Rejection measurements

Rejection measurements with reference molecules like dextrans, proteins or polyglycols are often used by membrane manufacturers. A parameter extensively used for membranes characterisation is the cut-off value, which is 

Recent studies based on comparison between gel permeation chromatography and ultra/micro-filtration [119] have shown that whatever the chemical nature and shape of the model macromolecule used, it is possible to predict the cut-off value of a membrane by considering the hydrodynamic volume of the macromolecule. This parameter provides an appropriate definition of the effective solute size to be considered in hydrod)mamic models. 

---

Deviation How much each measurement differs from the mean is an important number and is called the deviation. A deviation is associated with each measurement, and if a given deviation is large compared to others in a series of identical measurements, this may signal a potentially rejectable measurement (outlier) which will be tested by the statistical methods. Mathematically, the deviation is calculated as follows ... 

Rejections of solute compounds measured during the concentrations generally corresponded well with those measured in the screening test. Rejection values determined in the concentration test are presented in Tables IV and V along with values measured in the screening test. Rejection measurements made at 8- and 40-fold volume reductions are reported with the value determined at the lower concentration factor listed first (e.g., 87/83, see Tables IV and V). 

Bennett J. W., Comarmond M. J., Clark N. R., Carras J. N., and Day S. (1999) Intrinsic oxidation rates of coal reject measured in the laboratory. Proc. Sudbury 99 Mining Environ. 1, 9-17. 

Table VI shows the rejection measured for total N2, polyphenols and sugars with DDS 800 and PA 300 membranes. Sugar rejections increased from about 20% with DDS 800 to 100% with PA 300. Table VII shows the rejections of various cations using the PA 300 membrane. The rejection for the majority of cations is higher than 97%. Only Cu" " and Zn" " permeate easily through the membrane. The rejections for these two cations is essentially zero. The reason is attributed to a high specific interaction of Cu and Zn with the polymeric materials forming the membranes and to Donnan equilibrium.

A third possible complicating factor in rejection measurements, especially with protein solution, is physical Interaction of the solute with the membrane surface. Solute adsorption, for example, could alter the parameters L, a or P and cause anamo-lous rejection. An earlier search O for adsorption effects by a cellulosic membrane was negative. Although other investigators have reported such effects with non-cellulosic membranes, the effects with Cuprophan fibers were not observed, and adsorption parameters were not included in this transport model. 

The effect of serum on rejection of solutes, as compared with rejection measured out of saline, can be addressed in light of data for cellulosic membranes. Figures 10 and 11 demonstrate an effect consistently observed for the solutes myoglobin and cytochrome C. The observed and corrected rejection values fell off at approximately J >0.6 x 10 cm/sec from a plateau when saline was the solvent, whereas in serum the R versus J graphs were as predicted from the Spiegler-Kedem equation. The relative effects of serum or BSA on R, , and on resultant values for a... 

The GPC analysis of feed and permeate solution is ideally suited for rapid simultaneous rejection measurements. Simulta-eous rejection is of great Importance in ultrafiltration practice. As an example, we show here a simultaneous measurement of rejection of proteins and of lactose in whey ultrafiltration (Figure 13). The membrane used was the ABCOR HFK membrane and the feed solution had a typical composition of a partially concentrated... 

Figure 13. Simultaneous rejection measurement by GPC. The GPC profiles ARI trace Cupperl and A iso trace Clowerl jor whey jeed and permeate obtained by UF through the ABCOR HFK membrane. The peak labelled IgG corresponds to whey immunoglobulines peaks labelled a. and fi correspond to oc -lactalbumin and fi-lac-...

Manufacturers tend to characterize membranes by means of rejection measurements with reference molecules, such as dextrans, proteins, or polyglycols. A parameter extensively used for membranes characterization is the cutoff value, which is defined as the lower limit of solute molecular weight for which the rejection is at least 90%. We must keep in mind, however, that rejection measurements always depend on the type of solute (shape and flexibility of the macromolecular solute), the membrane (its interaction with the solute), and the process parameters used (pressure, cross-flow velocity, geometry of the test cell, and concentration and type of solute). In particular, concentration or polarization, pore blocking, and fouling phenomena affect rejection measurements very significantly. 

The TFC-SR membrane (Figure 7.1 IB) shows a similar pattern, the sodium rejection drops a litde between pH 5.5 and 6.5 (the isoelectric point was measured to be about 5). The low rejection measured at pH 7-8 was not confirmed here. 

The two actions, accept measurements (co) or reject measurements (oh), have different consequences, according to the prevailing unknown state acceptable errors (Eq) or unacceptable errors (Er). Let the consequences be quantified by the following loss structure ... 

In summary, solute rejection measurements provide a very simple technique for indicating the performance of a given membrane. For this reason they are very frequently used for... 

The time-domain algorithm operates directly on displacement data and can make use of singlefrequency excitation. In the formulation presented here the coefficients are assumed to be constant during the course of an experiment but time-dependence and geometric non-linearities can be accommodated if necessary. Perhaps the most powerful feature of the technique is that it is extremely effective at rejecting measurement noise and thus can operates on small displacement amplitudes. 

---