Current turnaround: 21 days from receipt of your samples. Due to high demand, HPLC-UV peptide testing and Certificates of Analysis are taking longer than usual.

Independent Research Testing Center · 875 Innovation Drive, Cambridge, MA 02139, USA For all independent testing inquiries, email admin@novagenlaboratories.com
Analytical methods

Blank Injections and Carryover: Finding a Contaminated System

A blank run contains no sample, so any peak in it describes the instrument, the solvents or the preparation. Reading that correctly is what keeps a system artefact from being reported as an impurity.

Novagen Analytical Labs

A blank injection is a chromatographic run made with no sample analyte in it. Its purpose is to show what the system produces by itself: the baseline, the solvent response, and any peaks that come from the instrument or the preparation rather than the material being tested. Carryover is one thing a blank can reveal. It is residual analyte from an earlier injection appearing in a later run. A peak found in a blank belongs to the system, not to the sample, and it must never be reported as a sample impurity.

This article explains the kinds of blank, how carryover arises and is measured, how it differs from other ghost peaks, and how a laboratory reduces it. It also states the limits of the approach, because a blank can only answer the questions it was designed to ask.

What a blank injection is for

A chromatogram of a sample contains peaks. Some of those peaks come from the sample. Others come from the solvents, the vial, the tubing, the column or a previous run. From the trace alone the two groups look the same. A blank separates them.

The logic is subtraction by evidence. If a peak is present at a given retention time in a run with no sample, it cannot be a component of the sample. If the same peak is absent from the blank and present in the sample, it is a candidate sample component. That candidate still needs identification, but the blank has removed one large class of false candidates.

A blank is therefore a control. It does not measure the sample. It measures the conditions under which the sample was measured.

The kinds of blank

The terms are used with some variation between laboratories and methods, so a report should say which blank was run. The common types are these.

  • Solvent blank. The pure solvent, often the mobile phase or the strong solvent used in the method, injected on its own. It shows the response of the solvent and the baseline behaviour of the gradient.
  • Diluent blank. The liquid used to dissolve or dilute the sample, prepared and injected as for a sample but without analyte. It shows peaks that come from the diluent, including impurities in the diluent and anything it picks up from the vial or cap.
  • Reagent blank. The full preparation carried out without the sample, including every reagent, buffer or additive. It shows contributions from the preparation steps themselves.
  • Matrix blank. A sample of the matrix known to be free of the analyte, prepared in the normal way. This is mostly used in bioanalysis, where the matrix itself produces signals.
  • Double blank. In bioanalytical work, a processed matrix sample with neither analyte nor internal standard. It is distinguished from a blank that contains the internal standard only. The term is not used consistently outside that setting.

Each type answers a different question. A solvent blank that is clean while a diluent blank shows a peak points at the diluent or its handling. A diluent blank that is clean while a reagent blank shows a peak points at a reagent. The sequence of blanks narrows the source.

What carryover is

Carryover is residual analyte from a previous injection that appears in a later run. The analyte was not in the later vial. It was left behind in the flow path and released during or after the next injection.

The usual locations are the parts that contact the sample at high concentration.

  • The outside and inside of the injector needle.
  • The needle seat and the injection port.
  • The sample loop and connecting tubing.
  • The rotor seal of the injection valve, where scratches and wear create small volumes that retain liquid.
  • Fittings with poorly seated ferrules, which leave dead volume.
  • The column, where strongly retained material may not elute within one gradient and appears in a later one.

Peptides are well known for adsorbing to surfaces, and some sequences are sticky enough that carryover is a routine concern. That is a general property of the analyte class. The degree depends on the sequence, the hardware and the method, and it has to be measured for the system in use.

How carryover shows up

The signature of carryover is a peak in a blank, or in a low concentration sample, at the retention time of an analyte that was injected earlier. Three features help to recognise it.

  • Retention time matches the earlier analyte. The carryover peak elutes where the preceding injection's main peak elutes.
  • It shrinks over successive blanks. Residue is washed out progressively, so the peak area tends to fall with each repeat blank. A constant peak across many blanks often points to a different source.
  • It scales with the preceding injection. A high concentration injection leaves more behind than a low one. A blank that follows a dilute sample may be clean, while one that follows a concentrated sample is not.

These are tendencies, not guarantees. Carryover from a worn rotor seal can be irregular. The practical conclusion is that a single clean blank does not prove the absence of carryover under all sequences, and a single contaminated blank does not identify the cause.

How carryover is quantified

The standard test is simple. Inject a high concentration standard, then inject a blank. Measure the peak area in the blank at the analyte retention time. Express it as a percentage of the analyte peak area in the preceding standard.

For example, if the blank shows a peak area equal to 0.1 percent of the preceding standard's area, the carryover is 0.1 percent. The figure is specific to that analyte, that concentration, that wash programme and that hardware condition.

Published guidance for bioanalytical method validation uses a related approach, compared against the lower limit of quantitation. In that convention, carryover in a blank injected after the highest calibration standard should not exceed 20 percent of the analyte response at the lower limit of quantitation, and 5 percent of the response for the internal standard. This appears in regulatory guidance for bioanalytical method validation, including ICH M10. It is a convention written for that application. It should not be applied to a purity assay or any other method without considering what the method is trying to detect.

The reason is practical. The acceptable amount of carryover depends on the smallest signal the method must report. A method that reports impurities at 0.05 percent of the main peak needs carryover well below that level, or the carryover is indistinguishable from a real impurity.

Carryover versus other ghost peaks

A ghost peak is an unexpected peak that does not come from the injected sample. Carryover is one source. There are others, and the distinction matters because the remedies differ.

  • Mobile phase contamination. Impurities in water, organic solvent or additives. In a gradient, trace contaminants in the weak solvent accumulate on the column during equilibration and are released as the organic proportion rises. The result is a peak that appears even when the injection volume is zero.
  • Gradient baseline drift. A rising or shifting baseline caused by the weak and strong solvents absorbing ultraviolet light differently as their proportions change. It is a response of the solvents to the gradient, not an injected compound.
  • Diluent or vial contributions. Leachables from caps, septa or vials, or impurities in the diluent.
  • Matrix components. Signals from the sample background that resemble the analyte.
  • System peaks. Disturbances to the baseline when the injected liquid differs from the mobile phase, often near the void volume.

One useful test separates injection-related sources from mobile-phase sources. Run the gradient with no injection at all. If the peak still appears, the injector and the preceding sample are not responsible. If the peak appears only after an injection, the injection path or the diluent is implicated. If the peak area increases with a longer equilibration time before the gradient, accumulation from the mobile phase is likely.

None of this identifies the compound. It identifies the part of the system that produced the signal.

How a laboratory reduces carryover

Reduction starts with finding where the residue is held, then changing one thing at a time and repeating the standard-then-blank test.

  • Needle wash. Most autosamplers can wash the needle with one or two solvents. A strong wash solvent dissolves what the weak one leaves behind. For peptides this often means a wash with a higher organic proportion, sometimes with an acid or another additive. The right composition depends on the analyte.
  • Wash volume and cycles. A larger volume, more cycles, or a wash both before and after the injection give more opportunity to clear the surface.
  • Backflush or valve switching. Flushing the flow path in the reverse direction, or switching the valve so that the loop is flushed with strong solvent during the run, removes residue held in the loop.
  • Maintenance. Worn rotor seals, scratched needle seats and poorly seated fittings retain liquid. Replacing them often resolves carryover that no wash change fixes.
  • Sequence design. Placing blanks after high concentration samples, and running samples from low to high concentration where possible, limits the effect on the following injection.
  • Column flushing. A strong solvent step at the end of the gradient, with adequate re-equilibration, limits release of retained material in later runs.

Blanks between samples are mainly a diagnostic. They reveal carryover rather than remove its cause. A blank placed after a high concentration sample does take up some of the residue that would otherwise reach the next real sample, so it limits the effect on that sample without fixing the source. Treat a blank as a safeguard, not a repair.

The relationship to system suitability

System suitability testing checks, before and during a run, that the system is performing adequately for the method. A blank injection is often part of it. A clean blank at the start shows the system is not already contaminated. A blank at the end, or between high concentration injections, shows that the sequence did not leave residue that would affect the results.

Where a blank fails its criterion, the results that depend on it are in doubt. The correct response is to find and fix the cause, then repeat the affected injections. Subtracting the blank peak from the sample result is sometimes done for a small, reproducible and well-characterised signal, but it is a correction with a stated basis and not a way to rescue a contaminated run.

What this means for reporting

A peak in a blank is a property of the system. If the same peak appears in a sample chromatogram at the same retention time, the defensible reading is that it is at least partly from the system, and it cannot be reported as an impurity of the sample on that evidence alone. Doing so would attribute to the sample something the sample did not contribute.

The opposite error also exists. A real impurity can co-elute with a ghost peak. A clean blank shows only that the system contributed nothing at that retention time under those conditions. It does not show that the sample peak is a single compound. That question belongs to peak purity and mass spectrometric identification.

A certificate describes the sample that was submitted and nothing beyond it. Purity is a statement of composition, not of safety, and every measurement has limits. The blank is part of how those limits are established. It tells the reader which signals the system is able to produce on its own.

Novagen is a third-party analytical laboratory. It reports what was measured in the sample submitted, and it publishes a verifiable certificate register at novagenlaboratories.com. Current quoted turnaround is 21 days from receipt of samples. Questions can be sent to admin@novagenlaboratories.com.

Common questions

What is a blank injection in HPLC?

A blank injection is a run made with no sample analyte in it. The injected liquid is usually the diluent, the mobile phase or a prepared matrix without the analyte. It shows what the system produces on its own, so that signals from the system can be separated from signals from the sample.

What is the difference between carryover and a ghost peak?

Carryover is residual analyte from a previous injection appearing in a later run. A ghost peak is any unexpected peak that does not come from the injected sample. It may arise from the mobile phase, the gradient, the vial or the diluent. Carryover is one possible cause of a ghost peak, but not the only one. The retention time of a carryover peak matches the analyte that was injected before it.

How do you calculate HPLC carryover?

Inject a high concentration standard, then a blank, and measure the peak area at the analyte retention time in the blank. Carryover is that area divided by the area of the analyte peak in the preceding standard, expressed as a percentage. Where a method is validated against a specific limit, the blank area is instead compared with the area at the lower limit of quantitation.

How much carryover is acceptable?

It depends on the method and its purpose. In bioanalytical method validation guidance, including ICH M10 and earlier regulatory guidelines, carryover in a blank after the highest calibration standard should not exceed 20 percent of the analyte response at the lower limit of quantitation, and 5 percent of the response for the internal standard. That is a convention for that setting. It is not a universal limit for every method.

How do you reduce carryover in an HPLC system?

Common measures are a stronger needle wash solvent, a larger wash volume or more wash cycles, washing before and after the injection, backflushing the flow path, and replacing worn rotor seals and needle seats. A blank injection placed after a high concentration sample does not remove the source, but it protects the next sample by taking up some of the residue. Each change should be confirmed by repeating the standard-then-blank test.

Is a peak in a blank a sample impurity?

No. A peak in a blank injection cannot be an impurity of the sample, because no sample was injected. The same peak appearing in a sample chromatogram must be treated as a system or preparation artefact until shown otherwise, and it should not be reported as a sample impurity.