Miniaturized Liquid Sample Preparation for Environmental Analysis

How do miniaturized liquid-phase extraction techniques, such as dispersive liquid-liquid microextraction (DLLME) or hollow fiber liquid-phase microextraction (HF-LPME), affect compatibility with modern chromatographic instruments, such as liquid chromatography tandem mass spectrometry (LC–MS/MS)?¹

Sample preparation based on liquid- and solid-phase extraction began about three decades ago with solid-phase microextraction (SPME) and liquid-phase microextraction (LPME) and was later expanded to include several other techniques. This miniaturization developed in parallel with the miniaturization of separation techniques, such as LC (from high-performance liquid chromatography [HPLC] to micro-LC, nano-LC, u-HPLC, and others) and gas chromatography (GC) (from large-bore packed columns to small-bore open-tubular columns). As analytical instrumentation has been miniaturized, extraction techniques have followed suit to ensure compatibility between the two worlds of a typical modern analytical workflow: sample preparation and analytical (qualitative and quantitative) determination. The decrease in sampling flow rate in automated systems and the use of much less sample and solvent, as in SPME-GC (a solventless technique), have made compatibility between sample preparation and analytical determination much easier. It is worth noting that electrospray ionization (ESI), the most popular method for interfacing LC with MS/MS, was demonstrated by John Fenn using miniaturized liquid chromatography coupled to mass spectrometry.² Thus, miniaturizing the sample preparation step will make it easier to couple with both liquid chromatography and mass spectrometry.

What are the primary challenges to automating miniaturized liquid extraction methods for routine chromatographylaboratories?

Various sorbent- and solvent-based miniaturized techniques are available in multiple formats and can be easily automatedusing robotic, column-switching, microfluidic, and other approaches. These approaches reduce chemical use and mitigate safety risks. They also improve analytical throughput and robustness, yielding more reliable results. Ongoing efforts aimto develop more practical and less polluting extraction processes. For sorbent-based microextraction techniques, severalmaterials are being synthesized with a focus on green sample preparation methods, including biosorbents derived from various natural sources (agricultural by-products such as peels and sawdust, microorganisms such as algae and yeast, andbiopolymers like cellulose, alginate, and chitosan), ionic liquids (ILs), graphene and other carbon-based materials, magnetic materials, and metal−organic frameworks (MOFs) and covalent organic frameworks (COFs).

Although a wide variety of sorbents for solid-based miniaturized sample preparation techniques is available, options for green solvents in counterpart miniaturized liquid-based microextraction techniques are much more limited. Among the available alternative solvents, the most investigated are supramolecular solvents (SUPRAS), liquid ionics, alternativehalogenated solvents, and deep eutectic solvents (DES). However, the chemicals and reactions involved in synthesizing these solvents have raised serious questions about their compatibility with the principles of green sample preparation (GSP).

Regarding the automation of miniaturized liquid extraction-based methods, they follow the same approaches as solid-phase extraction techniques, with robotic approaches being the most successful.

How significant are matrix effects when using green solvents in miniaturized extraction procedures, and what strategiesdo you recommend to mitigate them during chromatographic analysis?

In some analytical techniques, particularly LC–MS/MS using ESI for the analysis of complex samples such as food andbiological samples, including plasma and serum, some matrix components may coelute with the analyte under determination, interfering with the ionization process and causing the signal to be raised (enhancement) or lowered(suppression). This is generally termed the matrix effect and may be confirmed by comparing the signal obtained with theanalyte(s) in the matrix to that of the same analyte in a neat solution (without the matrix). This effect is of significantconcern, as it may have several deleterious consequences in the analysis of complex matrices, including reduced accuracy(false quantitative results, as the measured signal does not correspond to the actual one) and reduced sensitivity (as itchanges the limit of detection [LOD] and limit of quantification [LOQ]).

In miniaturized extraction procedures, the use of green solvents will not avoid the matrix effect. To mitigate this effect,several precautions must be taken.

Matrix-matched calibration is one of the most effective approaches to investigating the matrix effect. For that purpose, asolution containing a high-purity analytical standard of the target analyte(s) is added to a blank matrix (which does notcontain the analyte under investigation). This matches the matrix of the actual sample, and the potential interferent inthe matrix will be similar in both procedures. For instance, if one wants to investigate the effect of the apple matrix on theanalytical signal of a certain pesticide in this matrix, we could obtain a sample of organic apples from a reliable source. After extraction, a solution containing a certain amount of the pesticide(s) (“spiking solution”) is added to this extract (pesticide solution in matrix). To prepare the neat standard solution, a known volume of the analytical standard solutionof known concentration is added to a known volume of the pure, matching solvent. The determination of the analyte(s)signal in the spiked matrix and the neat solution will inform on the occurrence of a matrix effect. If both signals are similar,this suggests the absence of this effect for the investigated pesticide(s) in apple samples.

Another effective way to mitigate the matrix effect is to use appropriate internal standards (IS). In several cases, particularly with complex matrices and when determining multiple analytes simultaneously (for example, multiple pesticide residues with distinct chemical characteristics in food samples), selecting the appropriate analytical standard isnot straightforward. At a minimum, the IS should have a chemical structure similar to the target compound and, whenpossible, be isotopically labeled, such as with deuterium. This will make it easier to normalize the signal. If any matrix component affects the target compound's signal, the same (or very similar) effect should also occur with the IS.

Sound sample-preparation practices can help minimize matrix effects. When using more universal extraction procedures, such as conventional liquid-liquid extraction (LLE), QuEChERS, and others, several analytes that can cause matrixeffects are extracted along with the target compounds. One way to mitigate this problem is to use more selective extraction techniques, such as RAM (restricted-access media), MIPs (molecularly imprinted polymers), ILs, and other selective extraction approaches.

Finally, it should be remembered that, before the analysis, a method must be developed, optimized, and validated, duringwhich several improvements may be introduced to eliminate interferents that contribute to the matrix effect. In the case of an LC–MS/MS method, the LC gradient may be improved; a more appropriate mobile phase may be selected to providebetter separation and eliminate interferents; a more selective column may be chosen; smaller stationary-phase particlesmay be used; and MS conditions may be optimized (from instrument settings to ionization mode; depending on theanalyte, APCI, APPI, EI, and other ionization modes may work better than ESI). These measures help mitigate the presence of interferents during MS/MS analysis and, consequently, minimize the matrix effect.

Among the emerging green solvents (ILs, bio-derived solvents, SUPRAS), which show the greatest promise forseamless integration into automated, chromatography-compatible sample preparation workflows?

These solvents, along with several others, such as alternative halogenated solvents and DES, show great potential forminiaturized liquid-based extraction techniques. However, the “green solvent” designation for most of them remains underdebate, since their precursor reagents and the reactions involved in their synthesis do not always strictly align with greenchemistry principles. Even so, most of them offer clear advantages over conventional solvents in terms of greenness.Each class of emerging green solvents has its own characteristics and niche applications. Considering the currentpicture, I would imagine that SUPRAS and DES are the most promising solvents for straightforward integration into an automated sample prep-chromatography-MS/MS workflow.

What specific advancements in automation do you consider critical to expanding the application of miniaturized extraction in high-throughput environmental chromatography?

Within a short to medium time frame (5–10 years), I believe the main advances in lab automation aimed at expanding the applicability of miniaturized online extraction coupled to chromatography for high-throughput, environmentally friendly analytical standards will focus on optimizing simpler robotic technologies and column-switching approaches. In the first case, low-cost robotic systems, such as Cartesian robots micro-controlled via Arduino or similar platforms, will be readily available from instrument vendors or built in the lab, given that they are typically open source.3-6 In the column-switching approach, a first (extraction) column uses a selective sorbent for extraction, and a valve transfers the extract online to an analytical column for analyte determination. Both technologies already exist, particularly in research laboratories. To be useful in routine chromatography laboratories, these technologies must be simplified and made more affordable for these settings. Over a longer time frame, it is difficult to anticipate which technology will be predominant. Certainly, the use of miniaturized platforms with microdevices that integrate sample extraction, chromatography, and MS would be the ideal world for an analyst. However, the science behind this technology has been available for decades, yet its adoption in routine laboratories still seems far off.

How does the choice of extraction technique influence the sensitivity and selectivity in chromatographic analyses ofcomplex environmental samples?

Although often neglected, extraction techniques are key to the success of the entire analysis. This is particularly true when analyzing complex matrices such as environmental, food, and plasma samples. If the analyte is not extractedproperly, the accuracy of the results will be compromised. If proper cleanup is not performed (either during samplepreparation or afterward), many compounds will be extracted along with the target analyte, compromising sensitivity due to the matrix effect. A similar effect occurs when an inappropriate sorbent is chosen in techniques such as SPE, SPME, stir bar sorptive extraction (SBSE), and other microextraction methods. Using “generic” phases, such as dimethylpolysiloxane, the most frequently used sorbent in these microextraction techniques, results in poor extraction selectivity. Consequently, many matrix compounds are extracted, leading to a strong matrix effect when LC–MS/MS is used post-extraction. In the final step, accuracy will be compromised, yielding an incorrect result, as discussed earlier. On the otherhand, a better choice of the sorbent (more selective for the analyte of interest) will produce more accurate results, as interferents will be eliminated.

It is worth remembering that many analysts place too much confidence in the mass spectrometer, particularly given its cost. In several cases, a simple liquid-liquid extraction is performed, and the extract is transferred to LC–MS/MS (and, insome cases, directly to MS/MS) without regard for the extracted interferents. Because LC–MS/MS, especially with ESI interfaces, isa soft ionization technique that generates few ions and provides limited analyte identification, selective extraction wouldgreatly help eliminate interfering ions and improve the signal-to-noise ratio, thereby increasing analytical sensitivity.

What recommendations would you give to chromatographers aiming to implement these miniaturized, green, andautomated extraction method into existing analytical workflows?

The first suggestion I offer to those interested in implementing a miniaturized, green, and automated extraction method is to take time to understand the basic principles of the techniques, the instrumentation, and how to interpret results. For instance, SPME is a non-exhaustive sample preparation technique, unlike classical LLE, which is exhaustive. While LLE does not require any instrument and is an off-line technique, SPME may be directly coupled to either GC or LC and, in the latter case, requires an interface. LLE is an extraction technique only, requiring further off-line transfer of an aliquot ofthe extract to the chromatograph; SPME may extract, perform sample enrichment, and clean up at once, making thesample ready for chromatography sample introduction. Once the main principles and instrumentation are understood, process optimization becomes straightforward. This approach is also valid for most miniaturized sample preparation techniques. Reading the appropriate scientific literature can be a shortcut to obtaining valuable information on themethods to be implemented for the selected technique. A wide range of solid-phase- and liquid-phase-based techniques canbe easily found in the literature, providing an excellent starting point for developing a new method for a specific application.

Due to their perceived advantages over traditional sample preparation techniques such as LLE, Soxhlet, and others, miniaturized, green, and automated extraction methods are already a reality in many research laboratories worldwide and are becoming, albeit slowly, a trend in routine analytical chemistry laboratories.

References

1. Martins, R. O; Will, C.; Magalhães, M. F. A.; Lanças, F. M. Advances in Miniaturized Liquid Sample Preparation Techniques for Environmental Analysis: A Special Look Towards Green Solvents and Automation. J Chrom Open2025, 8, 100240. DOI: 10.1016/j.jcoa.2025.100240
2. Schmidt, A.; Kara, M.; Dülcks, T. Effect of Different Solution Flow Rates on Analyte Ion Signals in Nano-ESI MS, or: When Does ESI Turn into Nano-ESI? J Am Soc Mass Spectrom 2005, 14 (5), 492–500. DOI: 10.1016/S1044-0305(03)00128-4
3. Medina, D. A. V.; Rodríguez Cabal, L. F.; Lanças, F. M. Santos-Neto, Á. J. Sample Treatment Platform for Automated Integration of Microextraction Techniques and Liquid Chromatography Analysis. HardwareX 2019, 5, e00056. DOI: 10.1016/j.ohx.2019.e00056
4. Medina, D. A. V.; Maciel, E. V. S.; Lanças, F. M. Modern Automated Sample Preparation for the Determination of Organic Compounds: A Review on Robotic and On-Flow Systems. TrAC Anal Chem 2023, 166, 117171. DOI: 10.1016/j.trac.2023.117171
5. Medina, D. A. V., Cabal, L. F. R.; , Titato, G. M.; Santos-Neto, Á. J.; Lanças, F. M. Automated Online Coupling of Robot-Assisted Single Drop Microextraction and Liquid Chromatography. J Chrom A2019, 1595, 66–72. DOI: 10.1016/j.chroma.2019.02.036
6. da Silva, L. F.; Medina, D. A. V.; Lanças, F. M. Automated Needle-Sleeve-Based Online Hyphenation of Solid-Phase Microextraction and Liquid Chromatography. Talanta 2021, 221, 121608. DOI: 10.1016/j.talanta.2020.121608