How should carryover and ghost peak troubleshooting in peptide HPLC analysis be structured?
Start with the observation, not the suspected part. Record retention time, relative retention time, area, area percent, a short peak-shape note, wavelength, and reporting channel. State where the signal is seen: diluent blank, mobile-phase gradient blank, unused vial of the same container lot, only after a concentrated peptide cycle, or in the sample alone. Carryover is material retained from an earlier cycle that returns later, usually at a retention time already observed. A ghost peak is a signal that does not come from the nominal contents of the solution under test. The classes meet when a strongly retained peptide from an earlier gradient returns in a later run and is mistaken for an unknown related substance. They do not meet when the signal is already present in a blank that has never contacted that peptide. Raval and colleagues reviewed commonly observed HPLC troubleshooting problems, which is a reminder to file stray peaks with baseline, pressure, and retention faults instead of forcing a single cause (DOI:10.21276/ijprhs.2020.04.02). Educational notices on the basic principles of HPLC and HPLC system troubleshooting make the system point: the trace combines solvent delivery, sample introduction, the column, and the detector, so the check has to follow that order (DOI:10.1080/10739149908085875; DOI:10.1080/10739140009709740). Peptide methods tighten the file. Hydrophobic sequences and lingering ion-pair reagent can leave a small memory peak, and 214 nm detection will record amide absorbance from analyte and from some contaminants. The file should name method identity, column identity, gradient end composition and hold, needle-wash solvents, vial and cap type, any autosampler temperature setpoint, and the full order of cycles. Without it, an unidentified row on a certificate of analysis cannot be assigned to the lot or to the sequence.
Which blank and sequence tests separate carryover from a sample peak?
Use a sequence, not one chromatogram. After unreported equilibration cycles, run a diluent blank in the sample vial type, a reference or retention marker, an immediate blank, the research samples, a mid-sequence blank, and a closing blank after the most concentrated or most hydrophobic solution. Where the method allows, a zero-volume gradient shows peaks that belong to the solvent timetable rather than to the autosampler. A peak that exists before any peptide has entered the system is not carryover of that peptide. A peak absent from the opening blank and present only after a high-concentration cycle, at the same retention time, makes carryover the leading hypothesis. Define the arithmetic in the SOP. A usual figure is blank-peak area at the analyte retention time divided by the preceding sample-peak area, as a percentage. Laboratories may also compare that blank area with the area that matches the purity method's reporting threshold. A failed carryover check means later purity statements are not reportable. It does not invent a related-substance result for the lot. Repeat the blank after a longer high-organic column wash, and separately after extra needle-wash cycles. Shrinkage only after the longer column wash points to column memory or a late eluter. Shrinkage after the needle and seat wash, with the column wash unchanged, points to the autosampler path. Match every blank to the sample diluent, fill volume, vial, cap, and tray residence time. Water does not control an acidified organic diluent. A fresh vial does not control a solution that has waited. For hydrophobic peptides the closing blank is often the informative one, because adsorption and incomplete elution accumulate. Keep the sequence table, vial identities, and method identity with the batch report so a reviewer can see which blank followed which sample.
Where in the fluid path do peptide carryover and late eluters arise?
More than one part can hold peptide. Needle exteriors transfer solution from the septum into the seat or the next cap. Needle interiors, loops, and unswept fittings hold a film of hydrophobic sequence when the wash solvent is weaker than the diluent. A worn rotor seal leaks a trace across ports. Metering syringes repeat the problem when a concentrated reference is followed by a blank without enough wash volume and enough wash cycles. The method should state wash composition, volume, and whether the wash is external, internal, or both. Rescaling the chromatogram does not remove retained peptide. The column is the other reservoir. If the gradient ends at too little organic, or holds there too briefly, strongly retained material remains for the next run. It may return near the analyte or as a late peak that a short run never displays. A longer high-organic segment, followed by a defined return to the starting composition, is a chromatographic control. Guards store the same residue; replacing the guard and repeating the blank is a direct test. A column kept for strongly retained research sequences reduces cross-talk when unrelated peptides share an instrument. Column identity and installation date belong in the batch file. Proportioning can imitate carryover. Low-pressure mixing errors and contaminated lines produce peaks in every gradient, including blanks. Those peaks follow the timetable, not the previous vial. A mixer, degasser, or filter that has seen concentrated peptide can release material slowly. The check is a no-sample gradient after a column wash, compared with a fresh solvent batch. Flow-cell fouling is less often the cause in purity work, but a baseline that stays high after the column is bypassed is a hardware note. These checks do not alter the lot. They decide whether its chromatogram can be read.
How can vial leachates and diluent chemistry create a ghost peak?
Lin and colleagues described a ghost peak of clofazimine as a solution degradation product formed by nucleophilic substitution with nitrite leaching from certain glass HPLC vials (DOI:10.1016/j.jpba.2017.11.052). That finding is specific to clofazimine. It is not a peptide method, and it does not show that research peptides undergo the same reaction. The transferable point is that an artefact can form in the vial after preparation, from material leached by the container. Washing the column will not stop a peak that each waiting vial creates again. Split one solution across the routine glass vial, another glass lot, and a polymer vial if the method allows. Analyse one portion at once and another after the residence time the sequence will use. A peak that grows with time in one container only is a container or solution artefact. A peak that is immediate and the same in every container fits the sample or the diluent better. The blank must match container, cap, septum, and time. Septum fragments and cap extractables can add signals that a mobile-phase blank never shows. Peptide solutions add risks that are not the nitrite reaction above. Basic residues can bind untreated glass, so an early withdrawal is low and a later wash releases material that looks like carryover. Acidified diluents, organic co-solvents, and ion-pair reagents change solubility and what the glass releases. Polymer vials can lower glass adsorption and still add extractable peaks at 214 nm. Compare containers under the actual method and write the chosen container into the sample-preparation SOP. Do not copy a vial from an unrelated assay. If an aged solution grows new peaks, the certificate should separate those solution-stability artefacts from impurities measured in a fresh analytical solution. That note stays in the analytical record. It records container residence time and solution history beside the peak table so a later reviewer can separate an aged-vial signal from a peak measured in a freshly prepared analytical solution.
How should certificates of analysis treat carryover, ghost peaks, and weak citations?
Area-normalised purity moves when an artefact is kept or dropped. Counting a blank ghost peak as a related substance lowers the main-peak percentage without a lot cause. Dropping a real related substance because it was called carryover raises that percentage. Set the rule before transcription. Sample peaks that match a representative blank in retention time and approximate area may be marked blank-related and excluded only when the SOP allows that comparison and the blank is matched. Peaks above the blank, or absent from it, stay in the table as specified impurities or unidentified peaks, with relative retention time and area percent. A carryover peak only in a post-sample blank is a system-suitability failure, not an impurity of the preceding vial. The peak table and the chromatogram must show the same times, areas, and labels. Manual deletion, a change from drop-baseline to skim, or a higher threshold that hides a small peak needs a reason code and a reviewer. Solvent-front signals should be named so they are not summed as related substances. A short run that cuts off late peaks is not comparable with a run that holds at high organic. A signal at 214 nm that is absent at 280 nm may be non-aromatic, a solvent artefact, or a peptide without tyrosine or tryptophan. If 214 nm is the reporting channel, that pattern does not justify omitting it. Australian research readers should be able to see method identity, date, reviewer, sample name versus lot, vial identity, sequence position, blank identifiers, reporting threshold, and exclusions. If carryover stays above the written limit, report that purity was not reportable. Do not smooth the percentage. Citations need the same restraint. A 1991 Toxicon item titled Troubleshooting in der HPLC marks HPLC troubleshooting as a published topic; it does not supply a peptide limit (DOI:10.1016/0041-0101(91)90094-8). A 1997 education announcement on basic principles of HPLC and HPLC system troubleshooting is a training notice, not a results paper (DOI:10.1080/10739149709351479). Mahanta and colleagues used Canberra distance on a ghost-peak problem in power-analysis peak distribution (DOI:10.4018/ijisp.2018070103). That ghost peak is not chromatographic, and the distance metric is not an integration parameter. Header, vial, and lot identifiers must still match, and a failed carryover check must remain visible. The compiled file is a research chromatography record and is not for human use.
Frequently asked questions
What is the difference between HPLC carryover and a ghost peak?
Carryover is material left in the autosampler, valve, loop, or column that returns in a later cycle, usually at a retention time already seen. A ghost peak is a signal that does not come from the nominal contents of the solution being measured. Sources include the gradient timetable, container leachate, cap extractables, and late eluters. They overlap only if retained peptide is mistaken for a new impurity. A signal already present in a matched blank is not carryover of that peptide.
Should a blank ghost peak be included in peptide HPLC purity?
No. A peak that is fully accounted for by a matched blank is a system or container signal, not a related substance of the lot, provided the SOP allows blank comparison and the blank matches diluent, vial, cap, and time. Including it lowers area-normalised main-peak percentage without a lot cause. A sample peak larger than the blank, or absent from the blank, stays in the impurity table. Carryover seen only after a sample is a system-suitability failure and makes later purity figures not reportable.
Can a glass HPLC vial create a ghost peak during peptide analysis?
A published case showed a clofazimine ghost peak as a solution degradation product formed by nucleophilic substitution with nitrite leaching from certain glass HPLC vials. That chemistry is not evidence of the same reaction in peptides. It does show that container leachate can create a peak after preparation. For peptide work, compare glass lots and a polymer vial, and compare a fresh portion with a portion held for the sequence time. Report container artefacts separately from impurities in a freshly prepared analytical solution.
What should a research certificate of analysis show about carryover?
A readable HPLC package should identify the method, date, reviewer, sample name, lot, vial, and sequence position, and it should identify the blanks. It should state the reporting threshold and which peaks were excluded as blank-related. The chromatogram and peak table should agree. If carryover exceeds the laboratory's written limit, the report should say purity was not reportable for that sequence rather than present a smoothed percentage. These fields support research documentation only. The package is an analytical record and is not for human use.
Does every paper about a ghost peak apply to peptide HPLC?
No. Database searches mix chromatographic ghost peaks with unrelated uses of the same words. One provided source addresses a ghost-peak problem in power-analysis peak distribution using Canberra distance. That work is information security, not chromatography, and the distance metric is not an HPLC integration setting. Peptide investigations stay with matched blanks, container comparisons, fluid-path washes, and audit-trailed integration. Educational notices on HPLC system troubleshooting support training structure; they do not by themselves set a peptide carryover limit.
Why can a late-eluting peptide look like a new impurity in the next run?
If the gradient does not reach enough organic, or does not hold there long enough, strongly retained peptide remains on the column. The next gradient can elute it near the main peak or as an unidentified late peak. An opening blank run before any sample will not show that memory; a blank immediately after the concentrated solution will. Extending the high-organic segment and repeating the blank distinguishes column memory from a vial artefact. The run time used for purity reporting should be long enough to display those late signals.
References
- DOI:10.21276/ijprhs.2020.04.02 — Review on Common Observed HPLC Troubleshooting Problems — International Journal of Pharma Research and Health Sciences — 2020
- DOI:10.1080/10739149908085875 — Basic Principles of HPLC and HPLC System Troubleshooting — Instrumentation Science & Technology — 1999
- DOI:10.1080/10739140009709740 — BASIC PRINCIPLES OF HPLC AND HPLC SYSTEM TROUBLESHOOTING — Instrumentation Science & Technology — 2000
- DOI:10.1016/j.jpba.2017.11.052 — “Ghost peak” of clofazimine: A solution degradation product of clofazimine via nucleophilic substitution by nitrite leaching from certain glass HPLC vials — Journal of Pharmaceutical and Biomedical Analysis — 2018
- DOI:10.1016/0041-0101(91)90094-8 — Troubleshooting in der HPLC — Toxicon — 1991
- DOI:10.1080/10739149709351479 — Education Announcement: Basic Principles of Hplc and Hplc System Troubleshooting — Instrumentation Science & Technology — 1997
- DOI:10.4018/ijisp.2018070103 — Improving Power Analysis Peak Distribution Using Canberra Distance to Address Ghost Peak Problem — International Journal of Information Security and Privacy — 2018
Research use only
This article is provided for laboratory research and educational purposes only. Products referenced are not for human or veterinary use. ClaraScience makes no therapeutic, medical, or efficacy claims, and nothing here constitutes medical advice.
