Fast Low-Pressure Chiral GC for Pesticides in Wine

Many pesticides are chiral, meaning they exist in two (or more) mirror-image versions (called enantiomers), much like a left and right hand. The different enantiomers can behave very differently: often only one does the job of killing pests, while the others may be useless or even harmful to people and other organisms. Even so, most chiral pesticides are sold as a mix of both, and food safety rules measure only the total amount of a pesticide rather than each version separately. That means current testing could underestimate the real health risks.

Wine is a particularly tricky product to test because pesticide residues show up in tiny amounts within a complicated mix of other compounds. Earlier studies on pesticides in wine used standard methods that cannot tell the mirror-image versions apart, even though many of the pesticides they looked for were chiral.

Separating these twin molecules is possible with dedicated gas chromatography (GC) columns, but the process is usually slow, because speeding it up tends to blur the separation. To get around this, a team of researchers combined a chiral column with a technique called low-pressure gas chromatography (LPGC), which uses short, wide columns to run analyses faster while keeping results sharp and sensitive. They paired this with a highly selective mass spectrometer, the standard tool for detecting trace pesticide residues. The result is a new method that can screen white wine for 107 pesticides at once, both chiral and non-chiral, distinguishing the mirror-image forms while cutting analysis time. The authors believe it is the first method of its kind, and it could help regulators and routine monitoring programs get a more accurate picture of what consumers are exposed to in their wine.

LCGC International spoke to Mariosimone Zoccali, corresponding author of the article that resulted from this study,¹ about his team’s work.

The method presented in your paperis described as the first combining a chiral wide-bore column with low-pressure gas chromatography (GC) conditions for pesticide analysis. What specific bottleneck in conventional chiral GC (the resolution/speed/capacity trade-off) does this combination solve, and how?

Conventional chiral GC generally requires a compromise between enantiomeric resolution, analysis time, and sample capacity. Long narrow-bore columns can provide adequate resolution, but they usually require relatively long analysis times and tolerate only limited injected amounts. Conversely, shortening the column and increasing the carrier-gas velocity may accelerate the separation at the expense of resolution and detectability. The proposed strategy addresses this bottleneck by combining a short wide-bore γ-cyclodextrin column with low-pressure (LP) outlet conditions.¹ LPGC increases the gas-phase diffusion coefficients and permits the use of higher optimum linear velocities, while the wide-bore column provides higher sample capacity. Consequently, the method maintains acceptable enantiomeric resolution while reducing the analysis time and improving signal-to-noise ratios. Using the selected conditions, all 107 pesticides were analyzed in a 36 min cycle, with data acquisition completed within 27 min, moreover, acceptable separation was obtained for seven chiral pesticide pairs.

Why choose a wide-bore chiral column specifically for LPGC rather than adapting a narrow-bore chiral column to low-pressure conditions—what practical advantages does the wider internal diameter offer in terms of sample capacity and robustness for routine wine analysis?

A wide-bore chiral column was selected because LPGC requires the analytical column to operate under sub-ambient pressure conditions. In LPGC–mass spectrometry (MS), the vacuum generated by the mass spectrometer is transmitted upstream from the MS interface through the analytical column. The extent to which this pressure reduction propagates along the column depends strongly on its internal diameter. A conventional narrow-bore column has a much higher flow resistance, consequently, the pressure drop is concentrated near the MS outlet, and sub-ambient pressure conditions cannot be effectively established along the entire column length. By contrast, a short wide-bore column, typically with an internal diameter of approximately 0.53 mm, permits the vacuum conditions to extend over essentially the whole analytical column, enabling the characteristic high linear velocities and rapid separations of LPGC.

A restrictor is therefore required at the inlet side of the system. In the present work, a short, uncoated and deactivated capillary was installed between the injector and the chiral analytical column. Its function is to provide the necessary flow restriction between the pressurized injector and the analytical column operating under sub-ambient pressure. In this configuration, the restrictor maintains the injector and its pressure- or flow-control system under above-atmospheric pressure conditions, while preventing the low pressure established in the analytical column from propagating back into the inlet. It therefore enables the simultaneous use of a conventional split/splitless injector and vacuum-outlet chromatography. This configuration derives from the approach proposed by de Zeeuw and Peene, in which a deactivated restriction capillary was positioned between the inlet and a wide-bore column.²

In the present method, a 0.5 m × 0.1 mm i.d. uncoated capillary was connected directly to the injector and to the wide-bore analytical column using a low-dead-volume union. The injector could consequently be operated in high-pressure injection mode (240 kPa for 1 min) despite the sub-ambient pressure conditions prevailing in the chiral column. This configuration allowed a 5 μL splitless injection while protecting the inlet and the carrier-gas control system from unstable or negative-pressure conditions.

The choice of a wide-bore column offered additional advantages for routine wine analysis. Its higher sample capacity allowed a relatively large volume of concentrated extract to be introduced without column overloading, thereby increasing the amount of analyte transferred to the stationary phase and improving detectability. This was particularly important for pesticide residues occurring at trace concentrations in wine. The wide-bore column also contributed to the operation of the method over approximately 450 injections.

The main disadvantages of wide-bore columns: lower chromatographic efficiency, potentially higher stationary-phase bleed, greater carrier-gas consumption, and the need for an additional restrictor connection, were acceptable in this application because the method was a targeted analysis based on highly selective triple-quadrupole tandem mass spectrometry (QqQ-MS/MS) detection rather than an untargeted separation. These considerations are consistent with previous LPGC studies, which identify the high sample capacity and rapid analysis as the principal advantages of wide-bore columns.

Since enantiomeric discrimination on cyclodextrin phases relies on weak, temperature-sensitive host-guest interactions, how did you optimize the temperature program to preserve enantioselectivity while still capitalizing on the speed advantages of low-pressure conditions?

Because enantiomeric discrimination on cyclodextrin phases is governed by relatively weak and temperature-dependent host–guest interactions, excessively steep temperature programs can reduce the difference in retention between the two enantiomers. The temperature program was therefore optimized together with column length and carrier-gas velocity rather than treated as an independent parameter.

Different temperature gradients were evaluated. The steeper gradients generally produced unsatisfactory results because the analytes experienced excessively high intra-column temperatures, and several enantiomers eluted during the final hold at 180 °C, where the enantioselectivity of the cyclodextrin phase is reduced. The selected method used a 5 °C min⁻¹ gradient from 50 to 180 °C, providing a suitable compromise between enantioselectivity, analyte elution below the column temperature limit, analysis time, and signal-to-noise ratio.

You highlight column length, gas linear velocity, and temperature program as key optimization parameters—which of these had the largest individual impact on achieving adequate enantiomeric resolution, and were there any counterintuitive interactions between them?

Among the three parameters, the temperature program had the largest direct impact on enantiomeric resolution because it controlled the extent to which the temperature-sensitive cyclodextrin interactions were preserved. The 2 °C min⁻¹ program produced the highest average resolution, approximately 2.0 under the selected 5 m and 150 cm s⁻¹ conditions, whereas the 5 °C min⁻¹ program gave an average resolution of approximately 1.6.

However, the parameters showed important interactions. Increasing the gas velocity shortened retention and reduced the temperature experienced by late-eluting analytes, which helped prevent elution during the 180 °C hold, but it could also slightly reduce resolution. Similarly, shortening the column from 10 to 5 m accelerated the separation and facilitated the elution of higher-boiling pesticides, but excessive shortening or higher velocity increased band broadening and could compromise resolution and detectability. The final choice 5 m, 150 cm s⁻¹, and 5 °C min⁻¹ was therefore based on a global compromise rather than on the maximum resolution obtained under any single condition.

With 107 chiral and non-chiral pesticides spanning a wide volatility range, how did you ensure adequate chromatographic behavior and detectability across such a broad set of physicochemical properties within a single run?

The broad analyte panel was accommodated through a combination of chromatographic optimization, selective detection, and matrix-matched calibration. The final target list was selected by balancing analytical relevance, commercial standard availability, and compatibility with the thermal limits of the γ-cyclodextrin column.

The chromatographic range was assessed using a C₇–C₃₀ alkane mixture. Under the two principal optimization strategies, compounds up to C₂₇ or C₂₅ eluted before the 180 °C hold, demonstrating adequate coverage of a broad volatility range. The use of a short wide-bore column under LPGC conditions improved the elution of higher-boiling compounds, while the 5 °C min⁻¹ program and 150 cm s⁻¹ linear velocity improved detectability through narrower peaks and higher signal-to-noise ratios.

QqQ-MS operated in multiple reaction monitoring (MRM) mode provided compound-specific selectivity and sensitivity across the panel. Two MRM transitions were monitored for each pesticide. Matrix-matched calibration was applied to compensate for matrix-related response effects, with coefficients of determination ranging from 0.9253 to 0.9975.

Given that maximum residue limits (MRLs) are still set on total residue levels rather than individual enantiomers, what's the practical argument for investing in enantioselective methods now, ahead of regulatory catch-up—does this method aim to build the evidence base needed to shift that framework?

The practical justification is that total-residue measurements can conceal substantial differences between enantiomers in biological activity, toxicity, environmental fate, and degradation. The relevance is not merely theoretical: in one commercial wine sample, one fipronil enantiomer exceeded the applicable MRL whereas the other remained below it. A total-residue result could therefore provide a less informative description of the actual contamination profile and potentially obscure enantiomer-specific exposure.

The method can consequently be regarded as part of the evidence base required for a more refined regulatory framework. It provides the analytical capability needed to determine enantiomeric composition, quantify individual enantiomers, and investigate whether regulatory decisions based on total residues are sufficiently protective. The paper does not propose an immediate replacement of the current MRL system, but it clearly supports the generation of occurrence and risk-assessment data that could inform future regulatory developments.

Wine presents challenges like pigments, sugars, and ethanol that can affect both extraction efficiency and chromatographic performance. What sample preparation or injection strategies were critical to maintaining method robustness in this matrix specifically?

A modified QuEChERS procedure based on ethyl acetate extraction was used. Ten grams of wine were extracted with 10 mL of ethyl acetate in the presence of magnesium sulfate, sodium chloride, and citrate salts. The extract was then concentrated approximately 100-fold under a gentle nitrogen stream, without complete evaporation, to minimize losses of volatile analytes.

An important practical finding was that an additional dispersive-SPE clean-up step was not necessary. Although primary-secondary amine reduced the gravimetric amount of co-extracted material, it did not improve the analytical performance and was associated with possible interfering compounds from the clean-up materials. Omitting this step simplified and accelerated the procedure, reduced material consumption, and was compatible with good system stability.

The wide-bore column was essential at the injection stage because it allowed a 5 μL splitless injection. This approach combined enhanced analyte transfer with controlled solvent and matrix loading, while the wide-bore column provided the capacity required to tolerate the injected extract.

Why was QqQ-MS in MRM mode chosen as the detection technique for this application, and how did operating under low-pressure GC conditions (with potentially different peak widths and elution timing) affect MRM transition scheduling and dwell times?

QqQ-MS in MRM mode was selected because it provides high selectivity and sensitivity for targeted multiresidue analysis in complex matrices such as wine. It also offers broad quantitative working ranges and reliable confirmation through the simultaneous monitoring of quantifier and qualifier transitions.

Under LPGC conditions, the higher linear velocity and vacuum outlet produce faster elution and narrower chromatographic peaks. These features improve peak height and signal-to-noise ratio but reduce the time available for monitoring each analyte. The method therefore used scheduled compound-specific MRM acquisition, with two transitions per pesticide and an acquisition frequency of 14 Hz. This ensured that enough data points were collected across the narrow LPGC peaks while maintaining adequate dwell times and transition coverage.

How did the method's precision, accuracy, and recovery data compare to current EU performance requirements for routine pesticide monitoring, and were there specific compounds that proved more challenging to validate than others?

The method generally met the performance expectations for routine pesticide-residue analysis described in the Health and Food Safety (SANTE) framework. All reported LoQs were equal to or below the corresponding MRLs, with values ranging from 2.1 to 74.4 μg kg⁻¹. Intra-day precision was consistently below 15% for the investigated compounds, with an average CV of approximately 6%, and therefore remained within the commonly applied SANTE criterion of 20%.

Accuracy was generally satisfactory, ranging from 73 to 125% at 0.01 mg kg⁻¹, from 85 to 118% at 0.05 mg kg⁻¹, and from 89 to 107% at 0.1 mg kg⁻¹. Recoveries were more variable and relatively low on average: approximately 56% at 0.01 mg kg⁻¹ and 42% at 0.1 mg kg⁻¹, while the associated precision complied with the SANTE requirement.

In practical terms, how much faster is this LPGC chiral method compared to conventional chiral GC approaches for the same analyte panel, and what does that speed improvement mean for its feasibility in high-throughput programs like the Multiannual Control Programmes for Pesticide Residues (MACP)?

The selected LPGC method required a total run time of 36 min, with analytical data acquisition completed after 27 min. During method development, the slower 2 °C min⁻¹ program required approximately 75 min, whereas the selected 5 °C min⁻¹ program reduced the effective analytical cycle to 27 min while providing higher average signal-to-noise ratios, approximately 6000 compared with 3342 for the slower method. A comparison with a conventional chiral approach was not performed. However, conventional chiral GC analyses are generally conducted using a temperature-programming rate of approximately 2 °C min⁻¹.

The practical advantage is nevertheless clear: the method combines enantioselective analysis with a sub-30 min acquisition window, high sample capacity, and QqQ-MS selectivity, making it substantially more compatible with high-throughput monitoring than a conventional long-column chiral workflow.

For programs such as the MACP, this shorter cycle time increases daily sample throughput and reduces the analytical penalty associated with adding chiral information to routine multiresidue testing. The method is therefore particularly valuable as a screening and monitoring platform, although broader interlaboratory validation and application to additional wine types and food matrices would be required before considering full routine regulatory implementation.

In closing, I would also like to express my sincere gratitude to all the other co-authors, Giorgia Rinaldi, Peter Q. Tranchida, and Luigi Mondello, for their invaluable contributions to the conception, experimental work, data interpretation, and preparation of this manuscript. I am also particularly grateful for the many enjoyable and intellectually stimulating hours we spent together at the instrument, processing data, discussing the results, and writing this work (MZ).

References

1. Rinaldi, G.; Zoccali, M.; Tranchida, P. Q.; Mondello, L. Enantioselective Low-Pressure Gas Chromatography Combined with Triple Quadrupole Mass Spectrometry for the Analysis of 107 Chiral and Achiral Pesticide Residues in Wine. Anal. Chim Acta 2026, 1421, 346060, DOI: 10.1016/j.aca.2026.346060

2. de Zeeuw, J.; Peene, J.; Jansen, H. G.; Lou, X. W. A simple way to speed up separations by GC-MS using short 0.53 mm columns and vacuum outlet conditions. J. High Resolut. Chromatogr. 2000, 23, 677–680. DOI: 10.1002/1521-4168(20001201)23:12<677:AID-JHRC677>3.0.CO;2-L.