Ion Suppression in LC–MS: Causes, Diagnosis and Solutions

What ion suppression is — and what it is not

A matrix effect is any change in analyte response caused by other components of the sample. The response may decrease (suppression) or increase (enhancement). The term is therefore broader than ion suppression. In LC–ESI–MS, suppression is the most familiar expression because ESI turns a continuously changing liquid mixture into a population of charged droplets. If the composition of those droplets changes, ion yield changes with it.

Mechanistic work by King and co-workers showed that suppression depends strongly on the concentration and surface activity of co-eluting compounds, not simply on whether an interference shares the analyte’s mass-to-charge ratio [1]. Annesley’s practical review made the central laboratory point explicit: salts, ion-pair reagents, endogenous substances, drugs, metabolites and proteins can suppress response even when the selected reaction monitoring trace appears selective [2].

Ion suppression is not the same as low recovery. Recovery describes how much analyte survives extraction. Suppression describes how the remaining analyte responds in the source. A method can have 90% recovery and severe suppression, or 45% recovery and almost no matrix effect. Those are different failures and require different corrections.

Why ESI is vulnerable

In ESI, the LC effluent is nebulised into charged droplets. Solvent evaporation raises charge density until repeated droplet fission and ion release produce gas-phase ions. Matrix components can interfere at several stages:

• Surface competition: surface-active molecules occupy the region from which ions are preferentially emitted.
• Charge competition: abundant ionic species take a disproportionate share of available charge.
• Changed physical properties: salts and endogenous components alter conductivity, surface tension, viscosity and evaporation.
• Gas-phase chemistry: proton-transfer equilibria can favour a competing component over the target analyte.
• Source loading: too much total mass enters the source, so small changes in injection volume or matrix concentration cause nonlinear response.

APCI or APPI can be less susceptible for suitable compounds because ionisation occurs mainly in the gas phase, but switching source type is not a universal repair. Polarity, volatility, thermal stability and achievable sensitivity still determine whether the alternative is appropriate [9].

The matrix components that most often cause trouble

Phospholipids deserve special attention in plasma and serum. Protein precipitation removes proteins but leaves many phospholipids in solution. Studies comparing extraction strategies consistently show that cleaner extracts reduce matrix effects more reliably than a fast crash alone [5, 8].

How to measure ion suppression correctly

1. The three-set experiment

The most useful quantitative design separates matrix effect, recovery and overall process efficiency. Matuszewski and co-workers formalised the comparison using three sample sets [3]:

• Set A — neat standard: analyte in the final reconstitution solvent.
• Set B — post-extraction spike: blank matrix is extracted first, then analyte is added.
• Set C — pre-extraction spike: analyte is added to matrix before the entire extraction.

An MF of 1.00 means no net matrix effect, 0.70 means 30% suppression, and 1.20 means 20% enhancement. For quantitative work, calculate the internal-standard-normalised matrix factor:

The normalised value matters because a stable-isotope-labelled internal standard can compensate even when both analyte and internal standard are suppressed. What it cannot correct reliably is differential suppression caused by imperfect co-elution, isotope effects, cross-talk, insufficient isotopic purity or concentration-dependent source competition.

2. Post-column infusion

Post-column infusion answers a different question: when does the matrix alter ionisation? A fixed analyte solution is infused into the LC stream after the analytical column while an extracted blank matrix is injected. The infused signal should be constant. Any depression or increase maps a matrix-effect zone in retention time. Stahnke and co-workers used post-column infusion to compensate matrix effects in multiresidue analysis [6], and González et al. later demonstrated its value as an ongoing QC tool rather than a one-time development experiment [11].

3. Relative matrix effect across matrix lots

A pooled matrix can hide the real problem: variability between individuals, species, food commodities or environmental sites. Evaluate post-extraction spikes from independent lots at low and high concentration. For regulated bioanalysis, ICH M10 defines the matrix effect as altered analyte response caused by interfering matrix components and calls for testing low and high QCs from at least six independent sources or lots, with case-specific assessment of haemolysed or lipaemic matrix where relevant [ICH M10 / FDA].

4. Dilution and standard addition

If a two-, five- and ten-fold dilution produces a disproportionately larger analyte response after correction for dilution, the original extract is outside the robust source-loading range. Dilution is therefore both a diagnostic and a possible solution. Standard addition is useful when a blank matrix is unavailable, but it increases workload and does not remove the underlying source contamination or sensitivity loss.

A systematic troubleshooting workflow

Step 1: prove that the problem is matrix-dependent

Compare a neat standard, a post-extraction spike and a pre-extraction spike at the same nominal concentration. If the post-extraction spike is already low, the problem is ionisation or another post-extraction effect. If only the pre-extraction spike is low, investigate recovery, adsorption, degradation or incomplete release from the matrix.

Step 2: locate the suppression window

Run post-column infusion with at least one representative matrix lot. Overlay the analyte retention time. If the analyte sits in a deep suppression zone, the most powerful correction is usually to move one of them: retain the analyte more strongly, change stationary-phase selectivity, alter gradient timing, or remove the interfering class before injection.

Step 3: change one variable at a time

Do not simultaneously replace the SPE cartridge, change the column, halve the injection volume and raise source temperature. A better peak would not reveal which change worked. Use a short experimental sequence:

1. Dilute the final extract while keeping the injected analyte amount known.
2. Reduce injection volume.
3. Compare the current preparation with a more selective cleanup.
4. Move the analyte away from the suppression zone chromatographically.
5. Test an alternative ionisation mode only if analyte chemistry supports it.

How to reduce ion suppression

Sample preparation: remove mass before it reaches the source

Cleaner extracts normally give the largest robustness gain. Protein precipitation is fast and often acceptable for high-concentration analytes, but it is not a lipid-removal step. Liquid–liquid extraction can be highly selective when the analyte has a suitable neutral form and partition coefficient. Polymeric reversed-phase SPE is broader, while mixed-mode cation- or anion-exchange SPE can separate ionisable analytes from neutral matrix more aggressively. Dedicated phospholipid-removal products are useful for plasma and serum, but recovery must be checked for every analyte class rather than assumed from a product claim.

Chambers et al. compared multiple preparation and chromatographic strategies and showed that reduction of co-extracted matrix is a central part of controlling LC–MS/MS matrix effects [5]. Ahmad et al. reported substantially reduced phospholipid-related matrix effects using a selective phospholipid-removal workflow [8].

Chromatography: move the analyte out of the suppression zone

MS/MS selectivity does not prevent ionisation competition. Chromatographic resolution still matters. Increase retention of very early peaks, because the void region often contains salts and other polar co-extractives. Change stationary-phase chemistry when the analyte and suppressor move together under every gradient change. A higher-efficiency UHPLC separation can narrow peaks and reduce overlap, but a faster method can also compress more matrix into a shorter time window; speed alone is not the objective.

Mobile phase and source conditions

• Use volatile LC–MS-grade buffers such as ammonium formate or ammonium acetate at the lowest concentration that still gives acceptable chromatography.
• Avoid non-volatile salts entering the source. Divert the void and late wash to waste where the instrument configuration permits.
• Screen formic acid, acetic acid and buffer identity rather than treating them as interchangeable.
• Use trifluoroacetic acid cautiously in positive ESI; it can strongly reduce response for many analytes.
• Optimise desolvation temperature, gas flow and spray voltage after the LC and sample preparation are stable. Source tuning cannot rescue an overloaded, matrix-rich spray indefinitely.

Internal standards and calibration

The preferred internal standard is a stable-isotope-labelled analogue added before extraction and co-eluting with the analyte. A structural analogue may correct recovery but experience a different local matrix effect. Matrix-matched calibration compensates average bias but does not guarantee that an unusual patient, food or environmental sample behaves like the calibrator. The final method must therefore combine compensation with direct assessment of relative matrix effects [13, 14].

Analyte-specific starting points

Materials and columns: direct manufacturer links

These are examples for method screening, not universal endorsements. Select format, sorbent mass, particle size, dimensions and pH range for the analyte, matrix, loading capacity and instrument pressure limit.

Worked example: a plasma assay that passes standards and fails samples

A basic drug gives a sharp peak and excellent calibration in solvent. In extracted plasma the analyte/internal-standard ratio varies by donor, the low QC is biased low, and repeat injections gradually lose sensitivity.

1. Three-set experiment: recovery is 82%, but the post-extraction matrix factor ranges from 0.48 to 0.83 across six donors. The primary failure is variable suppression, not extraction loss.
2. Post-column infusion: the analyte elutes on the leading edge of a broad suppression region. A phosphocholine trace indicates substantial lipid-related material in the same window.
3. Dilution test: five-fold dilution improves the dilution-corrected response, confirming excessive matrix load.
4. Cleanup test: a phospholipid-removal workflow greatly reduces the suppression region, while a conventional protein crash does not.
5. Chromatographic test: a stationary-phase change shifts the analyte away from the remaining suppression minimum.
6. Final control: a co-eluting isotope-labelled internal standard and matrix-matched calibration reduce residual bias. Low and high QCs pass across independent plasma lots.

The important point is the order. The method first identifies the mechanism and retention-time overlap, then removes and separates the cause, and only then relies on compensation.

Common mistakes

• Trusting a clean MRM trace: non-isobaric matrix components can suppress the target without appearing in the target transition.
• Calling low response “poor recovery”: recovery and matrix factor must be measured separately.
• Testing one matrix pool: relative matrix effects are often more damaging than average suppression.
• Assuming the internal standard fixes everything: it only compensates reliably when it experiences the same local effect.
• Shortening the gradient without a suppression map: faster LC can compress matrix into the analyte window.
• Using more additive automatically: stronger buffering may improve peak shape while reducing MS response or adding source load.
• Optimising the source before cleanup: tuning can temporarily increase signal without making the method robust.

Action checklist

• Run neat, post-extraction-spiked and pre-extraction-spiked samples.
• Calculate analyte and IS-normalised matrix factors.
• Evaluate at least low and high levels across independent matrix lots.
• Map suppression by post-column infusion during development.
• Check whether the analyte elutes in the void, lipid window or wash region.
• Compare dilution, lower injection load and a cleaner extraction.
• Screen an orthogonal stationary-phase chemistry when gradient changes move analyte and matrix together.
• Use volatile LC–MS-grade solvents and additives.
• Add the stable-isotope-labelled internal standard before extraction whenever possible.
• Recheck the matrix effect after every meaningful change in matrix, anticoagulant, collection tube, sample preparation, column or source.

Limitations

A matrix factor is method-specific. It depends on analyte concentration, matrix lot, extraction, injection volume, LC conditions, source design and tuning. A value measured on one platform cannot be transferred as a universal property of the analyte. Post-column infusion maps the net effect of matrix at each retention time but does not identify every responsible compound. Phospholipid monitoring is useful in plasma, yet phospholipids are not the only suppressors. Finally, compensation can improve quantitative accuracy while the absolute loss of sensitivity remains; this still matters near the LLOQ.

FAQ

Can ion suppression occur when the chromatogram is clean?

Yes. The suppressor does not need to produce the monitored precursor/product transition. It only needs to co-elute and alter ionisation.

Is protein precipitation enough for plasma?

Sometimes, especially for abundant analytes and modest sensitivity requirements. But it does not reliably remove phospholipids. Demonstrate adequacy experimentally with matrix factors and a suppression profile.

Does a stable-isotope-labelled internal standard eliminate ion suppression?

No. It can compensate residual suppression if it co-elutes and behaves like the analyte, but it does not restore lost absolute sensitivity or prevent source contamination.

Should I switch from ESI to APCI?

Only if the analyte is compatible with APCI. Source switching is a chemistry decision, not a generic troubleshooting step.

Is dilution a valid solution?

Yes, when sensitivity permits and dilution integrity is demonstrated. Dilution often reduces matrix faster than it harms quantification, but it raises the effective LLOQ.

What is the best single test?

There is no single test. Post-extraction spiking quantifies the effect; post-column infusion locates it. Together they are far more informative than either alone.

References

1. King RC, Bonfiglio R, Fernandez-Metzler C, Miller-Stein C, Olah TV. Mechanistic investigation of ionization suppression in electrospray ionization. J Am Soc Mass Spectrom. 2000;11(11):942–950. doi:10.1016/S1044-0305(00)00163-X.
2. Annesley TM. Ion suppression in mass spectrometry. Clin Chem. 2003;49(7):1041–1044. doi:10.1373/49.7.1041.
3. Matuszewski BK, Constanzer ML, Chavez-Eng CM. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC–MS/MS. Anal Chem. 2003;75(13):3019–3030. doi:10.1021/ac020361s.
4. Taylor PJ. Matrix effects: the Achilles heel of quantitative high-performance liquid chromatography–electrospray–tandem mass spectrometry. Clin Biochem. 2005;38(4):328–334. doi:10.1016/j.clinbiochem.2004.11.007.
5. Chambers E, Wagrowski-Diehl DM, Lu Z, Mazzeo JR. Systematic and comprehensive strategy for reducing matrix effects in LC/MS/MS analyses. J Chromatogr B. 2007;852(1–2):22–34. doi:10.1016/j.jchromb.2006.12.030.
6. Stahnke H, Kittlaus S, Kempe G, Alder L. Compensation of matrix effects by postcolumn infusion of a monitor substance in multiresidue analysis with LC–MS/MS. Anal Chem. 2009;81(6):2185–2192. doi:10.1021/ac802362s.
7. Trufelli H, Palma P, Famiglini G, Cappiello A. An overview of matrix effects in liquid chromatography–mass spectrometry. Mass Spectrom Rev. 2011;30(3):491–509. doi:10.1002/mas.20298.
8. Ahmad S, Kalra H, Gupta A, Raut B, Hussain A, Rahman MA. HybridSPE: a novel technique to reduce phospholipid-based matrix effect in LC–ESI–MS bioanalysis. J Pharm Bioallied Sci. 2012;4(4):267–275. doi:10.4103/0975-7406.103234.
9. Peters FT, Remane D. Aspects of matrix effects in applications of liquid chromatography–mass spectrometry to forensic and clinical toxicology — a review. Anal Bioanal Chem. 2012;403:2155–2172. doi:10.1007/s00216-012-6035-2.
10. Furey A, Moriarty M, Bane V, Kinsella B, Lehane M. Ion suppression: a critical review on causes, evaluation, prevention and applications. Talanta. 2013;115:104–122. doi:10.1016/j.talanta.2013.03.048.
11. González O, Dubbelman AC, Hankemeier T. Postcolumn infusion as a quality control tool for LC–MS-based analysis. J Am Soc Mass Spectrom. 2022;33(6):1077–1080. doi:10.1021/jasms.2c00022.
12. Zhu P, Dubbelman AC, Hunter C, et al. Development of an untargeted LC–MS metabolomics method with postcolumn infusion for matrix-effect monitoring in plasma and feces. J Am Soc Mass Spectrom. 2024;35(3):590–602. doi:10.1021/jasms.3c00418.
13. Fu Y, Li W, Picard F. Assessment of matrix effect in quantitative LC–MS bioanalysis. Bioanalysis. 2024;16(12):631–634. doi:10.4155/bio-2024-0047.
14. Zhou W, Yang S, Wang PG. Matrix effects and application of matrix effect factor. Bioanalysis. 2017;9(23):1839–1844. doi:10.4155/bio-2017-0214.
15. International Council for Harmonisation. ICH M10: Bioanalytical Method Validation and Study Sample Analysis. 2022. Official guideline PDF.

Editorial note. Product and column links point directly to manufacturer pages for reader convenience. They are examples for method screening, not proof that a product is suitable for a particular analyte. No published figure has been copied; all diagrams in this article are original reconstructions of the underlying analytical concepts.