Reference Solutions: How Long They Last and Why It Matters

Every chromatographic assay contains a hidden claim: that the reference solution was the same concentration when the last injection ran as it was when the first one did. That claim is rarely stated and often untested. When it fails, the result moves, and the movement is attributed to the batch.

Solution stability is worth understanding because it sits upstream of almost every number on a certificate. It also explains a class of result that otherwise looks like a process problem: an assay that reads two points low with every other test normal.

Why the clock starts at dissolution

In the solid state, a molecule is protected by its own crystal lattice and by the low mobility of everything around it. In solution, it is exposed to dissolved oxygen, to light, to the container surface and to whatever the diluent does to its ionization state. Degradation that would take years in a drum can take hours in a vial.

The practical consequence is that the interval that matters is not the run time. It is the time between dissolution and the last injection, including the waiting that happens while a queue clears, while a column equilibrates, or while an analyst finishes another task.

Four routes by which a solution changes

Route	What happens	How it shows up
Adsorption	Analyte binds to glass, especially at low concentration	Response falls, impurity profile unchanged
Photodegradation	Ultraviolet exposure cleaves or isomerises	New peaks appear, response falls
Evaporation	Solvent leaves a loosely capped vial	Response rises, concentration is higher than labelled
pH drift	Buffer capacity exhausted or gas exchange	Retention shifts, peak shape changes

Only the second and third routes are visible in a single injection. Adsorption produces a clean chromatogram with a smaller peak, which is indistinguishable from a lower-concentration standard or a lower-assay sample. That ambiguity is why solution stability studies compare response and profile together, over several time points, against a freshly prepared control.

What a solution stability study should contain

A defined concentration, because adsorption is concentration-dependent and a study at 1 mg/mL says little about a working solution at 0.05 mg/mL. A defined diluent, because the same molecule in water and in a methanol-water mixture has different stability. A defined container, because borosilicate glass and polypropylene behave differently. A defined storage condition, including light. And time points that bracket the intended use with margin, not margin that ends at the intended use.

The output is an interval, and the interval should be shorter than the longest realistic day. A validated period of twenty-four hours that is used on a shift that runs to twenty-six is not a validated period, it is an expired one.

The mobile phase problem

Mobile phases age faster than solutions and are watched less. An aqueous-organic mixture loses its volatile component to the headspace, grows organisms in the aqueous fraction if it sits at room temperature, and changes pH as dissolved carbon dioxide equilibrates. None of these produce a warning. They produce retention drift, which is often corrected by extending the run rather than by preparing fresh mobile phase.

The connection to the result is indirect but real. Retention drift changes selectivity, which changes where an impurity elutes relative to the main peak, which changes whether an integration decision is defensible. This is the same territory as integration and its audit trail, one step upstream.

Why this looks like a batch problem

A degraded reference solution understates the analyte response and therefore understates the calculated assay. The batch reads 97.8% instead of 99.6%. Every other test is normal, the impurity profile is unchanged, and the natural conclusion is a low batch.

The difference between assay and purity is the diagnostic. If purity by area normalisation is unchanged and assay has dropped, the sample did not lose material. Either the reference did, or the preparation did, or the injection did. A batch that genuinely degraded would show both moving.

This is also why system suitability is checked before the sample is blamed. Suitability tests the system: repeatability of injections, resolution, tailing, theoretical plates. A reference solution that has lost 2% to adsorption passes every one of those tests, because the system is behaving perfectly with a smaller amount of analyte.

What a supplier should be able to state

The concentration and diluent used for the reference solution on a given certificate. The validated usable period and the storage condition it assumes. Whether the reference material is a pharmacopoeial standard or an in-house qualified standard, and if in-house, how it was qualified against a primary source. And whether the working solutions in routine use sit inside the validated interval or outside it on busy days.

The last item is the one that separates a system from a habit. Very few laboratories can answer it without checking, and the answer, once checked, is usually "sometimes".

We treat solution stability as part of the method rather than as a laboratory custom, and we describe it in those terms when a customer asks how an assay number was produced. The general expectations for reference standards and for method documentation sit in the USP general chapters, the Ph. Eur. framework maintained by EDQM and the British Pharmacopoeia.

The question worth asking is not whether the method is validated. It is: on the day this batch was assayed, how long was the reference solution in use before the last injection, and was that inside the interval you validated?

Designing a study that survives review

A solution stability study is small and easy to do badly. The version that survives review has four properties: a freshly prepared control at every time point, time points that bracket the intended use with margin, a stated storage condition including light, and an acceptance criterion expressed relative to the control rather than to the specification.

The control is the part that gets skipped. Comparing a twenty-four-hour-old solution against its own initial injection measures the solution and the instrument together, so a drifting detector and a degrading solution look the same. Preparing a fresh control at each time point separates the two.

The acceptance criterion is usually written as a percentage of the initial response, with a common band of ninety-eight to one hundred and two percent, and a requirement that no new peak above the reporting threshold appears. Both halves are necessary. A solution can hold its main-peak response while growing a degradation product that co-elutes poorly, and the response alone would not show it.

Solution stability is not forced degradation

The two studies are often confused because both involve holding a solution and looking at it later. Forced degradation is a stress experiment designed to demonstrate that the method can separate degradants from the analyte. It uses heat, acid, base, oxidation and light at levels chosen to produce ten to twenty percent loss, and its purpose is method specificity.

Solution stability uses the storage conditions the method will actually experience, for the intervals it will actually experience, and its purpose is an expiry time for the preparation. A forced degradation study that shows good separation says nothing about whether a solution can sit on a bench for a day.

Running one and calling it the other is a common shortcut, and it produces a method with an unstated solution life. The laboratory then applies a habit, usually one working day, which is not a validated interval.

The concentration trap

Adsorption to glass is concentration-dependent in a way that surprises people. A study performed at one milligram per millilitre can show no loss over twenty-four hours, while the working solution at fifty micrograms per millilitre loses several percent in the same period, because the surface area available per molecule is far larger.

The correct study uses the working concentration, in the container the method specifies, with the diluent the method specifies. A study at a convenient higher concentration is a study of a different solution.

What belongs in the method

The method should state the concentration, the diluent, the container, the storage condition and the validated usable period. It should also state whether the period applies to the standard solution, the sample solution, or both, because they can differ: a sample extract may carry matrix components that stabilize or destabilize the analyte.

Where the period is stated, the certificate's assay becomes portable. A customer who re-tests the same material can reproduce the number within the method's precision, rather than finding a difference that turns out to be a two-day-old standard.

We state the prepared-solution life in the method rather than in a laboratory habit, and we describe the study behind it when a customer asks how an assay figure was produced. The compendial expectations for reference standards and for method description are maintained by the USP, EDQM and the British Pharmacopoeia.

Frequently asked questions

When does the solution stability clock start?

At dissolution, not at weighing and not at the start of the chromatographic run. Once the material is in solution it is exposed to light, oxygen, temperature and the container surface, and the degradation that follows begins immediately. A solution prepared in the morning and injected in the afternoon has been ageing for both intervals.

How is solution stability actually demonstrated?

By comparing a freshly prepared solution against the same solution held under defined conditions and injected at intervals. The comparison looks at the analyte response and at the impurity profile, because a solution can lose analyte to an artefact that is not a degradation product, and gain nothing visible in the impurity window.

What causes loss of analyte in a stored solution?

Four routes. Adsorption onto glass, particularly at low concentration. Photodegradation, for molecules that absorb in the ultraviolet. Evaporation from a loosely capped vial, which concentrates rather than loses. And pH drift in a buffered mobile phase that is no longer buffered.

Does the mobile phase have its own shelf life?

Yes, and it is often shorter than the solution's. Aqueous-organic mobile phases lose the volatile component, grow organisms in the aqueous portion, and shift pH as dissolved gases equilibrate. A mobile phase prepared on Monday and used on Friday is a variable that shows up as retention drift, not as a result change.

What should a certificate or method say about it?

The method should state the solution concentration, the diluent, the container, the storage condition and the validated usable period. If it states none of these, the laboratory is applying a habit rather than a validated interval, and the habit is usually a working day.

Could this make a batch look out of specification?

Yes. A degraded reference solution lowers the measured response and depresses the calculated assay, which reads as a low batch rather than a laboratory problem. That is why the first step in an out-of-specification investigation should be to confirm the reference preparation, not to re-test the sample.