HPLC troubleshooting works best as a controlled, evidence-led investigation: first define the observed symptom, protect the original data, check the system and method conditions in a fixed order, and document the cause before making changes. This guide helps pharmaceutical QC analysts and method developers investigate common chromatographic problems without turning routine testing into unapproved method adjustment.
Use this guide as an educational aid. It is not an approved procedure, regulatory advice, or a substitute for the instrument manual, validated analytical procedure, laboratory SOP, or Quality Unit direction. In a GMP laboratory, follow approved procedures and change control. If a result may be out of specification (OOS), follow the laboratory’s formal investigation process and preserve all original data.
HPLC troubleshooting: a controlled first-response workflow
In HPLC troubleshooting, distinguish planned method development from testing with an approved procedure before adjusting flow, wavelength, integration, mobile-phase composition, or column conditions. Development experiments should follow an approved protocol where required. For routine GMP testing, do not change a validated method to obtain a passing result; document the observation and use the applicable deviation or laboratory investigation process.
- Describe the signal. Record what changed: system suitability failure, pressure, baseline, retention, peak shape, response, missing peak, carryover, or calibration behaviour. Compare with the method’s predefined criteria and recent valid runs.
- Secure the evidence. Retain the complete sequence, raw data, chromatograms, audit-trail information, calculations, instrument messages, sample and standard preparations, and contemporaneous notes under site procedures. Do not delete, overwrite, selectively reintegrate, or conceal an unexpected result.
- Check simple causes first. Verify method version, sequence and vial positions, sample identity, preparation calculations, solution age, mobile-phase identity and preparation, column identity, instrument status, leaks, and system suitability.
- Change one factor at a time. For method development, make controlled, recorded changes within the approved experimental plan. In routine testing, obtain required authorization before any procedural change and assess impact through change control.
- Confirm the cause. Use a justified check or repeat experiment only when permitted by the procedure or investigation plan. A symptom disappearing after an adjustment does not by itself establish root cause or invalidate prior data.
- Record the outcome. Document the evidence, impact assessment, corrective action, and approval. Escalate unresolved issues to the laboratory supervisor, instrument service provider, or Quality Unit as appropriate.
HPLC troubleshooting by symptom
HPLC troubleshooting: poor resolution or overlapping peaks
Check: peak identity and elution order, system suitability, column chemistry and history, mobile-phase composition and pH, gradient programme, temperature, flow rate, injection solvent, sample load, and extra-column tubing. Confirm that the right method and column configuration were used.
In method development: use a predefined design to study selectivity-driving variables such as stationary phase, organic modifier, pH where analyte chemistry allows, gradient slope, and temperature. Evaluate resolution and critical-pair behaviour against method-specific goals. In routine analysis: do not alter the approved method to force separation. Investigate setup or system causes, document the result, and follow the controlled procedure.
2. Tailing, fronting, or split peaks
Review sample concentration and injection volume, sample solvent strength relative to the initial mobile phase, sample pH and solubility, column condition, fittings and connections, guard-column or inlet-frit blockage, and the method’s integration settings. Tailing can also reflect analyte interactions with active sites or unsuitable pH; fronting can be consistent with overload, among other causes. A split peak can arise from injection, sample, connection, or column problems, so use the chromatogram and method context rather than assuming a single cause.
Compare a suitable standard and sample only when the method or investigation permits. Inspect for leaks or voids only using approved procedures. Do not reverse-flush, wash, or replace a column unless its instructions and site procedure allow it; record column use and any intervention.
3. Noisy, drifting, or unstable baseline
Check solvent grade, mobile-phase preparation, filtration requirements, degassing, solvent age, reservoir level, proportioning and mixing, pump delivery, leaks, detector flow cell, lamp status, wavelength, temperature stability, and whether the baseline behaviour is expected for the gradient. Bubbles, contamination, inadequate equilibration, or detector issues can contribute, but the investigation should use the instrument diagnostics and procedure.
Compare the baseline with a blank or mobile-phase run if permitted. Prepare fresh mobile phase only under the approved method and document the preparation. Follow the manufacturer’s instructions for detector maintenance and flow-cell cleaning; do not use an improvised cleaning solvent or procedure.
4. Retention-time shift or drift
Confirm equilibration time and initial composition, gradient table, flow rate, column temperature, mobile-phase proportions and pH, pump mixing and delivery, leaks, column history, and the correct column. Check whether the shift affects all peaks similarly or is selective; this pattern can help distinguish flow or composition changes from analyte-specific or column-related effects. Compare system suitability and reference-standard behaviour with established criteria.
Do not add an internal standard or change retention windows as an ad hoc correction. Use only the approved method’s identification and calculation rules. If the method is being developed or revised, assess retention variability in the planned study and justify the final procedure.
5. High, low, or fluctuating backpressure
For HPLC troubleshooting involving abnormal pressure, compare the trend with the method and instrument history, mobile-phase viscosity and temperature, inline filters, guard column, analytical column, tubing, fittings, and pump operation. A sudden increase may indicate a restriction; a sudden decrease may reflect a leak, empty reservoir, loss of prime, or a connection issue. Localize the source using the instrument manufacturer’s approved procedure.
Safety: HPLC flow paths operate under pressure. Stop the run and depressurize as instructed by the instrument manual and site safety procedure before inspecting or disconnecting fittings. Never loosen a pressurized connection.
6. Missing peaks, unexpected peaks, or carryover
Verify sample and vial identity, injection sequence, preparation, analyte stability, detector wavelength or settings, integration events, and whether the expected peak is within the validated reportable range. For unexpected or “ghost” peaks, assess blanks, solvent and vial cleanliness, carryover, sample contamination, mobile-phase contamination, and the prior injection sequence.
Use blank injections or a wash programme only as written or authorized. Do not remove a peak from processing solely because it is inconvenient. Assess all relevant peaks and retain the original processing and any authorized reprocessing history.
7. Irregular calibration, recovery, or response
Review reference-standard identity, potency or correction factors, weighing and dilution records, volumetric equipment, solution stability, mixing, pipette technique, sequence design, detector range, injection precision, and the specified calibration model and range. Confirm that calculations, units, weighting, and integration rules match the approved procedure. Avoid forcing a calibration model or excluding a point after seeing the results unless the procedure and a documented scientific rationale permit it.
Low response can also reflect sample loss, adsorption, degradation, detector settings, or a system issue. Compare standards, controls, and samples as defined in the protocol. If precision or recovery criteria fail, stop and investigate before using the data for a reportable decision.
System suitability for HPLC troubleshooting and OOS results
System suitability criteria are part of the procedure’s defined controls. If a suitability requirement fails, follow the procedure’s instructions for stopping, troubleshooting, and documenting the event; do not proceed to report sample results as valid unless the procedure and investigation support that decision. FDA’s CGMP laboratory-controls Q&A states that suitability checks should use qualified reference materials appropriate to the method and that all data, including failing, passing, and suspect data, belong in the CGMP record and need appropriate review.
An OOS result is not resolved by repeating tests until one passes. FDA’s OOS guidance describes a scientifically sound investigation that considers laboratory and production causes, with retesting or resampling justified and controlled by the investigation. Follow your approved OOS procedure, preserve the original result, and involve the Quality Unit. This article is not a substitute for that procedure.
Method development versus routine GMP testing
Method development is an intentional study to select and understand procedure conditions. Routine testing applies the approved analytical procedure to make a defined quality decision. A change that is reasonable during development may be unacceptable during routine testing without assessment and authorization. When development findings lead to a procedure change, document the scientific rationale, evaluate performance and validation impact, use formal change control, and obtain required approvals before implementation.
FDA’s ICH Q14 Analytical Procedure Development guidance describes science- and risk-based approaches across the analytical procedure lifecycle. FDA guidance represents the agency’s current thinking and is generally nonbinding; applicable statutes and regulations remain binding.
HPLC troubleshooting record checklist
- Observed symptom, date and time, sequence and injection, method and version, instrument ID, column ID and use history.
- Sample and standard identities, preparation worksheets, solution age, mobile-phase lots and preparation details.
- System suitability results, raw chromatograms, processing method, integration events, calculations, audit-trail records, and instrument messages.
- Checks performed, one-factor changes or interventions, applicable authorization, and the rationale for each step.
- Assessment of affected samples, batches, prior data, method performance, and possible product impact.
- Root-cause evidence or a clear statement that the cause remains undetermined; corrective actions, approvals, and follow-up.
Frequently asked questions
What should I check first when an HPLC run fails?
Check the observed failure against the method’s system suitability criteria, confirm the method version and sequence, and review sample and mobile-phase preparation, instrument status, column identity, and the complete data record. Follow the approved troubleshooting and deviation procedure before changing conditions.
Can I change the mobile phase during routine testing to fix poor resolution?
Not as an undocumented adjustment. A GMP test must follow the approved method. Record the issue, investigate using the procedure, and route any proposed method change through the required scientific assessment, change control, and approval.
What commonly causes retention-time drift?
Potential contributors include insufficient equilibration, changes in mobile-phase composition or pH, flow or mixing problems, temperature variation, leaks, and column changes or deterioration. Check the pattern of affected peaks and compare with system suitability and historical runs before concluding the cause.
How do I investigate an HPLC OOS result?
Use the site’s approved OOS investigation procedure and FDA’s current OOS guidance where applicable. Preserve the original data, assess potential laboratory and manufacturing causes, and justify any retest or resampling through the investigation. Do not test into compliance.
Does a passing system suitability result prove that every sample result is valid?
No. System suitability evaluates defined aspects of procedure or system performance. Sample preparation, identity, stability, calculation, chromatographic specificity, and other method controls still require review under the approved procedure.
Authoritative references and related HPLC guides
- FDA: ICH Q14, Analytical Procedure Development — nonbinding, risk-based development and lifecycle guidance.
- FDA: Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production — current guidance for OOS investigations.
- FDA: Questions and Answers on CGMP Requirements — Laboratory Controls.
- 21 CFR 211.160 — binding US CGMP requirements for laboratory controls.
- 21 CFR 211.194 — binding US CGMP requirements for laboratory records.
- FDA: Pharmaceutical Quality Control Labs inspection guide — inspection perspective on laboratory procedures, equipment, raw data, and investigations.
Continue with our HPLC method transfer guide, HPLC method validation article, HPLC column selection guide, and mobile-phase selection guide.
