Flux experiments, rejection

To represent mass transfer at wall, the idea is to use a set of data (flux and rejection) as realistic (i.e., issued from laboratory experiments) and as easy-to-obtain (i.e., with a reduced number of experiments) as possible, and then to select by trial-and-error the simplest model (from constant transmission rate and flux if possible to more complex variation of these indicators vhth time if necessary). 

Preliminary experiments were carried out with the TFC-SR membrane to test the impact of background salt composition on flux and rejection. Rejection results are summarised in Table 7.9. 

Two membranes of the same type (TFC-S), but different batches were tested for the effect of their significantly different pure water flux on rejection. Rejection of organics decreased from 80 to 65% for a flux increase from 33 to 74 Lm Tr bar. This is an expected trend due to the increase in wall concentration with an increased fltrx. Therefore a large error can be expected due to variations in membrane quality. However, these problems can be minimised by measuring the water flux of each membrane prior to experiments. 

Air Gap Membrane Distillation (AGMD) experiments were performed for seawater desalination using prepered hydrophobic membrane. Seawater desalination aims to obtain fresh water with free salt adequate for drinking. In our work seawater treated is collected from SIDI MANSOUR Sea, located at Sfax (Tunisia). Measurements of p>ermeate flux and rejection rates were carried out by AGMD as a function of the temperature. The feed side temperature was thus varied from 75 °C to 95 C, while keeping the cooling system... 

The composite membrane was subjected to the permeation experiments in which the volume flux of water and the rejection of polymer solutes, defined by... 

The available transport models are not reliable enough for porous material with a complex pore structure and broad pore size distribution. As a result the values of the model par ameters may depend on the operating conditions. Many authors believe that the value of the effective diffusivity D, as determined in a Wicke-Kallenbach steady-state experiment, need not be equal to the value which characterizes the diffusive flux under reaction conditions. It is generally assumed that transient experiments provide more relevant data. One of the arguments is that dead-end pores, which do not influence steady state transport but which contribute under reaction conditions, are accounted for in dynamic experiments. Experimental data confirming or rejecting this opinion are scarce and contradictory [2]. Nevertheless, transient experiments provide important supplementary information and they are definitely required for bidisperse porous material where diffusion in micro- and macropores is described separately with different effective diffusivities. 

The inconsistency between experiment and prediction must lead to the rejection of the model used to describe the system. In the case of oxidative phosphorylation this has led to a refined model, in which the chemiosmotic coupling is visualized as taking place within units of one (or a few) respiratory chain(s) plus ATP synthase, while the pumped protons have only limited access to the bulk phase inside and/or outside the mitochondrion [42]. This more refined model can again be tested by deriving from it flux-force relations according to the MNET approach. A discussion of the refined model can be found in Ref. 43. 

Polysulfone membranes were prepared from 12.5, 13.75, and 15% (wt. %) polysulfone solution in dimethylformamide and formed on the surface of porous, sintered polymethyl methacrylate bars. An effective surface of each membrane was 49.2 cm. The effect of some casting parameters (composition and the temperature of the casting solution, time of solvent evaporation) and the pressure applied on the transport and separation properties of the membranes were analyzed. The experiments were carried out in a 1.2 dm pressure apparatus with continuous circulation of the permeate between feeding tank and the apparatus. It was found that membranes cast from 12.5% polysulfone solution of a temperature of 298 K with no solvent evaporation displayed the best properties. After 160 hours of operation at 0.18 MPa, the membranes in question showed an ability of a 97 to 99% rejection of 781.2 molecular-weight dye. The volume flux of the dye solution varied from 0.6 to 0.8m /m per day. 

The influence of metal oxide derived membrane material with regard to permeability and solute rejection was first reported by Vernon Ballou et al. [42,43] in the early 70s concerning mesoporous glass membranes. Filtration of sodium chloride and urea was studied with porous glass membranes in close-end capillary form, to determine the effect of pressure, temperature and concentration variations on lifetime rejection and flux characteristics. In this work experiments were considered as hyperfiltration (reverse osmosis) due to the high pressure applied to the membranes, 40 to 120 atm. In fact, results reproduced in Table 12.3 show that these membranes do not behave as h)qjerfiltra-tion membranes but as membranes with intermediate performances between ultra- and nanofiltration in which surface charge effect of metal oxide material plays an important role in solute rejection. 

Ultrafiltration experiments were performed with an Amicon 8050 cell at 25 0 using a stirrer speed of approximately 700 rpm. Water fluxes (hydraulic conductivities) were measured at Ap = 1 psi dextran rejection was measured for feed solutions containing 0.2% T40, 0.2% TIO and 0.1% T500 (see Materials) under conditions of low concentration polarization. Transmembrane fluxes of dextran solutions were of the order of 0.2 x 10 cm/s at Ap = 1 psi. Feeds and permeates were analyzed by size-exclusion chromatography as described in Reference 9, and the chromatographs were used to calculate the rejection curves (Figure 1). 

The low ethanol rejection and the instability of the hollow-fiber NS-100 membranes preclude the use of this membrane for practical ethanol enrichment. Nevertheless, for the purpose of demonstrating the concept of CCRO using hollow-fiber membranes, CCRO experiments were conducted at the reduced ethanol concentration of 10 vol%. The permeate fluxes of NS-100 modules were measured at 250 psi in the absence and presence of recirculation with a 10-volX ethanol solution. The results were varied recirculation brought about flux increases ranging from 5% to about 20%. The limited flux increase may again be explained in terms of the formation of a polyamine gel during NS-100 membrane fabrication. Nevertheless, the flux increase shows that the hollow-fiber geometry is a viable one for CCRO operation. 

Sugar rejection was always zero. Polyphenol rejection increased from 9% to 57% with decreasing membrane cut-off. Difference in polyphenol rejection have been observed when the must is ultr filtered in the first UF step with the BMR 100515 and after with the BMR 021006, as reported in table III. Fig.3 shows the typical behaviour of the ultrafiltrate flux observed in these experiments. A constant flux was generally obtained after two hours. Table IV shows the final must ultrafiltrate flux values. All the experiments were carried out at the same axial velocity and at the same applied pressure. Table V shows results obtained with tubular membranes (Abcor-USA). 

Flux decays and rejection changes in the UF steps, depending upon the feed concentration and experiment history, must be mainly attributed to concentration polarization phenomena with consequent gel layer formation of the pressurized membrane surface, for the presence of proteins, colloids and in general high molecular species in the feed (3). 

Rejection Coefficients for Protein Ultrafiltration Experiments. Since the data for ultrafiltration rates seem to imply the condition of uniform wall flux, the calculation procedure for finding R for these experiments was identical with that used for saline solutions. No doubt the dlffusivity of the small solutes within... 

To determine CTg and Pg values it is necessary to measure the rejection at different volume flux, Jv, that means it is necessary to perform experiment at different pressures. And at different pressures Rg changes, when the bulk concentration is kept constant. To keep Rg value constant at different pressure it is necessary to change bulk concentration or the feed velocity. In our experiment the latter was kept constant, while the bulk concentration was adjusted to keep Rg constant. Even so it is hard to keep Rg precisely constant. Therefore finally the experimental relations between Rg and rejection, R, were obtained by plotting data on the graph and R values were read at appropriate Rg by Interpolation. 

Typical results of an ultrafiltration experiment also reflect the presence of concentration polarization. This phenomenon, l.e. accumulation of solute in front of the membrane, was described in great detail by others (Refs. 3, 4). A consequence of concentration polarization is a strong dependence of measured rejection coefficients on transmembrane fluxes. An illustration of the effect is presented in Figure 9, which shows the measured "apparent" rejection coefficients (Rg) as a function of transmembrane flux for two water-soluble polymers (Tetronic 707 and Carbowax 4000). It is clear from Figure 9 that if we want to minimize the effects of concentration polarization, we have to conduct experiments at very low values of transmembrane flux. 

Two experimental procedures were carried out. In the first, a 100 1 (water and dye filtrate) dilution was concentrated to one-tenth its initial volume. Rejection based on color absorbance (HlO nm) and electrical conductivity, flux, pressure, temperature, and crossflow rate were measured at intervals during the concentration experiment. In the second, a slightly diluted dye filtrate (2 3) was used and the hyperfiltration at steady state was evaluated as in the first procedure. The test was repeated at dilutions reaching (100 1), with pH and temperature excursions at a dilution of 3 1. 

Equation (26.46) shows that the water flux increases strongly with the pressure difference AP, and the selectivity increases also, since the salt flux does not depend on AP. Experiments confirm these trends, but the salt rejection with cellulose acetate is not as high as predicted. The water content C is about 0.2 g/cm , and tracer tests show 10 cmVs- Diffusion tests of NaCl in dense polymer films indicate = 0.035 and = 10 cm s- The fluxes and cannot be predicted accurately for an asymmetric membrane, because the skin thickness z is not known. However, the ratio of fluxes is independent of z, and the predicted salt... 

Sourirajan s partner, Sidney Loeb, began to experiment with laboratory UF membranes (not fully asymmetric) made by Schleicher and Schuell of cellulose acetate (CA). Loeb heated the membranes under water (annealing) to temperatures between 80° to 90°C, thereby increasing the salt rejection from 0 to 92%, but the water flux decreased to unacceptable levels. 

Osmotic phenomena have been observed since the middle of the eighteenth century. The first experiments were conducted with animal membranes and it wasn t unitl 1867 that artificial membranes were employed. In the early 1950 s, research workers at the University of Florida demonstrated, with thick films, that cellulose acetate possessed unique salt and water transport properties which made it potentially attractive as a reverse osmosis desalination membrane. During the 1960 s, Loeb and others at the University of California at Los Angeles developed techniques to prepare cellulose acetate membranes with an economical water flux and salt rejection at moderate driving pressures. With this development, reverse osmosis became a practical possibility. 

In other respects, it also shared the favorable characteristics of the NS-100 membrane. That is, it was resistant to pH 3 to 12, and showed far better compaction resistance than cellulose acetate. Also, it possessed the capability to operate at elevated temperatures, though some irreversible flux decline could still occur.34 Rejections of various organics were also good, as shown in Table 5.1 These were in sharp contrast to organic rejection data on cellulose acetate membranes. Initially, PA-300 was also postulated to possess good chlorine resistance.31 Subsequent experience showed it to be equally sensitive to chlorine as NS-100. 

A typical recipe for an interfacially formed aromatic polyamide composite membrane comprised a 2.0% aqueous solution of the aromatic diamine and a 0.1% nonaqueous solution of trimesoyl chloride. This recipe was extraordinarily simple, and ran quite contrary to experience with piperazine-based membranes. For example, surfactants and acid acceptors in the aromatic diamine solution were generally not beneficial, and in many cases degraded membrane performance by lowering salt rejection. In contrast, surfactants and acid acceptors were almost always beneficial in the NS-300 membrane system. In the nonaqueous phase, use of isophthaloyl chloride as a partial replacement for trimesoyl chloride had relatively little effect on flux, but tended to decrease salt rejection and increase susceptibility to chlorine attack. 

Effects of operating pressure on permeate flux value and rejection value during the total circulation experiments are shown in Figure 22.6. Operating pressure had little effect on the rejection value. Permeate flux value increased linearly with operating pressure up to 4 MPa. Once the operating pressure exceeded 4 MPa, permeate flux did not increase linearly. Therefore, suitable operating pressure for separation of anserine and carnosine with the nanofiltration membranes was determined to be 4 MPa. Similar results were obtained with other membranes. 

A summary of results obtained in the total circulation experiments with the 13 different nanofiltration membranes is listed in Table 22.2. The purpose of the nanofiltration treatment is to improve the purity of anserine and carnosine contained in the feed solution by removing impurities such as creatinine and sodium chloride into the permeate (Figure 22.8). Therefore, a membrane that shows higher rejection ability against anserine and carnosine and low rejection ability against creatinine and sodium ion is preferred for this purpose. Furthermore, higher permeate flux value implies that a process with the membrane will require smaller membrane area and lower initial cost. Based on these criteria, four membranes (NFT-50, DRA-4510, Desal DK, and Desal DL) were chosen as suitable membranes for purification and concentration of anserine and carnosine from chicken extract. 

The same system as described in the MF section and shown in Figure 4.1 was used for all rejection, fouling, and fractionation ultrafiltration experiments. The balance was connected to a PC for flux data collection. 

Flux decline is therefore the decline in pure water flux of the membranes before and after the experiments. The majority of experiments performed were recycle experiments. The apparent rejection was calculated using the following equation. Cpiis the concentration of permeate sample i taken directly from the membrane and not from the permeate container. It is thus the permeate concentration averaged over the filtration interval required to collect the sample. The concentration in the batch cell was measured at the end of each experiment. 

Experiments were carried out in the presence of electrolyte solution and pH 8 (in this case to provide a baseline for the aggregation experiments with organics). The rejection of the 75 nm colloids was complete, as shown in Table 5.3 (no 3). Figure 5.4 shows that after a egation the flux ratio is high... 

The aggregates were mechanically strong and did not break at the membrane surface in the standard (lOOkPa) experiments. Breakage of some of the aggregates occurred at a relatively high pressure of 300 kPa, with a 20 to 25% lower rejection. Results are shown later in the SPO section (Figure 5.11). The flux results are not shown, but the flux decline was very similar (50%) to that obtained in other experiments where 100 kPa was applied. 

---