LC-MS vs GC-MS: Choosing for Impurity Analysis

LC-MS vs GC-MS comes down to one question: can your analyte be vaporised without decomposing? GC-MS is the better choice for volatile, thermally stable, relatively small molecules such as residual solvents, alkyl halides and small nitrosamines, and it produces reproducible electron ionisation spectra that can be searched against libraries. LC-MS is the better choice for polar, non-volatile, thermally labile or larger molecules, including most drug substances, their degradation products and drug-substance-related impurities. For many trace impurity problems in pharmaceuticals, the two are complementary rather than competing, and the right answer depends on the analyte, the matrix and the limit you need to reach.

How GC-MS and LC-MS work

GC-MS

In GC-MS, the sample is vaporised in a heated inlet (or sampled from the headspace above it), carried by an inert gas through a capillary column and separated mainly by boiling point and polarity. Eluting compounds enter the ion source in the gas phase. Electron ionisation (EI) is the standard mode: it is a hard ionisation technique that fragments molecules in a highly reproducible way, so spectra can be compared with commercial reference libraries. Chemical ionisation (CI) is softer and is used when the molecular ion is weak or absent under EI.

LC-MS

In LC-MS, separation happens in the liquid phase, so the analyte never needs to be volatile. The column eluent is converted into gas-phase ions at atmospheric pressure, most commonly by electrospray ionisation (ESI) for polar and ionisable compounds, atmospheric pressure chemical ionisation (APCI) for less polar, moderately volatile compounds, and atmospheric pressure photoionisation (APPI) for non-polar species. These are soft techniques that mainly produce protonated or deprotonated molecules (for example [M+H]⁺), so structural information usually comes from tandem MS (MS/MS) or accurate-mass measurement rather than from a single spectrum.

LC-MS vs GC-MS: side-by-side comparison

Criterion	GC-MS	LC-MS
Analyte requirement	Volatile (or made volatile by derivatisation) and thermally stable	Soluble in a suitable solvent; volatility not required
Typical analytes	Residual solvents, alkyl halides, small nitrosamines, low-molecular-weight genotoxic impurities, volatile extractables	APIs, degradants, polar and ionic impurities, drug-substance-related nitrosamines, non-volatile leachables
Ionisation	EI (hard, reproducible); CI (soft)	ESI, APCI, APPI (soft, at atmospheric pressure)
Identification of unknowns	Strong: EI spectra are library-searchable	Relies on MS/MS fragmentation and accurate mass; libraries less transferable
Main source of error	Thermal degradation or artefact formation in the inlet; active sites	Ion suppression or enhancement from co-eluting matrix (matrix effects)
Mobile phase constraints	Carrier gas only	Volatile buffers and additives only (for example formic acid, ammonium formate or acetate); no phosphate
Sample introduction options	Direct liquid injection, static headspace, thermal desorption	Direct injection; solid-phase extraction or dilution for complex matrices
Best quantitation mode for trace work	SIM, or MRM on a triple quadrupole	MRM on a triple quadrupole, or high-resolution accurate mass

Key takeaway

Start with the analyte’s volatility and thermal stability, not with the instrument that happens to be free. If the compound survives vaporisation and is small enough to elute from a GC column, GC-MS usually offers simpler, more robust trace analysis. If it is polar, ionic, large or heat-sensitive, or if heating the sample could create the very impurity you are measuring, choose LC-MS.

Choosing by analyte class

Residual solvents

Residual solvents controlled under ICH Q3C are the classic GC application, normally run by static headspace so that the non-volatile drug matrix never enters the column. FID is the routine detector; MS is valuable for confirming the identity of unexpected peaks. See our guide to residual solvents analysis by GC headspace.

Mutagenic (genotoxic) impurities

Many potentially mutagenic process impurities are small electrophiles: alkyl halides, epoxides and some sulfonate esters. These are often well suited to GC-MS, sometimes after derivatisation to improve volatility or selectivity. More polar or reactive species, such as hydrazines, some aromatic amines and larger drug-like intermediates, generally need LC-MS, again occasionally with derivatisation to improve retention or ionisation.

Nitrosamines

Small nitrosamines such as NDMA and NDEA can be determined by either technique. FDA has published validated methods for sartans using headspace GC-MS, GC-MS/MS and LC-HRMS. Nitrosamine drug-substance-related impurities (NDSRIs), which carry most of the parent molecule, are generally too large and polar for GC and are analysed by LC-MS/MS. Thermal artefacts are a real risk: FDA’s LC-HRMS method for NDMA in ranitidine was developed because GC-based methods had been observed to raise measured NDMA levels through degradation of the drug itself. Our nitrosamine impurities testing guide covers risk assessment and acceptable intake limits in more detail.

Degradation products and unknown impurities

Degradants of small-molecule APIs usually share the parent’s polarity and molecular weight, so they are found and tracked by HPLC and characterised by LC-MS. Accurate mass and MS/MS fragments narrow down possible structures; definitive assignment often needs NMR on an isolated or synthesised sample.

Extractables and leachables

Here the two techniques are deliberately combined. GC-MS (including headspace) captures volatile and semi-volatile compounds such as residual monomers and some antioxidant breakdown products, while LC-MS covers non-volatile additives, oligomers and polar species. Relying on only one will leave gaps.

Sensitivity: working back from the limit

The required sensitivity should be derived from the specification, not from the instrument brochure. Under ICH M7(R2), a threshold of toxicological concern (TTC) of 1.5 µg/day, corresponding to a theoretical excess lifetime cancer risk of 1 in 100,000, can be used for a mutagenic impurity without compound-specific data.

Worked example. For a drug with a maximum daily dose of 500 mg:

Concentration limit = 1.5 µg/day ÷ 0.5 g/day = 3 µg/g, that is 3 ppm relative to the API.
If the method uses a test solution of 10 mg/mL, 3 ppm corresponds to 30 ng/mL of impurity in solution.
A common practice is to aim for a quantitation limit at or below 30% of the limit (here about 1 ppm, or 9 ng/mL), so that results near the limit are well within the validated range.

Both GC-MS in SIM or MRM mode and LC-MS/MS in MRM mode can usually reach levels like this. The deciding factor is more often matrix: a high API load can foul a GC inlet or suppress ionisation in an LC-MS source. Sample preparation, divert valves (to send the API peak to waste), headspace sampling and stable-isotope-labelled internal standards are the usual tools for managing this.

Practical pitfalls and good practice

Artefact formation in GC. Hot inlets can degrade labile APIs into the analyte of interest or into interfering peaks. Check by spiking, varying the inlet temperature and comparing with an orthogonal LC-MS result.
Matrix effects in LC-MS. Assess ion suppression with post-extraction spikes or post-column infusion. Use isotope-labelled internal standards or matrix-matched calibration where suppression is significant.
Non-volatile buffers. An HPLC method with phosphate buffer cannot be transferred directly to LC-MS. Redevelop with volatile additives and re-check selectivity.
Isobaric interferences. At unit resolution, different compounds can share a nominal mass. Chromatographic resolution, MRM transitions with qualifier ions or high-resolution MS reduce false positives. Confirm identity with retention time and ion ratios, not mass alone.
Carry-over and contamination. Trace work exposes contamination from solvents, vials, septa and gloves. Run blanks around high-level samples and investigate any blank response before reporting.
Validation. Trace MS methods are validated like any other quantitative impurity procedure, with specificity, response, accuracy, precision and a confirmed quantitation limit; see our guide to method validation under ICH Q2(R2).

A quick decision checklist

Is the analyte volatile, or can it be made volatile by a simple derivatisation? If no, use LC-MS.
Is it thermally stable at inlet temperature, and is the matrix free of heat-labile compounds that could generate it? If no, use LC-MS.
Is it a residual solvent or other small volatile in a non-volatile matrix? Headspace GC or GC-MS is usually simplest.
Is it polar, ionic or larger than a typical GC analyte (for example an NDSRI or degradant)? Use LC-MS.
Do you need to identify an unknown? GC-MS with library search for volatiles; LC-MS/MS or accurate mass, followed by NMR, for non-volatiles.
What limit must be met, and does the proposed method’s quantitation limit sit comfortably below it in the real sample matrix?

How SDS Labs can help

SDS Labs operates both LC-MS, with ESI, APCI and APPI ionisation interfaces, and GC-MS, with electron ionisation and chemical ionisation on capillary columns, alongside HPLC, GC with FID, NPD and TCD detectors, and 400 MHz NMR for structure confirmation. This lets us choose the technique to suit the analyte rather than the other way round, and use one technique to confirm the other where artefacts are a concern.

To discuss an impurity or trace analysis requirement, share your molecule, matrix and target limit via our contact page, call +91 90295 20144 or email info@sdslabs.in.

Frequently asked questions

What is the main difference between LC-MS and GC-MS?

GC-MS separates compounds in the gas phase, so analytes must be volatile and thermally stable, and it usually uses hard electron ionisation that gives library-searchable spectra. LC-MS separates compounds in the liquid phase and uses soft atmospheric-pressure ionisation such as ESI or APCI, so it handles polar, non-volatile, thermally labile and larger molecules, including most drug substances and degradants.

Is LC-MS or GC-MS more sensitive?

Neither is universally more sensitive. Both GC-MS in SIM or MRM mode and LC-MS/MS in MRM mode can reach low parts-per-million or lower limits relative to the drug substance. Practical sensitivity depends on how well the analyte ionises, how much matrix is injected and how well it is managed, so the method should be chosen and validated against the actual limit in the real sample.

Should nitrosamines be tested by GC-MS or LC-MS?

Small volatile nitrosamines such as NDMA and NDEA can be measured by headspace GC-MS, GC-MS/MS, LC-MS/MS or LC-HRMS, and FDA has published methods using GC-MS, GC-MS/MS and LC-HRMS. Nitrosamine drug-substance-related impurities are usually too large and polar for GC and need LC-MS/MS. Where the drug can degrade on heating to form nitrosamines, as with ranitidine, LC-based methods avoid that artefact.

Why is GC-MS better for identifying unknown volatile compounds?

Electron ionisation at a standard energy fragments molecules in a highly reproducible way, so a GC-MS spectrum can be matched against large commercial reference libraries. LC-MS soft ionisation mainly gives the molecular ion, and fragmentation depends on instrument settings, so identifying unknowns by LC-MS relies more on MS/MS interpretation, accurate mass and confirmation by NMR or reference standards.

Can an HPLC method be transferred directly to LC-MS?

Often not. Many HPLC methods use non-volatile buffers such as phosphate, which contaminate the mass spectrometer source and suppress ionisation. The method usually needs to be redeveloped with volatile additives such as formic acid or ammonium formate, after which selectivity must be rechecked and the method validated for its new purpose.

LC-MS vs GC-MS: How to Choose for Impurity and Trace Analysis