Conference Agenda

Historical textiles of functional and decorative significance represent an essential part of our cultural heritage. Their scientific analysis, conducted using advanced analytical techniques, complements historical research and provides a deeper understanding of their material composition, production methods, cultural background, and changes they undergo due to aging and environmental exposure.

A key aspect of studying historical textile is analysing their colouring matter. In the case of the most commonly used natural dyes – mordant dyes – this includes examining both the organic colouring compounds from dyes and metals introduced through mordants. Before the dyeing process, yearns are treated with a mordant – an aqueous solution of aluminium salts or transition metals (e.g., copper, chromium, iron, tin). This treatment facilitates the bonding of dye molecules to the fibre via metal cations deposited on the fibre surface, forming stable colour complexes. The type of mordant used affects the distribution and electron density within the complex, ultimately determining the final hue of the colourations. Therefore, a comprehensive understanding of the nature of colour produced by mordant dyes is essential for accurately reconstructing the original appearance of historical objects. This requires studying both the organic and inorganic components of the colouring matter.

The study of mordant metals has often been neglected or limited to non-invasive techniques that provide only superficial information. Although such techniques are preferred and favoured in the study of heritage objects, invasive and destructive techniques often yield more detailed, accurate, and unequivocal information about historical artifacts, such as the identification of organic dyes in fabrics. However, the small amount of material available for analysis – often less than 1 mg – poses a significant challenge to the use of conventional analytical methods. This obstacle highlight the need to develop more effective and reliable analytical procedures tailored for analysing small sample quantities, particularly for the identification of mordants in historical textiles.

In this study, we employed inductively coupled plasma optical emission spectroscopy (ICP-OES) for this purpose. This technique guarantees multi-element quantitative analysis of both main and trace components within a sample in a single analytical run. To adapt ICP-OES for the analysis of precious historical objects, it was necessary to modify the conventional continuous sample introduction system. Specifically, a novel system combining flow injection analysis (FIA) and multimode sample introduction system (MSIS) was developed. FIA reduces the required sample volume to just 20 µL, while the MSIS chamber integrates the advantages of classical pneumatic nebulization (PN) with hydride generation of the analysed elements (HG). This combination allows for the detection of easily excitable elements as well as those requiring enhanced sensitivity – such as tin, a mordant metal that is particularly challenging to analyse.

The developed FIA-PN/HG-ICP-OES system was employed to establish an analytical procedure for the determination of mordant metals in objects of historical significance. The method development involved optimizing both instrumental parameters and hydride generation conditions, with particular focus on improving the efficiency of tin hydride formation. To ensure the reliability and accuracy, the method was validated by analysing of certified reference materials (CRM) – rye grass (ERM®-CD281) and marine sediment (PACS-2). Following validation, the method was successfully applied to characterize mordants in samples collected from ancient textiles, demonstrating its suitability for use in archaeometric studies of heritage artifacts.

Plasma-mediated vapor generation (PMVG) is a recent and advanced sample introduction technique compatible with atomic spectrometric detection of elements at trace levels. The plasma, typically generated in a dielectric barrier discharge (DBD) reactor, produces various reactive species responsible for analyte conversion into a gas phase. PMVG does not require chemical reagents and has been proven highly efficient in several applications [1].

Although PMVG has been applied to a considerable range of analytes and sample matrices, it remains predominantly limited to liquid samples. In such system, PMVG from liquid samples is believed to proceed via interaction of analytes with reactive species produced especially in the plasma-liquid interface, including hydrated electrons and hydrogen radicals, which have a high reduction potential, being thus likely responsible for analyte reduction in solution and production of volatile species [1]. Direct PMVG from solid samples has not yet been reported. Solid samples are typically treated by digestion or extraction to release the analyte from the matrix into the solution prior to introduction into the PMVG reactor. However, these sample preparation methods are often time-consuming and require multiple reagents. To overcome this limitation, the present work investigates the possibility of direct treatment of solid samples by PMVG for mercury determination.

A lab-made 3D-printed PMVG reactor was constructed in a DBD configuration, i.e., using two copper electrodes separated by quartz barriers. For sample introduction, approximately 20 mg of raw fish tissue (shark) was placed on a small piece of fiberglass paper (10 x 10 mm), which was inserted into the reactor. Argon was used as a discharge and carrier gas at 50 mL min^-1. The electrodes were connected to a lab-made high-voltage sinusoidal power supply, delivering 3 kV. The reactor's gas output was connected to a cold dryer (-2 °C) for water vapor condensation, followed by an amalgamator tube filled with ca. 30 mg of gold-coated alumina to preconcentrate Hg released from the solid sample during the PMVG step. The amalgamator was kept at room temperature during the PMVG stage, while it was subsequently electrically heated to 700 °C to release the trapped Hg. The gas output was connected to a T-shaped quartz absorption cell (unheated), positioned in the optical path of an atomic absorption spectrometry (AAS) instrument.

The calibration curve was linear across the entire concentration range tested (10 to 1000 µg L^-1 of MeHg⁺, 20 µL), with a coefficient of determination of 0.9996. Peak height of the transient signal was used as the measurement criterion rather than peak area due to the lower relative standard deviation (< 10%). The limit of detection was 3 µg kg^-1 (50 pg Hg absolute). The use of the amalgamator results in significant improvement of the limit of detection, since all the Hg gradually volatilized from the solid sample during the PMVG step (180 s) was rapidly released from the amalgamator, resulting in a narrow peak (full width at half maximum of 4 s). A comparison between standard solutions containing 100 µg L^-1 of Hg²⁺ or MeHg⁺ showed no significant difference in the signal intensities. This demonstrates the capability of the reactor to convert both species into free Hg atoms. The total Hg in fish tissue determined by PMVG-AAS (0.73 ± 0.05 mg kg^-1) was in agreement with results obtained by a conventional method using a single-purpose mercury analyzer AMA-254 (0.67 ± 0.14 mg kg^-1). Additionally, approximately 10 mg of certified reference material (DORM-4; fish protein) was analyzed by the proposed PMVG based method, yielding 0.39 ± 0.05 mg kg^-1, corresponding perfectly to the certified value of 0.41 ± 0.04 mg kg^-1 Hg.

Although the results are still preliminary, the proposed method demonstrates a proof of concept of direct application of a PMVG reactor to solid samples. The developed system was able to volatilize Hg from a fish tissue and release it in the form of cold Hg vapor without the need for additional reagents or extra atomizer, and using a low power plasma reactor. The whole apparatus setup consisting of a PMVG reactor, its power supply source, the cold dryer and the amalgamator with temperature control, were low-cost and lab-made, with easy operation. In summary, this work presents the first successful use of PMVG directly to an untreated solid sample.

Acknowledgements

Financial support from the Czech Science Foundation (23-05974K), Institute of Analytical Chemistry (RVO: 68081715) and MŠMT ČR (LM2023039) are gratefully acknowledged.

Reference

[1] A. D'Ulivo, R.E. Sturgeon, Vapor Generation Techniques for Trace Element Analysis – Fundamental Aspects, Elsevier, Amsterdam, 2011.

Besides the economic relevance of the oil and gas industry, they are of high environmental concern, due to the generation of waste, such as oily sludge and drill cuttings. These wastes are of high complexity and contain potentially toxic organic compounds, metals, and non-metals that are of significant environmental concern ¹. However, the determination of metals and halogens such as Br and I in environmental samples is still an analytical challenge. Moreover, the determination of these elements by inductively coupled plasma mass spectrometry (ICP-MS), presents additional limitations due to plasma-generated interferences and matrix effects. Thus, sample preparation techniques such as alkaline extraction, pyrohydrolysis, and microwave induced combustion (MIC) have been efficiently applied for the further determination of Br and I ². However, alkaline extraction is an interesting alternative since it can be assisted by microwave and/or ultrasound, and it is simple, low-cost, and efficient. Normally, the alkaline extraction of halogens has been performed by tetramethylammonium hydroxide (TMAH) combined with heating ³. Although this extraction is not selective and various compounds from the matrix may be co-extracted, the resulting extract is significantly less complex than the original sample, making it particularly advantageous for ICP-MS analysis.

In light of the analytical challenges associated with the determination of Br and I and highly complex samples, we assume that an alkaline extraction procedure combined with the advantages of ICP-MS analysis can overcome the challenges inherent in quantifying these elements in wastes from the oil and gas industry. In this study, the simplicity and efficiency of TMAH extraction combined with the versatility and sensitivity of a quadrupole-based ICP-MS were used to overcome the challenges associated with the determination of Br and I in oily sludge and drill cuttings from oil and gas exploration wells. The parameters of the ICP-MS, which include plasma radiofrequency power and the nebulizer gas flow rate, were properly optimized. The sample preparation procedure using extraction with TMAH was optimized using the Doehlert design, evaluating the effects and interactions of the selected variables by the multiple response. The optimal extraction conditions were achieved using 100 mg of sample (drill cuttings or oily sludge), 500 µL of 25% (w/v) TMAH, an extraction temperature of 75 °C, and an extraction time of 4h. The accuracy of the proposed method was checked by the analysis of certified reference materials (CRMs) and by comparing the results with those obtained after pyrohydrolysis sample preparation. The determined Br and I concentrations in the CRMs were in good agreement with the certified values (t-test, 95% confidence level). In addition, the results obtained after pyrohydrolysis showed good agreement (97 to 107%) with the certified concentrations of the analytes, and no significant differences (t-test, 95% confidence level). The precision of the proposed method was evaluated by relative standard deviation (RSD), and good precision (RSD < 10 %) and low detection limits (0.01–0.03 mg kg⁻¹) were obtained, making this method suitable for the analysis of drill cuttings and oily sludge. A short-term stability study was performed to evaluate the signal drift and/or carbon deposition onto the cone´s surface. However, after 120 min of continuous sample introduction and under the optimized instrumental condition, the analytical signals did not significantly change, and no carbon deposits were observed.

The proposed method was applied to the determination of Br and I in 14 samples of oily sludge and 17 samples of drill cuttings from onshore and offshore oil and gas wells. In the analyzed drill cuttings, the concentration of Br ranged from 0.62 to 589 mg kg⁻¹, and I ranged from < 0.01 to 3.13 µg kg⁻¹. For oily sludge samples, the concentration of Br ranged from 0.64 to 5.69 mg kg⁻¹ and I from 0.55 to 2.69 mg kg⁻¹. Additionally, for drill cuttings, the obtained Br and I concentrations, their mineralogical composition, and their distribution across different sampling depths were evaluated by two-way hierarchical clustering analysis. These clusters indicate that the drill cuttings´ sampling depth influences the availability of certain elements. The proposed method offers a reliable approach for the determination of challenging elements (Br and I) in complex oily samples from the oil and gas industry.

Acknowledgments: The authors are thankful to CNPq, CAPES, the Federal Institute for Materials Research and Testing (BAM) and Alexander von Humboldt Foundation for their financial support

References

[1] E.E. Cordes et al., Front. Environ. Sci., 4 (2016) 58.

[2] F.S. Rondan et al., J. Food Compos. Anal. 66 (2018) 199–204.

[3] M.F. Mesko et al., J. Anal at Spectrom 31 (2016) 1243–1261.