters as much as the sample’s.

System suitability: checking the instrument first

Before an HPLC result can be trusted, the system has to show that it is working as the method requires on that day. This is the system suitability test, run before or alongside the samples. Typical checks include:

1. Repeatability: replicate injections of a standard give areas within a set relative standard deviation.
2. Resolution: the main peak is adequately separated from its nearest neighbor, so the two areas do not merge.
3. Tailing: peaks are close enough to symmetrical that integration is reliable.
4. Sensitivity: where low-level impurities matter, a dilute standard still produces a measurable peak.

A run that fails system suitability is not reported. Its records should still exist, because a failed check that disappears from the record is itself a data integrity problem.

What HPLC cannot see

The most common detector, ultraviolet absorbance, responds only to compounds that absorb light at the chosen wavelength. Water, most inorganic salts and many residual solvents produce no signal. Two compounds that elute together form one peak, and the method cannot tell them apart unless the detector adds information. Coupling HPLC to a mass spectrometer, written LC-MS, addresses both limits by recording molecular mass as each peak emerges, which strengthens identity and can reveal components hiding under a single peak.

None of this makes HPLC unreliable. It makes it specific. A chromatogram answers the question it was designed to answer, under the conditions it was run, for the components it can detect. A certificate that reports HPLC purity alongside water content, residual solvents and an orthogonal identity test is using the method for what it does well and covering what it misses.