DETEKCIJA KOVINSKIH KOMPLEKSOV IN ORGANOKOVINSKIH SPOJIN V VZORCIH IZ OKOLJA S SPEKTROMETRIJO TERMIČNIH LEČLeja Goljat
, 2019, doctoral dissertation
Abstract: Environmental pollution is one of the greatest challenges that the world is facing today. Toxic compounds, such as pesticides, allergens, pharmaceuticals, toxins and heavy metals are widely present in the air, water and soil, and can affect the health of people and animals even in small quantities, as well as they may cause long- or short-term damage in plants [Hill, 1997].
Heavy metals (mercury, arsenic, cadmium…) are widely spread in the environment. They derive from a number of sources, including mining, industrial wastes and vehicle emissions [Tchounwou et al., 2012]. They are easily incorporated into biological molecules and exert their toxic effects by displacing essential metals of a lower binding power in biologically active molecules or by acting as noncompetitive inhibitors of enzymes, affecting neurological, reproductive, renal and hematological systems [Sunil D’Souza et al., 2003; Heavy-Metal Pollution, 2018]. Metals form countless compounds (e.g. metal complexes and organometallic compounds) which are essential for living organisms (vitamin B12, hemoglobin, chlorophyll) and/or have a wide range of applications in industry and other areas, including analytical chemistry. Because of the potential risk which toxic metals represent to the living organisms and also because of the importance of some essential metals, different analytical techniques and detection methods have been developed for studies of their occurrence, fate and concentration in the environment and in organisms. However, providing a required sensitivity for determination and speciation of different metals and their compounds, especially in small- volume samples is still a challenge.
Therefore, general objectives of this dissertation were development of novel analytical methods for sensitive, reliable and fast determination of metal species, based on highly sensitive optothermal technique thermal lens spectrometry (TLS), which can be used as detection tool following colorimetric reaction of a selected metal ion or for direct detection of colored organometallic compounds.
This dissertation is composed of the following chapters: introduction, research goals, theoretical background, results and discussion, conclusion and references. The core of this dissertation is presented in the fifth chapter (results and discussion), which is divided into three parts. They separately cover development of methods for determination of iron redox species, pyoverdine and Fe-pyoverdine complexes and mercury. Pyoverdine is a siderophore, excreted by a certain bacteria in order to scavenge iron in the environment and is closely related to the chemistry of iron in such biological systems. Therefore, the first two parts are closely related.
Procedures for batch mode thermal lens microscopy (TLM), flow-injection thermal lens sprectrometry (FIA-TLS) and µFIA-TLM (flow injection and TLS detection in microspace) were developed for Fe(II) and Fe(III) determination, based on colorimetric reaction of Fe(II) with 1,10-Phenanthroline. All these procedures were focused on cloudwater examination with a tendency to minimize sample consumption but at the same time preserve low limits of detection (LOD) and limits of quantification (LOQ). TLM measurements with highly collimated probe beam were performed in a 100 μm optical path length cell (40 µL volume), which resulted in a considerably smaller sample volume requirement (500 µL in total) and consumption, as compared to UV-Vis spectrophotometry, which required at least 25 mL of sample due to large volume (almost 30 mL) of the 10 cm optical path-length sample cell. LODs for mode-mismatched TLM were 0.16 and 0.14 µM for Fe(II) and Fe(total) (sum of Fe(II) and Fe(III) concentrations), respectively, while LODs for UV-Vis spectrophotometry were 0.01 µM for both Fe(II) and Fe(total). By using the mode mismatched TLM we were able to detect concentrations corresponding to absorbances as low as 1.5 × 10-5, while the lowest absorbance detectable on the UV-Vis spectrophotometer corresponded to 1.1 × 10-3, despite the use of the 10 cm optical path-length cell.
Another important step in the development of new methods for Fe(II) and Fe(III) determination was the use of TLS detection in FIA (FIA-TLS). By injecting 50 µL of the sample into the FIA-TLS system, cca. 10 times lower LODs were achieved (1 × 10-3 µM for Fe(II) and 8 × 10-4 µM for Fe(total)), as compared to the UV- Vis spectrophotometry.
Nevertheless, the development of μFIA-TLM method, with on-line colorimetric reaction for Fe(II) and Fe(III) determination is considered as the most important achievement of this study. The results show that despite 100 times shorter optical path length and low sample consumption (3 µL of each sample/injection) compared to UV-Vis spectrophotometry, LODs for µFIA-TLM were 0.10 and 0.07 μM for Fe(II) and Fe(total) respectively, which is sufficiently for cloudwater analysis, since concentrations, lower than 0.1 μM are not expected [Parazols et al., 2006; Deguillaume et al., 2014]. Linear range for Fe(II) and Fe(III) determination by μFIA-TLM was between 0.1 and 70 µM. To test the accuracy of this method, artificial cloudwater was prepared, spiked with different amounts of Fe(II) and Fe(III) and analyzed for iron content by µFIA-TLM and UV-Vis spectrophotometry. Good agreement was observed between the two methods. To ascertain the ruggedness of the method 7 (or more) replicate determinations at two different concentrations for both, Fe(II) and Fe(total) in artificial cloudwater were carried out on day 1 (replicates were measured instantly after fortification), day 2 and day 5. A student’s t-test (p=0.05) was applied to compare 3 sets of obtained data (day 1, day 2 and day 5) and showed that sets are not significantly different from each other. Considering very low sample volume requirement of µFIA-TLM, this should be the method of choice for determination of Fe(II) and Fe(III) in investigations of processes in cloudwater, where multiparameter analysis is desired (determination of other ions, ligands, microbial counts, etc.). When larger sample volumes are available, FIA-TLS can be used for accurate determination of iron species at lowest concentration levels.
High performance liquid chromatography (HPLC) was applied for separation and detection of pyoverdine (PVD), produced by Pseudomonas fluorescens 36b5, a bacterial strain isolated from the aqueous phase of clouds at the Puy de Dôme station (1465 m, France). Reversed-phase (RP) chromatography (RP-18 chromatographic column Hypersil gold), hydrophilic interaction liquid chromatography (HILIC) (ZIC®-Hilic column) and three different detection systems
(diode-array (DAD), spectrofluorimetry (FLD) and TLS) were tested for their performance in separation and determination of pyoverdines and corresponding complexes of pyoverdine with iron (Fe(III)-PVDs).
PVDs and Fe(III)-PVD complexes could not be separated and quantified by applying HILIC technique, therefore it was concluded, that HILIC is not suitable for HPLC-DAD and also not for HPLC-TLS, since the method should offer a simultaneous sensitive detection of free PVDs as well as Fe(III)-PVD complexes in a single chromatographic run.
Since pyoverdine standards were available only as a mixture of several different forms of PVDs, whereby the exact composition was unknown, the quantification of each of the four major specie (two fluorescent PVDs and two nonfluorescent Fe(III)-PVDs) in the standard, which was obtained from Université Clermont Auvergne, Institut de Chimie de Clermont-Ferrand, was performed. When applying Hypersil gold column, a linear correlation between fluorescence intensity and absorbance of each component was observed in a concentration range 3–24 µg/mL, whereby LODs were estimated to be 0.03–0.04 µg/mL for each of the major PVD species (HPLC-DAD). Even though HPLC-FLD method provided cca. 100 times lower LODs, it is not the method of choice for determination of PVD species in cloudwater, because it does not allow detection of PVD complexes with Fe(III). When comparing HPLC-TLS and
HPLC-DAD, LODs were 5 to 8 times lower in case of HPLC-TLS, which was a significant improvement. Furthermore, recoveries (89–111 %) at two concentration levels of four PVD species in two independent samples, showed good reliability of the method.
Almost all mercury in uncontaminated drinking-water is thought to be in the form of Hg2+ [WHO, 2010]. Therefore, the method for Hg2+determination based on colorimetric reaction with triamterene, described originally by Al-Kady and Abdelmonem was further investigated in this study, as well as the possibilities of application of this reaction for Hg2+ determination by TLS. The stoichiometry of the complex formation was determined by the method of continuous variations and saturation experiment, suggesting formation of the complex with the formula Hg2-triamterene. The obtained value of the molar absorption coefficient was 9988 Lmol-1cm-1 at 403 nm, which significantly contradicts the existing data in literature, which reports the molar absorption coefficient of 5.32 × 104 Lmol-1cm-1 [Al-Kady and Abdelmonem, 2013]. Even though the spectrophotometric results were not encouraging for triamterene as colorimetric reagent for Hg2+ determination, it was further investigated for its performance in TLS system. Fe(II)-1,10-phenanthroline (ferroin) was used for comparison, because it was well studied for TLS applications previously. The results showed that Hg2-triamterene in solutions was degraded when it was exposed to the light of the excitation beam. Due to the lower molar absorptivity than reported in literature, fotodegradation and unfavorable complex stoichiometry, triamterene was not confirmed as a suitable colorimetric reagent for highly sensitive Hg2+ determination by TLS.
In summary, this dissertation investigates alternative approaches for analysis of metal complexes and organometallic compounds in small-volume environmental water samples. Methods, which were developed in this research, could potentially serve as improvements of existing technologies, to facilitate analysis of such samples, by offering simple handling of samples and superior sensitivity over the UV-Vis spectrophotometry.
Keywords: thermal lens spectrometry, thermal lens microscopy, high performance liquid chromatography, microfluidics, metal complexes, organometallic compounds, iron, pyoverdine, mercury
Published in RUNG: 05.09.2019; Views: 3570; Downloads: 152
Full text (3,65 MB)