LC-MS sample preparation methods and when to use them

LC-MS sample preparation covers the steps used to make a real sample compatible with liquid chromatography-mass spectrometry while preserving the analyte signal. The right approach depends on the matrix, analyte chemistry, required detection limit, throughput, and regulatory context. A dilute-and-shoot workflow may be enough for simple aqueous samples. Plasma, food extracts, environmental samples, and complex biological matrices often need protein removal, liquid-liquid extraction, solid-phase extraction, phospholipid cleanup, derivatization, or a combination of these steps.

The goal is not to make every extract perfectly clean. It is to produce a reproducible extract that meets the method’s purpose without creating avoidable ion suppression, carryover, instability, or sample loss. You can also explore more in sample preparation.

For more related workflows and laboratory content, see the sample preparation section.

Why sample preparation controls LC-MS data quality

LC-MS instruments are highly selective, but they are still affected by the sample matrix. Salts, proteins, phospholipids, surfactants, plasticizers, endogenous metabolites, buffer components, and co-extracted compounds can alter chromatographic retention, electrospray ionization, source cleanliness, and detector response. In quantitative LC-MS/MS, this often appears as ion suppression or ion enhancement. In untargeted analysis, matrix effects can also distort feature abundance, reduce coverage, and complicate compound identification.

Regulatory and professional guidance reflects this risk. FDA bioanalytical guidance and ICH M10 treat matrix effects, selectivity, carryover, recovery, stability, and dilution integrity as method characteristics that may need evaluation in regulated bioanalysis. CLSI C62, used by clinical laboratories as an LC-MS reference, also emphasizes evaluation of interferences and assay performance. These documents do not prescribe one universal extraction method. They point to the same practical principle: sample preparation must fit the matrix and the intended use.

That matters because each preparation choice changes the balance of performance. A cleaner extraction may reduce ion suppression but add time, cost, solvent use, and opportunities for analyte loss. A simpler workflow may support higher throughput but leave more matrix behind. Method development is therefore a controlled trade-off, not a search for one universally superior technique.

The main LC-MS sample preparation options

Most LC-MS sample preparation strategies combine several unit operations: homogenization, dilution, pH adjustment, protein removal, extraction, cleanup, concentration, reconstitution, filtration, and sometimes chemical derivatization. The methods below are commonly used alone or in combination.

Dilute-and-shoot

Dilute-and-shoot is the simplest workflow. The sample is diluted with a compatible solvent or mobile phase, then injected after mixing and, when needed, centrifugation or filtration. It is attractive for urine, simple aqueous samples, some environmental waters, and screening methods where analyte concentrations are high enough and the matrix is not too aggressive.

The limitation is cleanup. Dilution reduces the concentration of matrix components, but it also dilutes the analyte. If the method needs low detection limits, or if the matrix contains strong ion suppressors, dilute-and-shoot can create sensitivity and robustness problems. With dirty matrices, it may also increase contamination of the inlet, column, and ion source.

Protein precipitation

Protein precipitation is widely used for plasma, serum, and other protein-rich biological samples. A common approach is to add an organic solvent such as acetonitrile or methanol, mix, centrifuge, and analyze the supernatant. The workflow is fast, inexpensive, and compatible with 96-well plate automation.

Its main limitation is selectivity. Protein precipitation removes much of the protein burden, but it often leaves phospholipids, salts, and small endogenous molecules in the extract. For many drug-like small molecules, this may be acceptable when stable isotope-labeled internal standards and suitable chromatography are used. For trace analysis, or for analytes affected by strong matrix suppression, additional cleanup may be necessary.

Liquid-liquid extraction

Liquid-liquid extraction separates analytes by partitioning them between an aqueous phase and an organic phase. It can provide cleaner extracts than protein precipitation when analyte pH, ionization state, and solvent choice are well controlled. LLE is often useful for neutral or ionizable small molecules with favorable partition behavior.

LLE can also concentrate analytes if the organic extract is evaporated and reconstituted in a smaller volume. However, it may be less convenient for very polar analytes, mixed-acid/base panels, or high-throughput workflows. Emulsion formation, solvent handling, evaporation losses, and manual variability are practical concerns.

Solid-phase extraction

Solid-phase extraction uses a sorbent to retain the analyte, remove interferences, and then elute the analyte in a cleaner solvent. Reverse-phase, ion-exchange, mixed-mode, polymeric, and affinity-style sorbents are selected based on analyte chemistry. SPE can produce clean extracts with strong selectivity, making it valuable for low-level quantitative assays and difficult matrices.

The main drawbacks are cost, method development time, and the need to control conditioning, loading, washing, drying, and elution steps. SPE is powerful, but it is not automatically better. The sorbent chemistry and wash conditions must match the analyte and matrix.

Filtration, centrifugation, and ultrafiltration

Filtration and centrifugation protect the LC system from particles and precipitates. They are often used after extraction rather than as complete sample preparation methods. Ultrafiltration can separate small molecules from proteins or remove macromolecules based on membrane cutoff, but membrane binding should be checked for hydrophobic or highly adsorptive analytes.

These steps are easy to underestimate. A poor filter choice can introduce extractables, adsorb analytes, or change solvent composition. A short centrifugation step may leave fine particles that shorten column life. Consumables should be tested as part of the method, especially when moving from development to routine use.

Derivatization

Derivatization modifies an analyte chemically before LC-MS analysis. It may improve ionization, retention, stability, chromatographic peak shape, or separation of isomers. It is often considered for very polar compounds, small molecules with poor ionization efficiency, carbonyl compounds, amino acids, steroid panels, and other challenging analytes.

The benefit must be weighed against added complexity. Reaction yield, reagent purity, byproducts, incubation time, pH, temperature, and quenching can all affect reproducibility. Derivatization should be treated as part of the analytical method, not as an informal pre-step.

How to choose a preparation method

A practical LC-MS sample preparation decision starts with five questions.

• What is the matrix? Plasma, serum, tissue, food, wastewater, soil extract, cell culture media, and formulated products contain different interferences.
• What is the analyte chemistry? Polarity, pKa, solubility, protein binding, volatility, stability, and functional groups influence extraction choice.
• How low is the required detection limit? Trace-level methods usually need better cleanup, concentration, or both.
• How many samples must be processed? A research method for a small batch can tolerate more handling than a routine production or clinical workflow.
• What evidence is required? Regulated drug bioanalysis, forensic testing, clinical assays, and routine research have different documentation expectations.

The table below summarizes common selection patterns. It is a development guide, not a validation substitute.

Preparation method	Typical strengths	Common limitations	Often considered when
Dilute-and-shoot	Fast, low cost, minimal handling	Limited cleanup, analyte dilution, more matrix in system	Matrix is simple and target levels are not extremely low
Protein precipitation	Rapid protein removal, plate-friendly, economical	Residual phospholipids and small-molecule interferences	Biological fluids need high-throughput preparation
Liquid-liquid extraction	Good cleanup for suitable analytes, possible concentration	Solvent use, emulsions, less ideal for very polar analytes	Analyte partitioning can be controlled by pH and solvent
Solid-phase extraction	Selective cleanup, cleaner extracts, concentration potential	Higher cost and more optimization	Low-level quantitative assays require better matrix removal
Derivatization	Can improve ionization, retention, or stability	Adds reaction variables and possible byproducts	Analytes ionize poorly or chromatograph poorly without modification

For many methods, the best answer is a hybrid. A plasma method may use protein precipitation followed by phospholipid-removal plates. A food method may use solvent extraction, salt-assisted partitioning, cleanup sorbents, and concentration. A tissue method may require homogenization, extraction, centrifugation, and SPE. The workflow should be judged by measured performance, not by the apparent sophistication of the technique. See also: buying guides.

Matrix effects, recovery, and stability should be evaluated separately

Three terms are often mixed together during LC-MS method development: matrix effect, extraction recovery, and process efficiency. They are related, but they answer different questions.

• Matrix effect asks whether co-eluting matrix components change the MS response of the analyte.
• Recovery asks how much analyte is recovered through the extraction process.
• Process efficiency reflects the combined result of extraction recovery and matrix effect.

A method can have high recovery but poor matrix effect if it extracts both the analyte and many ion-suppressing compounds. Another method may have moderate recovery but excellent reproducibility and low ion suppression, making it stronger for quantitation. Recovery alone is not enough to select a preparation method.

Internal standards are essential in many quantitative LC-MS workflows, especially stable isotope-labeled internal standards that behave similarly to the analyte. They can correct for variability in extraction, injection, and ionization. However, they do not compensate for every cleanup problem. If an internal standard does not co-elute closely with the analyte, or if a matrix effect is severe and inconsistent, correction may be incomplete.

Stability also needs independent attention. Sample collection, anticoagulant choice, freeze-thaw cycles, bench-top time, autosampler conditions, extract storage, pH, light exposure, and evaporation can change analyte concentration. FDA and ICH bioanalytical guidance both emphasize stability in the intended matrix and under relevant handling conditions. In practice, a sample preparation method is not robust if it produces clean extracts but allows the analyte to degrade before injection.

Workflow details that often decide success

Two LC-MS methods can use the same headline preparation technique and still perform very differently. The details below often determine whether the workflow holds up in routine use.

Solvent and pH compatibility

Extraction solvent should dissolve the analyte, precipitate or remove unwanted components, and be compatible with the LC starting conditions after reconstitution. pH can control analyte ionization and partitioning, but it can also accelerate degradation. A strong mismatch between injection solvent and initial mobile phase can cause peak distortion, especially for early-eluting polar compounds.

Phospholipid and salt management

In electrospray LC-MS, phospholipids are a common source of ion suppression in plasma and serum extracts. Protein precipitation alone may leave them behind. Better chromatographic separation, phospholipid-removal products, mixed-mode SPE, or alternative extraction conditions may be needed. Salts are also problematic because they can suppress ionization and contaminate the source; desalting or dilution may be required depending on the method.

Container and filter effects

Vials, tubes, plates, caps, and filters can contribute contamination or analyte loss. Low-level LC-MS methods are particularly vulnerable to adsorption onto polypropylene, glass, or membrane surfaces. During development, labs should compare consumables when unexplained low recovery, carryover, or background peaks appear.

Evaporation and reconstitution

Evaporation can concentrate analytes and improve sensitivity, but it can also cause losses of volatile compounds, incomplete drying, residue effects, and longer cycle time. Reconstitution solvent should bring the analyte back into solution without creating precipitation or chromatographic mismatch. Vortexing, sonication, and controlled temperature can improve consistency, but they should be standardized.

Building a defensible LC-MS sample preparation workflow

A defensible workflow is built through controlled comparisons. During development, compare at least two plausible preparation routes using the same matrix lots when possible. Track recovery, matrix effect, precision, carryover, peak shape, background, sample cycle time, solvent volume, consumable cost, and instrument maintenance impact. The cleanest extract is not always the best routine method if it is slow, fragile, or difficult to transfer.

For quantitative work, matrix lots are especially important. Testing only one blank matrix can hide donor-to-donor or lot-to-lot variation. Regulated bioanalysis commonly evaluates multiple individual matrix sources, including special matrices where relevant. For nonregulated research, the same concept still applies: a preparation method should be challenged with realistic matrix variation before conclusions are trusted.

Documentation should capture the exact sample volume, reagent volumes, mixing times, centrifugation conditions, temperature, order of addition, plate or tube type, evaporation settings, and reconstitution conditions. These details may look minor, but they often explain why a method transfers poorly between analysts, instruments, or laboratories.

Automation should be considered after the chemistry is understood. Automating an unstable or poorly selective preparation method makes it faster, not better. Once the extraction is proven, automation can improve throughput, reduce repetitive manual steps, and improve timing consistency.

Common mistakes to avoid

• Choosing speed before selectivity. A five-minute preparation step can become expensive if it causes frequent source cleaning, reinjection, or failed batches.
• Assuming protein precipitation removes all important interferences. It removes proteins well, but many small endogenous compounds remain.
• Optimizing recovery without checking matrix effect. High recovery is not useful if the analyte signal is suppressed unpredictably.
• Ignoring analyte stability during preparation. Degradation during thawing, extraction, or autosampler storage can look like poor recovery.
• Using generic filters or plasticware without testing. Adsorption and extractables can be significant at low concentrations.
• Changing sample preparation after validation without assessment. Even small workflow changes can affect recovery, matrix effect, and stability.

Frequently asked questions

What is the simplest LC-MS sample preparation method?

Dilute-and-shoot is usually the simplest because it involves dilution and minimal cleanup. It works best for relatively clean matrices and methods with sufficient sensitivity. For complex biological, food, or environmental samples, simple preparation may not provide enough matrix control.

Is SPE always better than protein precipitation?

No. SPE often provides cleaner and more selective extracts, but it costs more and requires more optimization. Protein precipitation may be suitable when the method meets precision, accuracy, sensitivity, matrix effect, carryover, and stability requirements. The better method is the one that meets the analytical purpose with acceptable robustness.

How can matrix effects be reduced in LC-MS?

Matrix effects can be reduced by improving sample cleanup, changing extraction chemistry, diluting the sample, improving chromatographic separation, removing phospholipids or salts, using suitable internal standards, and optimizing the ion source conditions. The most appropriate approach depends on whether the interference is chemical, chromatographic, or matrix-specific.

When is derivatization worth considering?

Derivatization is worth considering when the analyte has poor ionization, weak retention, poor peak shape, instability, or difficult isomer separation. It should be validated carefully because the derivatization reaction introduces additional variables that can affect reproducibility.

What should be recorded in an LC-MS sample preparation procedure?

The procedure should record sample amount, reagent identity and volume, pH, mixing time, temperature, centrifugation speed and time, extraction format, consumables, evaporation settings, reconstitution solvent, storage conditions, and autosampler conditions. Detailed documentation improves transferability and troubleshooting.

The practical takeaway

LC-MS sample preparation should be selected by evidence, not habit. Dilute-and-shoot, protein precipitation, LLE, SPE, filtration, and derivatization all have valid roles, but each brings trade-offs in cleanup, recovery, sensitivity, cost, and throughput. A strong workflow controls matrix effects, protects analyte stability, fits routine operations, and can be documented clearly enough for another analyst to reproduce. For most laboratories, that means comparing practical options early, measuring matrix effect and recovery separately, and treating sample preparation as a central part of LC-MS method quality rather than a pre-instrument formality.