Solid-Phase Microextraction: The Theory Behind a Small but Powerful Sampling Tool
Sample preparation is often the least glamorous part of an analytical workflow, yet it can determine whether an analysis succeeds or fails. Before an analyte reaches a gas chromatograph, liquid chromatograph, or mass spectrometer, it may need to be isolated from interfering compounds, concentrated to a detectable level, and transferred into a form compatible with the instrument. Conventional approaches accomplish this using multiple handling steps and relatively large volumes of organic solvent.
Solid-phase microextraction, better known as SPME, takes a different approach. A small amount of extraction phase is exposed directly to a sample or its headspace. Analytes migrate into or onto that phase, after which they are transferred to an analytical instrument for measurement. Sampling, extraction, cleanup, and preconcentration are therefore combined into a compact device and a remarkably simple workflow.
Since its introduction by Arthur and Pawliszyn in 1990, SPME has grown from a coated fused-silica fiber for GC analysis into a family of fibers, thin films, coated blades, in-tube devices, and miniature probes. These formats are now used in environmental monitoring, food and aroma analysis, forensic science, pharmaceutical research, metabolomics, and even direct sampling from living organisms (Arthur & Pawliszyn, 1990; Reyes-Garcés et al., 2018).
How SPME works
A conventional SPME device contains a thin extraction coating supported on a fiber or metal substrate. During direct-immersion SPME, the coating is placed in a liquid sample. In headspace SPME, it is suspended in the gas phase above the sample. Once sampling is complete, the device is removed and the analytes are desorbed into a chromatographic or mass-spectrometric system.
The extraction phase may behave as an absorbent, an adsorbent, or a combination of both. In an absorptive coating, such as polydimethylsiloxane, an analyte partitions into the entire volume of a polymer. In a porous adsorptive coating, analytes accumulate mainly on active surfaces and within pores. The distinction matters because adsorption has a finite number of sites, making competition and displacement more important when complex mixtures are analysed.
Unlike liquid–liquid extraction, SPME is normally non-exhaustive: only a fraction of the analyte is removed from the sample. This is not necessarily a disadvantage. Modern detectors can measure very small extracted amounts, while limited analyte removal allows repeated sampling and, in biological systems, causes less disturbance to the system being investigated.
The equilibrium theory
The simplest model treats the sample and extraction coating as two phases. At equilibrium, an analyte distributes between them according to its relative affinity for each phase. This affinity is described by the distribution coefficient: the ratio of the analyte concentration in the coating to its concentration in the sample once equilibrium has been reached. A large distribution coefficient means that the analyte preferentially enters the coating, whereas a small value means that it largely remains in the sample.
Combining this partitioning relationship with mass balance explains much of SPME method development. The amount collected depends on the analyte's initial concentration, its affinity for the extraction phase, the volume of the coating, and the volume of the sample. Increasing the coating volume or selecting a coating with stronger affinity for the target analyte generally increases the extracted amount.
In many applications, the sample is effectively much larger than the extraction coating. Under these conditions, the amount collected is directly proportional to the analyte's original concentration and to the coating volume and distribution coefficient, but it becomes largely independent of the total sample volume. This is particularly useful in field or in vivo sampling, where the precise volume of the surrounding water, air, tissue, or biological fluid may not be known (Reyes-Garcés et al., 2018; Yu et al., 2022).
Headspace SPME adds a third phase. Analytes distribute among the sample, headspace, and coating. Volatile compounds can move rapidly through the headspace, while nonvolatile matrix components remain in the sample. Consequently, headspace sampling protects the coating and reduces contamination, although it discriminates against compounds with low volatility or strong affinity for the sample matrix. Zhang and Pawliszyn's foundational study demonstrated the analytical potential of this three-phase arrangement for volatile organic compounds (Zhang & Pawliszyn, 1993).
Equilibrium is only half the story
A method may be governed by equilibrium thermodynamics, but the time required to approach equilibrium is controlled by mass transfer. An analyte must move from the bulk sample across a boundary layer surrounding the device and then diffuse through or adsorb onto the coating.
The boundary layer is especially important in liquid samples. Without agitation, analytes near the coating become depleted and must diffuse across a relatively thick stagnant region. Stirring, shaking, or increasing flow reduces the effective boundary-layer thickness and accelerates extraction. This is why agitation rate must be controlled carefully when sampling is stopped before equilibrium.
Temperature has competing effects. Heating normally increases diffusion and helps volatile compounds enter the headspace, shortening extraction time. However, it may also reduce the analyte–coating distribution coefficient and promote desorption. The optimum temperature therefore reflects a balance between faster transport and lower coating affinity.
Coating thickness introduces another trade-off. A thick coating offers greater capacity but produces a longer diffusion path. Thin films improve mass transfer by offering a high surface-area-to-volume ratio, while larger thin-film devices can provide more extraction phase than traditional fibers. Device geometry is consequently not just an engineering detail: it affects sensitivity, equilibration time, mechanical strength, and compatibility with automated handling (Reyes-Garcés et al., 2018).
Importantly, equilibrium is not mandatory. Reproducible quantitative measurements can be made in the kinetic region if extraction time, temperature, agitation, device geometry, and sample conditions are rigorously controlled. Kinetic calibration methods, including approaches based on preloaded isotope-labelled standards, are especially valuable when equilibration would take too long for practical field or in vivo sampling (Yu et al., 2022).
Choosing the extraction mode and coating
Direct immersion provides intimate contact with the sample and is suitable for compounds ranging from moderately polar to hydrophobic. Its main weakness is exposure to the full matrix. Proteins, lipids, dissolved organic matter, and particles can foul the coating or change analyte mass transfer.
Headspace SPME is generally preferred for volatile and semivolatile compounds. It is widely used for aromas, off-flavours, environmental volatiles, and microbial metabolites because nonvolatile matrix material cannot reach the fiber. Adding salt, changing pH, adjusting temperature, and controlling headspace volume can alter analyte release from the sample.
Coating chemistry should match the analytical question rather than simply maximize the total signal. Nonpolar polymeric phases favour hydrophobic compounds, while mixed and porous coatings can retain a wider volatility range. More recent research has explored molecularly imprinted polymers, ionic liquids, carbon nanomaterials, metal–organic frameworks, and other materials designed to increase selectivity, capacity, or resistance to fouling. Although promising, a novel coating must also demonstrate reproducible fabrication, mechanical stability, controlled batch-to-batch behaviour, and effective desorption before it becomes useful for routine analysis (Zheng et al., 2023).
Applications across analytical chemistry
Environmental analysis was one of SPME's earliest major application areas. Fibers can collect volatile organic compounds from water or air and concentrate hydrophobic contaminants such as pesticides and polycyclic aromatic hydrocarbons. Field samplers allow extraction to begin at the sampling location, reducing the changes caused by transport and storage. Because SPME measures the fraction that exchanges with the coating, it can also provide information related to freely dissolved or bioavailable contaminant concentrations rather than only total concentration.
In food science, headspace SPME–GC–MS has become a standard approach for profiling aromas in fruits, beverages, oils, spices, and fermented products. Its selectivity can nevertheless reshape the apparent aroma profile: different coatings, temperatures, and extraction times may produce different relative peak patterns. Competitive adsorption can also occur when abundant volatiles occupy sites or displace less abundant compounds. Comparisons between samples are therefore meaningful only when the sampling protocol is tightly standardized (Reyes-Garcés et al., 2018; Zacharis & Tzanavaras, 2020).
Biomedical applications exploit SPME's small dimensions and matrix-cleanup capability. Biocompatible coatings can exclude macromolecules while extracting drugs, metabolites, and signalling molecules from blood, tissue, or other biological samples. Coupling these probes with LC–MS supports both targeted quantification and untargeted metabolomics.
The most intriguing development may be in vivo SPME. A miniature probe can be inserted temporarily into a plant, animal tissue, or other living system, allowing small molecules to be captured with limited removal of sample. Extraction onto the coating can stabilize reactive or short-lived metabolites that might otherwise disappear during excision, freezing, homogenization, or storage. Repeated measurements can reveal temporal changes within the same organism, potentially reducing biological variability and the number of organisms required (Yu et al., 2022).
Limitations researchers should not overlook
SPME is solvent-saving, but it is not automatically free from bias. Matrix effects can alter partitioning and diffusion. Porous adsorbents can become saturated, and competition may produce nonlinear responses. Carryover can occur if analytes are not completely desorbed, while fibers can age through thermal stress, fouling, or physical damage.
Quantification therefore requires appropriate calibration. External calibration may work for controlled matrices, but matrix-matched standards, standard addition, internal standards, equilibrium calibration, or kinetic calibration may be necessary for complex samples. Method validation should include extraction-time profiles, linearity, repeatability, coating lifetime, carryover, and tests of changes in temperature, agitation, salinity, pH, and sample composition.
SPME's greatest strength is not simply that it removes solvent. It changes sampling from a separate preliminary operation into an integrated part of the analytical measurement. When partitioning, mass transfer, coating selectivity, and calibration are understood, a very small extraction phase can provide information that conventional exhaustive preparation may lose—including spatial distributions, temporal changes, freely dissolved concentrations, and unstable metabolites. That combination of theory, miniaturization, and application continues to make SPME one of analytical chemistry's most versatile sample-preparation platforms.
References
Arthur, C. L., & Pawliszyn, J. (1990). Solid phase microextraction with thermal desorption using fused silica optical fibers. Analytical Chemistry, 62(19), 2145–2148. https://doi.org/10.1021/ac00218a019
Reyes-Garcés, N., Gionfriddo, E., Gómez-Ríos, G. A., et al. (2018). Advances in solid phase microextraction and perspective on future directions. Analytical Chemistry, 90(1), 302–360. https://doi.org/10.1021/acs.analchem.7b04502
Yu, M., Roszkowska, A., & Pawliszyn, J. (2022). In vivo solid-phase microextraction and applications in environmental sciences. ACS Environmental Au, 2(1), 30–41. https://doi.org/10.1021/acsenvironau.1c00024
Zacharis, C. K., & Tzanavaras, P. D. (2020). Solid-phase microextraction. Molecules, 25(2), 379. https://doi.org/10.3390/molecules25020379
Zhang, Z., & Pawliszyn, J. (1993). Headspace solid-phase microextraction. Analytical Chemistry, 65(14), 1843–1852. https://doi.org/10.1021/ac00062a008
Zheng, J., Kuang, Y., Zhou, S., et al. (2023). Latest improvements and expanding applications of solid-phase microextraction. Analytical Chemistry, 95(1), 218–237. https://doi.org/10.1021/acs.analchem.2c03246