Studies by chimiometric methods of the interaction between Pb(II) and humic acids

SILVIA ORSETTI; ESTELA ANDRADE; FERNANDO MOLINA

Tenerife, Islas Canarias, España

Congreso; 15th Meeting of the International Humic Substances Society; 2010

International Humic Substances Society

1. Introduction Humic substances are important components of natural organic matter in groundwaters and soils, where they have a fundamental role in the fate of polutants such as heavy metals. In particular, retention and transport of hazardous substances by humic and fulvic acids have been frequently studied in the last few years. Fluorescence spectroscopy, combined with multi – way data analysis techniques, is a powerful tool to obtain information regarding humic substances and their interaction with heavy metals, such as Pb(II). In this kind of analysis, excitation emission matrices (EEM) are used: they are obtained by combination of emission spectra measured at different excitation wavelengths. In general, humic acids show a broad emission peak between 300 and 700 nm and a broad excitation peak between 300 and 450 nm. However, the location and shape of these peaks strongly depend of the origin of the humic material. EEMs provide wide information, and until recently EEMs characterization techniques were focused on visual identification and shape of peaks, leading to qualitative analysis. Recently, multi – way data analysis techniques have been introduced in the study of fluorescence signals of natural organic matter. One of these techniques is parallel factor analysis (PARAFAC), which is able to decompose the fluorescence signal into the underlying fluorescent individual phenomena [1,2]. PARAFAC models three – way data by using ec. 1, fitting the equation by minimizing the sum of squares of the residuals (ijk): (1) xijk is an element of the three – way array with dimensions I, J and K. In the case of EEMs, xijk is the fluorescence intensity of sample i, measured at the emission wavelength j and excitation wavelength k. The term ijk accounts for the unexplained signal (experimental error). The output of the model are the parameters a, b and c. These represent, ideally, the concentration, emission spectra and excitation spectra of the underlying fluorophores, respectively, and they are usually referred as scores (a) and loadings (b, c). In this way, important information about the number of fluorophores, their fluorescence profiles and their behavior in presence of the metal is obtained; the last subject, from the relative concentration of each fluorophore in each individual sample, with a characteristic Pb(II) concentration [3]. The aim of this work is to study EEMs of humic acids (HA) at several Pb(II) concentrations, analyzing them with PARAFAC as a whole to estimate the number of individual components of the sample (fluorophores), their emission and excitation fluorescence spectra, as well as the relative concentration of those in each sample. The effect of pH value and ionic strength is also analyzed. 2. Materials and Methods The EEMs of two different HA where measured in absence and presence of several concentrations of Pb(II). These measurements were conducted for Elliot Soil humic acid (EHA, reference material of the IHSS) and other commercially available from Fluka, previously purified (FHA). Experiments were performed at pH 4.0 and 5.5, and at two values of ionic strength: NaClO4 0.1 and 0.02 M. The samples were prepared by dissolving the humic acid in the NaClO4 (aprox concentration of 30 mg L-1) using a minimum quantity of NaOH, and then the pH was adjusted with HClO4. This sample was separated in 25.0 ml aliquotes, and different volumes of a stock solution of Pb(II) were added to these. The EEMs of those aliquotes were measured, as well as the EEM of the humic acid sample without Pb(II). In all cases, the samples were under N2 atmosphere. Data processing: after elimination of the scattering zones in the EEMs, PARAFAC was used indicating number of components from 2 to 5, adding a non negativity constrain for the b and c loadings (which account for emission and excitation spectra respectively). From the values of core consistency and SD residuals, the number of components was estimated following Bro and Kiers [4], considering that a high core consistency value (over 50%) means that the model presents high spectral resolution of components. From there a, b and c were obtained. 3. Results and Discussion A total of 3 independent components were estimated for both humic acids. In Fig. 1 the relative concentrations of fluorophores (a score) in each set of samples is plotted for EHA. From Fig. 1 it can be noticed that all 3 components have a different behavior towards the increment of Pb(II) concentration, being the component A the most affected in its fluorescence intensity. The difference between components B and C is more marked at higher ionic strength (a and b of Fig. 1), component C showing less sensitivity towards Pb(II) at pH = 4.0 and I = 0.10 M (a in Fig. 1). The fluorescence decrease does not correlate with the fraction of carboxilic and phenolic groups bound to Pb(II) ions as predicted by the NICA – Donnan model, thus Pb(II) induced aggregation is proposed as the cause of fluorescence quenching. Fig. 1: a scores (Fluorescence Intensity Scores, in arbitrary units) are shown as a function of Pb(II) concentration. squares: component A; circles: component B; triangles: component C. pH 4.0 and I = 0.10 M (a); pH 5.5 and I = 0.10 M (b); pH 4.0 and I = 0.02 M (c) and pH 5.5 and I = 0.02 M (d). Fig. 2: emission (continous line) and excitation (dotted line) spectra of components for EHA (a) and FHA (b), both at pH = 5.5, I = 0.1 M. Black: component A; red: component B; blue: component C The loadings (b and c of PARAFAC) for EHA at pH = 5.5, I = 0.1 M are shown in Fig. 2. The shape and location of these emission and excitation spectra are typical of humic substances [5,6]. The spectra are shifted to the red (not shown) as the ionic strength decreases, specially at pH = 5.5. This was observed for most components. 4. Conclusions In both EHA and FHA 3 fluorophores can be estimated; one of them is strongly deactivated in presence of Pb(II) wile the others show less sensitivity. All present emission and excitation spectra typical of humic substances. The fluorophores which are most affected have their emission and excitation spectra shifted to the red, thus might be attributed to groups containing condensed rings. It is considered that humic molecules would aggregate with the metal as a bridge, presenting no–radiative deactivation in the aggregated form and thus quenching of their fluorescence intensity. The deactivation depends on the ionic strength and pH, especially for the components B and C. Regarding the effects of pH and ionic strength values, it was observed that emission and excitation spectra shifted towards the red with lower ionic strength, being this shift higher at superior pH. References 1. C. M. Andersen and R. Bro, J. Chemometrics, 17, (2003) 200. 2. C. A.Stedman and R.Bro, Limnol. Oceanogr.: Methods, 6 (2008) 572. 3. T.Ohno, A. Amirbahman and R. Bro, Environ. Sci. Technol., 42 (2008) 186. 4. R. Bro and H. A. L. Kiers, Journal of Chemometrics, 17 (2003) 274 5. T. Ohno and R. Bro, Soil Sci. Soc. Am. J., 70 (2006) 2028. 6. M. M. D.Sierra, M. Giovanela, E.Parlanti and E. J. Soriano–Sierra, Chemosphere, 58 (2005) 715