On the nature of organic matter from natural and contaminated materials : isolation methods, characterisation and application to geochemical modelling

Natural organic matter (NOM) is the material that is formed after the natural decomposition and transformation of dead plant and animal matter. The fresh organic matter (e.g. plant leaves or animal debris) is decomposed and transformed by microbial activity. As such, NOM is found everywhere in the natural environment; in soils, surface water and oceans. Due to its abundance at the earth’s surface, the production and decomposition of NOM plays an important role in the global carbon cycling. In soil systems, NOM has an extremely important influence on essential properties like soil structure, water retention, nutrient availability and binding of contaminants. In water systems (e.g. surface or river water), NOM is important in many (bio)-chemical processes and the geochemical cycling of elements/nutrients. Due to the importance of NOM in soils for agricultural production, research on its chemical and physical properties and classification dates back many centuries. Numerous laboratory procedures and classifications have been developed to relate organic matter properties to plant growth and soil fertility. However, since the growing public and political awareness of environmental pollution in the 1960s, the properties of NOM (especially humic substances, see below) have increasingly been investigated in the context of its interactions with potentially toxic compounds such as heavy metals and pesticides. Although the first phenomenological observations that natural organic matter can bind heavy metals have already been made in the 19th century, scientists started to gain more extensive data on the strong interaction of heavy metals with NOM in natural soils and aquatic environments during the past decades. At present, the important influence of NOM on the mobility of such potentially toxic compounds in the environment is widely acknowledged by the scientific community. Waste materials do often also contain organic matter, and in addition, high contents of potentially toxic contaminants such as heavy metals. In many countries, waste materials are increasingly being recycled in construction works (e.g. in road foundations, embankments and sound barriers). The potential environmental risk associated with the re-use of waste materials in such applications depends on the extent to which contaminants can be released as a result of “leaching”. Leaching is the release of contaminants from the solid phase (e.g. a waste material) to the water phase with which the material may be in contact (e.g., percolating rainwater). The leaching of contaminants such as heavy metals can be strongly enhanced by complexes with (soluble) organic matter. Therefore, it is important to characterise the properties of organic 168 Matter matter in (contaminated) waste materials in the context of long-term risk assessment of waste applications in construction. In addition, this knowledge can contribute to the development of waste treatment technologies, e.g., to improve the leaching properties of waste materials. Knowledge of the fundamental binding properties of NOM with respect to contaminants can also be specifically used for the development of models to predict the (long-term) leaching behaviour of waste materials. Natural organic matter is known to include a broad spectrum of organic constituents, many of which have their counterparts in biological tissues, and each with different chemical and physical properties. Two major types of compounds can be distinguished: Humic substances: a series of unidentifiable organic compounds of relatively high-molecular-weight. They are brown to black in colour and formed by secondary synthesis reactions. The term “humic substances” is used as a generic name, but many scientist discriminate between humic- and fulvic acids (HA and FA, respectively) based on their dissolution properties in alkaline and acid solutions. Humic acids are generally dark-brown to black in colour and have a relatively high molecular weight (several thousand to several hundred thousand atomic mass units (AMU), depending on the applied analysis techniques). Fulvic acids are light yellow to golden brown in colour and have molecular weights ranging from several hundred to about ten thousand AMU. The basis of this classification is also used throughout this thesis. Non-humic substances: all identifiable (biochemical) organic molecules that can be placed in one of the categories of discrete compounds such as sugars, amino acids, fatty acids etc. This thesis focuses on the development of methods for the isolation and characterisation of natural organic matter in general and organic matter in (contaminated) waste materials in particular. The overall purpose of this research is to develop a better understanding of the role of NOM with respect to the mobility of contaminants such as heavy metals in the environment. The main difficulty of studying humic substances is that these substances are very heterogeneous by their nature. Various classifications of humic substances are used in the scientific literature and these are all based on operational definitions. There is still no consensus today as to how HA and FA should be operationally defined. It is important to note that the terminology that is being used does not represent pure compounds. Each class of the humic substances consists of highly complex and heterogeneous mixtures of organic molecules. The study of chemical properties of NOM often requires isolation and purification of different fractions of the organic matter. The advantage of purification is that it reduces the heterogeneity in the properties of NOM. Numerous isolation and purification methods have been developed over the past decades to enable the further characterisation of organic matter properties. Although these procedures are well established and widely used by scientists, they all share the disadvantages that they are very laborious and primarily aimed on purification rather than quantification of the different fractions. Therefore, this thesis has a strong focus on developing analytical methods that enable an improved characterisation and quantification of the different fractions that are important for metal binding, preferably in a more time- efficient manner. In Chapter 2, a Competitive Ligand Exchange-Solvent extraction (CLE-SE) method was used to measure Cu binding to DOC in leachates from municipal solid waste incinerator (MSWI) bottom ash. The copper binding properties of dissolved organic carbon (DOC) were investigated with specific attention for the identification and quantification of the organic ligands. Evidence was found for an important role of fulvic acids (FA) in the strongly enhanced leaching of Cu from MSWI bottom ash. This work implies that the complexation of contaminants with natural organic matter is an important process in these relatively inorganic waste materials. Chapter 3 describes the development of an automated procedure to isolate and purify HA and FA from various materials. The conventional (manual) isolation and purification procedures that are widely used by scientists, share the disadvantage of being very laborious and time consuming. These disadvantages are largely overcome by automation and enabled to gather an extensive set of purified HA and FA samples from diverse origin for further characterisation purposes. The main objective for automation of the conventional procedure was to save a significant amount of labour and total throughput time in the performance of HA and FA isolation and purification. The novelty of the method lies in the automated handling of the multiple liquids and columns, required in the isolation/purification procedure, in both forward and back elution mode. By automating the procedure, better standardisation of HA and FA isolation and purification methods is feasible. The automated procedure significantly reduces the total throughput time needed, from 6–7 days to 48 h, and the amount of labour to obtain purified HS for further characterisation. 170 In chapter 4, the development is described of a rapid batch method for the In chapter 4, the development is described of a rapid batch method for the experimental characterisation and quantification of HS in natural and contaminated systems. In principle, the quantification of HS in the environment can be studied with the manual or the developed automated isolation and purification procedure (chapter 3). However, the established models for calculation of metal binding to HS require absolute concentrations of HA and/or FA as input. The conventional isolation and purification procedures are too elaborate (and mainly focussed on purification) for this type of application. Because the principles of the new batch method are essentially the same as those of the well-known conventional isolation and purification procedures, the HA and FA properties identified in this study are, therefore, of general importance for the interpretation of the occurrence and behaviour of HS in the environment. The novelty of this method lies in the fact that it greatly facilitates the analysis of HA and FA concentrations (e.g. for use in geochemical modelling, chapter 5). The new method can be performed within 1.5-4 hours per sample and multiple samples can be processed simultaneously, while the conventional procedures typically require approximately 40 hours for a single sample. Based on results from previous chapters, “multi-surface” geochemical modelling of heavy metal leaching from MSWI bottom ash was used to develop a mechanistic insight into the beneficial effects of accelerated aging of MSWI bottom ash on the leaching of copper and molybdenum (Chapter 5). Therefore, the rapid batch procedure described in Chapter 4 is used to characterise DOC quantitatively in terms of humic, fulvic and hydrophilic acids over a wide pH range. Important processes controlling the solid/liquid partitioning of humic and fulvic acids and their role in the effects of aging on contaminant leaching are identified. In addition, a new approach is developed to model the pH dependent leaching of fulvic acids from MSWI bottom ash based on adsorption to reactive Fe/Al-(hydr-)oxides. This chapter shows that accelerated ageing results in enhanced adsorption of FA to reactive Fe/Al- (hydr-)oxides, leading to a significant decrease in the leaching of both FA and associated Cu. In Chapter 6, the carbon speciation (inorganic, organic and elemental carbon) in MSWI bottom ash samples is studied to identify the amount and properties of the different carbon species present in MSWI bottom ash. The carbon speciation was quantitatively measured (partly based on the method described in Chapter 4), and its relation with the leaching of Cu, in fresh and carbonated MSWI bottom ash. Results show that up to only 25% of loss on ignition (LOI) consists of organic carbon (OC), while about 15% of organic carbon (OC) in the three samples consists of HA and FA. Since only these small reactive carbon Summary and synthesis 171 fractions contribute to enhanced metal leaching from MSWI bottom ash, fractions contribute to enhanced metal leaching from MSWI bottom ash, it is concluded that LOI measurements are insufficiently discriminative for a quantitative assessment of environmentally relevant organic carbon species in MSWI bottom ash. The results of this study imply that dedicated methods, focusing on specific carbon fractions, are more appropriate for assessment of environmentally relevant organic carbon species than the measurement of LOI. These methods may greatly improve the assessment of the long-term environmental properties of bottom ash in utilisation or disposal scenarios. Chapter 7 describes a comparative study regarding the proton binding properties of the previously isolated and purified HS (Chapter 3). Binding models for HS, such as the NICA-Donnan model, have so far been developed and calibrated against organic matter from natural origin (e.g. soils and surface waters). In Chapter 5, the NICA-Donnan model was found to perform well when applied to a contaminated (waste) material, i.e. municipal solid waste incinerator (MSWI) bottom ash. However, the proton binding properties of HS originating from waste environments have not been analysed directly and demonstrated to fall within the observed range for natural materials. The aim of this study is to analyse the proton binding properties of humic and fulvic acid samples originating from secondary materials, waste materials and natural samples in order to assess whether the charge development of these HS can be described with generic NICA-Donnan parameters. New proton binding parameters are presented for these HS and are shown to be similar to those of HS originating from natural environments. These results suggest that the NICA-Donnan model and generic binding parameters are adequate to describe proton binding to HS in both natural and contaminated materials. This finding widens the range of environments to which the NICA-Donnan model can be applied and justifies its use in geochemical speciation modelling of metal mobility in contaminated (waste) materials.