Integrating in-vial thin film microextraction using polysiloxane-based adhesive tapes with low-temperature plasma ionization mass spectrometry: A solvent-free approach for determining cocaine and methamphetamine in saliva samples

Abstract

The direct combination of sample preparation and mass spectrometry (MS) arose as an alternative to classical chromatographic methods, reducing solvent consumption while increasing the analysis throughput. In this combination, selectivity relies on the efficient isolation of the analytes from the sample matrix and the discrimination power of the instrumental technique. Low-temperature plasma-based ionization has demonstrated potential for these couplings, typically including a thermal desorption (TD) step to transfer the analytes to the gaseous phase. Thermal-resistant adhesive tapes (TRT) are promising materials in microextraction-TD couplings because they are thermally stable and can interact with target analytes thanks to their polysiloxane-based adhesive. In this paper, a novel in-vial microextraction technique based on TRT is described, and its direct coupling to Soft Ionization by Chemical Reaction In Transfer (SICRIT®) MS is explored. The TRT is attached to a commercial headspace metallic cap from which the septum has been previously removed. The cap is finally screwed in a 5 mL vial containing the sample where the extraction takes place. After the extraction, the cap containing the TRT is analyzed by a TD-SICRIT-MS/MS in a dedicated interface. This first approach has been evaluated using the determination of cocaine and methamphetamine in saliva samples as proof of concept. The limits of detection (LODs) were fixed at 1.5 μg L⁻¹, so the sensitivity fits well with the concentration level of these drugs in biofluids. The intra-day and inter-day precision, both expressed as relative standard deviation (RSD), were better than 15.9 and 16.7%, respectively. The accuracy, expressed as relative recovery, was in the range between 85–119% for both analytes. Once extracted, the analytes were stable in the tapes for at least two weeks opening the door to on-site extraction workflows. The low price of the tapes (1000 TRT segments cost ca. 7.5 $) is a positive aspect for popularizing this microextraction technique. Furthermore, the method is free of organic solvents and does not require additional gases (such as N₂ or Ar) to carry out the desorption and ionization of the analytes, which makes the process environmentally friendly, simple and safe.

---

Green foundation

1. We propose an innovative microextraction based on thermal-resistant adhesive tapes, which is directly combined with mass spectrometry (MS) by thermal desorption. It avoids chromatography, thereby reducing solvent consumption and the need for expensive auxiliary gases. The sample preparation utilizes inexpensive and commercially available elements, making it accessible to all laboratories.

2. The achievement is the development of a solventless analytical method in the bioanalytical context where liquid chromatography-MS is the gold standard. It avoids chromatography (typically operates with mobile phases at the mL min⁻¹ range), and the use of auxiliary gases (in electrospray-MS, the nebulizing gas is required in the L min⁻¹ range).

3. An open thermal desorption unit is used in this research. The design of a closed interface, where the heat would be more efficiently used in the thermal desorption of the analytes, would reduce energy consumption, thus improving the greenness.

Introduction

Ambient ionization mass spectrometry (AIMS) is an interesting alternative for the direct analysis of a wide range of samples, including biological,^1–3 environmental,^4–6 or food^7–9 ones. However, the inherent complexity of some of these matrices often requires a previous sample preparation to isolate and preconcentrate, if possible, the analytes before their final determination.¹⁰ In those cases where sensitivity is not a constraint, direct infusion mass spectrometry (DI-MS) can be used. In this alternative, no special interfaces are required, but the whole sensitivity enhancement of the sample preparation is not exploited since only a fraction of the extract is actually injected into the spectrometer. Many analytes have been analyzed following this strategy, ranging from drugs of abuse in biological samples^11,12 to pharmaceutical products in environmental samples,¹³ among others. By contrast, the integration of sample preparation and MS analysis in the same device can provide a better analytical performance as, in principle, the total amount of the extracted analyte can be potentially transferred to the spectrometer for its analysis. Solid phase microextraction (SPME) and derived techniques have been extensively used in these combinations thanks to their miniaturized character, which makes the development of dedicated interfaces simpler.^14–16 In this context, thin film microextraction (TFME) is especially useful due to its higher sorption capacity derived from the higher contact surface of planar sorptive phases.^17–19 The type of sorptive phases in the TFME-MS couplings depends on the elution strategy, in such a way that polysiloxanes are typically selected when thermal desorption is intended to be used.^20,21

In-vial microextraction has emerged in the sample preparation context as a more straightforward and compact alternative to traditional techniques. It is based on the integration of the sorptive phase in the vessel where the sample is placed. In most cases, this phase is immobilized on the vessel walls. Up to date, a few in-vial microextraction devices have been proposed based on metal–organic frameworks (MOFs),^22,23 silver nanoparticles (AgNPs),²⁴ polydimethylsiloxane/divinylbenzene composite (PDMS/DVB),²⁵ polycaprolactone²⁶ or polystyrene/oxidized carbon nanotubes films²⁷ as sorptive phases. Interestingly, in 2024, Bianchini et al. used natural material (biochar obtained from orange peel) to coat the walls of a Falcon® tube using silicone as an adhesive.²⁸

Adhesive tapes (ATs) are highly available materials. Thanks to their performance and low cost, ATs are good candidates for direct MS analysis, acting both as sampling devices and substrates in Direct Analysis in Real Time (DART),²⁹ substrate spray,^30,31 or thermal desorption-dielectric barrier discharge ionization (TD-DBDI)³² applications. In addition, ATs have been used to support samples³³ or different sorptive phases, such as polymeric particles^34–36 or polymers³⁷ in MS analysis. Among the different ATs commercialized, temperature-resistant tapes (TRTs) consist of a polyimide film coated with a polysiloxane adhesive, both components being stable at high temperatures as those used in thermal desorption experiments. In addition, polysiloxanes can interact with target analytes by hydrophobic interactions. In fact, polydimethylsiloxane (PDMS), is a paradigmatic sorptive phase in microextraction, and it has been extensively used in fiber-based SPME and TFME in biomedical^38,39 and environmental^40,41 analysis.

The use of raw adhesive tapes as sorptive phases in TFME has not yet been widely explored. In this article, we evaluate the potential of TRT for the extraction of two drugs of abuse (cocaine and methamphetamine) from saliva samples. The adhesive capacity of the tapes plays a double role. The glue is used as the actual sorptive phase but also allows the simple integration of the tape into the sample vessel, giving rise to an in-vial microextraction technique. The thermal stability of the TRT also allows the direct coupling to Soft Ionization by Chemical Reaction In Transfer (SICRIT) MS by means of a dedicated interface. In fact, this commercial interface in combination with thermal desorption has demonstrated its potential in the direct analysis of swab samplers.⁴² The proposed method can be considered green and safe for the operator. In addition, the method stands out for its low cost of materials and simplicity, being an organic solvent-free method that does not require additional gases to carry out the desorption and ionization step. All this makes this coupling a good candidate for the analysis of these two compounds in a simple and environmentally friendly way.

Experimental section

Chemicals, materials, and samples

Unless specifically mentioned, all reagents were obtained from Sigma Aldrich (Madrid, Spain). A stock solution of analytes (cocaine and methamphetamine) was prepared at 1 g L⁻¹ in methanol and stored in the freezer at −20 °C. From these stock solutions, a methanolic standard containing both analytes at 20 mg L⁻¹ was prepared and used as an intermediate solution for building the working solutions. Cocaine-d3 and methamphetamine-d5 served as internal standards (ISs) and were acquired as methanolic solutions of 1 g L⁻¹. An intermediate IS methanolic solution at 20 mg L⁻¹ was prepared.

The extraction was performed in typical headspace 5 mL-glass vials with a metallic screw cap. TRTs containing a polysiloxane glue over a polyimide film were purchased from online markets and used as sorptive phases. According to the supplier (Youmile) the TRT presents a maximum isothermal and gradient temperatures of 250 °C and 300 °C, respectively.

A pool of blank oral fluid was created by mixing samples from various volunteers to cover the inter-personal variability. This saliva pool was used for optimization and validation purposes. Before analysis, the samples were spiked, when necessary, and centrifugated at 10 000 rpm (11 069g) for 3 min to remove any dirt. Then, the pH of the sample was adjusted to 10 using an aqueous solution of ammonium hydroxide (30/70 ammonium hydroxide/water). This adjustment is essential to keep the analytes in their neutral form, thus promoting their interaction with the polysiloxane adhesive.

All experiments were performed in accordance with the Helsinki declaration and were approved by the Research Ethics Committee of the Province of Córdoba (SICEIA-2023-000030) accredited body that depends on the Andalusian Health Service (Regional Government of Andalucia). Each volunteer provided written informed consent to participate in this study following the guidelines established by the regulators (Real Decreto 1716/2011).

In-vial microextraction and TD-SICRIT-MS/MS analysis

Typical headspace 5 mL glass vials were used to hold the sample during the microextraction procedure. The septum of the metallic screw cap was removed, and a segment of TRT was glued to the outside face of the cap, as shown in Fig. S1.† The TRT maintains the tightness of the vial, and the glue exposed in the central circular section of the cap (diameter of 0.8 cm) acts as the sorptive phase.

For the extraction, 1.5 mL of the sample was placed in the vial, closing it with the modified cap as indicated in the left panel of Fig. 1. The vial was turned upside down to put the sample in contact with the tape and stirred for 30 min at 750 rpm. Finally, the cap was removed and washed directly with water, as indicated in the central panel of Fig. 1.

Fig. 1 Scheme of the complete analytical procedure. The left panel shows how the adhesive tape is adapted to the vial through a metallic cap. The central panel represents the extraction procedure including the incubation and washing step. The right panel indicates how the caps are finally analyzed by the TD-SICRIT-MS/MS.

After the extraction, the analytes are determined by TD-SICRIT-MS/MS using a dedicated interface that is presented in Fig. S2.† The design is inspired by an interface recently reported by us for the direct analysis of particulate solid samples.³² The new interface is adapted to analyze metal vial caps, thus integrating the sample preparation of biofluids with SICRIT-MS/MS analysis. The interface was built using a metal Swagelok® reducer that is screwed to the SICRIT ionization source operated at a defined frequency (15 000 Hz) and an amplitude (1400 V). A hot air gun (SEEKONE Heat Gun SDL-2816) with a continuous emission temperature of 270 °C kept the Swagelok® reducer hot to assist the thermal desorption of the analytes. The interface was pre-heated for 5 min before starting the first analysis and remained heated during the working session. As indicated in the right panel of Fig. 1, the cap containing the analytes was placed in the Swagelok® reducer, where it was maintained for 30 s while the analytes were desorbed. The cap does not seal the interface, so room air can enter the interface (the flow rate is defined by the vacuum of the MS inlet, typically being 2 L min⁻¹), thus transferring the analytes to the SICRIT-MS/MS. The ionized analytes are determined in a Thermo LTQ hybrid mass spectrometer (Thermo Fisher Scientific, San Francisco, CA, USA) using the Ion Trap analyzer. The MS parameters are specified in the ESI (Table S1†). Finally, an external magnet was used to remove the cap to ensure the safety of the operator. A margin of 20 s was established between samples to avoid cross-contamination, thus providing an analysis throughput of 1 sample per min. The TD-SICRIT-MS/MS procedure is clearly shown in Movie S1 (ESI†).

Optimization of the extraction parameters

In AIMS, the ionization of the analytes is produced in open sources, which may compromise the reproducibility of absolute analytical signals. For this reason, the optimization of the extraction parameters was done using a typical extraction-chemical elution workflow, the final extracts being analyzed in an Agilent 1260 Infinity liquid chromatography (LC) system (Agilent, Palo Alto, CA, USA) equipped with an Agilent 6420 Triple Quadrupole MS. A direct infusion tandem mass spectrometry (DI-MS/MS) analysis was used for analytes determination. The description of the DI-MS/MS is presented in ESI (Table S2†). The signal trends observed in optimization are independent of the analytical system.³⁵

In this case, the extraction procedure comprised the steps shown in the central panel of Fig. 1, including a final elution. This elution was done by applying 500 μL of methanol inside the cap. For better elution, this volume was aspirated and released 20 times over the tape. The final methanolic extract was transferred to an HPLC vial and analyzed.

Results and discussion

Polysiloxanes are interesting polymers due to their high thermal stability, chemical inertness, and non-polar nature, among other properties. They have been extensively used in analytical chemistry, for instance, as coatings in solid-phase microextraction fibers and gas chromatographic columns. Commercial TRTs are composed of a polyimide substrate coated with a polysiloxane adhesive, both polymers maintaining their properties at high temperatures. Fig. S3† shows the ATR-IR spectrum of the adhesive layer, which fits with the spectra reported for polysiloxane polymers in the literature.^43–45 Polysiloxanes exhibit two characteristic bands at 1060 and 1016 cm⁻¹ corresponding to the Si–O–Si stretching. Also, C–H asymmetric and symmetric stretching bands can be observed at 2963 and 2902 cm⁻¹, respectively. Asymmetric (1412 cm⁻¹) and symmetric (1255 cm⁻¹) Si–CH₃ bending, Si(CH₃)₂ stretching (841 cm⁻¹), Si–C symmetric stretching (790 cm⁻¹), and Si–C stretching (693 cm⁻¹) are also easily observed in Fig. S3.† Energy Dispersive X-ray spectroscopy (EDX) measurements, performed at the Central Services for Research Support (SCAI) of the University of Córdoba, were also used to investigate the composition of the surface. The spectrum of both sides of the tape (polyimide film and coated adhesive) and their composition are shown in Fig. S4 and Table S3,† respectively. The presence of Si in the adhesive coating corroborates the chemical nature of the glue. In addition, the absence of N (present in the polyimide structure) on the adhesive coating indicates a complete coverage of the polyimide film with the glue.

Although incubating (by immersion) the tapes in the sample may seem the simplest way to use them as sorptive phases, the adhesive capacity of the tapes makes this process difficult to reproduce. The tapes wrinkle and stick together, so maintaining the same sorptive phase/sample contact area is very difficult. However, the adhesive capacity is more of an advantage than a disadvantage, as it allows the tapes to be easily immobilized on a surface. In this work, the TRTs are stuck to metallic caps, thus always exposing the same area (defined by the central ring of the cap, see Fig. S1†) to the samples. In this configuration, it is essential to assess the watertightness of the extraction devices. Watertightness was gravimetrically evaluated. For this purpose, three vials were filled with 1.5 mL of water, closed with caps modified with TRTs, and agitated upside down for 24 h. Negligible water losses were observed, thus indicating that the vials were tightly sealed during the extraction.

Once the adhesive was chemically characterized and the watertightness demonstrated, the potential of TRT for the extraction of target compounds was evaluated. Considering the non-polar nature of the polysiloxane and the thermal desorption step integrated during the MS analysis, the target compounds should be: (i) non-polar or semipolar compounds to be extracted by the polymer and (ii) volatile or semivolatile to be easily desorbed in the interface. With these two criteria in mind and considering our recent experience in determining drugs, cocaine and methamphetamine were selected as model analytes (chemical properties indicated in Table S4†). Also to promote their transference to the adhesive tape, the extraction pH was fixed at alkaline conditions to have the analytes in their neutral form. To go beyond a simple proof of concept, in this work a method for determining both drugs in saliva samples has been optimized and fully validated.

The effect of different variables on the extraction performance was initially evaluated by using blank saliva samples spiked with the target analytes at a concentration of 50 μg L⁻¹. The inherent viscosity of the saliva can adversely affect the extraction by hindering the analytes diffusion to the sorptive phase. This disadvantage can be solved by diluting the sample with water after its centrifugation. However, this dilution decreases the concentration of the analytes negatively affecting the sensitivity. The dilution factor was studied at two levels (1/2 and 1/4 v/v) and the results were compared with those obtained for undiluted samples (Fig. 2a). The best results were obtained for undiluted saliva which was selected for further studies. This selection also simplifies and speeds up the process by reducing the number of steps. In this first study and to fully demonstrate the key role of the tape on the extraction of the analytes, the sorption capacity of both sides of the tape (polyimide and adhesive sides) and the cap were evaluated independently. The results showed that the contribution of the cap and the polyimide side to the extraction of the analytes was negligible.

Fig. 2 Effect of the (a) sample dilution, (b) ionic strength, (c) agitation rate and (d) extraction time on the absolute analytical signal. All the experiments were performed in triplicate and the standard deviation is presented as error bars. The starting conditions for the study of the first variable were: sample adjusted at pH 10 with no salt addition and agitated during 1 h at 150 rpm.

When hydrophobic interactions are involved in the extraction process, the salting-out effect may have beneficial effects by decreasing the solubility of analytes in the sample matrix and favoring their transference to the sorptive phase. For this reason, different amounts of sodium chloride (selected as model electrolyte) were added to saliva samples after its centrifugation to a final concentration of 1, 3, and 5% w/v. As can be observed (Fig. 2b), the ionic strength increases the extraction performance at 1% (due to a salting out effect) although a negative effect is observed at higher values (due to an increase in the viscosity of the sample). However, the effects are not too marked, and they can be normalized by using the corresponding isotopically labelled compounds as indicated in Fig. S5.†

The sample agitation enhances mass transference, and it was evaluated in an orbital shaker at five different rates (150, 300, 500, 750, and 1000 rpm). As indicated in Fig. 2c the extraction increased with time up to 750 rpm, which was selected as the optimum value, while a slight decay was observed at 1000 rpm probably due to an increase in the temperature of the sample by the strong agitation. At 750 rpm, the extraction rate was time dependent, as can be observed in Fig. 2d. Considering a compromise between analytical signal and sample throughput, 30 min was selected as the extraction time.

Once the variables affecting the extraction were studied, the method was validated by the TD-SICRIT-MS/MS following the ICH M10 guideline.⁴⁶ A 1/x weight matrix-match calibration was built analyzing blank saliva samples spiked with the target analytes at different concentration levels. ISs were also added to improve the reproducibility of the analytical signal by normalizing the variability of the extraction and analysis steps. Fig. 3 shows the desorption profile of cocaine and methamphetamine obtained for the consecutive analysis of six independent blank samples spiked at 10 μg L⁻¹. Each panel of Fig. 3 shows the MS transition of each analyte. Some conclusions can be obtained from the chronogram. On the one hand, each sample is analyzed in ca. 1 min, thus providing a high analysis throughput. On the other hand, the absolute signals for methamphetamine are highly reproducible between samples, while a higher variability in peak shape is observed for cocaine. This behavior can be ascribed to the lower polarity (less intense interaction with the adhesive) and higher volatility (easier thermal desorption) of methamphetamine. The latter aspect does not have an adverse effect on analytical performance since using internal standards, as shown in the validation, significantly reduces the variability.

Fig. 3 Chronograms for (a) cocaine and (b) methamphetamine obtained for the analysis of six independent blank samples spiked with the analytes at 10 μg L⁻¹.

The calibration curve and the validation results are presented in Fig. S6† and Table 1, respectively. The linearity was maintained from LOQ to 200 μg L⁻¹ with R² of 0.9928 and 0.9908 for cocaine and methamphetamine, respectively. The limit of detection (LOD) and quantification (LOQ) were calculated for a S/N of 3 and 10, respectively. LOD resulted to be 1.5 μg L⁻¹ while LOQ was 5 μg L⁻¹. The precision and accuracy were investigated using quality controls (QCs) at LOQ, 10 μg L⁻¹ (low QC), 75 μg L⁻¹ (medium QC), and 200 μg L⁻¹ (high QC). The intra-day precision was investigated by quintuplicate for each QC and the inter-day precision was studied using three independent extractions at each concentration levels on three different days. At LOQ the intra-day and inter-day precision were better than 7.6 and 11.3%, respectively. For the rest of QC, the intra-day and inter-day precision were better than 15.9 and 16.7, respectively. The accuracy, expressed as relative recovery, was studied in quintuplicate for all QCs. The relative recoveries were in the range between 96–119% at LOQ and in the range between 85–109 for the rest of QCs. During the analysis of the spiked samples, no carry-over between samples was observed. The use of an independent cap/tape for each sample and the continuous heating of the main body of the interface (avoiding the deposition of the analytes in the interface) can be the reason for this positive fact.

Table 1 Analytical features of cocaine and methamphetamine in saliva samples by in-vial TFME combined with TD-SICRIT-MS/MS

Analyte	LOD (μg L^−1)	LOQ (μg L^−1)	R ^2	Lineal range (μg L^−1)	RSD intra-day (%, n = 5)					RSD inter-day (%, n = 9^a)				Accuracy intra-day (% relative recovery, n = 5)
					5 μg L^−1	LQC μg L^−1	MQC μg L^−1	HQC μg L^−1	Dilution QC	5 μg L^−1	LQC μg L^−1	MQC μg L^−1	HQC μg L^−1	5 μg L^−1	LQC μg L^−1	MQC μg L^−1	HQC μg L^−1	Dilution QC
LOD, limit of detection; LOQ, limit of quantification; LQC, low concentration quality control; MQC, medium concentration quality control; HQC, high concentration quality control; RSD, relative standard deviation.a Inter-day precision was studied using three independent extractions at each concentration level on three different days (3 × 3 = 9).
Cocaine	1.5	5	0.9928	LOQ-200	7.6	10.8	11.3	15.9	21.0	11.3	13.8	16.7	14.1	96 ± 9	94 ± 11	85 ± 10	102 ± 16	126 ± 27
Methamphetamine	1.5	5	0.9908	LOQ-200	6.6	4.7	6.6	6.4	9.3	6.5	11.2	10.5	11.0	119 ± 6	105 ± 4	109 ± 7	108 ± 7	113 ± 10

---

To research the dilution integrity of the sample, a concentration five times greater than upper limit of quantification was prepared. The sample was diluted ten times with blank saliva and analyzed by the proposed methodology in quintuplicate and the intra-day precision and accuracy were investigated for each analyte. While the values obtained for methamphetamine were acceptable, the values obtained for cocaine slightly exceeded the limits allowed by the validation guide.

In-vial microextraction using polysiloxane adhesive tapes combined with TD-SICRIT-MS, a bigger picture

The novel analytical platform has been optimized and validated for determining two drugs (cocaine and methamphetamine) in saliva samples. These target analytes were preliminary selected according to their chemical properties, specifically their semivolatile character and intermediate polarity. The optimized method was finally applied to samples containing codeine and methadone to have a better idea of the application scope of the new platform. Codeine and methadone define a wider polarity range. In fact, methadone is less polar than cocaine and methamphetamine, while codeine is more polar than them (Table S4†). The extraction of samples containing these new drugs and their analysis by TD-SICRIT-MS/MS provided no signal for codeine, while the signal for methadone was not intense. A deeper study, using chemical elution of the tapes and the analysis of the eluates by LC-MS, demonstrated that codeine was not extracted in the tapes. This result is in line with literature where polysiloxanes (e.g., PDMS) are not indicated for the extraction of polar compounds. However, the situation behind methadone results was different. In fact, methadone was effectively extracted when chemical elution and LC-MS is used for its analysis. A retention of the analyte was observed not only in the tape but also in the metallic cap. We hypothesized that an inefficient thermal desorption could be the reason for the low sensitivity in TD-SICRIT-MS/MS. Movie S2 (ESI†) shows the evolution of the temperature on the TD interface before and after placing the cap with the TRT. The interface is maintained at a stationary temperature of ca. 145 °C, and this temperature increased up to 183 °C in less than 8 s when the cap is placed due to heat accumulation. Considering the expected strong interaction of methadone with the glue and its high boiling point, we can conclude that 183 °C is not enough to thermally elute the analyte and transfer it to the gaseous phase. Therefore, in its present form, the platform can be used to detect semivolatile compounds with intermediate polarity.

Commercial microextraction devices are typically expensive, especially those designed for AIMS. In this case, 30 m of the TRT used in this work costs around 9 $. Considering that ca. 2.5 cm is required for each extraction, 1000 sorptive phases cost around 7.5 $. This aspect increases the affordability of the extraction technique and is very positive for its popularization. The low cost and small size of the devices open the door to the on-site application of the extraction technique. In this sense, the stability of the analytes in the tape, once extracted, plays a critical role. This stability was investigated at two concentrations (10 and 200 μg L⁻¹), storing the phases at two temperatures (room temperature and 4 °C) for two weeks after the extraction. After the extraction, the caps were just covered with parafilm to avoid tape contamination. The relative analytical signals were used as analytical parameters, and their evolution was compared with those obtained on the same day of extraction (denoted as week 0). The results (shown in Fig. S7†) demonstrated an appropriate stability of the analytes on the tapes. However, for on-site applications, more extreme conditions should be evaluated to simulate ordinary mail shipping. In this regard, the sample storage, done here simply with parafilm, should be improved.

The in-vial microextraction allows the simultaneous incubation of 40 samples, a value that can be increased if a bigger orbital shaker is used. Considering that the incubation time was fixed at 30 min and accounting for the time required to develop the rest of the steps (e.g., pH adjustment, introduction of the samples in the vials, and washing the tapes after the extraction), an extraction of 40 samples per hour is reasonable. As is observable in Fig. 3, each cap is analyzed in one minute, providing a measurement throughput of 60 samples per hour. These values provide a potential sample throughput of 280 samples in a working session of 8 h. The complete automation of the analysis step would allow to reach this value.

The sustainability and practicality of the platform were evaluated using the Analytical Greenness Metric for Sample Preparation (AGREEprep)^47,48 and Blue Applicability Grade Index (BAGI)⁴⁹ metrics, respectively. The pictograms obtained for both metrics, presented in Fig. S8 and S9,† demonstrated the green character of the sample preparation and the applicability of the platform to solve the given analytical problem being, in turn, a solvent/gas-free method. The applicability is also demonstrated by comparing the proposed method with other counterparts reported in the literature for solving the same analytical problem.^50–56 This comparison, presented in Table 2, showed that the new method is competitive in analytical figures while requiring less time and resources to be developed. Although the sample throughput and reagent consumption are not specifically indicated in some of these articles, it can be inferred that the new method outperforms many of these alternatives, except for the one based on paper spray, in these aspects. Our experience indicates that paper-spray measurements require time to place the substrate in front of the mass spectrometer inlet, a step that, in our approach, is easy and fast to perform.

Table 2 Comparison of the proposed method with others reported in the literature

Analytes	Matrix	Pre-treatment	Instrumental technique	LOD (μg L^−1)	RSD intra-day (%)	RSD inter-day (%)	Accuracy (% RR)	Additional green character	Ref.
LOD, limit of detection; RSD, relative standard deviation; RR, relative recovery. Pre-treatment: TFME, thin film microextraction; PT-SPE: pipette tip-solid phase extraction; SPE: solid phase extraction; D-SPE; dispersive solid phase extraction; HS-SPME: headspace-solid phase microextraction; LLE: liquid–liquid extraction. Instrumental technique: MIP-PSI-MS, molecularly imprinted polymer assisted paper spray ionization mass spectrometry; Nano-ESI-mini-MS, nano electrospray mini mass spectrometry; IMS: ion mobility mass spectrometry; LC-MS/MS: liquid chromatography tandem mass spectrometry; GC-MS: gas chromatography mass spectrometry; DI-MS/MS: direct infusion-tandem mass spectrometry; TD-SICRIT-MS: thermal desorption Soft Ionization by Chemical Reaction In Transfer mass spectrometry.
Methamphetamine	Urine	TFME	MIP-PSI-MS	0.8	<9.5	<3.2	98.9–104.2	On-line elution and analysis	50
Cocaine	Urine	PT-SPE	nanoESI-mini-MS	0.25	<8.5	—	89.7–92.6	Direct analysis	51
Methamphetamine				1	<8.5	—	90.5–92.9
Cocaine	Saliva	SPE	IMS	18	<7	<12	81–100	Direct analysis	52
Methamphetamine	Saliva	D-SPE	LC-MS/MS	5.29	1.14	—	85.8–116		53
Cocaine	Urine	HS-SPME	GC-MS	0.6	<6.5	<8.0	93.0–106.7	Solvent-free method	54
Methamphetamine				0.2	<6.9	<8.6	92.0–117.6
Cocaine	Saliva	TFME	DI-MS/MS	1.5	<2.6	<13.3	92.0–102.7	Natural materials as precursors	55
Methamphetamine	Oral fluid	LLE	GC-MS	5	<2.87	<9.73	99.7–100		56
Cocaine	Saliva	TFME	TD-SICRIT-MS	1.5	<15.9	<16.7	85.2–101.9	Solvent-free method	This work
Methamphetamine				1.5	<6.6	<11.2	105.0–119.4

---

Conclusions

In this work, a novel in-vial thin film microextraction technique based on using commercial adhesive tapes as the sorptive phase is presented. TRT tapes consisting of a polysiloxane glue over a polyimide film are used due to the well-known sorption capacity of polysiloxanes. Their chemical nature makes TRT compatible with thermal desorption, and this aspect has been exploited to directly combine the microextraction technique with SICRIT-MS/MS using a dedicated TD unit. This combination has been used for determining cocaine and methamphetamine in saliva samples with appropriate validation parameters. The sensitivity levels, in the low μg L⁻¹ range, allows the determination of both drugs in saliva samples below for example the cut-off concentration levels defined by the European Workplace Drug Testing Society for both analytes in saliva samples.⁵⁷

A deep evaluation of the new platform indicates that, in the present form, it can be used for the determination of semivolatile compounds with intermediate polarity, thus providing a broad enough scope to solve many analytical problems. The extension of this scope should rely on using more polar coatings (for extracting polar substances) and applying higher desorption temperatures (for determining non-polar compounds strongly bound to the polysiloxane). Both aspects will be the focus of further research. The stability of the analytes in the tapes, preliminary evaluated in this article, opens the door to on-site extraction, extending the workflows at which the technique can be applied.

The potential of the platform is also supported by the high sample throughput that can be potentially achieved. In this context, the automation of the analysis step is identified as an interesting aspect for further research to improve the potential of the technique. Also, the affordability of the tapes (cost-effective and commercially available on the Internet) is a positive aspect for popularizing this microextraction technique.

The sustainable character and applicability of the new platform have also been evaluated, showing excellent indexes for both aspects.

Author contributions

C. Calero-Cañuelo: Investigation, methodology, writing – original draft. R. Lucena: Conceptualization, funding acquisition, writing – review & editing. S. Cárdenas: Project administration, funding acquisition, writing – review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data supporting this article have been included in the optimization (Fig. 2) and calibration (Fig. S6†) graphs, where the data with their corresponding variability (represented as error bars) are included. Table 1 summarizes the main analytical parameters obtained from the data presented in Fig. 4. Additional data are also included as part of the ESI.†

Acknowledgements

Grant PID2023-146313OB-I00, funded by MICIU/AEI/10.13039/501100011033, is gratefully acknowledged. Carlos Calero-Cañuelo expresses his gratitude for the predoctoral grant (Ref. FPU20/04765) from the Spanish Ministry of Science, Innovation, and Universities. This article is based upon work from the National Network for Sustainable Sample Preparation (RED2022-134079-T funded by MICIU/AEI/10.13039/501100011033) and the Sample Preparation Study Group and Network supported by the Division of Analytical Chemistry of the European Chemical Society.

References

1. W. Zhou, E. Nazdrajić and J. Pawliszyn, J. Am. Soc. Mass Spectrom., 2022, 33, 1187–1193.
2. Y.-N. Yao, D. Di, Z.-C. Yuan, L. Wu and B. Hu, Anal. Chem., 2020, 92, 6207–6212.
3. S. Mondal, U. Pandey, S. Chakrabarti, P. Pahchan, D. Koner and S. Banerjee, J. Proteome Res., 2025, 24, 2314–2323.
4. M. Zhang, Y. Niu, R. Shang, Y. Xu, H. Zhang, K. Yu, G. Kan, D. Song, Y. Jiang and J. Jiang, J. Environ. Chem. Eng., 2025, 13, 115938.
5. K. Hu, W. Zhou, C. Yang, Y. Wang, R. W. Jiang, Z. Zhang and J. Pawliszyn, J. Hazard. Mater., 2024, 472, 134609.
6. F. P. M. Jjunju, D. E. Damon, D. Romero-Perez, I. S. Young, R. J. Ward, A. Marshall, S. Maher and A. K. Badu-Tawiah, Sci. Rep., 2020, 10, 10698.
7. A. Geballa-Koukoula, A. Gerssen, M. H. Blokland and M. W. F. Nielen, J. Am. Soc. Mass Spectrom., 2022, 33, 2038–2045.
8. A. Kasperkiewicz, G. A. Gómez-Ríos, D. Hein and J. Pawliszyn, Anal. Chem., 2019, 91, 13039–13046.
9. K.-H. Chen, Y.-C. Li, F. Sheu and C.-H. Lin, Food Chem., 2021, 363, 130305.
10. J. Millán-Santiago, C. Calero-Cañuelo, R. Lucena and S. Cárdenas, J. Pharm. Biomed. Anal., 2025, 253, 116535.
11. A. M. Pedraza-Soto, C. Calero-Cañuelo, R. Lucena and S. Cárdenas, Microchem. J., 2024, 202, 110732.
12. R.-J. Raterink, Y. Witkam, R. J. Vreeken, R. Ramautar and T. Hankemeier, Anal. Chem., 2014, 86, 10323–10330.
13. I. Perkons, J. Rusko, D. Zacs and V. Bartkevics, Sci. Total Environ., 2021, 755, 142688.
14. M. F. Mirabelli, J.-C. Wolf and R. Zenobi, Anal. Chem., 2016, 88, 7252–7258.
15. M. C. Dumlao, L. E. Jeffress, J. J. Gooding and W. A. Donald, Analyst, 2016, 141, 3714–3721.
16. D. Rickert, G. A. Gómez-Ríos, V. Singh and J. Pawliszyn, Rapid Commun. Mass Spectrom., 2022, 36, e9388.
17. Y. A. Olcer, M. Tascon, A. E. Eroglu and E. Boyacı, TrAC, Trends Anal. Chem., 2019, 113, 93–101.
18. W. Zhou, Q. Lan, M. Dutt and J. Pawliszyn, Anal. Chem., 2024, 96, 16520–16524.
19. M. Tascon, G. A. Gómez-Ríos, N. Reyes-Garcés, J. Poole, E. Boyacı and J. Pawliszyn, J. Pharm. Biomed. Anal., 2017, 144, 106–111.
20. K. Murtada, V. Galpin, J. J. Grandy, V. Singh, F. Sanchez and J. Pawliszyn, Sustainable Chem. Pharm., 2021, 21, 100435.
21. M. N. Wieczorek, W. Zhou and J. Pawliszyn, Food Chem., 2022, 389, 133038.
22. I. Taima-Mancera, M. J. Trujillo-Rodríguez, J. Pasán and V. Pino, Anal. Chim. Acta, 2025, 1345, 343663.
23. I. Taima-Mancera, M. J. Trujillo-Rodríguez, J. Pasán and V. Pino, Adv. Funct. Mater., 2024, 34, 2402517.
24. A. Rodríguez-González, A. I. Jiménez-Abizanda, V. Pino and A. Gutiérrez-Serpa, J. Chromatogr. Open, 2024, 6, 100182.
25. X. Liu, Q. Hu, Y. Tong, N. Li, S. Ouyang, H. Yang, J. Xu and G. Ouyang, Anal. Chim. Acta, 2021, 1152, 338226.
26. F. Merlo, V. Cerviani, C. Fontàs, A. Speltini, A. Profumo and E. Anticò, Anal. Chim. Acta, 2025, 344164.
27. X.-M. He, G.-T. Zhu, J. Yin, Q. Zhao, B.-F. Yuan and Y.-Q. Feng, J. Chromatogr. A, 2014, 1351, 29–36.
28. P. Bianchini, F. Merlo, V. Quarta, L. Ferrari, C. Milanese, A. Profumo and A. Speltini, Adv. Sample Prep., 2024, 12, 100129.
29. K. Maeno, Sci. Rep., 2019, 9, 17789.
30. C. B. Nguyen, W. R. A. Wichert, D. O. Carmany, E. M. McBride, P. M. Mach, E. S. Dhummakupt, T. Glaros and N. E. Manicke, Anal. Chem., 2021, 93, 13467–13474.
31. S. Prunty, D. Carmany, E. S. Dhummakupt and N. E. Manicke, J. Forensic Sci., 2023, 68, 1615–1625.
32. C. Calero-Cañuelo, R. Lucena and S. Cárdenas, Talanta, 2025, 293, 128042.
33. D. Saigusa, R. Saito, K. Kawamoto, A. Uruno, K. Kano, J. Aoki, M. Yamamoto and T. Kawamoto, Anal. Chem., 2019, 91, 8979–8986.
34. P.-K. So, B.-C. Yang, W. Li, L. Wu and B. Hu, Anal. Chem., 2019, 91, 9430–9434.
35. C. Calero-Cañuelo, F. A. Casado-Carmona, R. Lucena and S. Cárdenas, Talanta, 2024, 272, 125774.
36. A. Belhameid, F. A. Casado-Carmona, A. Megriche, Á. I. López-Lorente, R. Lucena and S. Cárdenas, Microchim. Acta, 2024, 191, 513.
37. A. M. Pedraza-Soto, R. Lucena and S. Cárdenas, Anal. Chim. Acta, 2025, 1348, 343814.
38. I. Gazioglu, A. Kabir, O. S. Zengin, E. Kepekci Tekkeli, K. G. Furton, A. Tartaglia and M. Locatelli, Journal of Chromatography, B, 2021, 1163, 122520.
39. N. H. Godage and E. Gionfriddo, Journal of Chromatography, B, 2022, 1203, 123308.
40. A. Castiñeira-Landeira, L. Vazquez, A. M. Carro, M. Celeiro, A. Kabir, K. G. Furton, T. Dagnac and M. Llompart, Microchem. J., 2024, 196, 109542.
41. A. C. Heredia, J. Cabrera-Peralta, I. Ocaña-Rios and A. Peña-Alvarez, Chem. Pap., 2023, 77, 2039–2049.
42. T. P. Forbes, E. L. Robinson, E. Sisco and A. Koss, Anal. Chem., 2024, 96, 13352–13357.
43. J. Giese-Hinz, C. Kothe, C. Louter and B. Weller, Int. J. Adhes. Adhes., 2022, 117, 103019.
44. J. C. Steinbach, M. Schneider, O. Hauler, G. Lorenz, K. Rebner and A. Kandelbauer, Polymers, 2020, 12, 2473.
45. S. I. Salih, J. K. Oleiwi and H. M. Ali, IOP Conf. Ser.: Mater. Sci. Eng., 2018, 454, 012086.
46. Committee for Medicinal Products for Human Use, ICH guideline M10 on bioanalytical method validation and study sample analysis Step5, 2022.
47. W. Wojnowski, M. Tobiszewski, F. Pena-Pereira and E. Psillakis, TrAC, Trends Anal. Chem., 2022, 149, 116553.
48. R. Rajput and S. Desai, Green Anal. Chem., 2024, 11, 100167.
49. N. Manousi, W. Wojnowski, J. Płotka-Wasylka and V. Samanidou, Green Chem., 2023, 25, 7598–7604.
50. T. de Brito, D. de Aguiar, I. Pereira and B. Vaz, J. Braz. Chem. Soc., 2021, 32, 269–276.
51. J. Wu, Y. Zhang, M. Zhang, L. Dong, B. Zou, J. Jin, Y. Liu, D. Yu, Y. Xu, Y. Fan, Z. Ouyang, Q. Shen and H. Wang, J. Chromatogr. A, 2024, 1738, 465485.
52. A. Sorribes-Soriano, F. A. Esteve-Turrillas, S. Armenta, M. de la Guardia and J. M. Herrero-Martínez, J. Chromatogr. A, 2017, 1481, 23–30.
53. J. Anjali, K. Himanshu, N. Biswa Prakash, G. Kavita and S. Rajendra Kumar, Lett. Appl. NanoBioScience, 2024, 13, 100.
54. T. Baheri, Y. Yamini, M. Shamsayei and M. Tabibpour, J. Sep. Sci., 2021, 44, 2814–2823.
55. C. Calero-Cañuelo, R. Lucena and S. Cárdenas, Anal. Bioanal. Chem., 2025, 417, 1317–1326.
56. L. Bahmanabadi, M. Akhgari, F. Jokar and H. Sadeghi, Hum. Exp. Toxicol., 2017, 36, 195–202.
57. European Workplace Drug Testing Society (EWDTS), European Guidelines for Workplace Drug Testing in Oral Fluid, 2022.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5gc02488a

---