Environ. Sci. Technol. 2005, 39, 5319-5326
Effect of Humic and Fulvic Acid Concentrations and Ionic Strength on Copper and Lead Binding I S O C H R I S T L , * ,† A X E L M E T Z G E R , ILONA HEIDMANN, AND RUBEN KRETZSCHMAR Institute of Terrestrial Ecology, Swiss Federal Institute of Technology, Grabenstrasse 3, CH-8952 Schlieren, Switzerland
We investigated the influence of humic and fulvic acid concentration (in the range of 1-1000 mg/L) on the binding of the two trace metals Cu(II) and Pb(II). The ability of the non-ideal competitive adsorption (NICA)-Donnan model to correctly predict Cu and Pb binding at low humic or fulvic acid concentration and lower ionic strength (0.01 M NaNO3), based on model parameters obtained from experiments conducted at high humic or fulvic acid concentrations (∼1000 mg/L) and higher ionic strength (0.1 M NaNO3), was tested. The binding of Cu and Pb to humic and fulvic acid in 0.01 M NaNO3 was determined over wide ranges in proton and metal ion activities using three different methods: ligand exchange-adsorptive differential pulse cathodic stripping voltammetry at low humic or fulvic acid concentrations (1-3 mg/L), differential pulse anodic stripping voltammetry at intermediate humic or fulvic acid concentrations (10-20 mg/L), and ionselective electrodes at high humic or fulvic acid concentrations (∼1000 mg/L). The results demonstrate that binding isotherms for Cu and Pb can be measured at low humic or fulvic acid concentration using suitable voltammetric techniques. The binding isotherms for Cu and Pb to humic and fulvic acid obtained at constant pH values in the range of pH 4-8 are shown to be independent of humic and fulvic acid concentration. The NICADonnan model, calibrated for Cu and Pb binding using data measured at high humic and fulvic acid concentrations and an ionic strength of 0.1 M, accurately predicts Cu and Pb binding at low humic and fulvic acid concentrations and lower ionic strength (0.01 M). We conclude that NICADonnan parameters obtained by fitting experimental data measured with ion-selective electrodes at high humic or fulvic acid concentrations can be used for geochemical modeling of soils and aquatic environments with much lower concentrations of humic or fulvic acids.
Introduction Humic and fulvic acids act as important cation sorbents, contributing strongly to proton and trace metal cation buffering in soils and aquatic ecosystems (1). The proton and trace metal buffering capacity of humic and fulvic acids is related to their proton-reactive functional groups exhibiting * Corresponding author phone: +41-44-6336001; e-mail:
[email protected]. † Present address: Institute of Terrestrial Ecology, Department of Environmental Sciences, Swiss Federal Institute of Technology Zu ¨ rich (ETHZ), ETH Zentrum, CHN, CH-8092 Zu ¨ rich, Switzerland. 10.1021/es050018f CCC: $30.25 Published on Web 06/15/2005
2005 American Chemical Society
also a high binding affinity for trace metal cations. Chemical analysis of humic and fulvic acid samples of different origin revealed that humic and fulvic acids exhibit a variety of different proton-reactive functional group types, e.g., carboxyl groups and phenolic OH groups (2). Typically, the amount of proton-reactive functional groups ranges from ∼4 mol/kg to 14 mol/kg, with the tendency toward higher values for fulvic acids compared to humic acids (3, 4). In many studies on metal cation binding of humic and fulvic acids that have been reported in the literature, ionselective electrodes (ISEs) were used to measure free metal cation activities in solution in the presence of humic or fulvic acids. However, the working range of ISEs is limited to total metal cation concentrations above natural trace metal cation concentrations when studies are conducted in natural samples that are usually weakly buffered with respect to free metal cation activity (5, 6). For example, Xue and Sunda (7) reported that erroneously high Cu2+ activities were measured with a solid-state Cu2+ ISE in lake water samples below a Cu2+ activity of ∼10-7 M. Avdeef et al. (8) showed that the solid-state Cu2+ ISE responded linearly to the logarithm of Cu2+ activity down to 10-19 M in a solution buffered with ∼15 mM ethylenediamine. This observation suggested that the working range of ISEs could also be widened for humic and fulvic acid samples if the solutions contain high humic or fulvic acid concentrations, providing enough metal ion buffering. Studies on Cu2+, Pb2+, and Cd2+ binding to humic and fulvic acids demonstrated that free Cu2+, Pb2+, and Cd2+ activities can be measured down to log {Cu2+} ) -14, log {Pb2+} ) -10, and log {Cd2+} ) -10 with solid-state ISEs using humic or fulvic acid concentrations of at least 1 g/L (9-11). However, the need of high humic or fulvic acids concentrations required to study trace metal cation binding with ISEs (>1 g/L) entails that humic and fulvic acids must be concentrated because natural concentrations in lakes, rivers, soil, and groundwaters are usually several orders of magnitude lower (12). In addition, little attention has been given to the effect of changes in the conformation of humic and fulvic acids on metal cation binding. Variations in ionic strength and pH were observed to affect the apparent molecular size of humic and fulvic acids (13, 14). This phenomenon was consistent with the concept of random coiling of humic and fulvic acids being considered as macromolecules (15). Recent findings based on size exclusion chromatography (SEC) and nuclear magnetic resonance (NMR) spectroscopy indicate that dissolved humic and fulvic acids may be micelle-like, supramolecular assemblies of small entities rather than macromolecules (16-20). Considering the critical micelle concentration (CMC) values published by Hayase and Tsubose (21) for humic and fulvic acid in the range of 1-10 g/L, the apparent conformation of humic and fulvic acids investigated with ISEs at concentrations of ∼1 g/L may differ strongly from the conformation in more dilute, natural systems. If micelle-like structures form at high humic or fulvic acid concentrations, a portion of their reactive functional groups may be less accessible for trace metals as they might be entrapped in the more hydrophobic interior. Until now, only few studies investigated the effect of natural organic matter (NOM) concentration on metal cation binding. Li et al. (22) studied Cd2+ binding to Suwannee River NOM at pH 5.2 in 0.1 M NaClO4 using a Cd2+ ISE. They found that Cd2+ binding was not affected by increasing the NOM concentration from 95 mg/L carbon to 9500 mg/L carbon at free Cd2+ activities ranging from 10-8 M to 10-5 M. Fitch et al. (23) used a Cu2+ ISE to study Cu2+ binding to a soil humic VOL. 39, NO. 14, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5319
acid at pH 4 in 0.005 M KClO4 and concluded that Cu2+ binding affinity increased when humic acid concentrations were increased from 60 mg/L to 250 mg/L. They speculated that conformational changes of humic acid were caused by higher humic acid and Cu2+ concentrations resulting in increased retention of free Cu2+ through entrapment. Perdue and Carreira (24) investigated Cu2+ binding by Suwannee River DOM at pH values of 4-7 at an ionic strength of 0.01 M using a Cu2+ ISE. For pH 4, 5, and 6, the binding isotherms for 50 and 1000 mg/L Suwannee River DOM were identical, whereas at pH 7, the data indicated a slightly higher Cu2+ binding affinity at 1000 mg/L DOM, compared to 50 mg/L DOM. Although no conclusive result on the effects of humic or fulvic acid concentrations on metal cation binding is available, there is a strong need for such information if organic sorbents such as humic and fulvic acids are intended to be considered in environmental modeling. Quite often, binding constants used in advanced models such as, for example, the non-ideal competitive adsorption (NICA)-Donnan model (25-27) are derived from data determined at fairly high humic or fulvic acid concentrations. For the prediction of trace metal fate in natural systems exhibiting typically low humic or fulvic acid concentrations, these constants only hold if they are unaffected by variations of humic or fulvic acid concentration. The objectives of this study are (a) to investigate the consistency of copper and lead binding isotherms for humic and fulvic acids derived from measurements with ligand exchange-adsorptive cathodic stripping voltammetry (AdSV) at low humic and fulvic acid concentrations (1-3 mg/L), anodic stripping voltammetry (ASV) at intermediate humic and fulvic acid concentrations (10-20 mg/L), and ionselective electrodes (ISEs) at high humic and fulvic acid concentrations (1000 mg/L) and (b) to explore whether NICA-Donnan model parameters calibrated with data measured at high humic and fulvic acid concentrations (1000 mg/L) and an ionic strength of 0.1 M can be used to predict metal binding to humic and fulvic acid under conditions relevant for natural freshwater and pore water samples, i.e., low humic or fulvic acid concentrations (1-20 mg/L) and lower ionic strength (0.01 M).
Experimental Section Humic and Fulvic Acids. For this study, we used humic and fulvic acids extracted from a well-humified organic horizon (H) of a Humic Gleysol at Unterrickenzopfen (Switzerland, 7° 50′ E/47° 13′ N; 490 m above sea level). The extraction and purification of humic and fulvic acids were conducted following the standard procedure of the International Humic Substances Society (28) with minor modifications as described in detail by Christl et al. (29). Both purified humic and fulvic acid extracts were stored in darkness at 3 °C. Humic and fulvic acid samples were not affected by storing them for more than five years. Characterization of the humic and fulvic acids by Fourier transform infrared (FT-IR) spectroscopy, SEC, ultraviolet-visible (UV-vis) spectroscopy, and fluorescence spectroscopy did not show any differences, compared to the characterization results reported previously (29). Determination of Cu2+ and Pb2+ with Ion-Selective Electrodes. Data on Cu2+ and Pb2+ binding to fulvic acid determined with ion-selective electrodes (ISEs) in 0.01 M NaNO3 at high fulvic acid concentration (∼1000 mg/L) were taken from Heidmann et al. (30). The binding of Cu2+ and Pb2+ to humic acid was investigated in a 0.01 M NaNO3 background electrolyte with the same titration technique as used by Heidmann et al. (30) for fulvic acid. Metal titrations of solutions containing 1000 mg/L humic acid were conducted at pH 6 and pH 8 for copper and pH 4 and pH 5.5 for lead. All experiments were performed at 25 ( 1 °C. Proton 5320
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 14, 2005
and free metal ion activities in solution were monitored using a pH glass electrode (6.0123.100, Metrohm), ISEs for Cu2+ (Orion 9429) and Pb2+ (Orion 9482), and an Ag/AgCl reference electrode (6.0733.100 Metrohm). Cu2+ and Pb2+ ISEs were calibrated by titrating metal salt solutions with ethylenediamine (11). During titrations, the solutions were stirred for 4 min after each addition of titrant. After metal addition, the pH value was automatically readjusted to the desired pH value by base addition and kept constant for 20 min within a tolerance range of ( 0.007 pH units (0.4 mV). Voltage readings of electrodes were recorded when the potential drift was
