Aquatic colloids: Concepts, definitions, and current challenges

Colloidal phases in natural waters may be important to various environmental questions, especially those concerning the cycling of vital and toxic trace chemicals. Current treatments of the role of colloids in chemical speciation largely rely on operational definitions of phases such as l,OOO-Da ultrafilter and 0.45pm filter cut-offs. Defining chemical phases exclusively by a physical parameter such as size is contributing to a situation where the observed filterable vs. unfilterablc distribution coefficients, D, are not well predicted from thermodynamically derived sorbed vs. solute equilibrium constants, K. Achieving the goal of relating the natural distributions of chemicals to theoretical expectations is contingent upon progress in development of a functionally meaningful colloid definition and interpretation of observed distributions of trace substances in terms of the relevant physicochemical propertics of the system. We assess the phase status of typical components in natural waters from a “chemcentric” point of view (i.e. one whose motivation is to understand the cycling of trace chemicals in the environment). As a result, we define colloids so as to provide a thermodynamic grounding for evaluating chemical speciation and a hydrodynamic framework distinguishing phases that are transported with the solution from those that are not. These constraints lead one to define an aquatic colloid as any constituent that provides a molecular milieu into and onto which chemicals can escape from the aqueous solution, and whose movement is not significantly affected by gravitational settling. Such a definition allows development of mass balance equations, suited to assessing chemical fates, that reflect processes uniquely acting on dissolved, colloidal, or settling particle phases. For aquatic scientists concerned with the behavior and effects of trace chemicals, one important process is the partitioning of those trace constituents between the dissolved and bound pools. The resulting speciation affects the extent to which the chemical participates in various transport and transformation processes. For example, one may anticipate that the truly dissolved trace molecules may participate in homogeneous solution phase reactions, while this is not true for their counterparts sorbed to colloids (particles “immune to gravity”; Graham 1861) or settling particles (henceforth referred to as gravitoids). Such sorbed species may be less bioavailable and may exhibit different photoreactivity than their solution-phase counterparts. Likewise, for transport processes, one may anticipate that dissolved and colloidbound molecules are carried with the moving fluid (e.g. in sediment porewater irrigation), while the gravitoid fraction may fall out of a mixed water body to loci below. Distinguishing among these functionally distinct forms is essential if we are to elucidate the cycling and effects of trace chemicals in natural waters. Recently, efforts to quantify the roles of colloids in the cycling of trace compounds have been confounded by the realization that results obtained by the most common sampling technique, cross-flow ultrafiltration (CFF), are operator and equipment dependent (Buesseler et al. 1996). CFF also appears to cause undesired fractionation of colloidal com