Apparatus for rapid casting of tubular cellulose acetate reverse osmosis and ultrafiltration membranes

An improved apparatus is described for making tubular cellulose acetate membranes for reverse osmosis and ultrafiltration applications. The incorporation of an adjustable centering-bob and a sleeve in the design of the casting-bob housing, and the inclusion of an automatically controlled electrical water probe at the bottom of the casting-bob are the novel features of the apparatus. The adjustable centering-bob offers the capability of regulating the passage for the flow of the casting solution during film casting; this capability makes the casting-bob housing useful for a wide range of casting solution viscosities necessary for making both reverse osmosis and ultrafiltration membranes. The sleeve incorporated in the design makes it possible to switch from one tube-casting to the next immediately without any need for the intermediate time consuming operation of cleaning up the casting-bob system. Thus a single casting-bob housing is sufficient for making a plurality of membranes, one after another, with little loss of time between castings. The water probe maintains in the casting tube any desired length of air-zone for the freshly cast membrane. The operation of the apparatus is amenable to a high degree of automation. These features make the apparatus particularly suitable for industrial utilization.     

Formation of asymmetric cellulose acetate membranes

Cellulose acetate membranes were prepared from casting solutions containing dioxane as a solvent and varying concentrations (up to 6%) of maleic acid as an additive. Coagulation took place in water at different temperatures. The effect of these variables on membrane structure and membrane properties is related to two phenomena of phase separation in the system cellulose acetate/dioxane/water, viz. gelation in the toplayer and liquid-liquid phase separation in the sublayer of the membrane. We adopted a solution transport model which correlates membrane flux with the skin thickness and membrane salt rejection with the compactness (microstructure) in the skin. Effects of variables such as maleic acid concentration and coagulation temperature on the position of curves in the phase diagram and on membrane properties are discussed. It is concluded that more work should follow on the kinetics of the phase changes mentioned.     

Phenol transport in cellulose acetate membranes

Past detailed studies of solute transport through reverse-osmosis membranes have been conducted only with simple salts. The present work with phenol was undertaken largely because of the practical observation that the transport of low molecular weight organics is much more rapid than that of the salts. Studies of phenol sorption from dilute aqueous solution indicate that the diffusion coefficient for phenol in water-saturated 39.8 wt.-% acetyl cellulose acetate is 9.6 × 10^-10 cm.²/sec., and the equilibrium distribution coefficient between the acetate phase and water is 42. Thus, the diffusion coefficient is quite close to that measured for sodium chloride, and the higher permeability of the membranes to phenol can be attributed entirely to their greater sorption of this solute. In direct osmosis experiments performed with significant water flow a measurable interaction or positive coupling between water and phenol flows has been observed. Further evidence of flow coupling is derived from reverse osmosis experiments in which significant negative solute rejection is observed; i.e., the permeate is enriched in phenol by as much as 20%. It is shown that a solution-diffusion transport model is not adequate to rationalize the results, and a more complex transport model is apparently required.     

High flux cellulose acetate ultrafiltration membranes

Conditions for casting high flux tubular and flat cellulose acetate ultra-filtration membranes are given. The usefulness of some of the tubular membranes cast under the above conditions for ultrafiltration applications in industrial water purification and reuse is illustrated. Using ethanol-water mixture as the gelation medium in the temperature range-20° to 30°C, ultrafiltration membranes giving water fluxes in the range <4 to >400 m³/m² day at the operating pressure of 689.5 × 10³ Pa (100 psig) have been obtained.     

Modified cellulose acetate membranes for desalination

Residual hydroxyl groups of commercially available cellulose acetate (39.9% acetyl) have been reacted partially with phenyl isocyanate. The characteristics of these modified polymers have been studied. Membranes have been cast from these polymers to study their potentiality as reverse osmosis membranes. The work has been further extended to investigated the stability of these modified CA under the influence of high temperature and γ irradiation.     

Sulfonamide release from cellulose acetate membranes

Summary Cellulose acetate membranes were chemically modified by esterification with oxalyl chloride which was used as a hydrolytically labile spacer. Then five sulfonamides were attached to the acyl chloride present on the halfesterified membranes to give drug-polymer conjugates. The rate of the hydrolytic cleavage of the spacer-drug linkage was measured in simulated gastric fluids at 37°±0.5°C by determining the amount of drug released by spectrophotometric analysis. The results showed that in all cases the process followed a zero-order kinetics.     

Ion movements in cellulose acetate membranes

The behavior of ions in a cellulose acetate membrane was discussed from the standpoint of mobilities obtained from the membrane potentials. The mobilities of univalent ions larger in radius or divalent ions relative to Na⁺ ion mobility in the membrane are much different from those in aqueous solutions. The order of the mobilities of ions in the membrane is shown quantitatively. The mobilities of the ions with radii larger than a certain value in the membrane decrease with increase in their size. This could be explained by the physical friction between the ion and the membrane wall. This friction is of importance as the ionic radius comes close to the intermolecular gaps between polymer chains. The mobilities of Na⁺ anyd 1^? ions in the membrane are (2–3) × 10^?7 cm²/sec-V and are about three to four orders of magnitude less than those in aqueous solutions.     

Permeability studies with cellulose acetate membranes

Summary The diffusive permeability of cellulose diacetate membranes was measured using NaCl as solute and distribution coefficients were calculated. Water and salt permeability coefficients derived from dialysis-osmosis experiments were also given. The results obtained were discussed with the aid of scanning electron micrographs and chlorine distribution maps.     

DRY-RO membranes from cellulose acetate carbamates

Cellulose acetate carbamates (CACs) are the polymers which result when organic isocyanates are reacted with the free hydroxyl groups of cellulose acetate (CA). CACs are more hydrolytically stable and exhibit physical properties which are superior to those of their CA mixed ester analogs. Two synthetic approaches to CACs have been utilized in this study: (1) preformation, i.e., separate synthesis of such polymers prior to their inclusion in solutions for membrane casting; and (2) in situ formation, i.e., the inclusion of blocked isocyanates in standard dry process casting solutions of CA followed by thermal activation of the resultant dry membranes leading to regeneration of free isocyanate and subsequent CAC formation. Preformed CAC polymers have been prepared utilizing phenyl-, 3-chloropropyl-, 3-bromopropyl-, and 3-bromopropyl-(isothio)-, isocyanates. Polymers containing omega-halocarbamate moieties were quaternized with dimethylbenzylamine to produce ionogenic (QCAC) polymers containing quaternary ammonium groups. DRY-RO membranes from the QCACs exhibit flux/rejection values varying between 6–8 gfd at 98% rejection and 20 gfd at 90% rejection (0.5% NaCl feed at 400 psi and 25°C). In situ formation of CAC membranes has been effected with tolylene- and hexamethylene-diisocyantes, with quaternized isocyanate monomers employed for the preformed CAC polymers, and with specially tailored diisocyanates containing ionogenic groups. Crosslinking rendered all of the membranes acetone insoluble. Inasmuch as in situ formation substitutes the easy synthesis of blocked isocyanate monomers for the more difficult separate synthesis of preformed CAC polymers, it is anticipated that the former will replace the latter.     

Improvement of porous cellulose acetate reverse osmosis membranes by change of casting conditions

The effects of temperature of casting solution in the range ?10° to 15°C, that of casting atmosphere in the range 10° to 30°C, relative humidity of casting atmosphere in the range 35% to 75%, and solvent evaporation period in the range 0.5 to 3 min were studied on shrinkage temperatures, solute separations, and product rates of Loeb-Sourirajan-type cellulose acetate membranes in reverse osmosis experiments. The composition of casting solution used was as follows: cellulose acetate, 17; acetone, 69.2; magnesium perchlorate, 1.45; and water, 12.35 wt-%. Best performance was obtained with membranes cast under the following conditions: temperature of casting solution, 10°C; temperature of casting atmosphere, 30°C; relative humidity of casting atmosphere, 65%; and solvent evaporation period, 1 min. For a 90% level of solute separation, the productivities of the above type of membranes were 22.9, 61.4, and 64.5 gallons/day-ft² at 250, 600, and 1500 psig using 3500 ppm NaCl–H₂O, 5000 ppm NaCl–H₂O, and 28395 ppm NaCl–H₂O feed solutions, respectively. In all cases, the feed flow rates corresponded to a mass transfer coefficient of 45 × 10^?4 cm/sec on the high-pressure side of the membrane. The general specifications of the above type of membranes are given for the operating pressures of 250, 600, and 1500 psig. The effects of the above casting condition variables on the surface pore structure during film formation are discussed.     

Effect of casting conditions on the performance of porous cellulose acetate membranes in reverse osmosis

Several sets of porous cellulose acetate membranes were made using the same casting solution composition and gelation conditions but varying the casting solution temperature and solvent evaporation conditions. The films were tested in reverse osmosis experiments at 250 psig using aqueous feed solutions containing 3500 ppm NaCl. The results show that the product rate obtained at a given level of solute separation is independent of evaporation time in the range tested and, for a given casting solution composition, the temperature of the casting solution and conditions of solvent evaporation during film formation together constitute an important interconnected variable governing the porous structure of the resulting membranes. These results offer a new approach to the problem of developing more productive reverse osmosis membranes and have led to a new class of porous cellulose acetate membranes capable of giving product rates 100% to 150% higher than those of the best membranes reported, at any given level of solute separation under the experimental conditions used. These results are of practical importance in low-pressure reverse osmosis applications.     

Permeabilities of coagulated cellulose acetate dialysis membranes

The diffusive flux of NaCl and the hydraulic flux of pure water through coagulated cellulose acetate membranes are examined. Coagulated cellulose acetate membranes (without densification by heat treatment or drying) possess higher permeability than what may be expected from the permeabilities of the dry polymer. Their overall hydraulic permeability (ultrafiltration rate of water) is greatly dependent upon the membrane casting conditions and the resulting asymmetry of the membrane. On the other hand, the asymmetry of a membrane does not play as great a role in the diffusive permeability of a solute. With homogeneous membranes, higher diffusive flux is always accompanied by higher hydraulic permeability. With asymmetric membranes, this is not always true. The diffusive permeability of NaCl and the hydraulic permeability of water through coagulated cellulose acetate membranes can be controlled nearly independently. Consequently, high diffusive (NaCl) permeability with low hydraulic water permeability and vice versa can be obtained by varying the casting conditions and also by partially saponifying the denser portion of the membrane.     

Immobilization of polymers on cellulose acetate membranes

Using models of dead-end filtration theory, the kinetics of forming dynamic layers of sulfate lignin (SL) and sodium carboxy-methylcellulose (Na-CMC) onto CA membranes during cross-flow filtration of dilute solutions of polymers was studied. It was found for both polymers (compact SL and linear Na-CMC), that the polymer layer with the least hydraulic resistance, which yields a small reduction in membrane water permeability (10–20%), but a significant increase in salt rejection, is formed, if the process kinetics corresponds to J-V linear dependence predicted by the model of 'blocking a pore by a single particle's. The results obtained may be used to define the optimum conditions for immobilizing the available catalytic active polymers on regular semipermeable membranes during the membrane filtration process.     

Transport coefficients of asymmetric cellulose acetate membranes

Using the linear relations of thermodynamics of irreversible processes, the transport coefficients l_p,l_π,l_πp and σ were measured for an asymmetric cellulose acetate membrane with NaCl, Na₂SO₄, CaCl₂, NaF, and saccharose over the concentration range (0–0.5 mol/l or l mol/l) at 20°C or 25°C. The experimental findings manifest a strong dependence of the three transport coefficients l_p,l_π, and linπp on solute concentration. This strong dependence on concentration can be attributed to a concentration gradient within the porous sublayer of the asymmetric membrane. Thereby, it is shown that the transport coefficients of an asymmetric membrane depend on the solute concentration on both sides of the membrane instead of the mean concentration c?_s, as would be the case for a homogeneous membrane.     

Membrane potential measurements on cellulose acetate membranes

Membrane potentials and apparent transport numbers of the cation are reported for cured cellulose acetate membranes bounded by HCl, NaCl, KCl and MgCl₂ solutions, using Ag/AgCl electrodes and a flow-cell method. Membranes cured at 70°, 80° and 90° are used. Bounding solution concentrations vary from 0.005 to 0.05 M at the high concentration side (bounding the dense side of the membrane), and are kept constant at 0.002 M for the low concentration solution. In the KCl 90° membrane case the low concentration solution is varied as well, from 0.0001 to 0.002 M. Results show that cured cellulose acetate membranes are permselective towards univalent cations. This is interpreted as resulting from a low cation-exchange capacity of the dense layer of the cured membrane. The permselectivity increases with increased curing temperature. Addition of a non-electrolyte to the low concentration side reverses the osmotic flow and leads to higher apparent transport numbers of the cation. It is concluded that diffusion in small pores contributes significantly to the transport of ionic solutes through uncompacted membranes.     

Improved cellulose acetate membranes for reverse osmosis

Performance of cellulose acetate membranes in reverse osmosis varies with the conditions under which they are cast. By varying casting solution composition and holding time in a systematic way, improvement in water flux at a given level of salt rejection has been obtained. Statistically designed experiments have been helpful in optimizing these two variables. A phase diagram of the cellulose acetateformamide-acetone casting system has been determined which gives the region of natural solubility of this three component system.     

Studies on the radiolytic degradation of cellulose acetate membranes

The effect of γ-irradiation on the performance of wet cellulose acetate membranes in the dose range of 2.5–10 Mrads was investigated using a ⁶⁰Co source. Changes in transport properties and inherent viscosity of the membranes suggested continued degradation as a result of irradiation. Solubility and specific gravity changes accompanying irradiation indicated some sort of structural aggregation occuring at higher doses. Consumption of dissolved oxygen during irradiation and the extent of deacetylation of the membrane polymer were determined to study the kinetics of the degradative process. Analysis of the end products of irradiation was attempted by UV spectroscopy. ESR spectra of membrane polymer after irradiation were analyzed to identify the free radicals generated. A tentative mechanism of radiolytic degradation causing the observed performance failure is proposed.     

Movements of univalent ions in cellulose acetate membranes

Membrane potentials across the asymmetric membranes of cellulose acetate with various salt rejection properties have been measured for univalent ions. The behavior of ions in the membranes is discussed from the viewpoint of relative ionic mobilities calculated from the membrane potentials. The relationship between ionic mobilities and ionic radii in the membranes having salt rejections lower than 80% is almost the same as that in aqueous solutions. This implies that the ions in these membranes behave as if they exist in bulk water. However, the ionic mobilities in the membranes having salt rejections higher than 86% differ significantly depending on the ionic radii. It seems probable that the bound water influences the ionic mobilities in these membranes.