GB2182581A - Fabrication of separation media - Google Patents

Abstract

Process for fabricating ultra thin membranes for gas separation, comprises the following steps: (1) A polymer is dissolved in a suitable solvent and allowed to spread on a water surface. (2) The solvent is allowed to evaporate and to leave a monolayer of the polymer on the water surface. (3) A porous substrate e.g a polymer sheet or hollow fibre, oriented perpendicularly to the film is passed at a fixed speed through the air water interface and then in the reverse direction. Layers of the polymer are thereby deposited onto the substrate so as to build up the ultra thin membrane.

Description

SPECIFICATION Fabrication of separation media The present invention relates to gas separation and liquid separation and more particularly relates to a process for the fabrication of separation media.

The separation ofgas mixtures into their individual components has numerous applications and the development of membranes processes for these separations has become of increasing importance. The use of membrane processes for gas separation has certain advantages over alternative techniques for example those based on adsorption, absorption and liquefaction. These advantages include potential energy efficiency, compactness, relative simplicity and ease of operation. The modular nature of membrane technology also enables scaling up or reduction of capacity when necessary.

The membranesforthe gas separation process are desirably (a) highly permeable to components ofthe mixture to be separated (b) highly selective for specific components ie the specific components have a high permeability relative to other constituents in the gas mixture (c) chemically inert and physically stable and (d) continuous iefree from defects such as pinholes. The desired product may be the permeate ornon-permeate or both.

Certain polymers appear to have characteristics making them suitable for use in the form of gas separation membranes. However conventionally prepared polymeric membranes generally have low gas permeation rates and poor selectivity often resulting in uneconomic separation in the case of the large gas flows.

The present invention relates to ultra-thin gas separation membranes capable of relatively large permeation rates and relatively high selectivity and a process forfabricating the membranes.

Thus according to the present invention there is provided a process for fabricating a composite separation medium comprising the steps of (a) dissolving a polymer in a suitable solvent, (b) spreading the solution on the surface of an aqueous liquid, (c) allowing the solvent to evaporate to leave afilm of a polymer on thewater surface, (d) passing a porous substrate oriented substantially perpendicularly to the film through the film, preferably at a fixed speed, through the air/water interface, and (e) passing the substrate through the film in the reverse direction. Either or both steps (d) and (e) may deposit a layer of polymer onto the surface ofthe porous substrate. Preferably the aqueous liquid is water.It is desirable that the polymer solvent combination spreads evenly overthe surface ofthe aqueous liquid.

The invention also includes composite separation media or ultra thin membranes whenever prepared by a process as hereinbefore described.

Steps (d) and (e) may be repeated as necessary to build up successive layers of polymer on the porous substrate. The deposited film on the porous substrate preferably has a thickness of the order 100 Angstroms when a number of layers have been built up.

Byfilm is meant a layerofthe polymer built upfrom one or more monomolecularthicklayers. It is desirable to ensure that the monolayer is continuous on the water surface and preferably the film is compressed to a specific surface pressure and maintained at that pressure or a pressure sufficient to maintain the continuous nature of the film. This may be achieved by the use of a Langmuirtrough such as that supplied byJoyce-Loebl Ltd.

Preferably the polymer comprises hydrophilic groups attached to a hydrophobic backbone or chain.

Suitable polymers include cellulosic polymers such as cellulose acetate and ethyl cellulose, polyimides, acrylic polymers such as polymethyl methacrylate, and vinyl polymers such as polyvinyl acetate.

Suitable polymer/solvent combination include polyvinyl acetate (PVA) and polymethyl methacrylate (PMMA) dissolved in chloroform. Blends or mixtures of polymers in suitable solvents may also be used.

Suitable porous substrates include polymer sheets or hollow fibres and have a pore size generally not greaterthan 0.5 microns in diameter in the surface to be coated.

The speed or dipping rate ofthe substrate into the film may be in the range 0.1 to 60 cm/min and is preferably 1 to 5 cm/min.

The invention will now be described by way of example only with reference to Figures 1 to 3 ofthe drawings. Figure 1 is a plan view ofthe Langmuirtrough used in the experiments.

The Langmuirtrough employed for the experiments was Joyce-Loebl unit constructed from a glass U-section 10 contained within a metal frame 11. Glass end plates 12 are clamped to the U-section (sealing being affected by PTFE compression strips) by the framework. The approximate volume of the trough was 18 litres, the maximum working surface are being about 1000 cm2 and the maximum dipping height approximately 9 cm. The trough rested on a table of adjustable height which facilitates filling of thetrough and zero adjustment of a Wilhelmy plate.

The working area of the water surface of the trough was defined not by the edges of the trough, but rather by a continuous band of PTFE coated tape 13 mounted on a system of six PTFE rollers 14, and dipping into the water surface 15 (Figure 1). The rollers were secured to two mobile overarms driven symmetrically inwards or outwards as required to decrease or increase the area of the working surface, whilst maintaining the PTFE tape taut. The speed of compression or expansion of the surface may be adjusted by means of potentiometer mounted at the front of the central control and readout unit.

Surface pressures were measured by means of a filter paper Wilhelmy plate (approx 1 cm wide) suspended byafinethread from the beam of a Cahn model microforce balance. The microforce balance provided an inputforthe electronics control unit, on which the measured weight is displayed in milligrams. The balance output zero and calibration may be adjusted at the electronics unit.

Surface area measurement is by means of a ten-turn potentiometer connected to the barrier drive motor.

The outputfrom the potentiometer was Oto + 1 V nominal and could be measured by means ofan independent chart recorder.

The trough has a feedback mechansim by which the surface pressure is maintained constant. Avoltage corresponding to the required surface pressure is set on the 'Reference' control of the electronics unit. The barrier may be driven automatically so asto maintain this surface pressure.

The Langmuirtrough also has a dipping head,withinthecabinetand mounted above the trough, onwhich the substrate to be coated can be mounted. The dipping head was adapted to pass the substrate through the water surface at a chosen speed, the extent of travel being limited by microswitches which reverse the direction of travel on closure. The number of immersions into and extractions from the subphase can be preset on the electronics unit.

The microforce balance, area transducer, barrier and dipping head drive motors were initially calibrated to enable determination of surface pressure-area isotherms and Langmuir-Blodgettfilm deposition.

The subphase comprised deionised, distilled water with a pH of about 5.3, at a temperature of 22D + 2 C.

Dilute solutions of a polymer such as PVA in a solvent such as chloroform were spread on the clean water surface. A period of approximately 10 min was allowed for solvent evaporation prior to compressing the surface (and recording the surface pressure isotherm) in orderto ensure good monolayerfilm quality.

Deposition of the Langmuir-Blodgett (L-B) film was conducted continuously at a constant speed on a given substrate. The surface pressure was maintained constant at a preset value, the trough being operated with a fixed barrier speed. At the end of each dipping cycle the contact between the water surface and the substrate was broken, thus providing an opportunity for any entrained waterfilm to drain.

Poly(vinyl acetate) (PVA) was dissolved in chloroform to give a solution of concentration 2.5 mg/cm3. 20 microlitres ofthis solution was applied dropwise to the surface of the pure water sub-phase allowing sufficient time for spreading between each drop. The solvent was then allowed to evaporate, leaving a monolayerfilm of PVA on the water surface. A reference surface pressure of 21.2 m Nm#1was set on the control unit and the feedback mechanism switched on. The barriers then compressed the film until this surface pressure was achieved and then held the surface pressure constant atthis value.

A9 cm square section of a poly(sulphone) ultra-filter (molecular weight cut-off 2000 - 3000) was mounted in the dipping mechanism of the Jpyce-Loebl trough. The speed of the dipping motor was set to 1 .24cm/min and the PS substrate driven through the PVAfilm into the water sub-phase. After8 cm of substrate has been dipped, the direction of dipping was automatically reversed by a micro-switch. The PS substrate then passed through the PVAfilm going from water sub-phase to air. The direction of travel was then automatically reversed when the substrate had leftthe sub-phase completely and was 1 cm above the sub-phase. This cycle was repeated until 21 insertions and 21 extractions of the substrate had been performed.

Deposition of PVA onto the PS substrate was monitored using the movement ofthe barrier required to maintain the surface pressure at a constant value via thefeedback unit. The barrier motion was coupled to a lineartransducerand monitored againsttime using a chart recorder. The area change ofthefilm on the sub-phase during an insertion of extraction ofthe substrate could then be evaluated via a previous calibration. The deposition was characterised using the so-called deposition ratio,T, defined by; change ofarea offilm on sub-phase during a dip geometrical area of substrate The deposition ratios measured for extraction of the substrate were consistently closed to unity indicating good monolayer deposition.The insertion deposition ratios were consistently close to zero indicating that very little or no deposition was taking place. A 21-layer L-B film was produced.

The composite membrane so formed was transferred to a gas permeation cell for the measurement of gas permeabilities.

Measurements of Cm2 and CH4 permeability at a range of upstream pressures were made for both uncoated PS ultra-filters and PS ultra-filters used as substrates in PVA L-B film deposition. The results are shown in Figures 2 and 3. The graphs for coated and uncoated substrates are radically different indicating that the L-B film is acting as a dense ultra-thin barrier to gas permeation. For the coated and uncoated substrates, the carbon dioxide permeabilities were 5.1 x 110-8 and 1.2 x 1 #-# and the methane permeabilities were 0.6 x 10-8 and 1.7 x 1 0-7 respectively. The u nits of permeability are cm3(STP).cm-'.(cm Hg)-l.s-1.The quoted values were obtained by extrapolating the linear portion of the permeability curve to zero pressure.

A measure of the separation performance of a membrane is given by the ideal separation factor, a*, for gas Aand Bdefined by: or, = permeability of pure gas A permeability of pure gas B a* varies with pressure but a convenient reference point is zero pressure. The uncoated substrate had an ideal separation factor (CO2/CH4) of 0.7, at zero pressure indicating that very little separation of Cm2 from CH4will take place. The substrate on which a 21-layer PVAfilm had been deposited had a similarly defined separation factor of 8.5, indicating that the permeate gas will be enriched in CO2. Deposition of the PVA L-B film thus affords gas separation properties to a substrate which previously exhibited a very low separation factor.

Claims (1)

1. Process for fabricating a composite separation medium comprising the steps of (a) dissolving a polymer in a suitable solvent, (b) spreading the solution on the surface of an aqueous liquid, (c) allowing the solvent to evaporate to leave a film of polymer on the water surface, (d) passing a porous substrate oriented substantially perpendicularlyto the film through the film through the air/water interface, and (e) passing the substrate through the film in the reverse direction.

2. Process according to claim 1 in whicht the polymer comprises hydrophilic groups attached to a hydrophobic backbone or chain.

3. Process according to claim 1 or claim 2 in which the polymer is a cellulose polymer, a polyimide, an acrylic polymer, a vinyl polymer, or a polymer mixture.

4. Process according the claim 2 in which the cellulosic polymer is ethyl cellulose or cellulose acetate.

5. Process according to any of claims 1 to 3 in which the solvent in chloroform, benzene, toluene, xylene, tetrahydrofuran, acetone, dioxane, a cyclic ether, N-methyl pyrrolidone orformamides.

7. Process according to any of the preceding claims in which steps (d) and (e) are repeated so as to build up successive layers of polymer on the porous substrate.

8. Process according to any of the preceding claims in which the porous substrate is a polymer sheet ora hollow fibre.

9. Process according to any of the preceding claims in which the porous substrate has a pore size which is generally not greater than 0.5 microns.

10. Process according to any of the preceding claims in which the speed of passing the porous substrate throughthefilm is from 0.1 to 60 cm/min.

11. Process according to claim 10 in which the speed of passing the porous substrate through the film is from 1 to 5 cm/min.

12. Process forfabricating a composite separation medium as hereinbefore described and with reference to the accompanying drawings.

13. Composite separation media whenever fabricated by a process according to any of claims 1 to 12.