In situ formation of biocidal metal nanoparticles on thin-film composite reverse osmosis membranes for biofouling mitigation

Abstract

The present invention relates to a method including reacting a solution of a salt of a biocidal metal with an active layer of water purification membrane, discarding the biocidal metal salt solution such that a thin layer of the biocidal metal salt solution remains on the membrane surface, reacting a reducing agent solution with the active layer of the membrane and the thin layer of the biocidal metal salt solution thereby forming a biocidal metal nanoparticle-modified membrane, removing the reducing agent solution, and rinsing the biocidal metal nanoparticle-modified membrane.

Claims

1. A method for in situ formation of biocidal metal nanoparticles on a reverse osmosis membrane, said method comprising: a) contacting a solution of a salt of a biocidal metal with an active layer of the membrane; b) discarding a portion of the biocidal metal salt solution such that a thin layer of the biocidal metal salt solution remains on the membrane surface; c) reacting a reducing agent solution with the active layer of the membrane and the thin layer of the biocidal metal salt solution thereby forming a biocidal metal nanoparticle-modified membrane; d) removing the reducing agent solution; and e) rinsing the biocidal metal nanoparticle-modified membrane.

2. The method according to claim 1, wherein the biocidal metal is at least one selected from the group consisting of copper, gold, silver, selenium and zinc.

3. The method according to claim 1, wherein the reducing agent is one or more selected from the group consisting of borohydrides, sulfites, phosphites, oxalic acid, formic acid, hydrazine and ascorbic acid.

4. The method according to claim 1, wherein the biocidal metal salt solution has a concentration of 0.1 to 100 mM.

5. The method according to claim 1, wherein the duration of the reacting of a) is in the range of from 0.5 minutes to 30 minutes.

6. The method according to claim 1, wherein the reducing agent solution has a concentration of 0.1 to 200 mM.

7. The method according to claim 1, wherein the duration of the reacting of c) is in the range of from 0.5 minutes to 60 minutes.

8. The method according to claim 1, wherein the rinsing of e) is performed with deionized water.

9. The method according to claim 1, wherein the duration of the rinsing of e) is in the range of from 1 second to 10 minutes.

10. The method according to claim 1, wherein each of a) through e) is performed at ambient conditions.

11. The method according to claim 1, wherein the biocidal metal is silver, the biocidal metal salt solution is AgNO.sub.3, and the biocidal metal nanoparticle-modified membrane has a silver loading with a linear relationship (R.sup.2=0.99) to the concentration of the AgNO.sub.3 solution, wherein the AgNO.sub.3 solution has a concentration of 0.1 to 100 mM.

12. The method according to claim 1, wherein the biocidal metal is silver, and the water permeability of the silver nanoparticle-modified membrane is reduced by no more than 17% as compared to the unmodified membrane before performing a) through e).

13. The method according to claim 1, wherein the biocidal metal is silver, and the silver nanoparticle-modified membrane and the unmodified membrane before performing a) through e) have about the same salt rejection and salt permeability coefficient.

14. The method according to claim 1, wherein the biocidal metal is copper, and the biocidal metal salt solution is CuSO.sub.4.

15. The method according to claim 1, wherein the biocidal metal is copper, and the copper nanoparticle-modified membrane and the unmodified membrane before performing a) through e) have about the same salt rejection and salt permeability coefficient.

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a schematic diagram of in situ formation of Ag-NPs on a TFC RO membrane: pristine RO membrane (A) is covered with AgNO.sub.3 solution (yellow; 1-5 mM) for 10 minutes (B), then, the AgNO.sub.3 solution is discarded leaving a thin layer of the AgNO.sub.3 solution on the surface (C), next, the membrane is recoated with NaBH.sub.4 solution (1-5 mM) for 5 minutes (D) to form Ag-NPs that uniformly cover the membrane surface (E).

(2) FIG. 2 illustrates the membrane surface color after in situ formation of Ag-NPs: pristine (A) and in situ Ag-NPs modified membranes using concentrations of 1:1 (B), 2:2 (C) and 5:5 (D) AgNO.sub.3:NaBH.sub.4 (mM), respectively.

(3) FIG. 3 is SEM micrographs of (A, C) pristine and (B, D) in situ Ag-NP modified active layer of TFC membrane at different magnifications as indicated (solutions of 5 mM AgNO.sub.3 and 5 mM NaBH.sub.4 (5:5) were used during the in situ formation reaction).

(4) FIG. 4(A) is an XPS spectra of pristine (red) and functionalized TFC membrane with in situ formed Ag-NPs with different concentrations of Ag and BH.sub.4 (the curves of 5:5 (blue), 2:2 (pink) and 1:1 (green) represent membranes that were in situ modified with different Ag:BH.sub.4 solution concentrations (in mM); the peaks at 365 eV are related to the presence of silver). FIGS. 4(B) and 4(C) are lower and higher resolution, respectively, of TEM micrographs of the cross section of TFC-RO membrane polyamide selective layer after in situ formation of Ag-NPs.

(5) FIG. 5 illustrates an elemental surface analysis by XPS of pristine and 5:5 (Ag:BH.sub.4 in mM) in situ Ag-NPs modified membrane.

(6) FIG. 6 is an XPS spectra of in situ Ag-NPs modified membrane (5:5) with no sonication (A) and after 5 mM sonication in DI water (B) (the observed silver peaks demonstrate that the Ag-NPs remained on the membrane surface after sonication).

(7) FIG. 7(A) illustrates water permeability coefficient, A, and salt permeability coefficient, B, of pristine and in situ Ag-NPs modified membranes prepared with different Ag:BH.sub.4 solution concentrations (in mM; 1:1 is 1 mM Ag and 1 mM BH.sub.4, etc.) (experimental conditions: pressure of 27.5 bar (400 psi), crossflow velocity of 21.4 cm/s, temperature of 25.0?0.5? C., and 50 mM NaCl). FIG. 7(B) illustrates residual silver loading on in situ Ag-NPs modified membrane (5:5) after dissolution for several days in 10 mL of 5 mM NaHCO.sub.3 solution (pH 8.3).

(8) FIG. 8(A) illustrates membrane surface charge represented by zeta potential; and FIG. 8(B) illustrates membrane surface hydrophilicity represented by contact angle measurements (notations 1:1, 2:2 and 5:5 mean Ag:BH.sub.4 concentrations (in mM) during the in situ Ag-NPs formation reaction).

(9) FIG. 9(A) illustrates silver loading on the membrane surface after in situ formation reaction with different Ag:BH.sub.4 concentrations (1:1, 2:2, and 5:5 in mM, Pris. means pristine membrane). FIGS. 9(B), 9(C) and 9(D) illustrate silver loading, elemental surface coverage, and membrane intrinsic transport properties, respectively (water permeabilityA, salt permeability coefficientB), for in situ Ag-NPs modified membranes when silver concentration was constant (2 mM) and different BH.sub.4 concentrations were applied (0.5, 2, and 5 mM).

(10) FIG. 10 illustrates the number of attached live bacteria colonies (CFU) on pristine (grey) and in situ Ag-NPs modified membranes (green) for gram negative (E. coli and P. aeruginosa) and gram positive (S. aureus) bacteria (experimental conditions: pristine and in situ modified (2:2 Ag:BH.sub.4) membranes were contacted and incubated with isotonic solution (0.15M NaCl; 20 mM NaHCO.sub.3 as buffer, pH 8.2) with bacteria (OD.sub.600: 0.15?0.01) for 5 h at 37? C.; after incubation the membranes were sonicated with sterile isotonic solution and the detached bacteria were plated on LB agar).

(11) FIG. 11 illustrates the number of attached live bacteria on pristine (green) and membrane with in situ formed Ag-NPs (gray) for E. coli bacteria (1:1, 2:2, 5:5 means Ag:BH.sub.4 concentrations in mM, respectively) (experimental conditions: pristine and in situ Ag-NPs modified membranes were contacted with isotonic solution (0.15M NaCl; 20 mM NaHCO.sub.3 as buffer, pH 8.2) with bacteria (OD.sub.600: 0.15) for 2 h at 37? C.; after incubation the membranes were sonicated with sterile isotonic solution and the detached bacteria were plated on LB agar).

(12) FIG. 12 illustrates three-dimensional AFM images of: (A) control polyamide membrane, and (B) in situ Ag-NPs modified membrane (5:5) (note that the X and Y dimensions are both 15 ?m (2.5 ?m/division), while the Z-scale is 500 nm (250 nm/division)).

(13) FIG. 13 illustrates SEM micrographs of bacteria morphology: E. coli on pristine membrane (A) and on 2:2 in situ Ag-NPs modified membrane (B); P. aeruginosa on pristine membrane (C) and 2:2 on in situ Ag-NPs modified membrane (D); S. aureus on pristine membrane (E) and on 2:2 in situ Ag-NPs modified membrane (F); experimental conditions were 5 h incubation of bacteria solution with the membranes (initial OD.sub.600: 0.15?0.9, 0.15M NaCl; 20 mM NaHCO.sub.3 as buffer, pH 8.2).

(14) FIG. 14 illustrates CLSM 3-D images of P. aeruginosa biofilm structures grown over 24 h on pristine (A) and on 5:5 in situ Ag-NPs modified membrane (B) (biofilms were stained with Con A (blue), SYTO? 9 (green) and PI (red) dyes specific for EPS (polysaccharides), live and dead cells, respectivelynote all axes units are in micrometers), and (C) corresponding dead, live and EPS biovolumes calculated from CLSM image stacks of a pristine and 5:5 in situ Ag-NPs modified membrane (asterisks above the bars represent significant (P<0.05) difference between groups).

(15) FIG. 15 is a schematic diagram of in situ formation of Cu-NPs on a TFC RO membrane: pristine RO membrane (A) is covered with CuSO.sub.4 solution (pale blue; 50 mM) for 10 minutes (B), then, the CuSO.sub.4 solution is discarded leaving a thin layer of the CuSO.sub.4 solution on the surface (C), next, the membrane is recoated with NaBH.sub.4 solution (20-50 mM) for 5 minutes (D) to cover the membrane surface with Cu-NPs (E).

(16) FIG. 16 is SEM micrographs of (A, C) pristine and (B, D) in situ Cu-NP modified active layer of TFC membrane at different magnifications as indicated (solutions of 50 mM CuSO.sub.4 and 50 mM NaBH.sub.4 (50:50) were used during the in situ formation reaction).

(17) FIG. 17 is an XPS spectra of pristine (red) and functionalized TFC membrane with in situ formed Cu-NPs (green) with 50:50-Cu:BH.sub.4 solution concentrations (in mM); the peaks at 932 eV are related to the presence of copper.

(18) FIG. 18 illustrates water permeability coefficient, A, salt permeability coefficient, B, and R, salt rejection of pristine and in situ Cu-NP modified membranes prepared with different Cu:BH.sub.4 solution concentrations (in mM; 50:20 is 50 mM Cu and 20 mM BH.sub.4, etc.) (experimental conditions: pressure of 27.5 bar (400 psi), crossflow velocity of 21.4 cm/s, temperature of 25.0?0.5? C., and 50 mM NaCl).

(19) FIG. 19(A) illustrates membrane surface hydrophilicity represented by contact angle measurements; and FIG. 19(B) depicts membrane surface charge represented by zeta potential (notations 50:50 mean Ag:BH.sub.4 concentrations (in mM) during the in situ Cu-NPs formation reaction).

(20) FIG. 20 illustrates residual copper loading on in situ Cu-NPs modified membrane (50:50) after dissolution for several days in 10 mL of 5 mM NaHCO.sub.3 solution (pH 8.3).

(21) FIG. 21 illustrates the number of attached live bacteria colonies (CFU) on pristine (black) and in situ Cu-NPs modified membranes (Blue) for gram negative E. coli bacteria (experimental conditions: pristine and in situ modified (50:50 Cu:BH.sub.4) membranes were contacted and incubated with isotonic solution (0.15M NaCl; 20 mM NaHCO.sub.3 as buffer, pH 8.2) with bacteria (OD.sub.600: 0.15?0.01) for 2 h at 37? C.; after incubation the membranes were sonicated with sterile isotonic solution and the detached bacteria were plated on LB agar).

DETAILED DESCRIPTION OF THE INVENTION

(22) The present invention relates to a method for the in situ formation of biocidal metal nanoparticles on a water purification membrane. More specifically, the method includes reacting a solution of a salt of a biocidal metal with an active layer of the membrane, discarding the biocidal metal salt solution such that a thin layer of the biocidal metal salt solution remains on the membrane surface, reacting a reducing agent solution with the active layer of the membrane and the thin layer of the biocidal metal salt solution thereby forming a biocidal metal nanoparticle-modified membrane, removing the reducing agent solution, and rinsing the biocidal metal nanoparticle-modified membrane. The method is most preferably performed at ambient conditions.

(23) Any biocidal metal can be used in the method of the present invention, including, but not limited to, copper, gold, silver, selenium and zinc. Any salt of a biocidal metal can be used in the method of the present invention, including, but not limited to, nitrates, sulfates, chlorates, chromates, carbonates and oxides. Any reducing agent can be used in the method of the present invention, including, but not limited to, borohydrides, sulfites, phosphites, oxalic acid, formic acid, hydrazine and ascorbic acid. The rinsing solution can be, for example, deionized water.

(24) The method of the present invention will be described in greater detail below with respect to exemplary embodiments drawn toward the in situ formation of nanoparticles on membranes using particular biocidal metals, however, the scope of the present invention is not limited theretouse of other biocidal metals in the method of the present invention is encompassed herein.

(25) A. Silver Nanoparticles (Ag-NPs)

(26) i) In Situ Formation of Ag-NPs on TFC RO Membrane

(27) Dried sheets of seawater reverse osmosis membrane (Dow Filmtec SW30XLE) were received from OctoChem, Inc. The dried membranes coupons were wetted through immersion in 20% isopropanol and 80% DI water for 20 min. Then, the membranes were rinsed and soaked three times in DI water for 24 h and stored at 4? C. until use.

(28) In situ formation of Ag-NPs on the RO membrane was carried out in the following stages. The membrane was placed in between a glass plate and rubber frame (inner hole size: 7.5 cm?12 cm) designed to hold the solutions on the active layer side of the membrane. 10 mL of AgNO.sub.3 solution (1-5 mM) was reacted with the isolated active layer for 10 min (FIG. 1B). Subsequently, the AgNO.sub.3 solution was discarded, leaving only a thin layer (for instance, 300 ?m or less, or 200 ?m or less, or 100 ?m or less) of AgNO.sub.3 solution on the membrane surface (FIG. 1C). Then, the membrane active layer was reacted with a NaBH.sub.4 (1-5 Mm) solution for 5 min (FIG. 1D), forming silver nanoparticles. After formation of Ag-NPs, the NaBH.sub.4 solution was removed from the membrane (FIG. 1E), and the prepared in situ Ag-NPs modified membrane was rinsed for ?10 s with DI water. All in situ reactions were done at ambient conditions.

(29) The concentration of the above-noted AgNO.sub.3 solution can be from 0.1 mM to 100 mM, preferably from 0.1 mM to 50 mM, more preferably from 0.1 mM to 20 mM, even more preferably from 1 mM to 10 mM, and most preferably from 1 mM to 5 mM.

(30) The concentration of the above-noted NaBH.sub.4 solution can be from 0.1 mM to 200 mM, preferably from 0.1 mM to 100 mM, more preferably from 0.1 mM to 50 mM, even more preferably from 1 mM to 10 mM, and most preferably from 1 mM to 5 mM.

(31) The above-noted reaction time for reacting the AgNO.sub.3 solution with the isolated active layer can be from 0.5 minutes to 30 minutes, preferably from 1 minute to 25 minutes, more preferably from 2 minutes to 20 minutes, even more preferably from 5 minutes to 15 minutes, and most preferably about 10 minutes.

(32) The above-noted reaction time for reacting the membrane active layer with the NaBH.sub.4 solution can be from 0.5 minutes to 60 minutes, preferably from 0.5 minutes to 30 minutes, more preferably from 1 minute to 20 minutes, even more preferably from 1 minute to 10 minutes, and most preferably about 5 minutes.

(33) The above-noted rinsing time for rinsing the prepared in situ Ag-NPs modified membrane can be from 1 second to 10 minutes, preferably from 1 second to 5 minutes, more preferably from 1 second to 1 minute, even more preferably from 1 second to 30 seconds, and most preferably about 10 seconds.

(34) Scanning and transmission electron microscopy were used to characterize the in situ Ag-NPs modified and the pristine membranes. For SEM, the membrane was dried and sputter-coated (Denton Vacuum, DESK V) with 15 nm chrome prior to analysis (Hitachi Ultra-High-Resolution Analytical Field Emission Scanning Electron Microscope, FE-SEM, SU-70). For transmission electron microscopy (TEM) cross-section image, the in situ Ag-NPs modified membrane sample was fixed in an epoxy resin and cut using a Leica UC7 cryomicrotome into 100 nm slices at ?60? C. Then, the sample was loaded onto carbon coated copper microgrids via dry pickup. Bright-field TEM images were obtained by a FEI Tecnai T12 Spirit Twin TEM/STEM operated at 120 kV. X-ray photoelectron spectroscopy (XPS; Surface Science Instruments model SSX-100SPECS) measurements for surface elemental analysis were done with monochromatized Al-K? X-ray source (h?=1486.6 eV).

(35) ii) Quantifying of Silver Loading

(36) Ag-NPs modified membrane coupons (3.8 cm.sup.2) were placed in glass vials containing 10 mL DI water with 0.1 mL of 70% HNO.sub.3. The vial with the membrane was agitated for one day to dissolve the metals from the membrane. Subsequently, the dissolved silver concentration in the vial solution was quantified with inductive coupled plasma mass-spectrophotometer (ICP-MS, ELAN DRC-e ICP Mass Spectrometer, Perkin Elmer).

(37) In order to examine the residual silver loading on the membrane after dissolution, 3.8 cm.sup.2 coupons of in situ Ag-NPs modified membranes were placed in glass vials filled with 10 mL of 5 mM NaHCO.sub.3 solution (pH 8.3; without HNO.sub.3). The vial with the membrane was agitated. At predetermined time intervals, the solution was removed, and the empty glass vial containing the membrane was refilled with 10 mL DI water with 0.1 mL of 70% HNO.sub.3 to promote dissolution of the residual silver from the membrane. Subsequently, the dissolved silver concentration in the solution was quantified with ICP-MS.

(38) iii) Membrane Characterization

(39) The transport properties of the membrane were examined in a standard laboratory RO filtration system.sup.48. Two measurements were taken for each type of membrane. The membrane (area of 20.02 cm.sup.2) was compacted overnight under 31 bar (450 psi), and then the pressure was reduced to 27.6 bar (400 psi) to determine the pure water permeability measurement, A. The salt rejection was measured at 400 psi with 50 mM NaCl. The salt permeability coefficient, B, was calculated from the water fluxes and salt rejections according to the method described previously.sup.49. All the filtration experiments were carried out at 25.0?0.5? C. with a crossflow velocity of 21.4 cm/s. .sup.48 Hoek, E. M. V., Kim, A. S., Elimelech, M., 2002. Influence of crossflow membrane filter geometry and shear rate on colloidal fouling in reverse osmosis and nanofiltration separations. Environmental Engineering Science 19 (6), 357-372..sup.49 Yip, N. Y., Tiraferri, A., Phillip, W. A., Schiffman, J. D., Elimelech, M., 2010. High performance thin-film composite forward osmosis membrane. Environmental Science & Technology 44 (10), 3812-3818.

(40) Surface contact angle (VCA Video Contact Angle System; AST Products, Billerica, Mass.) measurements were performed to evaluate the membrane surface hydrophilicity. To account for variations in the measurements, eight measurements were performed on each of two samples. The zeta potential of the membranes was calculated from streaming potential measurements performed with an electro kinetic analyzer (EKA, Brookhaven Instruments, Holtsville, N.Y.) following procedures reported previously.sup.50. Prior to the measurements, the membrane samples were equilibrated with the testing solution (1 mM KCl) for at least 45 min. Measurements were performed at room temperature (23? C.) at pH 4-9, adjusted with aliquots of 1 M HCl and KOH. For each membrane type, two separate samples were characterized to account for the system variation. .sup.50 Walker, S. L., Bhattacharjee, S., Hoek, E. M. V., Elimelech, M., 2002. A novel asymmetric clamping cell for measuring streaming potential of flat surfaces. Langmuir 18 (6), 2193-2198.

(41) Surface roughness was determined using atomic force microscopy (AFM, Bruker, Digital Instruments, Santa Barbara, Calif.) in tapping mode with a silicon probe. The probe had a spring constant of 40 N/m, a resonance frequency of 300 kHz, a tip radius of 8?4 nm, and a cantilever length of 125?10 ?m. Micrographs were captured from three different spots with area of 15 ?m?15 ?m on each membrane sample. Three samples of each membrane type (pristine and modified) were characterized to calculate average roughness.

(42) iv) Antimicrobial Activity of Ag-NP Functionalized Membranes

(43) The antimicrobial properties of the membrane were assessed using three model bacteria: E. coli, P. aeruginosa, and S. aureus. An overnight culture of bacteria (1 mL) grown in LB was diluted in 24 mL fresh LB broth and grown for 2-3 hours to exponential phase. Viable cell attachment experiments were carried out to quantify the number of live bacteria attached to pristine and in situ modified (2 mM AgNO.sub.3: 2 mM NaBH.sub.4) membranes. The bacteria culture was centrifuged for 1 minute at 5,000 rpm (Eppendorf 5415D) in 1 mL centrifuge tubes, and resuspended in isotonic saline solution (0.15 M NaCl, 20 mM NaHCO.sub.3, pH 8.2) to an initial OD.sub.600 of 0.15?0.09. Circular membrane coupons (3.8 cm.sup.2) were placed in sterile plastic tubes with the bacteria in an isotonic (saline) solution. Only the membrane polyamide active layer was exposed to the bacteria suspension during incubation. The tubes were incubated for 5 h (37? C.). After incubation, the membranes were rinsed gently with DI water to remove unattached bacteria. Membrane coupons were then bath-sonicated for 7 min in 10 mL of isotonic solution to remove the attached bacteria. The suspension was diluted, plated on LB agar plates, incubated overnight following which bacteria colonies were counted. To evaluate the effect of different silver loadings on viable bacteria attachment, a similar procedure was conducted with only 2 hour incubation with E. coli, using 1:1, 2:2, and 5:5 (AgNO.sub.3:NaBH.sub.4 in mM) in situ Ag-NPs modified membranes.

(44) Membrane samples were also viewed with SEM (Hitachi Ultra-High-Resolution Analytical Field Emission Scanning Electron Microscope) after contact with bacteria to observe changes in cell morphology. After 5 h incubation, the membranes were gently rinsed, and the bacteria were fixed with 5% glutaraldehyde at ?20? C. overnight. The fixed membranes and bacteria were thawed at room temperature and the remaining liquid was removed. Membranes were dried and coated with 15 nm chrome (Denton Vacuum Desk V) before SEM imaging.

(45) v) Impact of In Situ Formed Ag-NPs on Biofilm Development

(46) The effect of in situ formed Ag-NPs on biofilm development was evaluated in two custom-designed flow cells.sup.51 equipped with a pristine or modified membrane coupon (7 cm.sup.2). P. aeruginosa was cultivated in an artificial sterile wastewater media enriched with 0.5% LB broth as a carbon source. Fresh wastewater was prepared by supplementing deionized water (Nano Pure II, Barnstead, Dubuque, Iowa) with 1.2 mM sodium citrate, 0.8 mM NH.sub.4Cl, 0.2 mM KH.sub.2PO.sub.4, 0.2 mM CaCl.sub.2.H.sub.2O, 0.5 mM NaHCO.sub.3, 8.0 mM NaCl, and 0.15 mM MgSO.sub.4. 7H.sub.2O (modified from Bar-zeev and Elimelech (2014).sup.52). .sup.51 Bar-Zeev, E., Berman-Frank, I., Girshevitz, O., Berman, T., 2012. Revised paradigm of aquatic biofilm formation facilitated by microgel transparent exopolymer particles. Proceedings of the National Academy of Sciences of the United States of America 109 (23), 9119-9124..sup.52 Bar-zeev, E., Elimelech, M., 2014. Reverse osmosis biofilm dispersal by osmotic back-flushing: Cleaning via substratum perforation. Enviromental science and technology letters.

(47) Initially, bacteria were inoculated for four hours by passing a synthetic wastewater solution with a concentrated (10.sup.9 cells mL.sup.?1) bacteria culture at flow rate of ?0.6 mL min.sup.?1 through both flow cells. During the experiment, the flow-cell was held vertically to avoid attachment of bacteria or particles by gravity to the glass slide. After 4 h, the circulated solution was changed to a sterilized synthetic wastewater without bacteria for promoting biofilm growth of the bacteria that were attached in the first stage. The second stage of biofilm growth lasted 24 h and during this stage the flow rate was maintained at ?2 mL min.sup.?1. The temperature was kept at 25? C. throughout the experiment.

(48) At the end of the run, two 1 cm.sup.2 sub-samples were cut from the membrane coupon and placed in separate petri dishes with 3 mL sterile wastewater. Samples were rinsed three times with sterile wastewater to remove loosely adhered bacteria. After rinsing, the biofilm was stained with SYTO? 9 and propidium iodide (PI) according to reagent manual (LIVE/DEAD? BacLight?, Invitrogen, USA) to identify live and dead cells (green and red, respectively). Concurrently, samples were stained with 50 ?M concanavalin A (Con A, Alexa Flour? 633, Invitrogen, USA) to determine biofilm EPS (blue). All samples were stained for at least 20 min in the dark. Unbound stain was removed by rinsing the samples three times with sterile WW.

(49) Confocal images were captured using a CLSM (Zeiss LSM 510, Carl Zeiss, Inc.) equipped with a Plan-Apochromat 20?/0.8 numerical aperture objective. A minimum of six Z stack random fields (635 ?m.sup.2) were collected for each sample, with a slice thickness of 2.14 ?m, using ZEN? (Carl Zeiss, Inc.) to obtain representative areas of biofilm. SYTO? 9, PI, and Con A were excited with three sets of lasers: 488 nm argon, 561 nm diode-pumped solid state, and 633 nm helium-neon laser, respectively. Image analysis was performed using Auto-PHLIP-ML (http://sourceforge.net/projects/phlip/), Image J software (http://rsbweb.nih.gov), and MATLAB? (The Mathworks?, Inc.). Thickness and biovolume were determined for the EPS polysaccharides (Con A staining), live, and dead bacterial cells (SYTO? 9 and PI staining) components of the biofilm for all samples. Total biovolume was calculated by summing the live, dead, and EPS biovolumes.

(50) vi) Chemicals

(51) Calcium chloride dihydrate (CaCl.sub.2.2H.sub.2O), copper sulfate (CuSO.sub.4), ethylenediaminetetraacetic acid (C.sub.10H.sub.16N.sub.2O.sub.8; EDTA), nitric acid (HNO.sub.3), potassium chloride (KCl), silver nitrate (AgNO.sub.3) and sodium dodecyl sulfate (C.sub.12H.sub.25NaO.sub.4S; SDS) were purchased from Sigma Aldrich (St. Louis, Mo.). Ammonium chloride (NH.sub.4Cl), hydrochloric acid (HCl), magnesium chloride hexahydrate (MgSO.sub.4.6H.sub.2O), potassium phosphate (KH.sub.2PO.sub.4), sodium chloride (NaCl), sodium citrate dihydrate (HOC(COONa)(CH.sub.2COONa).sub.2.2H.sub.2O), sodium hydroxide (NaOH), and 2-propanol (isopropyl alchohol, CH.sub.3CHOHCH.sub.3) were purchased from J. T. Baker (Phillipsburg, N.J.). Potassium bicarbonate (KHCO.sub.3), sodium bicarbonate (NaHCO.sub.3), and sodium borohydride (NaBH.sub.4) were obtained from Fisher Scientific (Fair Lawn, N.J.). Potassium hydroxide (KOH) was purchased from Aventor Performance Materials (Center Valley, Pa.) and ethanol from Decon Laboratories, Inc. (King of Prussia, Pa.). Bacto agar and Luria-Bertani (LB) broth were received from BD Biosciences (Sparks, Md.) and American Bioanalytical (Natick, Mass.). Unless specified, all the discussed solutions were prepared in deionized (DI) water treated by Milli-Q ultrapure water purification system (Millipore, Billerica, Mass.).

(52) vii) Model Bacteria Strains

(53) Kanamycin resistant Escherichia coli (E. coli) BW26437 was received from the Yale Coli Genetic Stock Center (New Haven, Conn.). Pseudomonas aeruginosa (P. aeruginosa) ATCC 27853 was obtained from the American Type Culture Collection, and Staphylococcus aureus (S. aureus) 8325 was kindly provided by Dr. Naomi Balaban. Cultures were maintained on LB agar plates and grown in LB media prior to experiments. LB agar plates for E. coli included 25 mg/L kanamycin sulfate (American Bioanalytical).

(54) viii) Results

(55) (a) In Situ Formation of Ag-NPs on the Membrane Surface

(56) The current in situ formation method requires only two chemicals for functionalizing RO membrane with Ag-NPs: silver nitrate and reducing agent. DI water, which was used as the solvent for the in situ formation procedure, is available in RO plants. Several advantages led to the selection of sodium borohydride (NaHB.sub.4) as the reducing agent. NaBH.sub.4 is a common reducing agent used in industrial scales for bleaching of wood pulp.sup.53. The strong reactivity of NaHB.sub.4 leads to rapid metal reduction on the membrane. In addition, under the conditions used, NaBH.sub.4 does not reduce carboxylic or amides.sup.54 which are constituents of the polyamide selective layer of TFC-RO membranes. Therefore, reduction by NaBH.sub.4 is expected to have minimal impact on the RO membrane salt selectivity and water permeability. .sup.53 Gulsoy, S. K., Eroglu, H., 2011. Influence of sodium borohydride on kraft pulping of eur pean black pine as a digester additive. Industrial & Engineering Chemistry Research 50 (4), 2441-2444..sup.54 Periasamy, M., Thirumalaikumar, P., 2000. Methods of enhancement of reactivity and selectivity of sodium borohydride for applications in organic synthesis. Journal of Organometallic Chemistry 609 (1-2), 137-151.

(57) The initial stage of the in situ formation procedure (FIG. 1) included contact of dilute silver salt solution (AgNO.sub.3) with the membrane active layer. During this stage, the hydrolyzed silver ions diffused toward the membrane surface. It is likely that the negatively charged carboxylic groups found on the surface of TFC-RO membranes.sup.55 form electrostatic pairs with silver ions and contribute to Ag-NP nucleation during the reduction stage.sup.56,57. The majority of the AgNO.sub.3 solution was removed (but could be reused), leaving only a residual thin film of solution covering the membrane active layer. The following reaction of the membrane surface with dilute NaBH.sub.4 solution led to immediate reduction of the free silver ions found in the residual thin film and to silver precipitation on the membrane surface, according to.sup.58:
Ag.sup.++BH.sub.4.sup.?+3H.sub.2O.fwdarw.Ag.sup.0+B(OH).sub.3+3.5H.sub.2(g)
Three types of solutions were initially examined during the in situ formation procedure: 1:1, 2:2, and 5:5 which refer to the concentrations of AgNO.sub.3:NaHB.sub.4 in mM. .sup.55 Tiraferri, A., Elimelech, M., 2012. Direct quantification of negatively charged functional groups on membrane surfaces. Journal of Membrane Science 389, 499-508..sup.56 Joly, S., Kane, R., Radzilowski, L., Wang, T., Wu, A., Cohen, R. E., Thomas, E. L., Rubner, M. F., 2000. Multilayer nanoreactors for metallic and semiconducting particles. Langmuir 16 (3), 1354-1359..sup.57 Li, Z., Lee, D., Sheng, X. X., Cohen, R. E., Rubner, M. F., 2006. Two-level antibacterial coating with both release-killing and contact-killing capabilities. Langmuir 22 (24), 9820-9823..sup.58 Song, K. C., Lee, S. M., Park, T. S., Lee, B. S., 2009. Preparation of colloidal silver nanoparticles by chemical reduction method. Korean Journal of Chemical Engineering 26 (1), 153-155.

(58) The first indication for successful reduction of silver on the membrane was a change of the membrane surface color from white to yellow-brown (FIG. 2). Higher AgNO.sub.3 and NaHB.sub.4 solution concentrations led to a darker yellow-brown membrane color, indicating a higher loading of silver on the membrane. In addition, the homogenous color indicates a uniform distribution of the reduced silver on the membrane. At relatively high SEM magnification of 45,000, there was no significant visual difference between in situ modified and pristine membrane (FIG. 3A, B). However, when the SEM magnification was increased to 100,000?, small particles covering the polyamide surface features were observed (FIG. 3C, D). It was important to understand the structure of the silver precipitates since formation of a continuous silver precipitate layer on the membrane surface would lead to surface blockage and severe reduction in membrane permeability, similar to membrane fouling by scaling. TEM micrographs confirmed that silver was not reduced as a continuous layer but rather as discrete spherical Ag-NPs with an approximate diameter of less than 15 nm (FIG. 4B, C). Thus, the yellow-brown color of the in situ modified membrane, observed by naked eye, originated from the typical plasmon adsorbance of Ag-NPs.sup.59. .sup.59 Solomon, S. D., Bahadory, M., Jeyarajasingam, A. V., Rutkowsky, S. A., Boritz, C., Mulfinger, L., 2007. Synthesis and study of silver nanoparticles. Journal of Chemical Education 84 (2), 322-325.

(59) TABLE-US-00001 TABLE S2 Elemental surface coverage (%) of pristine and in situ Ag-NPs modified membranes as measured by XPS. 1:1, 2:2 and 5:5 are the concentrations of AgNO.sub.3:NaBH.sub.4 (in mM) during the in situ formation of Ag-NPs reaction. Oxygen Carbon Nitrogen Silver Pristine 21.3 ? 1.9 70.3 ? 0.3 8.4 ? 1.6 0.0 ? 0 1:1 21.0 ? 1.3 70.3 ? 1.2 7.8 ? 0.1 0.89 ? 0.0 2:2 21.9 ? 0.5 68.8 ? 0.6 7.3 ? 0.6 2.0 ? 0.6 5:5 21.8 ? 1.0 68.2 ? 0.8 6.7 ? 0.3 3.2 ? 0.1

(60) XPS measurements were performed to analyze the elemental composition of the in situ modified membrane (FIG. 4A). Unlike the spectrum of the pristine membrane where no silver signal was observed, a signal at binding energy of 364 eV (Ag 3d) was clearly seen in the spectra of the in situ modified membranes. Higher AgNO.sub.3 and NaHB.sub.4 concentrations during the in situ formation reaction resulted in stronger XPS signal, and silver surface coverage analysis showed a percent coverage of 0%, 0.89?0.02%, 2.0?0.64%, and 3.2?0.14% for the pristine membrane and 1:1, 2:2, and 5:5 in situ modified membranes, respectively (Table S2). These observations confirmed the presence of silver on the membrane and the ability to control the silver loading on the membrane by changing the solution concentration. The other XPS peaks were related to oxygen, carbon, and nitrogen (peaks at 527, 396, 281 eV for O 1 s, N 1 s, and C 1 s, respectively), which appear in all the spectra (pristine and in situ modified membranes), as these elements are constituents of the polyamide layer.

(61) According to the surface elemental analysis (Table S2), the nitrogen/carbon ratio at the membrane surface was slightly reduced for the in situ modified membranes. This reduction in N/C ratio (up to 18%), likely due to masking of the polyamide amine group by the Ag-NPs, indicates a decrease in the nitrogen coverage and implies that nitrogen (from the precursor AgNO.sub.3 solution) had no significant content in the formed Ag-NPs. The oxygen/carbon ratio on the membrane surface exhibited a maximal increase of 5%, indicating possible minor addition of oxygen to the membrane surface due to Ag-NP loading. The tendency of silver to be oxidized leading to formation of thin oxidation layer on Ag-NPs was previously established.sup.60 and may explain the minor increase of oxygen on the in situ modified membrane. The absence of peaks typical to boron and sodium (B 1s188 eV, Na 2s63 eV, and Na 2p31 eV; FIG. 5) in the in situ modified membrane spectra indicates that these elements, which participated in the in situ formation procedure, were absent in the Ag-NPs. Therefore, by elimination, it can be concluded that the in situ reduced Ag-NPs on the membrane comprised only metallic silver potentially coated with a thin oxidized layer. .sup.60 Li, X. A., Lenhart, J. J., Walker, H. W., 2010. Dissolution-accompanied aggregation kinetics of silver nanoparticles. Langmuir 26 (22), 16690-16698.

(62) Because of the relatively high hydrodynamic shear inside the membrane module during operation, it is important to demonstrate the affinity of the in situ formed Ag-NPs to the membrane surface. A 5:5 in situ modified membrane was sonicated for 5 min in DI water and then analyzed by XPS. The visual yellow-brownish color that remained after sonication and the observed Ag related peaks in XPS spectra (FIG. 6) demonstrate that Ag-NPs were strongly attached to the in situ formed Ag-NPs to membrane surface.

(63) (b) Impact on Membrane Properties

(64) Many applications of RO, like seawater desalination, require membranes with very high salt rejection (?99%) and high water permeability.sup.61. In addition, membrane surface properties (e.g., roughness, hydrophilicity, and surface charge) will dictate the fouling propensity of the membrane.sup.62,63. Therefore, it was important to ascertain that the Ag-NP in situ formation on the membrane properties had negligible impact on membrane surface properties. .sup.61 Lee, K. P., Arnot, T. C., Mattia, D., 2011. A review of reverse osmosis membrane materials for desalination-development to date and future potential. Journal of Membrane Science 370 (1-2), 1-22..sup.62 Elimelech, M., Phillip, W. A., 2011. The future of seawater desalination: Energy, technology, and the environment. Science 333 (6043), 712-717..sup.63 Vrijenhoek, E. M., Hong, S., Elimelech, M., 2001. Influence of membrane surface properties on initial rate of colloidal fouling of reverse osmosis and nanofiltration membranes. Journal of Membrane Science 188 (1), 115-128.

(65) The water permeability coefficient (A) decreased from 2.41?0.14 for the pristine membrane to 2.12?0.20, 2.01?0.10, and 2.01?0.12 L m.sup.?2 h.sup.?1 bar.sup.?1 for the 1:1, 2:2, and 5:5 in situ modified membranes, respectively (FIG. 7A). These values reflect a maximum reduction in permeability of 17% due to the in situ formation reaction. Salt rejection values were 98.85?0.26%, 98.64?0.01%, 98.63?0.1%, and 98.33?0.2% for the pristine membrane and the 1:1, 2:2, and 5:5 in situ modified membranes, respectively. The calculated values of the salt permeability coefficient (B) decreased slightly from 0.28?0.05 for the pristine membrane to 0.22?0.01, 0.21?0.02, and 0.27?0.02 L m.sup.?2 h.sup.?1 for the 1:1, 2:2, and 5:5 in situ modified membranes, respectively (FIG. 7A).

(66) The decrease in membrane water permeability is likely attributed to the deposited Ag-NPs on the membrane, which lowered the effective membrane surface area for water flow. The observed moderate reduction in membrane water permeability while maintaining the membrane salt selectivity further indicates that the polyamide layer properties were not greatly affected by the in situ formation reaction. This minimal impact on the intrinsic properties of the polyamide layer is related to the selective reduction of silver by NaBH.sub.4 at the low concentrations used.

(67) Key membrane surface properties that impact fouling propensitysurface charge, hydrophilicity, and roughnesswere also not significantly affected by the in situ formation of the Ag-NPs on the membrane surface. The in situ modified membrane (5:5) and the pristine membrane exhibited comparable zeta potential versus pH curves (FIG. 8A). Also the measured contact angles for the in situ modified membranes were only slightly different than the pristine membrane (FIG. 8B). Due to the small size (?15 nm) of the in situ modified Ag-NPs, a minimal increase in surface roughness parameters was observed with root mean squared surface roughness (R.sub.rms) of 107?15, 139?26, 115?14, and 123?19 nm for the pristine and 1:1, 2:2, and 5:5 in situ modified membranes, respectively (values of other surface roughness parameter can be found in Table S5).

(68) TABLE-US-00002 TABLE S5 Surface morphology of pristine and in situ Ag-NPs modified membranes In Situ Reduction Control 1:1 2:2 5:5 R.sub.rms.sup.a (nm) 107 ? 15 139 ? 26 115 ? 14 123 ? 19 R.sub.a.sup.b (nm) 80 ? 12 107 ? 24 95 ? 7 95 ? 15 R.sub.max.sup.c (nm) 880 ? 130 1029 ? 65 1064 ? 128 954 ? 124 SAD.sup.d (%) 20.2 ? 6.2 28.4 ? 10.5 37.0 ? 10.2 30.9 ? 14.1 .sup.aRoot mean squared roughness (R.sub.rms): the RMS deviation of the peaks and valleys from the mean plane. .sup.bAverage roughness (R.sub.a): arithmetic average of the absolute values of the surface height deviations measured from the mean plane. .sup.cMaximum roughness (R.sub.max): the maximum vertical distance between the highest and lowest data points in the image following the plane fit. .sup.dSurface area difference (SAD): the increase in surface area (due to roughness) over a perfectly flat plane with the same projected area.

(69) (c) Ag-NPs Dissolution Tendency

(70) Since Ag-NPs dissolve in aqueous solutions, we assessed the potential long-term use of the in situ reduced Ag-NPs by examining the residual silver loading on the membrane after days of dissolution. Silver ion dissolution from Ag-NPs highly depends on the solution chemistry and pH.sup.64,65,66. We examined the dissolution of the Ag-NPs (5:5) membrane in a solution buffered by NaHCO.sub.3 at pH 8.2-8.3, similar to pH of seawater and wastewater. Our results indicate that the residual silver on the membrane after one week of dissolution remained roughly unchanged at ?3.5 ?g cm.sup.?2 (FIG. 7B). The amount of silver ions that were dissolved in the media was also constant all over the 7 days of dissolution, with values of 5.5?0.6% relative to the residual silver loading on the membrane. However, we emphasize that in RO plants the Ag-NP dissolution rate will likely be higher due to the more complex solution chemistry. Also the hydrodynamic conditions near the membrane surface will impact the Ag-NP dissolution rates. .sup.64 Levard, C., Mitra, S., Yang, T., Jew, A. D., Badireddy, A. R., Lowry, G. V., Brown, G. E., 2013. Effect of chloride on the dissolution rate of silver nanoparticles and toxicity to e. Coli. Environmental Science & Technology 47 (11), 5738-5745..sup.65 Li, X. A., Lenhart, J. J., Walker, H. W., 2010. Dissolution-accompanied aggregation kinetics of silver nanoparticles. Langmuir 26 (22), 16690-16698..sup.66 Liu, J. Y., Hurt, R. H., 2010. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environmental Science & Technology 44 (6), 2169-2175.

(71) (d) Variances of the In Situ Formation Method

(72) The loading time of Ag-NPs on RO-TFC membrane by the in situ formation procedure was relatively fast (15 min). This procedure was divided into two main stages. The first stage lasted for 10 min, in which AgNO.sub.3 solution was contacted with the membrane surface. This stage was designed to allow the diffusion of the silver ions (Ag.sup.+) to the vicinity of the membrane surface and possible partial complexion with the native carboxylic groups on the membrane.sup.67,68. In the second stage, which lasted for 5 min, the silver ions were reduced by NaBH.sub.4 to form Ag-NPs on the membrane. In this stage, after 1-2 minutes a clear brown-yellow coverage of the membrane was seen. Since the times of these two in situ reduction stages were arbitrarily determined, it should be possible to shorten even more the total Ag-NP loading time in the in situ reduction procedure. .sup.67 Joly, S., Kane, R., Radzilowski, L., Wang, T., Wu, A., Cohen, R. E., Thomas, E. L., Rubner, M. F., 2000. Multilayer nanoreactors for metallic and semiconducting particles. Langmuir 16 (3), 1354-1359..sup.68 Li, Z., Lee, D., Sheng, X. X., Cohen, R. E., Rubner, M. F., 2006. Two-level antibacterial coating with both release-killing and contact-killing capabilities. Langmuir 22 (24), 9820-9823.

(73) Three concentrations of AgNO.sub.3 and NaHB.sub.4 were examined in the in situ formation procedure: 1:1, 2:2, and 5:5 mM of AgNO.sub.3:NaHB.sub.4. As the concentration of AgNO.sub.3:NaBH.sub.4 increased, the measured silver loading on the membrane also increased, with values of 0.8?0.02, 1.5?0.03, and 3.7?0.4 ?g cm.sup.?2 for the 1:1, 2:2, and 5:5 in situ modified membranes, respectively (FIG. 9A). These silver loadings demonstrate a linear relationship (R.sup.2=0.99) between solution concentration and silver loading. However, since both the concentration of silver and sodium borohydride were varied, in order to understand whether silver concentration or borohydride concentration dictated the silver loading on the membrane, a fixed AgNO.sub.3 concentration of 2 mM was in situ reduced by 0.5, 2, and 5 mM NaBH.sub.4. The measured silver loadings on the membrane were 1.25?0.33, 1.5?0.03, and 1.48?0.22 ?g cm.sup.?2 for the 0.5, 2 and 5 mM NaBH.sub.4, respectively (FIG. 9B). The surface coverage as calculated by XPS was 2.84?0.74%, 2.04?0.64%, and 2.42?0.84% for the 2:0.5, 2:2, and 2:5 in situ modified membranes, respectively (FIG. 9D). In addition, the water permeability and salt selectivity of the in situ modified membranes were not statistically different for the three types of NaBH.sub.4 concentration used (FIG. 9C, Table S6). Consequently, it can be concluded that the silver salt concentration determines the silver loading on the membrane and not the NaHB.sub.4. This observation raises the opportunity for chemical saving by lowering the amount of NaBH.sub.4 used.

(74) TABLE-US-00003 TABLE S6 Transport properties of pristine and of in situ Ag-NPs modified membranes for different concentrations of BH.sub.4 solution: A - water permeability, R - salt rejection, B - salt permeability, coefficients. During the in situ formation reaction the concentration of AgNO.sub.3 solution was fixed to be 2 mM and the concentration of NaBH4 was changed to 0.5, 2 and 5 mM. (2:0.5 represent 2 AgNO.sub.3: 0.5 NaBH.sub.4 (in mM) etc.). A R B (L m.sup.?2 h.sup.?1 bar.sup.?1) (%) (L m.sup.?2 h.sup.?1) Pristine 2.41 ? 0.14 98.85 ? 0.26 0.28 ? 0.05 2:0.5 2.11 ? 0.04 98.38 ? 0.46 0.21 ? 0.09 2:2 2.01 ? 0.10 98.63 ? 0.1 0.21 ? 0.02 2:5 1.97 ? 0.10 98.55 ? 0.04 0.24 ? 0.01

(75) (e) Antibacterial Activity of In Situ Modified Membrane

(76) Modifying the RO membrane with Ag-NPs imparted a strong antibacterial activity to the membrane surface. After 5 hour of incubation with the 2:2 in situ modified membrane, the number of vial E. coli, P. aeruginosa, and S. aureus bacteria colonies (CFU) decreased by 78?12%, 91?8%, and 96?2.2%, respectively, compared to the pristine membrane (FIG. 10). In order to examine the effect of silver loading on the antibacterial activity, E. coli bacteria were contacted with pristine and in situ modified membranes for 2 h. A decrease of 90.7?3.8%, 91.2?2.6%, and 90.7?3.8% in the number of viable bacteria colonies for the 1:1, 2:2, and 5:5 in situ modified membranes compared to the pristine membrane was observed (FIG. 11). These observations demonstrate the strong antibacterial activity of the modified membrane on gram negative and gram positive bacteria in relatively low silver loadings. No differences in the adhesion propensities of the bacteria to the membrane were expected since the surface properties (hydrophilicity, surface roughness, and zeta potential) of the modified and pristine membranes were similar (FIGS. 8 and 12, Table S5). Thus, it can be inferred that the profound decrease in the number of attached live bacteria was mainly due to the toxicity of Ag-NPs. SEM micrographs validate the cytotoxicity of Ag-NPs by demonstrating changes in bacteria morphology on the modified membrane compared to the pristine membrane, especially for P. aeruginosa and E. coli (FIG. 13).

(77) In addition, biofilm development was significantly suppressed on the in situ Ag-NP membrane coupons indicated by the CLSM imaging and analysis of biovolume dimensions (FIG. 14). Biovolume on in situ Ag-NP membrane compared to pristine membrane reduced by 73%, 38%, and 25% for live, dead, and EPS (respectively), leading to total (live+dead+EPS) biovolume reduction of 41%. During 24 h of biofilm growth, the toxicity of the in situ formed Ag-NPs led to the killing and reduction of the number of live bacteria on the membrane, thus, the ability of the live bacteria to proliferate and to release EPS was suppressed, following by diminishing of overall biofilm formation.

(78) (f) Conclusion

(79) Applying Ag-NPs as biocides in RO processes requires thorough understanding of the Ag-NP loading method. While previous work demonstrated Ag-NPs strong antibacterial activity, their suggested Ag-NP loading methods did not address all of the constraints found in an industrial RO plant. These constraints require gentle, simple, rapid, and inexpensive procedure for Ag-NP loading that can be repeated without disassembling the membrane module. Additionally, the loading method must maintain the unique RO membrane properties. The in situ Ag-NP formation procedure demonstrated here is simple, rapid (15 min), very efficient in terms of chemical usage, and can be repeatedly applied directly inside the membrane module. Moreover, beside minor reduction in membrane permeability, the unique transport and surface properties of the membrane maintained. Both significant reduction in the number of attached live bacteria in static tests and suppression of biofilm development demonstrated the strong antibacterial activity by the in situ formed Ag-NPs. Therefore, this loading procedure greatly improves upon previously presented membrane modification methods, justifying future research that focuses on understanding the mechanisms and assessing the potential of Ag-NPs for mitigation of biofouling in RO processes.

(80) B. Copper Nanoparticles (Cu-NPs)

(81) i) In Situ Formation of Cu-NPs on TFC RO Membrane

(82) In situ formation of Cu-NPs on the RO membrane was conducted in a similar procedure as in the situ formation of Ag-NPs, but with copper salts instead of silver, different concentration of NaBH.sub.4, and longer reaction time. Initially, 10 mL of CuSO.sub.4 solution (50 mM) was reacted with the isolated active layer for 10 min (FIG. 15B). Subsequently, the CuSO.sub.4 solution was discarded, leaving only thin layer of CuSO.sub.4 solution on the membrane surface (FIG. 15C). Then, the membrane active layer was reacted with a NaBH.sub.4 (20-50 Mm) solution for 15 min (FIG. 15D), forming copper nanoparticles. After formation of Cu-NPs, the NaBH.sub.4 solution was removed from the membrane (FIG. 15E), and the prepared in situ Cu-NPs modified membrane was rinsed for ?10 s with DI water. All in situ reactions were done at ambient conditions.

(83) The concentration of the above-noted CuSO.sub.4 solution can be from 0.1 mM to 200 mM, preferably from 0.1 mM to 100 mM, more preferably from 1 mM to 80 mM, even more preferably from 25 mM to 75 mM, and most preferably about 50 mM.

(84) The concentration of the above-noted NaBH.sub.4 solution can be from 0.1 mM to 200 mM, preferably from 1 mM to 150 mM, more preferably from 5 mM to 100 mM, even more preferably from 10 mM to 75 mM, and most preferably from 20 mM to 50 mM.

(85) The above-noted reaction time for reacting the CuSO.sub.4 solution with the isolated active layer can be from 0.5 minutes to 30 minutes, preferably from 1 minute to 25 minutes, more preferably from 2 minutes to 20 minutes, even more preferably from 5 minutes to 15 minutes, and most preferably about 10 minutes.

(86) The above-noted reaction time for reacting the membrane active layer with the NaBH.sub.4 solution can be from 0.5 minutes to 60 minutes, preferably from 1 minute to 45 minutes, more preferably from 5 minutes to 30 minutes, even more preferably from 10 minutes to 20 minutes, and most preferably about 15 minutes.

(87) The above-noted rinsing time for rinsing the prepared in situ Cu-NPs modified membrane can be from 1 second to 10 minutes, preferably from 1 second to 5 minutes, more preferably from 1 second to 1 minute, even more preferably from 1 second to 30 seconds, and most preferably about 10 seconds.

(88) ii) Quantifying of Copper Loading

(89) Copper loading on the membrane was quantified with SEM and XPS using the same methodology as described above for the Ag-NP in situ formation.

(90) iii) Membrane Characterization

(91) Intrinsic membrane transport properties (A, B, and R) and surface characteristics (contact angle, zeta potential, and AFMsurface roughness) measurements were conducted following the same methodology as described above for the Ag-NP in situ formation.

(92) iv) Antimicrobial Activity of Cu-NP Functionalized Membranes

(93) The antimicrobial properties of the membrane were assessed using E. coli as model bacteria. An overnight culture of bacteria (1 mL) grown in LB was diluted in 24 mL fresh LB broth and grown for 2-3 hours to exponential phase. Viable cell attachment experiments were carried out to quantify the number of live bacteria attached to pristine and in situ modified (50 mM CuSO.sub.4: 50 mM NaBH.sub.4) membranes. The bacteria culture was centrifuged for 1 minute at 5,000 rpm (Eppendorf 5415D) in 1 mL centrifuge tubes, and resuspended in isotonic saline solution (0.15 M NaCl, 20 mM NaHCO.sub.3, pH 8.2) to an initial OD.sub.600 of 0.15?0.09. Circular membrane coupons (3.8 cm.sup.2) were placed in sterile plastic tubes with the bacteria in an isotonic (saline) solution. Only the membrane polyamide active layer was exposed to the bacteria suspension during incubation. The tubes were incubated for 2 h (37? C.). After incubation, the membranes were rinsed gently with DI water to remove unattached bacteria. Membrane coupons were then bath-sonicated for 7 min in 10 mL of isotonic solution to remove the attached bacteria. The suspension was diluted, plated on LB agar plates, incubated overnight following which bacteria colonies were counted.

(94) vi) Chemicals

(95) The used chemicals are described in the chemical list of Ag-NP in situ formation procedure.

(96) vii) Model Bacteria Strains

(97) Kanamycin resistant Escherichia coli (E. coli) BW26437 was received from the Yale Coli Genetic Stock Center (New Haven, Conn.). Cultures were maintained on LB agar plates and grown in LB media prior to experiments. LB agar plates for E. coli included 25 mg/L kanamycin sulfate (American Bioanalytical).

(98) viii) Results

(99) (a) In Situ Formation of Cu-NPs on the Membrane Surface

(100) Cu-NPs were successfully in situ formed on the face of the polyamide layer as can be clearly seen in SEM micrographs (FIG. 16). Since the Cu-NPs were observed even in 45K magnification, it can be inferred that the in situ formed Cu-NPs are larger than the 15 nm in situ formed Ag-NPs which were not seen at that magnification (FIG. 3B). XPS measurement reinforced the evidence for successful coverage of Cu-NPs on the membrane by indicating peaks related to the presence of copper at 932 eV, which were not seen in the pristine membrane spectra (FIG. 17).

(101) (b) Impact on Membrane Properties

(102) The intrinsic transport properties of the membrane were inconspicuously changed during the Cu-NPs in situ formation procedure (FIG. 18). At relatively low NaHB.sub.4 concentration of 20 mM and CuSO.sub.4 concentration of 50 mM, the water permeability coefficient (A; L m.sup.?1 h.sup.?1 bar.sup.?1), the salt permeability coefficient (B; L m.sup.?2 h.sup.?1), and the salt rejection (R; %) of Cu-NP in situ modified membrane were 2.40?0.21, 0.2?0.01, and 98.85?0.08 as compared to 2.41?0.41, 0.28?0.14, 98.85?0.26 for the pristine membrane, respectively. Increasing the NaHB.sub.4 concentration to 50 mM while maintaining the CuSO.sub.4 concentration (50 mM) during the in situ reduction, led to minute increase in water and salt permeability coefficients (A: 2.59?0.17 L m.sup.?1 h.sup.?1 bar.sup.?1, B: 0.341?0.17 L m.sup.?1 h.sup.?1) and decrease in salt rejection (R: 98.325?0.34%). Thus, although NaBH.sub.4 reported to be a selective reducer which is inert to the polyamide constituents (amide, carboxylic groups), at high concentration (such as 50 mM) the polyamide layer may be slightly damaged leading to its opening as indicated by the minimal increase of water and salt permeability coefficient.

(103) The surface traits of the membrane were insignificantly impacted by the Cu-NPs in situ formation procedure. The surface hydrophilicity, represented by contact angle measurements, indicated small increase from 45.4?1.7? for the pristine membrane to 59.8?3.9? for the 50:50 modified membrane (FIG. 19A). The zeta potential of the in situ 50:50 copper modified membrane were similar to the pristine membrane in all the examined pH range (FIG. 19B). SEM micrographs (FIG. 16) revealed that the typical predominant roughness features of the polyamide layer surface were maintained after Cu-NPs in situ modification, implying minimal impact on the overall surface roughness of the membrane.

(104) (c) Cu-NPs Dissolution Tendency

(105) In the essential simple solution examined here (DI water+5 mM NaHCO.sub.3 as buffer, pH 8.3), a prominent dissolution of the in situ formed Cu-NPs to dissolved copper ions was not observed during 7 days of experiment (FIG. 20). The surface copper loadings on the membrane were maintained and undulated around values of 8 ?g cm.sup.?2 (as Cu.sup.2+) during all the experimental days. The amount of dissolved Cu.sup.2+ remained the same during the experiments with portion between 10-15% of the total copper that were initially in situ reduced on the membrane. It is emphasized that the solution chemistry highly affects the dissolution rates, and hence, in the presence of more intricate solution such as with salts, NOM, or extreme pH values, the dissolution rates of the in situ formed Cu-NPs will likely be accelerated. Also, increasing the crossflow shear above the membrane may lead to higher Cu dissolution rates than the one observed here.

(106) (d) Antibacterial Activity of In Situ Modified Membrane

(107) The in situ formed Cu-NPs imparted strong antibacterial activity to the membrane surface leading to reduction of more than 90% in the number of live E. coli bacteria attached to the membrane after 2 h incubation, as compared to the pristine membrane (FIG. 21). This reduction may be explained either by toxicity effect or by weakened adhesion of bacteria to the modified membrane. The minute changes in the membrane surface properties (zeta potential, hydrophilicity, and roughness) during the in situ formation procedure, as discussed before, implies that the strong toxicity of Cu-NPs toward bacteria is likely the mechanistic explanation for the rendered antibacterial activity of the Cu-NPs in situ modified membrane.

(108) (e) Conclusion

(109) Copper, long known for its antibacterial activity and relatively low cost, is an attractive biocide with a proven history as a biofouling control agent in industrial applications. However, constraints on brine disposal quality and copper implementation costs require precise and efficient usage of copper for biofouling control in RO desalination. Localized loading of a copper source in the vicinity of the TFC-RO membrane surface, as demonstrated in the current study, can be an economically attractive method for localized protection of the membrane during the desalination process. The simplicity of the current Cu-NP synthesis, the facile membrane functionalization procedure, the insignificant impact on membrane transport properties, the ability to repeatedly recharge the Cu-NPs after depletion, and the imparted antibacterial activity, demonstrate the potential of the suggested methodology for biofouling in TFC-RO desalination processes.