1. Introduction
The organic matter in terrestrial and aquatic environments is being classified as particulate organic matter (POM) and dissolved organic matter (DOM). POM includes particles of natural organic matter (NOM) characterized by average sizes ranging from 50 μm to 2 mm.
[1]. DOM is the corresponding fraction of dissolved POM that can pass through a 0.45 mm mesh filter
[2]. In the marine environment this rather artificial distinction between POM and DOM is sometimes questioned and replaced by a broader definition where organic matter is considered as a continuum of gel-like polymers, replete with colloids cross-linked by polymer strings, sheets, and bundles while the size can reach hundreds of micrometers
[3]. An important part of DOM comes from the carbon mobility processes occurring in the environment. However, much of it is lost through various mechanisms such as substrate to soil microorganisms or adsorbed on soil sediments and eventually diffuses into the ocean. Much of the carbon DOM present in terrestrial and aquatic compartments is oxidized to CO
2 through microbial metabolism and then accumulates in the atmosphere
[4]. DOM is a parameter that responds to changes in ecological processes and is capable to promote the dissolution and transport of metals such as micronutrients and organic compounds, including pollutants
[5][6][7][8]. Organic compounds in DOM have been shown to affect the chemical reactivity of soil minerals
[9] and the complex process of podzolization of soil
[10]. The production of greenhouse gases such as methane, and soil denitrification is connected to the bacterial degradation of DOM
[11]. Furthermore, the localization of the DOM in the soil, for example, located in pores of various sizes, influences its mobility, ecological fate and availability
[2].
The major pools of DOM are molecules solubilized and leached from soil organic matter (SOM). Those can be biomolecules originating from both fresh materials, such as leaf and grass litter, root exudates, decaying fine roots, soil microorganisms etc. and microbiologically processed, humified, fractions such as humic substances (HS)
[12]. Due to the environmental conditions, the dissolved organic carbon cycle is relatively fast. The controlling factors for DOM biodegradability, and thus carbon dynamics, are influenced by several factors. Higher content of non-humified hydrophilic compounds such as carbohydrates, organic acids and proteins increase DOM degradability. On the contrary, the presence of aromatic and hydrophobic residues leads to a reduction of its biodegradability, due to the high resistance shown by the aromatic compounds towards degradation phenomena and to the inhibitory action of the enzymatic activity. Effects of heavy metal concentrations on DOM biodegradability is not clear yet since various studies reported conflicting results. As degradation is dependent on microbial activity, the type and concentration of microbial population must be taken into account as well
[13].
HS are the most abundant part of DOM and their chemical structure and even their concept is still a matter of discussion
[14]. For decades, it had been accepted that HS are macromolecules formed through a two-step process. In the first step, molecular building blocks are produced from the degradation of plant-, microbial-, and animal-derived organic systems. In the second step, secondary reactions are believed to occur, including condensation of resulting fragments into larger products, each hardly referable to the parent feedstock
[15]. This view has been revised and new models have been developed to explain the chemical-physical properties of HS related with their long-term persistence and their molecular architecture
[16]. From a chemical-physical point of view, HS are described as dark-coloured, chemically, and presumably biologically refractory, heterogeneous organic compounds playing the main role in soil fertility
[17].
This description does not coincide with a specific path of production of a set of organic compounds nor a characteristic molecular mass identified by a chemical reactivity (recalcitrant carbon fraction). However, it excludes insoluble organic matter present in sedimentary rocks (kerogen)
[18], and other complex mixtures of organic compounds produced by combustion processes (carbon black)
[19]. Strong debate still exists about the chemical nature of HS in various forms, i.e., as aqueous dispersions, within soils and at mineral interfaces, as well as on their role in carbon stabilization mechanisms
[20][21]. Recently the carbon stabilization via interaction with soil mineral surfaces is referred to as mineral-associated organic matter (MAOM)
[22]. In a small aggregate (40–50 µm), MAOM consists of low-molecular weight compounds potentially sorbed to mineral surfaces, via either incorporation into microbial biomass or extracellular enzymes
[23]. However, the new view of stabilisation mechanisms and observed longevity of MAOM
[24], is still unclear, because the organo-mineral interactions are easy alterable by root exudates
[25].
This is fundamentally a paradigm shift in how the interactions act between SOM components and soil minerals
[26], i.e., MAOM aggregates, since they are not permanent. Instead, the persistence of SOM assisted by both the microbiome and the physical structure of the soil is considered the main carbon stabilization mechanism
[27]. For example, according to Baveye et al.
[28], no significant advances have been made in the last century about the basic (bio)chemistry of HS and carbon stabilization. In fact, it appears that all the modern views about a better definition of HS or on a widely accepted operational extraction procedures do not provide all answer able to overcome the Waksman model as hypothesized in the early years of the Twentieth century
[28][29][30][31]. How much is this true? Certainly, the investigation on structure and aggregation of HS can provide useful information about the relevant environmental processes and models in which these natural substances are involved.
2. Models of the Structure of Humic Substances
Along all the research activities carried out to elucidate the nature of aggregation state of HS, various conceptual models have been proposed, among which researchers can begin to mention the oldest one indicated as the rigid sphero-colloid model elaborated by Ghosh et al.
[32]. In particular, those authors linked the structure of HS dissolved in solution to their concentration, being able to form either spherical or flexible linear colloids at high and low HS concentrations, respectively. Vapor-pressure osmometry (VPO) was carried out to investigate the aggregation of standard fulvic acid (FA)
[33]. Swift and coworkers argued that HS dissolved in aqueous solutions could consist of high molecular weight macromolecules capable of stabilizing a random coil configuration
[34][35]. The polymer gel theory was also invoked to interpret the supramolecular dynamics of HS in marine environment
[36]. Beginning in 1990, Wershaw launched the first hypothesis of molecular aggregation, in which HS was described as a self-assembled material consisting of a hydrophobic interior and a hydrophilic exterior
[37][38][39]. That hypothesis was also supported by Guetzloff et al.
[40][41], who demonstrated through surface tension measurements, that purified HA was able to form pseudo-micelles in alkaline aqueous solutions at concentrations greater than 7.2 g∙L
−1. Evidences for the formation of hydrophobic micro-environments following the dissolution of HS in water were formerly documented through fluorescence quenching studies. In particular, the HA-promoted reduction of fluorescence quenching of both pyrene
[42] and positively charged synthetic organic fluorophores
[43], provided the evidence that HS in aqueous solutions were capable of configuring domains with a micellar cage structure.
2.1. The Micellar Model
Differently from orthodox micellar solutions, humic pseudo-micelles were initially considered as a type of aggregates characterized by a broad distribution of both size and functional groups. According to that vision, pseudo-micelles still retain the ability of traditional micelles to sequester hydrophobic species that partition into the nonpolar microdomains
[44][45][46]. Using ultrasonic velocimetry Kučerík et al.
[47] demonstrated that the aggregation can be observed at concentrations as low as 1 mg∙L
−1 under various pH and ionic strength conditions. The micelle-like behavior of HS was also investigated by Piccolo et al.
[48] who verified through size exclusion chromatography (SEC), the effect of the concentration of organic acids on the HS dissociation into small-sized micellar aggregates. HS aggregation was also studied in aqueous solution by HPSEC
[49].
The formation of aggregates in solutions of HS was studied through capillary zone electrophoresis (CZE) for the first time by Fetsch et al.
[50]. In the late 1990s, Jones et al.
[51] published a review in which they surveyed the major findings reported up to that time on the colloidal properties of HS, by emphasizing the aggregation behavior in both artificial and natural environments. By varying the composition of the mobile phase at fixed ionic strength of a SEC system, Conte et al.
[52] were able to monitor the conformational changes of different types of dissolved HA and FA. In particular, through high- and low-pressure chromatography, the authors verified that HS complied with the model describing the association of small heterogeneous molecules into larger dimensional structures. The work of Piccolo et al.
[53] on HA and FA offered further direct evidence more in favor of the model based on the reversible self-association of small molecules rather than the polymer random coil picture. The reductions in HS dimensions that occurred as a result of variation of composition of the mobile phase were attributed by the authors to a drastic attenuation of weak intermolecular associations giving rise to separate smaller molecules rather than to the compaction of macromolecular chains. Other studies have rejected the idea that the micellar structure of HS could only be inferred by changing the retention volumes following the addition of organic acids.
According to Varga et al.
[54], secondary (ionic, hydrophobic) interactions should always be considered for the interpretation of SEC developed for HS. Young et al.
[55] interpreted the SEC and fluorescence experiments on several type of HA aqueous solutions in terms of the formation of ‘pseudo-micelles’ assisted under the influence of cations in solution.
The former concept of ‘pseudo-macromolecularity’ was proposed by Piccolo
[56] to suggest that the macromolecular properties of HS could be compatible with those produced by pseudo-micellar structures or by other molecular assemblies involving long-chain hydrocarbons, fatty acids, esters and suberin-like components. Subsequently, Sutton et al.
[57] published a new model on the molecular structure of HS based on supramolecular associations of light molecular mass components, stabilized by hydrophobic interactions and H-bonds, which self-assemble into micellar aggregates in suitable aqueous milieu.
This model was based on the use of several techniques, such as gel permeation chromatography (GPC), size exclusion chromatography at high pressure (HPSEC), and ultraviolet visible spectroscopy (UV-VIS). Indirect evidence in favor of the micellar model includes, (i) the extent of the reduction in surface tension of aqueous HS solutions observed with increasing HS concentration
[58], (ii) the improved solubility of hydrophobic organic compounds at following dispersion in aqueous solutions of HS
[37][38][39][40][41] and (iii) the preferred distribution of fluorescent probes such as pyrene, within the hydrophobic microdomain of HA aggregates in aqueous solutions
[42][45].
Criticisms against the micellar model can be summarized with the following questions: (i) how can HS lacking well-defined amphiphilic substances self-assemble in aqueous solutions in the form of colloidal aggregates? (ii) What are the typical sizes, shapes and internal structures of HS aggregates in aqueous media? (iii) and how do these parameters depend on the pH and Concentration of HS, dissolved salts and metal ions?
2.2. The Supramolecular Aggregate Model
Alongside the development of polymeric and micellar-like models, a flourishing line of research was developed on the use of the fractal theory in the study of the aggregation of HS
[59].
After the advent of polymeric and micellar models, more credits are given to the hypothesis that HS are supramolecular aggregates of small organic compounds, held together by weak dispersive forces other than covalent bonding
[60][61][62][63][64]. Hence, according to the concept of supramolecular structures, HS can be considered as relatively small and heterogeneous molecules of various origin, which self-organize in supramolecular conformations. Humic superstructures of relatively small molecules are now imagined as complex aggregates stabilized only by weak forces such as dispersive hydrophobic interactions (van der Waals, π–π, and CH–π bonding) and H-bonds, the latter being progressively more important at low pH values. The representation of HS according to the model based on the spontaneous association of molecular components through weak bonds, provides several important implications: (1) accumulation of organic matter in soils is driven by hydrophobic interactions; (2) any HS supramolecular structure can undergo interchange processes with hydrophilic molecules generated through the degradation of biological tissues; (3) the hydrophobic bonds also lead to the consequent stabilization of the organo-mineral complexes and the general structure of soil
[60].
Experimental evidences that HS have a hierarchical or “structure within a structure” architecture with at least two sequential levels of organization were provided by Chilom et al.
[65][66][67]. In details, they developed a method in which a given HA
0 sample was firstly subjected to Soxhlet extraction using a C
6H
6:CH
3OH azeotrope for 72 h to collect two main fractions, respectively, a lipid-like L
0 and a non-amphiphilic HA
1, corresponding to about 67% of the total organic carbon (TOC). Then, the L
0 fraction was separated into further two sub-fractions: HA
2 accounting for almost one-third of TOC (possessing amphiphilic properties and soluble in an alkaline aqueous solution) and L
1 (lipid-like and soluble in nonpolar solvents).
Differences in heat capacities between the pristine HA and the reconstituted physical mixtures (HA1 + L0 and HA1 + HA2 + L1) were compared. Thus, it was deduced a first level of organization correspondent to self-assembled nanoparticles formed by the amphiphilic HA2 and the lipid L1 fractions, which further interact with the components of HA1 yielding the overall HA0 composite (second level of organization). However, the interactions that trigger and stabilize the structure of these assembled molecular complexes are still unknown.
The diffusion coefficients of carbohydrates, carboxyl-rich alicyclic molecules and aliphatic compounds in DOM strongly depend on their concentrations indicating their gradual association
[68]. Based on this observation, Drastik et al.
[69] hypothesized that free hydrophilic molecules at lower concentration can play a role as hydrotropic compounds and stabilize and solubilize the aromatic aggregates in DOM while at higher concentration slowly aggregate forming physically stabilized gel-like structures.
Earlier results showing the different stability of HS aggregates at increasing temperatures
[70], led to a new model in which HS are described as a hydrophobic inner core stabilized by outer layers of amphiphilic and polar molecules that impart a hydrogel-like structure
[71]. Recent research developed by Wells et al.
[72], emphasized the role played by “strong” H-bonds among polar domains, in the formation of stable hydrogels-like structure. In such hypothesis, HS aggregates with size > 1 μm have been proposed to consist of “metachemical” hydrogels < 1 mm embedded within water dispersible porous physical hydrogel scaffolds
[73], which can form, dissipate, and reassemble along a sequence of alternating cycles of mechanical agitation and quiescence
[74].