Colloidal fraction separation from dissolved

Bulk techniques still have a place in the search for presolar components. Although they cannot identify the presolar grain directly, they can measure anomalous isotopic compositions, which can then be used as a tracer for separation procedures to identify the carrier. There are several isotopically anomalous components whose carriers have not been identified. For example, an anomalous chromium component enriched in 54Cr appears in acid residues of the most primitive chondrites. The carrier is soluble in hydrochloric acid and goes with the colloidal fraction of the residue, which means it is likely to be submicron in size (Podosck el al., 1997). Measurements of molybdenum and ruthenium in bulk primitive meteorites and leachates from primitive chondrites show isotopic anomalies that can be attributed to the -process on the one hand and to the r- and /7-processes on the other. The s-process anomalies in molybdenum and ruthenium correlate with one another, while the r- and /7-process anomalies do not. The amounts of -process molybdenum and ruthenium are consistent with their being carried in presolar silicon carbide, but they are released from bulk samples with treatments that should not dissolve that mineral. Thus, additional carriers of s-, r-, and/ -process elements are suggested (Dauphas et al., 2002). 

In the (aquatic) environment elements occur in particulate-, colloidal- and dissolved forms. These forms are usually distinguished by filtration or centrifugation. Traditionally, a 0.45 um (membrane)- filter separates the particulate from the dissolved forms. This may result in the passage of colloidal fractions through the filter, classifying colloidal matter incorrectly within the dissolved fraction. Although the interaction between dissolved and particulate (surface) fractions cannot be neglected, it is common in speciation studies to consider the "dissolved" fraction. The dissolved forms of trace elements are mainly present as ... 

In water studies it is standard practice to filter the sample soon after collection, usually through a 0.45p,m membrane disc (made of cellulose acetate, cellulose nitrate or polycarbonate). This process arbitrarily divides the sample components into soluble and insoluble fractions, but as shown in Table 2.3, the average size of different chemical species varies widely, and some differentiation between species can be obtained through using filter media of different pore sizes. For example, fully dissolved compounds can be separated from finer colloidal forms by using gel filtration and dialysis, and sub-division of the total content into fractions based on particle or molecular size (see Section 2.3) has been used for speciation of elements in waters. 

It is now well known that trace-element concentrations in continental waters depend on the size of the pore filters used to separate the particulate from the dissolved fraction. This is apparent in Table 1, where results from the Amazon and Orinoco are reported using two filtration sizes the conventional 0.2 p,m filtration and filtration with membranes of smaller cutoff size (ultrafiltration). These results suggest the presence in solution of very small (submicro-metric) particles that pass through filters during filtration. The view that trace elements can be separated into particulate and dissolved fractions can thus no longer be held this has led authors to operationally define a colloidal fraction (0.20 p.m or 0.45 p.m to 1 nm) and a truly dissolved fraction (<1 nm) (e.g., Buffle and Van Leeuwen, 1992 Stumm, 1993). The existence of a colloidal phase has a major influence on the speciation calculation schemes presented above (based only on aqueous complexation), as the apparent solubihty of trace elements will be enhanced by the presence of colloids. The dynamics of colloids also completely change... 

Filtration of water samples at collection removes from the water the suspended and colloidal fractions (for separate analysis, if desired) but some dissolved... 

The values of kj calculated by Bacon and Anderson (1982), and used in most models of Th scavenging, varied with particle concentration and ranged from 0.2 to 1.2 Such values are appreciably longer than expected from sorption rates onto particle surfaces. The discrepancy can be explained if dissolved Th is initially sorbed to surfaces of very small particles (colloids) that pass through the typical filters (0.1-0.4 im) used to separate dissolved from particulate fractions (Santschi et al. 1986). 

Filtration Traditionally, aquatic chemists have operationally defined dissolved species by filtration through a 0.45 pm pore size membrane. The term dissolved is erroneously used to describe the filterable metal fraction. Filterable and nonfilterable metals are more accurate and appropriate terms. For water samples, filtration through small pore size (0.01-0.45 pm) membranes can be used to separate small particulate and even colloidal species. Ultrafiltration that involves the use of small pore size filters and pressurized filtration has been used to separate macromolecules. Ultrafilters are available with molecular weight cutoffs (MWCO) typically between 1000 and 50 000. Cross-flow ultrafiltration, where water flow is parallel to the surface of the membrane is used for the efficient isolation of colloidal material from large volume samples. Adsorption losses to the large surface area membranes can be an issue and several methods employ correction factors to account for such effects. 

Foam fractionation and flotation are the terms given to the separation processes by which inorganic and organic ions, molecules, colloidal particles, and suspended solids are floated from aqueous solutions with a rising stream of bubbles. Flotation is sometimes used to mean the removal of particulate materials by bubbling, whereas foam fractionation indicates the foaming off of dissolved materials by adsorption on bubble surfaces. Overlap and inconsistency in terminology are often encountered. 

---