Patent Application
20150165389
Not applicable,
Since the 1960's there have been numerous developments in the use of Cellulose Acetate (CA), Cellulose Triacetate (CTA), and other cellulose polymer derivatives to make membranes that are suitable for the Reverse Osmosis (RO) desalination of sea/saline waters [I&EC Process Design and Development, Vol. 4, No. 2, April 1965, pp. 207-212; IBID, Vol. 6, No. 1, March 1967, pp. 23-32; Polymer Letters, Vol. 11, pp. 603-608 (1973); Desalination, 61 (1987) pp. 211-235; HWAQHAK KONGHAK Vol. 28, No. 5, October, (1990), pp. 602-611; U.S. Pat. No. 3,673,084; U.S. Pat. No. 3,807,571 and; U.S. Pat. No. 3,878,276].
In recent years there have also been modifications of CTA/CA polymers for use in Forward Osmosis (FO) membranes as disclosed in U.S. Pat. No. 7,445,712.
The current disclosure is a composition for forming reverse osmosis (RO), forward osmosis (FO), or nano or micro filtration (NF) membranes from a stable liquid blend of two of the following polymers: an oxygen polymer, a nitrogen polymer, and a sulfur polymer, where each polymer in a blend have matched solubility parameters; provided, that a nitrogen polymer when incompatible can be in the form of a powder; where the weight ratio of polymers in each blend can range from 1:99 to 99:1; where each polymer optionally can be halogenated; where any polymer can be dispersed in a solvent for forming a blend.
By “oxygen polymer”, we mean a polymer having as its main structure or repeating units, —CHO groups. By nitrogen polymer, we mean a nitrogen backbone polymer (—NHO repeating units) typified by special nylons, amines, amides, polyurethanes, and the like. By sulfur polymer, we mean a sulfur backbone polymer (—SHO repeating units) typified by polysulfides, polysulfones, polyethersulfones, and the like. Note: if the nitrogen polymer is insoluble, it may be incorporated as a powder into the oxygen polymer or sulfur polymer. Such polymers typically will be provided in a solvent or blend of solvents.
The method of forming such RO, FO, or NF membranes starts with casting a wet film or extruding a hollow fiber of a membrane composition comprising a stable liquid blend of two of the following polymers: oxygen polymer, a nitrogen polymer, and a sulfur polymer, where each polymer in a blend have matched solubility parameters; provided, that a nitrogen polymer can be in the form of a powder; where the weight ratio of polymers in each blend can range from 1:99 to 99:1; where each polymer optionally can be halogenated; where any polymer can be dispersed in a solvent for forming a blend.
Next, the solvent is evaporated from said the film or extruded hollow fiber, where low solvent evaporation times (e.g., seconds to a few minutes) produce an ultrafiltration or nanofiltration morphologies, medium solvent evaporation times (e.g., 3-5 minutes) produce FO morphology, and long evaporation times (e.g., 5-30 minutes) produce reverse osmosis morphology.
The evaporated cast film or extruded hollow fiber is water quenched, where the quench water optionally can contain one or more of inorganic or organic microparticles or nanoparticles; nonionic, anionic, cationic, zwitterionic polymers; or amino acids
The quenched cast film or extruded hollow fiber then is annealed and optionally microembossed.
The foregoing processing steps are represented in
For a fuller understanding of the nature and advantages of the present media and process, reference should be had to the following detailed description taken in connection with the accompanying drawings, in which:
The drawings will be described in further detail below.
The disclosed water treatment membrane platform technology is based on unique combinations of commercially available polymers and specialty materials to produce stable and efficient membranes for forward osmosis (FO), nanofiltration (NF), microfiltration (MF), and reverse osmosis (RO) applications. The factors considered in designing membranes with the desired characteristics are Hansen solubility parameters of the polymer blend, zeta potential and surface energies of the membrane, surface roughness as well as its hydrophilic/hydrophobic properties. Creating a balance of all these variables is difficult to achieve with a single or two component blends, thus the disclosed technology incorporates other polymer or inorganic materials and nanomaterials, as well as novel processing techniques for enhanced control of flux and salt rejection to design the final membrane system. These additives, as well as bulk and surface modification techniques, provide enhanced antifouling and chlorine resistance properties.
The first order design of a polymer blend starts with understanding the relationships between the Hansen solubility parameters associated with the different classes and structures that are being considered for the system. For example, cellulose acetate (CA) has the following dispersion, polar, hydrogen bonding solubility parameter values (MPa)1/2: δd=18.6; δp=12.7; δh, =11, respectively. In order to create a compatible polymer with CA, the solubility parameters for both the CA and the other polymer must be within 3 units of each other, for each property. If the individual solubility parameters of both polymers are significantly different (larger or smaller) then an unstable system and an incompatible blend would be produced. For example, a CA/Nylon/Polyamide membrane is a system where the individual solubility components of the CA closely match the solubility components of the P1 material and are compatible. A dynamic mechanical analysis of this blend showed a single glass transition peak (Tg) indicating that the two polymers are miscible. Incompatible or polymer dispersions of two different polymers with large differences in their solubility parameters can exhibit two Tg peaks indicating separate phases in the system.
The hydrophilic or hydrophobic nature of the polymer can be determined from its water sensitivity or % oxygen content [McGinniss Equation χOxygen; U.S. Pat. No. 4,566,906] contained in the backbone of the polymer structure. For example polyvinyl alcohol and low alkyl functional acrylics absorb water, or are sensitive to water, while non-oxygenated polymers, like polyolefins and polystyrenes, are significantly less sensitive to water. Therefore, the disclosed NF, MF, FO, and RO membranes are comprised of unique combinations of water-sensitive, oxygen-containing polymers (cellulosics, acrylics, polyesters); water-sensitive, nitrogen- or sulfur-containing polymers (nylons, sulfones); water soluble/dispersible anion and cation polymers; water soluble/dispersible nonionic/zwitterionic polymers; polymers with low sensitivities to water (elastomers, aromatic polymers); and such crosslinking polymer materials as epoxies, polyurethanes/amides, and melamine resins. The disclosed compositions of matter are different in that not all nitrogen containing polymers are compatible with oxygen or sulfur containing polymer structures and visa versa. In Table 1 are selected solubility compatibility listings for several examples of the types/classes of polymers suitable for this invention.
After downselection of polymer blends, based on the modeling output, membranes were produced from various combinations of polymers based on cellulose acetate, cellulose triacetate, polyamides, polysulfones, and other polymers/additives. Production variables included solvent evaporation time, water bath quench/annealing time, and temperature.
The first step in making the membrane is to take the membrane polymer/solvent solution and apply it to a glass or other composition plate and draw down a wet film with a draw down bar set at a thickness of, say, for example, 10 mils.
In a polymer/solvent cast flat sheet membrane or extruded hollow fiber membrane the solvent evaporation times are critical for creating the initial morphology of the membrane before it is water quenched and annealed into its final structure. For the polymers/solvents of this disclosure, solvent evaporation (static or forced air/room temperature or elevated to 100° C.) times less than a minute, produce ultrafiltration and nanofiltration membranes, while solvent evaporation times of 15-20 seconds or between 1 to 3 minutes and 3 to 30 minutes produce morphologies suitable for FO and RO membrane technologies respectively. Similar solvent evaporation times also would apply to hollow fibers after they are extruded. It also is possible to create membrane particles by spraying the polymer/solvent solutions form the initial desired morphologies and then quench them in water to maintain their porosity, so as to be used as a novel nano or course (millimeter) filtration media as a pretreatment process for FO or RO membrane processes.
After the initial morphology is created, the flat sheets or hollow fibers or particles are quenched (dipped or exposed to a water spray) in water (5 minutes at ice water or room temperature or less at elevated temperatures) to solidify the structure followed by an additional heat treatment (wet or dry) to anneal the system and lock in the final structure.
The thickness of the flat sheet membranes can have an active 0.1 to 0.2 micron size dense layer on a much thicker (1-30 mils) substrate layer, while hollow fibers can have outside diameters of 85 to 2000 microns and inside diameters between 42 to 200 microns.
The water bath quench process can be run at low temperatures (ice water), room temperature or elevated temperatures (50° C. or less) while the annealing step is usually run wet at 50° C.-80° C. for 5 to 10 minutes and cosolvents, such as, for example, methanol, also can be added to help the coagulation process. The quench water optionally can contain one or more of inorganic or organic microparticles or nanoparticles; nonionic, anionic, cationic, zwitterionic polymers; or amino acids. After downselection of polymer blends, based on the modeling output, membranes were produced from various combinations of polymers based on cellulose acetate, cellulose triacetate, polyamides, and polysulfones. Production variables included solvent evaporation time, water bath quench/annealing time, and temperature.
The quenched cast film or extruded hollow fiber then is annealed in a separate water bath at a temperature ranging between about 50° to 80° C. for about 5 to 10 minutes
The additives that can be incorporated into the membrane polymer formulation casting or extrusion solutions and water quench baths are as follows:
The additives listed above can be either added to the membrane polymer solutions as homogeneous mixtures or dispersions before casting or extruding or they can be added (solubilized or dispersed) into the water quench bath that incorporated the additives into the membrane during the coagulation of the polymer films to form the final membrane structure. The additives also can remain thermoplastic or converted into thermosetting structures if so desired.
Prepare FO or RO membrane solutions by first dissolving CA or CTA or other water-sensitive polymers in dioxane. Add acetone and mix to dissolve, methanol, and lastly monofunctional (lactic acid) or multifunctional acids. For hybrid solutions containing nylon or other water-sensitive nitrogen or sulfur-sensitive polymers or polymers with low sensitivity to water, add them (nylon) prior to the acids (lactic acid).
Draw down this solution onto a glass plate using a Gardener blade set at 10 mil thickness followed by immersion in room temperature tap water (note—another part of this disclosure is to include water soluble anionic, cationic or nonionic polymers in the water quench solution for improving the internal concentration polarization response of the membrane and improve its antifouling properties) for up to 5 minutes or until the membrane film separates from the glass and incorporates the water soluble membrane property enhancement polymers if so desired. Rinse with tap water and store in a zip lock bag containing 100 ml tap water.
RO composite membranes are assembled on a polysulfone/nonwoven fabric using a dilute poly(vinyl alcohol) adhesive layer to hold an FO membrane to the polysulfone support. Filter paper also can be used directly without an adhesive binder to support the membrane in the RO test cell. This composite RO membrane is then dried at 55° C. overnight.
Pressure-driven systems, like reverse osmosis (RO), have been studied a great deal because of their effectiveness in water purification. With the expanding need for clean water that can be produced economically, interest in forward osmosis systems has grown. Forward osmosis (FO) membranes operate similarly to RO membranes in that both membranes allow water to move across a semi-permeable barrier while inhibiting the flow of solutes. Unlike pressure-driven RO systems, FO systems utilize the osmotic pressure differential between the draw solution and feed solution, which is naturally occurring.
In this testing, deionized (DI) water and salt (NaCl) water will be used in an FO system to determine membrane efficiencies.
a. Ring stands and clamps
b. FO apparatus (graduated DI/NaCl cells)
c. Strap wrench
d. Motor operated stirring rods
e. DI water
f. NaCl solution
Reverse Osmosis (RO) membranes operate by using high pressure pumps to reverse naturally occurring osmotic pressure to remove contaminants in a permeate solution from a filtrate. In this testing, salt will be removed from a saltwater mixture (filtrate). Experimentation will be benchmarking commercial membranes, as well as testing the disclosed membranes, in the fabricated RO system with the intention of developing a superior RO membrane.
a. Syringe with Tygon® tubing
b. Glass vial (×2)
c. Vial stand
d. Salinity meter
The CDH45 from Omega is a portable hand held digital salinity meter, which displays the salinity of water in percentage (%) along with temperature. The CDH45 is designed for low-concentration salinity measurement. This is based on a principal that salt water conducts electricity much more easily than pure water and hence the salinity content of water can be calculated based on the electrical conductivity measurement. It has a probe (a pair of electrodes, which measures the electrical conductivity of water at a given temperature. Then it uses an in-built conversion table (factor) in order to convert the conductivity data into salinity data in % mass of the dissolved solid. It also automatically uses a temperature compensation factor, which accounts for the changes in conductivity with temperature.
Temperature Compensation: −5° to 60° C. (23° to 140° F.), automatic
0 to 0.9% (±0.1)
1.0 to 1.9% (±0.2)
2.0 to 2.9% (±0.3)
3.0 to 4.9% (±0.5)
5.0 to 7.9% (±1.0)
8.0 to 10.0% (±1.5, depending on measuring technique)
Calibration information is not provided in the instruction manual. Indirect method of calibration was used to calibrate the salinity meter. It was done by measuring a series of known concentration (salinity) NaCl solutions (in DI water) with the salinity meter and recording the data and its deviation (if any). The calibration data is tabulated below (Table 2):
The FO membrane formulation compositions for Example 1 are described in Table 3 and their FO testing results are shown in Table 4.
The FO membrane formulation compositions for Example 2 are shown in Table 5 and their FO testing results are shown in Table 6.
The starting column height for DI water is 11.5 inches. From the graph (
The disclosed membranes were 1 and 5 percent Nylon to 99 to 95 percent CA (Battelle Membranes) (99.9 to 100 percent salt rejection), while the unhatched bar was an HTI FO membrane and hatched RO bar was a DOW XL RO membrane. Both commercial membranes had no signs of salt detection in the DI waterside of the membrane.
The membrane compositions and FO testing results are shown in a combined Tables 9 and 10.
For example, prior art samples 1 through 5 have undesirable salt rejection or salinity readings in the DI water side of the membrane of 90 to 95 percent and total DI water transfer volumes of 2 to 8. Control samples 6 to 12 all have salt rejection values of at least 99% volume changes of between 2 and 6. Samples 6 through 11 of this disclosure (combinations of water-sensitive Nylons with water-sensitive CA/CTA polymers) have no signs of salt transfer to the DI waterside of the cell and DI water volume changes of 2 to 5. Sample 12 of this disclosure (combination of a water-sensitive Nylon with a water-sensitive nonionic polyvinyl acetate) had no signs of salt transfer to the DI water and a DI volume change of 5. Sample 13 is the same composition as 6, but was laminated with a polyester mesh for strength. The salt back transfer again was 0 or at least 99+% percent salt rejection but the change in DI volume was restricted to 0.5-1 from 2.
These results demonstrate the ability to significantly control the salt transfer properties of the blends of this disclosure over the control prior art membranes under the same experimental conditions.
Table 11 shows the results of RO testing for several commercial RO membranes with membranes of this disclosure.
Other Considerations—As part of this disclosure it was discovered that sulfonated polymers can be reacted with amines to make amides which can be used to control the flux, water wettability, salt rejection and antifouling properties of a membrane.
Historically, cellulose-based membranes were used for water desalination. Table 12 compares the performance of FO membranes (CTA and CTA/Polyamide or Nylon, where the Polyamide is ELVAMIDE 8061 DuPont) is a secondary polymer additive) with commercial FO membranes (HTI-NW and HTI-SS). The flux and salt rejection are in a comparable range, under the same testing conditions. It is noteworthy that the addition of a Polyamide polymer to the CTA membrane considerably increases the flux, without affecting the salt rejection. The performance of NF membranes can be enhanced by incorporation of a secondary polymer (Thermosetting Epoxy/Amine) system, where the flux increased almost four times, with a small increase in a salt rejection.
These particular water transport (flux) and salt rejection studies were carried out on a dynamic FO test apparatus where the membrane flow cell holder was a modified RO test cell which was fitted with two low flow rate peristaltic pump systems. Both the DI water flow rates across the membrane face and the draw solution (3.5 to 6% salt water) flow rates on the other side of the membrane could be adjusted to have equal flows across each side of the membrane or different flows across the two membrane sides. Table 13 describes the salt rejection and flux results for membranes run under Battelle's relatively low equal flow rates on both sides of the membrane and membranes run at very high equal flow rates on each side of the membranes at the Colorado School of Mines (CSM) test facility (Golden Colo.)
The first set of commercial RO membranes (Hydranautics 84200.SWC5J) were obtained from Nitto Denko Corp. and exposed to a commercial DuPont grout Sealer that contained a hydrophobic fluoropolymer (fluorinated acrylic copolymer) at 1% solids in propylene glycol monobutyl ether. The membranes were soaked with the grout sealer for 120 seconds at room temperature and air knifed to remove the excess solvent. These samples were then allowed to air-dry overnight to finalize the membrane modification process.
A second set of Nitto Denko membranes were coated with Olympic Water Guard waterproofing (hydrophobic) sealant (12% solids) [water acrylic resin (25035-69-2); polysiloxane (71750-80-6) and ethylene glycol (107-21-1)] which was diluted with DI water to form 1% and 5% solutions. These solutions were applied to the membranes and dried in an identical manner as the DuPont grout modified samples.
All three duplicate sets of RO membrane coated samples (DuPont grout—samples A, B; 1% and 5% Olympic Water Guard—samples B, C and D, E respectively) and their untreated controls (samples F and G) were sent to Battelle's Florida Marine Research Center (FMRC) where they were placed on holding racks and lowered into the Halifax River for 14 days of exposure testing to the very active marine fouling environment.
The 1% grout (samples A and B) and the Olympic coatings (1% and 5%) samples B, C, D, and E showed little or no signs of fouling (attachment of bioorganisms hydroids) while the untreated control samples F and G were completely covered with biological growth organisms.
Modification of the grout and the Olympic coatings to contain nonionic and cationic or zwitterionic polymers were also investigated in this example. Various amounts of these types of polymers were added to either the grout or Olympic coatings and the results of these studies on the Nitto Denko membranes are shown in Table 14.
The Florida immersion tests were run for 28 days and the order of least fouling to highest fouling was as follows:
Systems 9 and 11 (excellent—no sign of bioorganisms/films); Systems 5, 6, 7, 8, 10 (very good—1 to 3% growth coverage); Systems 1, 2, 3, 4 (good—3-5% growth on sample surface); system 12 (untreated control) unsatisfactory—30 to 100% of the surface was covered with biofilms or organisms that could not be removed with a water wash rinse.
A cellulose triacetate (CTA)/nylon/polyamide type polymer (ELVAMIDE 8061) blend in dioxane was prepared as previously described in Example 7 of this application. The normal way of processing these types of polymer blends into membranes is to allow the solvent to evaporate over a specified time period (short times—seconds to a few minutes for ultra and nanofiltration membranes); longer time periods (3 to 5 minutes) for FO membranes and even longer times (5 to 30 minutes) for RO membranes followed by immediate quenching in a water bath to lock in their morphological features that control the final flux and salt rejection properties of the membrane.
In this particular example different polymer materials were added at a 5% level to the water quench bath before immersing the polymer blend into it to create the final structure. The addition of different polymers in the quench bath interacted in a unique manner with the CA/nylon blend and created new composite polymer structures that could not be obtained by any other method.
These new composite polymer membranes had different flux capacities and excellent fouling resistance capabilities than their unmodified control membrane counterparts.
The results of these new polymer modified membranes and their transport properties are shown in Table 15.
One of the major problems in treating or coating preformed membrane structures for flux enhancement, salt rejection or fouling resistance is to not block the pore structures of the membrane during the coating or treatment process. Application of a high solids coating formulation can fill the pores of the membrane resulting in a significant decrease in the ability of the membrane to transport fluids through its structure (decrease in flux). We have found that a vacuum or forced air assist coating process that drives the coating and air through the pores of the membrane is very beneficial in just coating the membrane surfaces without filling or blocking the pores of the membrane. This special coating process leaves the desired treatment systems on the surface of the membrane without blocking/filling the pores and significantly decreasing the original flux capacity of the membrane.
Results for different coating processes to create fouling resistant hydrophobic surfaces on filtration membrane structures while maintaining their flux properties are shown in Table 16.
The primary features of this disclosure are to produce micro and nanostructure patterns on continuous polymer films or membrane surfaces using a roll-to-roll nanoimprint lithographic [R2RNIL] process. The methods used to create this technology are described as follows:
Silica on Silicon Wafers
13 test patterns, see diagram on right for label
Target Etch Depths vary according to pattern
A series of commercial membranes were modified by either a coating or chemical treatment and the resultant change in their surface energies (water drop contact angles) and water transport properties are shown in Table 17.
The salt rejection and flux experiments were carried out using the dynamic FO test equipment with 0.3 LPM flow rates on both sides of the membrane and the VWR conductivity meter.
In another dynamic FO experiment where the flow rate was higher on the DI water side (1 LPM) than the salt draw solution side of the membrane (0.3 LPM) the Sepro PS35 untreated control had a flux of 5 GFD while the 1% Epoxy modified Sepro PS35 had a flux rate of 18 GFD. Both systems had 99% salt rejection. Similar results were observed for Sepro PAN 400 treated with 1% Epoxy where the flux was −5 GFD for the untreated control sample but 4.12 GFD for the treated sample.
Combinations of 1% Epoxy and 1% of 0.13 um size Nylon 6 powder (KOBO TR-1, Toray, 3474 S. Clinton Ave., S. Plainfield, N.J. 07090) showed 98% salt rejection and 9.23 GFD flux values.
The Epoxy system used in these experiments was a 50/50 blend of Momentive's water dispersions EPI-REZ Resin 3510-W-60 and EPIKURE Curing Agent 6870-W-53 (EXEL LOGISTICS, Houston Tex.) which were applies at 1% solids coatings in water to the different membrane surfaces and allowed to air dry at room temperature for 2 days before running the dynamic tests.
The 1% Polyacid coating was 1 gm Polyacrylic acid/1800 molecular weight (9003-01-4) (Aldrich), 0.45 gms Cymel 1172 melamine crosslinking oligomer (Cytec), and 0.13 gms para-toluenesulfonic acid catalyst (6192-52-5) (Aldrich) in 98.42 gms water. The membranes were soaked in this solution for 120 seconds and oven cured at 125 C for 1 hour before testing.
There are several physical and chemical parameters that influence how fast a biological organism will attach to a polymer or membrane surface. The wetability of a surface (hydrophobic or hydrophilic) is one of the more critical parameters, as is the surface roughness of the polymer/membrane as well.
It is well known that very hydrophobic smooth surfaces, like fluorocarbon and silicone polymers, have very high water drop contact angles and can resist the attack by bioorganisms for very long extended time periods in a marine environment. If, however, these smooth surfaces become roughened or contaminated with dirt then these situations can lower the contact angles and lead to attachment and growth of the bioorganism.
Another set of parameters comes from the chemical nature of the polymer or formulation that makes it hydrophilic or ionic in nature (low water drop contact angles) or if there are chemical reactants like amino acids, dopamine derivatives, quaternary ammonium salts that are actually biocides which decrease the ability of the bioorganism to form a film and grow on the surface of the polymer or membrane exposed to a water source with high bioactivity.
The model proposed that best describes this disclosure is based on the differences between a pure hydrophobic surface, a pure hydrophilic surface, and a surface that contains some form of biocide activity. All of these model surfaces (hydrophobic, hydrophilic, ionic, or contains a biocide), assuming equal surface roughness characteristics, start off with no signs of fouling; but at some point in time an early biofilm growth induction period occurs which continues for a certain period of time, after which strong fouling of the substrate is observed.
With hydrophobic surfaces having contact angles between 90° and 150° and surface roughness values between 1.25 and 4.5 μm arithmetic mean height after static exposure to a marine fouling community start to show signs of decreasing their hydrophobic nature and lowering their contact angle values to between 30° to 70° at which point in time the first signs of biofilm growth appears (induction period). Extended periods of static exposure time for very hydrophobic surfaces but now with contact angels around 50° results in the start of extensive fouling processes.
A similar analogy can be formulated for hydrophilic surfaces with low contact angles of 5°, but over time (induction period) increase to 30° to 70° where the initial biofilms start to grow and continue until strong fouling is observed.
Ionic (nonionic, cationic, zwitterionic, anionic) and systems with biocides combined with either hydrophobic or hydrophilic polymers undergo the same decrease or increase in their contact angle values into the biofilm growth induction period region of 30° to 70° contact angle values for these surfaces but continue to resist attack of the bioorganisms for longer time periods because of the inherent biocide nature of the system.
Table 18 shows the results for a series of commercial RO membrane (Nito Denko SWC5J) structures that were treated with hydrophobic coatings, exposed to 14 days in a marine environment as an example of a postulated biofilm growth model.
1.5 (L)
1.5 (L)
64 (H)
63 (H)
All of the coated samples in Table 18 resulted in considerably less fouling than the uncoated controls. As long as the contact angles were above 90° did not matter what the surface roughness values (high or low) were for these systems as the hydrophobic nature of the surfaces controlled the initial biofilm growth process.
The surface roughness values for the 1% Olympic and DuPont grout sealer coatings did influence the lowering of the contact angles from the coated A side to the B coated sides of these membranes. The 5% Olympic coated membranes had a trend more similar to the uncoated control membranes where the A sides (coated and uncoated) had lower surface roughness values and lower contact angle values while the B side coated and uncoated sides had both high surface roughness values and high contact angle values.
Analysis of the data in Table 8 is described as follows:
Relative Fouling Resistance Ranking (A Side Coated Samples)=8.41E3 (Contact Angle)+0.13(% Coating Applied)−0.52 (Surface Roughness)+0.63 R2=0.99
Relative Fouling Resistance Ranking (B Side Coated Samples)=2.16E2 (Contact Angle)+9.04E2 (% Coating Applied)+1.21 (Surface Roughness)−7.42 R2=0.99
After about a month exposure time the hydrophobic coatings on the membranes started to show signs of biofilm formation resulting from the lowering of their contact angle values.
In another set of fouling experiments, a 98% Cellulose Acetate/2% Nylon (ELVAMIDE 8061) solution in dioxane was divided into 6 samples as described and tested for fouling resistance in Table 19.
These results in Table 19 show the improved advantage of incorporation of an additive in the water quench bath as opposed to putting the additives in the casting solution directly.
A general model that helps us understand the ability of different additives to enhance the fouling resistance of a membrane treatment or formulation/additive modification is shown in
The (y) response for the linear portions of the curves generated in
At t=between 5 and 10 days—equations 1 and 2 (pure hydrophobic/hydrophilic systems)—the start of the biofilm growth takes place and the linear portion of the curve—now defined as the biofilm growth induction period—progresses for a given membrane system up to 20 days. After 20 days the start of the hard fouling processes can occur.
At t=between 5 and 20 days—equations 3 and 4 (combinations of the hydrophobic or hydrophilic membrane systems but now combined with critical concentrations of ionic/nonionic/biocide additive material)—the start of the biofilm growth (induction period) gets extended out to 30 days after which the possibility of hard fouling is more likely to occur.
Note! For all membrane systems, we define that any membrane surface or structure that either starts out with a water drop contact angle of around 30° to 70° (for this model we chose 50°) or starts with a high contact angle (100° or higher/hydrophobic surface) or a hydrophilic/ionic/biocide (5° to 20°) and progresses downward or upward to the critical point of 50° will see biofilm growth during this progression of time. In the pure hydrophobic membrane system the contact angle maintained its 100° contact angle until 10 days into the test at which point the biofilms started to grow decreasing its value to 50° approximately linearly up to a 20 day time period (equation 1). The combination of the ionic or nonionic materials in the hydrophobic membrane structure started out at a somewhat lower hydrophobic surface due to the choice of materials (80° and maintained this value up to 20 days exposure after which the biofilm started to grow and reached the critical point (contact angle of 50°) after which hard fouling can occur at 30 days (equation 3). Both hydrophilic systems started to progress towards biofilm formation after 5 days (the pure hydrophilic membrane had starting contact angle values of while the ionic/nonionic blend had contact angle values of about 20°), but the pure hydrophilic system lasted 20 days before reaching the critical value of 50° while the blend of the ionic or nonionic materials with the hydrophilic membrane structure to 30 days to reach this value (equations 2 and 4 respectively).
These experimentally derived results and equations are designated as relatively early examples of how membranes can be made with a general set of materials that can be manipulated to change the ability of a biofilm to grow and attach to the surfaces of different membrane systems. We now can use the linear equations that define the biofilm growth induction period from the experimental data of
The use of nanosize TiO2 particles or micron size iron oxide/hydroxide (made as follows: FeSO4+2 OH−→Fe(OH)2+SO42−) in the disclosed CTA/Nylon formulations can be used to control the flux and salt rejection of the membrane's transport properties, as reported in Table 20.
It as been have discovered that if 10% by weight of an amino acid, such as, for example, cysteine (Aldrich), is added in the water quench bath used to prepare the CTA/Nylon membranes of this disclosure, then the amino acid becomes incorporated into the membrane structure and can be used to selectively remove arsenic (100 ppb sodium arsenide) from contaminated water. A membrane made using this additive in the quench bath process selectively decreased the 100 ppb sodium arsenide to 50 ppb over a 48-hour time period when exposed to the contaminated water.
As an alternative as shown in
The skilled artisan likely can evolve other advantages in using the pre-treating filter.
The size of the particles, density of the packed pre-treating filter; and other factors would determine the pressure drop and flow rate of the pre-treating filter. Much of the load would be taken from the expensive membrane and transferred to a less expensive (and easier to regenerate) pre-treating filter.
Referring to
While the device, compositions, and process have been described with reference to various embodiments, those skilled in the art will understand that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope and essence of the disclosure. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the essential scope thereof. Therefore, it is intended that the disclosure not be limited to the particular embodiments disclosed, but that the disclosure will include all embodiments falling within the scope of the appended claims. In this application all units are in the metric system and all amounts and percentages are by weight, unless otherwise expressly indicated. Also, all citations referred herein are expressly incorporated herein by reference.
This application claims benefit of U.S. provisional application 61/498,031 filed on Jun. 17, 2011.
