Adsorpciono ponašanje odabranih organskih ksenobiotika na sorbentima relevantnim za tretman voda

Abstract

Fokus ove doktorske disertacije je bilo izučavanje sorpcionog ponašanja tri herbicida koji se nalaze na listi prioritetnih supstanci prema Okvirnoj Direktivi o vodama EU (2000/60/EC) sa generalnim ciljem dobijanja podataka korisnih za prioritizaciju na osnovu novog fundamentalnog znanja o efikasnosti (ad)sorpcije za uklanjanje ovih supstanci iz vode u lokalno specifičnim uslovima. Odabrani su ksenobiotici različite rastvorljivosti, hidrofobnosti i donorsko-akceptorskih osobina, koji pripadaju različitim klasama herbicida (dinitroanilinski-trifluralin, hloroacetamidni-alahlor i triazinski- atrazin). Prva faza je obuhvatala karakterizaciju sorbenata i vodenih matriksa .U sintetičkom matriksu, trifluralin pokazuje veći afinitet na ispitivanim sedimentima u odnosu na alahlor i atrazin. I u slučaju organoglina, sorpcija je najizraženija za trifluralin. Alahlor i atrazin pokazuju slične sorpcione kapacitete na TMA-K, dok je na TMA-B, sorpcija izraženija za alahlor u odnosu na atrazin. U zavisnosti od prirodnog matriksa, na oba sedimenta, najveće Kd vrednosti su dobijene za trifluralin, dok su Kd vrednosti za atrazin i alahlor bile veoma slične ili pak veće za atrazin. U slučaju organoglina, vrednosti Kd su rasle sa porastom log KOW vrednosti herbicida na TMA-K u oba prirodna matriksa. Kada je u pitanju TMA-B, rast Kd vrednosti sa rastom log KOW vrednosti uočava se u površinskom matriksu pri Ce = 0,05 mg/l i 0,5 mg/l, i podzemnom matriksu pri najvećoj ravnotežnoj koncentraciji Ce = 0,5 mg/l. Kao najefikasniji sorbenti su se pokazali TMA-B (Kd = 10,5 - 483 l/kg) i model sediment (Kd = 16,4 - 761 l/kg). Odsustvo jasne linearne korelacije izmeĎu Kd i % TOC upućuje na važnost interakcija mineralne faze i herbicida. U slučaju sedimenata nisu utvrđene korelacije Kd sa log KOW vrednostima, dok kod organoglina one postoje. K d /KOW odnosi su najveći kod atrazina na svim sorbenti ma pa se pretpostavlja da on najviše učestvuje u uspostavljanju specifičnih interakcija zbog svojih donorsko-akceptorskih osobina. Što se tiče uticaja DOC, zaključuje se da poreĎenje efikasnosti sorbenata treba sprovoditi u prirodnim matriksima jer sorpcioni koeficijenti variraju u zavisnosti od koncentracije polutanata, tipa i sadržaja DOC. Kolonskim eksperimentima je pokazano da atrazin i alahlor prolaze kroz kolonu sedimenta, dok je trifluralin skoro u potpunosti adsorbovan i u vrlo maloj koncentraciji detektovan u efluentu što je u saglasnosti sa rezultatima šaržnih eksperimenata na ovom sorbentu. Procenjeni faktor retardacije pomoću modela TransMod za atrazin je veći u prirodnim matriksima (R d = 54 i Rd = 55 u podzemnoj i površinskoj vodi, redom) nego u sintetičkoj vodi (R d = 40). Kod alahlora nisu uočene značajne promene faktora retardacije meĎu matriksima (R d = 30-35). Ovi rezultati su suprotni očekivanjima na osnovu hidrofilnosti ksenobiotika i mogu biti rezultat upravo interakcija sa organskim materijama koje su bile izražene u kolonskim testovima, verovatno i više nego u šaržnim testovima gde se nije pojavilo povećanje Rd vrednosti u prirodnim matriksima u odnosu na sintetički matriks. Bez obzira što je alahlor hidrofobniji od atrazina, u uslovima datog matriksa i sorbenta ispoljava slične (šaržni testovi) ili slabije sorpcione karakteristike (kolonski testovi) koje upućuju na mobilnost kroz prvi zaštitini sloj. Treća faza je obuhvatala ispitivanje adsorpcije herbicida na aktivnim ugljevima u prahu. Ispitana je kinetika adsorpcionog procesa određivanjem koeficijenta za unutarčestični prenos mase herbicida u sirovim i ozoniranim prirodnim matriksima. Određena je efikasnost uklanjanja herbicida na razlilčitim ugljevima u prahu. Efikasnost uklanjanja se kretala u različitim opsezima u zavisnosti od odabranog uglja, vodenog matriksa i doze uglja. Na komercijalnim ugljevima najveća postignuta efikasnost (doza uglja od 15 mg/l) za alahlor iznosi u površinskoj vodi 95%, a u podzemnoj 76%. Za atrazin to je u površinskoj vodi 58% i u podzemoj vodi 56%, a za trifluralin u površinskoj vodi 87%, a u podzemnom matriksu 92%. Na uglju sitnih čestica koji se koristi u kombinaciji sa membranskom filtracijom ove vrednosti su i veće. Stepen uklanjanja DOC na komercijalnim ugljevima (pri dozi od 15 mg/l) iznosi do 57% za površinski matriks i do 51% za podzemni matriks. Slična efikasnost u uklanjanju DOC je postignuta i na uglju sitnih čestica u oba matriksa. Ozonizacija je ispoljila očekivan negativan uticaj, više ili manje izražen. U slučaju ksenobiotika, uticaji ozonizacije su bili različiti, u zavisnosti od primenjenog uglja, matriksa i supstance. Najizraženiji uticaj je kod alahlora, i to u negativnom smislu, zbog pada efikasnosti uklanjanja, verovatno zbog kompeticije sa organskom materijom ili efekta solubilizacije. U nekim slučajevima, ali mnogo manje, taj efekat je primećen i kod trifluralina. Na oba komercijalna uglja primećen je pozitivan uticaj ozonizacije na uklanjanje atrazina, ali u različitim matriksima, pa se može spekulisati pretpostavljanjem da dolazi do favorizovane raspodele u novim oblogama POM na uglju formiranim nakon ozonizacije koje su i dovoljno adsorbabilne, ali i sa dovoljno reaktivnim osobinama za asociranje sa polarnim supstancama, što je delimično i potvrđeno rezultatima adsorpcione analize.

The focus of this doctoral thesis was the study of the sorption behavior of three herbicides which are on the list of priority substances according to Water Framework Directive (2000/60/EC). The aim was to obtain the data useful for prioritization on the basis of the new fundamental knowledge about (ad) sorption efficiency for the removal of these substances from the locally specific water matrices. Selected xenobiotics were differing in solubility, hydrophobicity, and donor-acceptor properties, which belong to different classes of compounds(trifluralin as herbicide, alachlor as chloracetamide herbicide and atrazine as triazine herbicide). In the first phase of the thesis characterization of sorbents and water matrices was performed. The aim of the second phase was the testing of the sorption and desorption of organic xenobiotics in the sediments and organoclays. The equilibrium sorption parameters and removal efficiencies of herbicides were determined in different water matrices. In the synthetic matrix, trifluralin showed a higher sorption affinity onto tested sediments in comparison to the alachlor and atrazine. In the case of organoclays, sorption was mostly pronounced for trifluralin. Alachlor and atrazine have similar sorption capacities onto the TMA-K. For the TMA-B, the sorption was more pronounced for alachlor than atrazine. Depending on the natural water matrices, the highest Kd values were obtained for trifluralin onto both sediments. The Kd values for atrazine and alachlor were very similar or even greater for atrazine. In the case of organoclays, Kd values increased with increasing log KOW values of herbicides onto TMA-K in both natural matrices. When it comes to TMA-B, Kd valuesincreased with increasing log KOW values in the surface matrix for Ce = 0,05 mg/l and 0,5 mg/l, and in groundwater at the highest equilibrium concentration C e = 0,5 mg/l. TMA-B (Kd = 10,5 - 483 l/kg) and model sediment (kd= 16,4 to 761 l/kg) were the most efficient s orbents. The absence of the clear linear correlation between Kd and % TOC indicates the importance of the interaction of mineral phases and herbicides. Correlation between K d and KOW values did not existed in the case of sediments, contrary to the organoclays. Since K d /KoW ratios were the h ighest for atrazine onto all sorbents, it was assumed that atrazine mostly participate in specific interactions due to its donor-acceptor properties. Regarding the influence of DOC on sorption, it was concluded that a comparison of sorbents efficiencies should be carried in native matrices as sorption coefficients vary depending on the concentration of pollutants, the type and content of DOC. Column experiments showed that atrazine and alachlor passed through the column of the sediment, while the trifluralin almost completely was adsorbed and in very low concentrations detected in the effluent which is in accordance with the results of the batch experiments. Estimated retardation factors of atrazine were higher in natural matrices (R d = 54 and Rd = 55 in groundwater and surface water, respectively) than in the synthetic water (Rd = 40). In the case of alachlor, retardation factors were similar among water matrices (Rd = 30-35). These results are in the opposition based on the xenobiotic hydrophilicity, and they could be the consequence of interaction with the organic matter present in the column tests, probably more than in the batch tests. Estimations of Rd based on batch tests did not show an increase of Rd values in natural matrices, in comparison to the synthetic matrix. More hydrophobic alachlor, in the circumstances of the sorbent matrix exerts similar (batch tests) or lower sorption (column tests) characteristics that indicate mobility through the first protective layer. The third phase included herbicide adsorption tests onto powdered activated carbons. The adsorption kinetics was examined by determining the intraparticle mass transfer coefficients for herbicides in raw and ozonated natural water matrices. The removal efficiencies of herbicides for different powdered activated carbons were determined. Removal efficiencies were in different ranges depending on the selected carbon, water matrix and carbon dose. The highest removal efficiencies for commercial carbons (for carbon dose of 15 mg/l) for alachlor were 95% and 76% in surface and groundwater, respectively. The highest removal of atrazine was 58% in surface water and 56% in groundwater. In thecase of trifluralin, removal efficiencies were 87% in surface water and 92% in groundwater. The removal efficiencies for the carbon with fine particles (usually used in combination with a membrane filtration) were greater. The removal of DOC by commercial carbons (at the dose of 15 mg/l) was less than 57% in the surface water and 47% in groundwater. A similar efficiency in DOC removal was achieved for the carbon with fine particles in both water matrices. Ozonation showed the expected negative influence on DOC removal efficiencies, more or less pronounced. In the case of xenobiotics, ozonation influence was different, depending on the applied carbon and xenobiotic. The most important negative influence was in the case of alachlor probably due to competition with organic matter or solubilization effect. The same, but less pronounced effect was observed for trifluralin. The positive influence of ozonation was observed in removal of atrazine by both commercial carbons but in different water matrices. One can speculate that it comes to the favorable distribution in new NOM coatings on carbons formed upon ozonation, which are sufficiently adsorbable, but with enough reactive qualities for association with polar substances, which is partly confirmed by the results of adsorption analysis.