Interactions between humic acid and hematite and their effects on metal ion speciation

A.W.P. Vermeer

Research output: Thesisinternal PhD, WU

Abstract

<br/>The impact of toxic chemicals (like metal ions) on the environment is a phenomenon that has been recognised as a mayor problem over the last decades. The speciation of these chemicals determines whether or not a contaminated site has to be regarded as dangerous. The fate of the contaminants depends on the environment and on the local conditions (pH, electrolyte composition and composition of the natural system). For example: contaminants can leach into the groundwater and become a threat for the drinking water supply, they can have an adverse effect on the organisms living in or on the soil or the chemicals can be stored in the soil without direct adverse effects. The binding capacity of the environment is thus of great importance for predicting the mobility and the bio- availabifity of metal ions in the natural environment. To make a sound risk assessment of contaminated sites and to support decision makers with information on the need of recovery a specific site, knowledge of the interactions between the different components in the system is inevitable and has to be established.<p>To approximate the metal ion binding to the soil system, the composition of the soil has to be studied and the different components have to be characterised. Furthermore the proton and metal ion binding to the different components have to be measured, and the effects of the interactions between the different soil components have to be investigated. Once this is done we may be able to predict the speciation of the contaminating metal ions in polluted soils, based on the above described aspects, and make a statement about the risks associated with the contaminated sites.<p>Soil usually consists of a mixture of organic and inorganic constituents. The inorganic fraction contains silica (SiO <sub><font size="-2">2</font></sub> ), metal(hydr)oxides of iron, aluminium, and manganese, clay minerals, and may contain calcium carbonate. Clay minerals exhibit a "constant" negative charge, whereas the metal oxides and the organic fraction have a pH dependent charge. For metal(hydr)oxides this variable charge depends on the structure of the material, for humic substances a range of surface groups (carboxylic and phenolic type of groups) with different pK values can be observed. In this investigation we used a "model soil" consisting of a metal(hydr)oxide and a humic acid. Within this system, the metal ions can either bind to the mineral surface, which is presented by hematite particles, and to the organic fraction, a purified commercially available humic acid (PAHA).<p>A main aspect of studying metal ion binding to model systems is the relation between the metal ion binding in the model system and that in the natural system. Model systems are of most relevance if one can extrapolate directly the obtained results to the much more complicated natural systems. The possible application of laboratory experiments to estimate the speciation of metal ions in natural systems is improved by the experiments described in this thesis. The relative importance of the interaction between metal oxide and humic material on the metal ion binding in the environment depends strongly on the type of environment. So far some results have been obtained in predicting metal ion binding for systems where the above mentioned interaction is expected to be a second order effect. However, in numerous environmental systems some of the binding sites of the organic soil fraction may be masked or neutralised by the interaction with the metal oxide surfaces or other mineral components of the soil matrix. In my study, the effects of these interactions between the different soil components on the proton and cadmium binding to both the organic and the mineral species have been investigated in detail.<p>Summarising, the main goal of this study was to establish qualitative, and where possible, quantitative models that can be used to predict the metal ion speciation in the natural environment by doing laboratory experiments with a model soil. To achieve this goal we studied the proton and metal ion binding to the single components of our model soil as well as to some PAHA/ hematite complexes. To be able to investigate the effect of the adsorption of PAHA onto hematite on the metal ion binding, also these interactions were extensively investigated. The latter experiments were compared with theoretical calculations based on the Self Consistent Field theory for polyelectrolyte adsorption.<p>In general Humic substances are described as a mixture of naturally occurring, polydisperse, heterogeneous polyelectrolytes. Humic acids are predominantly negatively charged due to the abundance of carboxylic and phenolic type of groups. Due to the hydration of the charged groups and the electrostatic repulsion between these charges the humic acids can be described as having an extended conformation that adjusts itself to changes in the environmental conditions. A change in pH and salt concentration resulting in a reduction or increase of the electrostatic repulsion will lead to a more tightly or respectively more loosely coiled configuration.<p>"The" structure of humic acid does not exist, therefore the structure and the geometry of the molecules always have to be approximated. To study the adsorption of these natural polyelectrolytes and the conformation of the adsorbed molecules and that of the molecules in solution a model would be appropriate. Often cited models are those described by Ghosh and Schnitzer and Cameron et al. Based on the structural features proposed by these authors, humic acids molecules are often described as fairly flexible polyelectrolytes. Following this description it seems sensible to compare the properties of humic substances with the properties of simple polyelectrolytes as is done throughout this thesis.<p>The humic acid used in this study reflects many characteristics normally found for humic substances and it may be classified as a soil and/ or peat humic acid. The results obtained for the proton and cadmium binding to the humic acid as a function of pH and salt concentration show the following trends: increasing the salt concentration decreases the binding of cations to the humic acid by reducing the electrostatic attraction. Decreasing the pH increases the proton adsorption but decreases the metal ion adsorption due to the decreased electrostatic attraction and due to an increased competition for binding sites.<p>For the interpretation of the proton and cadmium adsorption isotherms the NICA- Donnan model was applied. The Donnan model is based on the polyelectrolyte properties of the humic matter and accounts for the electrostatic interactions within the humic acid domains. The Donnan volumes are obtained independently by measuring the hydrodynamic volumes of PAHA at the conditions at which the isotherms are measured, The Donnan volume can be obtained directly at high (0.1 M) salt concentration by viscosity measurements. At low salt concentrations (0.001 M) the measured gel volumes are to small too satisfy the Donnan model. This can be corrected by adjusting the volumes using the Debye length. The measured Donnan volumes correspond with a humic acid molecule that is much larger than its dry volume. This type of behaviour can be compared with the random coil model, commonly used for simple polyelectrolytes. The observed trends in the hydrodynamic volume of the humic acid molecules can be explained by electrostatic effects. With increasing electrostatic repulsion the hydrodynamic radius of the humic acid molecules increases. With increasing ionic strength the charge is shielded and consequently the hydrodynamic radius of the PAHA molecules decreases.<p>The proton and cadmium binding to the hematite particles show comparable trends upon pH changes as was observed for PAHA, however, some significant differences have to be mentioned. Whereas the humics are negatively charged over the entire pH range studied, the hematite particles are predominantly positively charged. Hematite is strongly positive at low pH, and becomes less positive until the p.z.c. (p.z.c. of hematite equals 8.9) is reached, where the surface is electrically neutral. The adsorption of cadmium ions at pH 4 and 6 occurs against the electrostatic repulsion between the positively charged cadmium ions and the hematite surface. Cadmium ions can, at pH values far below the p.z.c., only adsorb onto the bare hematite due to strong specific interactions. As a consequence the cadmium adsorption at pH 4 is hardly measurable, whereas at pH values around the p.z.c. the adsorption is increased by several orders of magnitude. This in contrast to the humic where the pH dependence of the cadmium adsorption is only small.<p>The next step in the evaluation of the effects of the interactions between the different components on the metal ion binding was to study the interactions between the humic and the hematite particles. Adsorption isotherms were measured as a function of pH, salt concentration and cadmium concentration. The adsorption of the humic material to the oxide increases with decreasing pH and increasing salt concentration. This is mainly due to charge differences between the negatively charged humic material and positively charged hematite. It was concluded that both coulombic and specific interactions were essential to describe the observed dependencies of the humic acid adsorption. Due to the specific adsorption energy superequivalent adsorption occurs.<p>Studying the adsorption of PAHA onto hematite again the humic acid molecules were assumed to behave as fairly flexible polyelectrolyte molecules that were able to adjust their conformation upon adsorption, The conformation of the adsorbed layer as well as the adsorbed amount was described using a description that was developed for polyelectrolyte adsorption. Within this model the adsorption of chain molecules is described in terms of "trains" and "loops and tails". At high pH and low salt concentration the PAHA molecules are adsorbed relatively flat on the surface, which was described by a relatively large fraction of "trains". At low pH and high salt concentration a large fraction of the adsorbed PAHA is not in direct contact with the surface. Due to this a significant amount of adsorbed polyelectrolyte can be described as adsorbed in "loops and tails", which results in a high adsorbed amount. These pH dependencies were also reflected in the layer thickness of the adsorbed PAHA as was measured by Dynamic Light Scattering. Because of the excess negative charge associated with the adsorbed humic acid segments extending from the surface, superequivalent adsorption gives rise to an electrostatic barrier. The developed electrostatic barrier reduces further adsorption, and causes also very long equilibrium times.<p>The effects of polydispersity on the shape of the adsorption isotherms and on the adsorption/ desorption hysteresis were studied theoretically. From experiments on humic acid/ fulvic acid mixtures it could be concluded that in general the higher molecular weight fraction of the humics adsorbs preferentially onto hematite and causes the gradual increase of the measured adsorbed amount that was observed with increasing humic acid equilibrium concentrations. It was shown that the adsorption/ desorption hysteresis upon dilution could be ascribed to the adsorption fractionation and is not an indication of irreversible adsorption.<p>Proton adsorption to mixtures of PAHA and hematite, as a function of the overall PAHA concentration, was compared with the proton adsorption to the single components. The proton binding to the humic acid/ hematite complex is influenced by the interactions between both components, and the proton adsorption isotherm is not a simple summation of the proton isotherms to the humic and the hematite separately. Both hematite and humic acid molecules have a variable charge that is affected by mutual interactions between these particles. At low pH values the proton binding to the mixed system was lower compared to the direct sum, whereas at high pH values it was higher. These effects were explained by two additional processes; (1) a decreased proton adsorption to the humic acid and (2) an increased proton adsorption to the hematite particles. Which of the two processes dominates the overall proton adsorption depends on the charge of the interacting particles. Due to the adsorption of the negatively charged humic to the mainly positively charged hematite the electrostatic potential in the vicinity of the hematite surface is changed considerably. The component that influences the potential decay most significantly determines which process dominates. If due to the presence of PAHA a negative potential is developed close to the surface, protons will accumulate in the vicinity of the surface sites resulting in an increased surface charge. If the potential is positive, the dissociation of the adsorbed humic is promoted strongly in the surface region.<p>An estimate of the differences in proton charge caused by these interactions is obtained by comparison of the proton charge of the individual samples and their complexes. From this comparison it can be concluded that in the adsorbed state some of the carboxylate groups of the humic acid are bound to protonated surface sites of hematite. These groups are less available for further proton binding. In contrast to this an increased proton adsorption on the oxide surface is deducted. The negative potential field of the adsorbed humic acid layer promotes protonation of the hematite surface. The specific adsorption of humic acid to the hematite surface has as additional effect a shift of the p.z.c. of the hematite surface to a higher pH value and an increase of the surface charge of the hematite particles over the entire pH range. Even at moderate coverage with humic acid, the shift of the p.z.c. is of the order of one pH unit, and therefore quite significant. The screening of the surface by adsorbed PAHA increases the effectiveness of the hematite in binding cations at a given pH value.<p>The effects observed for cadmium binding to the complex can be ascribed to similar processes as was done for the proton binding. The increased binding to the positively charged surface is the result of a lowering of the positive potential due to the negative adsorbed charge associated with the humic acid. This lowering of the positive surface potential will enhance the binding of cadmium to the reactive sites of the oxide surface. The humic acid adsorption can even overcompensate the positive charge of the oxide. The negative potential for the humic acid molecules not in direct contact with the oxide surface (loops and tails) is equal or less negative than for the free humic acid molecules under the same conditions leading to a measured lowering of the PAHA metal ion affinity as was expected based on the proposed model.<p>Next to these experiments some model calculations were done to increase our understanding of the complex system. The model calculations were performed with the self consistent field lattice theory for the adsorption of weak polyelectrolytes onto a surface with a variable charge. This theory is an extension of the lattice theory for polymer adsorption originally developed by Scheutjens and Fleer. The lattice in this model serves as discrete sites onto which polymer units, ions and solvent can take positions. The lattice sites in each layer, parallel to the surface, are indistinguishable so that a mean-field approximation should be applied.<p>The electrostatic double layer was described using a multi-Stern-layer model and the degree of dissociation of the chargeable polyelectrolyte segments was allowed to vary with the distance from the surface. The decay of the electrostatic potential as a function of the distance to the surface and the volume fraction profiles of the polyelectrolyte molecules was calculated numerically as a function of pH, ionic strength and segment-solvent/ segment- surface interaction parameters. Based on these quantities the adsorbed amount, the charge associated with the surface and the adsorbed polyelectrolyte were calculated. Furthermore the importance of different interaction mechanisms; pure electrosorption, specific binding and hydrophobic binding was evaluated.<p>The incorporation of a variable surface charge within the self consistent field calculations was shown to have a major effect on the description of the interactions. It was shown that both components are able to titrate each other ("internal" titration). Depending on the conditions the charge associated with the adsorbed polyelectrolyte molecules compensates or even overcompensates the surface charge. The electrostatic potential developed around the different components affects the overall electrostatic potential profile which vice versa influences the degree of dissociation of the single components. The segment type that is most important for the developed potential profile determines the degree of dissociation of the other segments. At low pH the surface charge is dominant and consequently the polyelectrolyte segments will mainly be affected via increased dissociation, whereas at high pH mainly the surface sites are affected resulting in an increase of surface charge. Based on these calculations it was possible to approximate the shape of the extra induced charge on the oxide surface in a qualitative way. Using this we could estimate the proton adsorption to the separate components in the PAHA/ hematite complexes.<p>The calculated effects of pH and salt concentration on the conformation of the adsorbed polyelectrolyte were in line with the above described experiments. With decreasing salt concentration and at low pH the surface charge/ polyelectrolyte charge is screened to a lesser extent, the adsorption is increased and the charge adjustment is even more profound. Due to these differences in the degree of dissociation of both components the following trends were observed. At high pH, low salt concentration and high adsorption energy the polyelectrolyte molecules are adsorbed relatively flat on the surface which can be described by a large fraction of train segments. The few segments of the molecule that are not in direct contact with the surface protrude relatively far into the solution due to lateral repulsion effects. Consequently the adsorbed layer formed is very dilute. At low pH and high salt concentration the adsorbed polyelectrolyte layer can be described by a large fraction of adsorbed segments in loops and tails. Due to the high fraction of adsorbed segments in loops the adsorbed amount per surface area is relatively high. The relatively small decrease of the hydrodynamic layer thickness with increasing pH as compared to that of the adsorbed amount is caused by the relatively pronounced reduction in loop segments. The length of the tails decreases only slightly. These observations compared well with the pH dependence of the measured layer thickness as compared to the measured adsorbed amount.<p>Considering the discussed mechanism controlling the speciation of metal ions in a ternary system, some conclusions can be drawn considering the speciation in the natural environment. It has been discussed that especially at low pH values binding sites are withdrawn from the system, which is attributed to the carboxylate groups of the adsorbed humic acid, whereas at high pH values an increased proton adsorption is observed. Based on the binding characteristics of the metal ions to the single components we may predict the overall adsorption in a qualitative way. The adsorption of metal ions that bind strongly to the organic soil fraction and are only slightly attracted by the oxides will be decreased due to the interactions between the different soil components. In contrast, for species that bind strongly to the mineral particles the adsorption will be increased.<p>Based on these considerations we can estimate the metal ion binding to a natural, complicated system if we know the binding characteristics of the single components, the composition and the conditions of the system. If the conditions are such that most metal ions bind to the mineral we may expect an increased metal ion adsorption as compared to the laboratory experiments with the single components. A good estimate for this increase can be obtained from the metal ion adsorption isotherm of the surface at about 1 pH unit above the pH of the actual system. If the conditions are such that most metal ions will bind to the organic fraction a decreased metal ion adsorption as compared to the laboratory experiments with the single components may be expected.<p>In conclusion; by combining laboratory experiments with model calculations further insight has been obtained into the binding of metal ions to mixtures of interacting components. Further it has been shown how this knowledge can be extrapolated to natural systems.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
Supervisors/Advisors
  • van Riemsdijk, W.H., Promotor
  • Koopal, L.K., Promotor, External person
Award date25 Nov 1996
Place of PublicationS.l.
Publisher
Publication statusPublished - 1996

Keywords

  • humus
  • humus horizons
  • litter (plant)
  • humic acids
  • chemical speciation
  • haematite
  • metal ions

Fingerprint Dive into the research topics of 'Interactions between humic acid and hematite and their effects on metal ion speciation'. Together they form a unique fingerprint.

  • Cite this