Filter size fractionation

The comparison of REE concentrations for the <0.2 pm fraction with the lower-filter-size fraction shows that there is no unique pattern of colloidal material when rivers of different pH and different environments are compared. Ultrafiltration experiments conducted by Deberdt et al. (2002) on rivers from the Amazon and Orinoco basins as well as on Cameroon Rivers show slightly depleted LREE patterns to flat REE patterns when the colloidal fraction is normalized to the bulk solution. The results obtained by Sholkovitz (1995) and Ingri et al. (2000) for rivers... 

Because a filter sample includes particles both larger and smaller than those retained in the human respiratory system (see Chapter 7, Section III), other types of samplers are used which allow measurement of the size ranges of particles retained in the respiratory system. Some of these are called dichotomous samplers because they allow separate measurement of the respirable and nonrespirable fractions of the total. Size-selective samplers rely on impactors, miniature cyclones, and other means. The United States has selected the size fraction below an aerodynamic diameter of 10 /xm (PMiq) for compliance with the air quality standard for airborne particulate matter. 

Filter aids should have a narrow fractional composition. Fine particles increase the hydraulic resistance of the filter aid, whereas coarse particles exhibit poor separation. Desired particle-size distributions are normally prepared by air classification, in which the finer size fractions are removed. 

Several studies have examined the partitioning of U on particles and colloids. Results from detailed sampling and particle separation in the Amazon estuary shows that most of the uranium at the Amazon River mouth is associated with particles (>0.4 im) and that >90% of the uranium in filtered water (<0.4 im) is transported in a colloidal phases (from a nominal molecular weight of 10 000 MW up to 0.4 im) (Swarzenski et al. 1995 Moore et al. 1996). Mixing diagrams for uranium in different size fractions in the Amazon estuary reveal that uranium in all size fractions clearly display both removal and substantial input during mixing. 

Andrews and Sundaresan (2005) have also extracted the filtered particle-phase viscosity from these simulations and found that at low particle volume fractions (0.0-0.25), the filtered viscosity varies nearly linearly with particle volume, and that it increases monotonically (and nearly linearly) with filter size. 

The use of filters and membranes of different pore size to accomplish a sequential size fractionation is in principle, and under certain circumstances, possible it was proposed (for literature see Buffle, 1988 and 1991) to estimate the size of the various colloids and macromolecules and to determine to which extent trace elements (particularly metals) are associated with various size categories of colloids and macromolecules. Such sequential size fractionation techniques need to be applied with extreme caution we list some of the reasons why these techniques may yield errorous results (for details consult Buffle, 1991) ... 

Operational problems in size fractionation by membrane filters... 

In another series of e q>eriments in this laboratory (U), it was found that the particle size of the coal fraction used affected the ash level in the filtered coal extract solution. In these e q>eriments, the coal sample, which is nominally < 200 pm, was divided into different size fractions by sieving, and each fraction digested using the normal procedure. Table 3 shows the results obtained with Point of Ayr and Calverton coals. Note that the Point of Ayr coal was a different batch to that used for the results described previously, with a simil analysis except for ash, which was higher at 19 %. 

Elutriation is important in most industrial fluidized beds and is generally thought of as a disadvantage. In addition to the small particles which may be present in the initial particle size distribution, fines may be created in the course of operation by the attrition of bed particles. Elutriated particles usually need to be collected and recovered either because they represent the loss of product particles of a given size, because they must be separated from the exhaust gas for environmental reasons, or because of safety concerns there is a considerable risk of a dust explosion with very fine particles and perhaps especially so with many food particulates. Therefore the fluidized bed plant will require ancillary gas cleaning equipment such as a cyclone, filter or electrostatic precipitator to separate the fines from the gas. The loss of a particular size fraction from the bed may change fluidized bed behaviour and it then becomes important to return the fines to the bed continuously. 

Improved control devices now frequently installed on conventional coal-utility boilers drastically affect the quantity, chemical composition, and physical characteristics of fine-particles emitted to the atmosphere from these sources. We recently sampled fly-ash aerosols upstream and downstream from a modern lime-slurry, spray-tower system installed on a 430-Mw(e) coal utility boiler. Particulate samples were collected in situ on membrane filters and in University of Washington MKIII and MKV cascade impactors. The MKV impactor, operated at reduced pressure and with a cyclone preseparator, provided 13 discrete particle-size fractions with median diameters ranging from 0,07 to 20 pm with up to 6 of the fractions in the highly respirable submicron particle range. The concentrations of up to 35 elements and estimates of the size distributions of particles in each of the fly-ash fractions were determined by instrumental neutron activation analysis and by electron microscopy, respectively. Mechanisms of fine-particle formation and chemical enrichment in the flue-gas desulfurization system are discussed. 

The particle sampler chosen for this study was the Stacked Filter Unit (SFU) described by Cahill e al (, ) Particle collection in two size fractions was achieved by placing two Nude-pore membrane filters in series. The first filter, with 8pm... 

Up to this point, the evidence for two particle populations has been based solely on the refractory specific activity. Additional confirmation is based on the observed particle size distribution by mass of an aerial filter sample. A portion of the aerial filter sample designated as 2 in Table I, was separated into size fractions, and the weight distribution of the fraction is shown in Figure 3. The ordinate values are simply ... 

In operationally defined speciation the physical or chemical fractionation procedure applied to the sample defines the fraction isolated for measurement. For example, selective sequential extraction procedures are used to isolate metals associated with the water/acid soluble , exchangeable , reducible , oxidisable and residual fractions in a sediment. The reducible, oxidisable and residual fractions, for example, are often equated with the metals associated, bound or adsorbed in the iron/manganese oxyhydroxide, organic matter/sulfide and silicate phases, respectively. While this is often a convenient concept it must be emphasised that these associations are nominal and can be misleading. It is, therefore, sounder to regard the isolated fractions as defined by the operational procedure. Physical procedures such as the division of a solid sample into particle-size fractions or the isolation of a soil solution by filtration, centrifugation or dialysis are also examples of operational speciation. Indeed even the distinction between soluble and insoluble species in aquatic systems can be considered as operational speciation as it is based on the somewhat arbitrary definition of soluble as the ability to pass a 0.45/Am filter. 

The type of approach described here is obviously more important for systems where the solution chemistry of the nuclide (charge, oxidation state and degree of complexation) is more complicated. Without supporting laboratory data, it is possible that significant retention values may be incorrectly interpreted as being due to radionuclide association with material in a particular size fraction. The components of the environmental sample might contribute to the separation process and retain species which on a size basis should readily pass through the filter membrane. 

Sampling devices range from simple filter holders to sequential configurations for automated routine air monitoring of many samples in series. Membrane filters can be obtained in different pore sizes, so that they can be used in series as particle size fractionators. 

Ultrafiltration was applied to examine the size fractionation of Al, Ca, Cu, Fe, K, Na, and Pb in white and red wines [91]. Metal determinations were performed on the unfiltered wine, the 0.45 p,m membrane-filtered wine and each ultrafiltrate fraction. Aluminum was determined by ET-AAS, while FAAS was employed for Cu and Fe. An electroanalytical technique, stripping potentiometry, was selected for Pb measurement, whereas flame photometry was chosen for K and Na quantification. Fractionation patterns were evaluated and discussed. Castineira et al. [92] combined on-line tangential-flow multistage ultrafiltration with a home-built carbon analyzer and ICP-MS for size fractionation of nonvolatile dissolved organic compounds and metal species in three German white wines. The study showed that the major part of the elements investigated (up to 25) were dissolved in the size fraction of < 1 kDa, with the exception of Ba, Pb, and Sr, which also appeared in other fractions. 

As an example of thin-foil analysis in the SEM, samples of the mineral standard kaolinite (<2 pm fraction, KGa-2, Source Clay Minerals Repository) were dispersed on TEM grids and analyzed using a JEOL JSM 6400 SEM under the control of an automated Noran Voyager EDS system equipped with a thin-window EDS detector.93 For comparison, EDS spectra were also collected using the same instrument for the kaolinite size fractions deposited on polycarbonate filters, fixed to SEM... 

---