Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Aug 4;5(32):20261–20269. doi: 10.1021/acsomega.0c02106

Application of Desalination Membranes to Nuclide (Cs, Sr, and Co) Separation

Hyung-Ju Kim †,*, Sung-Jun Kim †,‡, Seungmi Hyeon †, Han Hi Kang †, Keun-Young Lee †
PMCID: PMC7439396  PMID: 32832779

Abstract

graphic file with name ao0c02106_0011.jpg

Desalination and nuclide separation, with cesium (Cs), strontium (Sr), and cobalt (Co), using commercial polymeric membranes are investigated under room temperature (298 K) to elucidate the permeation mechanism and possibility of applying commercial membranes to the separation of radioactive nuclides. The physicochemical properties of membranes are characterized by multiple techniques. The thickness of the selective layer and the boundary between the layers of membranes are observed by scanning electron microscopy. The chemical structure of selective and support layers is assessed by direct Fourier transform infrared/attenuated total reflection measurements on membrane samples. Thermogravimetric analysis demonstrates the composition comparison between membranes, which describes the relative amount of selective layers consisting of polyamide. The separation performance of polyamide-based commercial membranes is tested on simulated seawater (35,000 ppm of NaCl) and single- and multi-component aqueous nuclide solutions (10 ppm). Nanofiltration (NF) membranes exhibit a high flux of 160–210 L m–2 h–1 with low 31–64% rejection on the permeation of simulated seawater, while reverse osmosis (RO) membranes display a low flux of 13–22 L m–2 h–1 with nearly 80% rejection. This reveals RO membranes to be more effective for the rejecting nuclides (Cs, Sr, and Co) in dilute aqueous solutions, and NF membranes have advantage on high throughput. RO membranes reject above 93% for single components and even higher for mixed nuclide separation (>98%), and NF membranes permeate high flux above 230 L m–2 h–1. This study indicates that the desalination membranes (NF and RO) can be potential candidates for nuclide separation with combination.

1. Introduction

Over the past few decades, clean water has become a significant issue for many countries.1,2 In nuclear power plants (NPPs), water (or seawater) is used as the moderator or coolant, which motivates them to be located near the coast.3,4 Because NPPs can sometimes be dangerous and unpredictable, the treatment of enormous wastewater containing nuclides is of significant importance and a much studied area. The actual radioactive wastewater contains radionuclides with an activity concentration of 10–103 Bq g–1 containing impurities such as organic complex agents and dissolved salts.5 The release of radioactive wastewater from nuclear facilities definitely causes the serious contamination of aqueous systems, and among the various nuclides released, 137Cs, 90Sr, and 60Co are considered the most significant species because of their high gamma radiation, long half-life (30.2 years, 28.79 years, and 5.27 years, respectively), and high solubility in water.6 Because of the potentially harmful nature of these fission products in nuclear facility, for instance, causing cancer in tissues and organs when entered into living organisms,5,6 there is great interest in developing improved techniques to purify aqueous media by removing hazardous materials. Further, IAEA regulated the maximum 137Cs, 90Sr, and 60Co concentrations allowed for safe discharge into the environment as 0.1 Bq g–1 (3.1 × 10–8 ppm), 1 Bq g–1 (1.8 × 10–7 ppm), and 0.1 Bq g–1 (2.5 × 10–9 ppm), respectively.7

The current handling technologies are evaporation and ion exchange to treat radioactive wastewater.8 However, the evaporation process requires high energy consumption and leads to corrosion issue. Ion exchange produces tremendous amount of spent resin, which must be transported and stored for a long time after solidification.9 As an alternative, membrane-based separation has been an increasingly popular technology because of its low-energy intensity, high efficiency, simple maintenance, continuous operation, and low person exposure for radiation.10 To surpass other techniques, membrane materials used for separation should possess high permeability (throughput) and high selectivity (process efficiency).11−13 Thus, membrane fabrication is one of the most critical factors in selective separation. Because wastewater treatment related to NPPs is the industrial operations and highly dependent on safety, a polymeric membrane, which exhibits stable processability, is more appropriate for the separation of nuclides. The polymeric membranes have been employed to the wide range of treatment applications by passing through raw water including impurities, such as macro- to micro-particles (e.g., sand, hair, and pollen), molecules (e.g., bacteria, asbestos, and viruses), and ions (e.g., salt and metal).14

There are several membrane processes such as microfiltration,15 ultrafiltration,16 nanofiltration (NF),9 reverse osmosis (RO),17 membrane distillation,18−21 and forward osmosis5,22 applied to purify radioactive wastewater. These techniques are classified by operation principle, membrane structure, separation mechanism, and target component. The selection of appropriate operation and membrane is the most important aspect in membrane separation community.23 NF membranes are widely applied for desalination and separation of wastewater to achieve low operation pressures and high permeation fluxes.24 Also, the RO membrane-based separation process is the key technology consistently providing clean water because of its high performance and cost efficiency.25

In this work, we describe the NF and RO techniques using commercial membranes for application in desalination and aqueous nuclide separation. Membranes used in NF and RO are studied to understand the physicochemical properties and permeation characteristics of aqueous nuclide solutions such as Cs, Sr, and Co in single and mixed streams. We demonstrate that the NF and RO membranes display decent permeation properties, highly rejecting nuclides considering their salt rejection at 298 K correlating with their properties of selective layer.

2. Results and Discussion

2.1. Membrane Characterization

Figure 1 exhibits the surface and cross-sectional high-resolution scanning electron microscopy (HR-SEM) images of the commercial membranes, NF90, NF270, TW-2540, and XLE-2540. The commercial membranes displayed the ridge-and-valley surface morphology, typically observed for the interfacial polymerization-assembled polyamide (PA) layer.26 A continuous and uniform polymeric layer was observed from the NF and RO membranes. As shown in Figure 1e–h, the thickness of the selective layer varied between membranes, which has a significant effect on the membrane flux. More specifically, the thickness of NF270 (70 ± 14 nm shown in Figure 1f) is thinner than that of NF90 (252 ± 52 nm shown in Figure 1e), which implies to have relatively higher flux. Besides, the thicknesses of TW-2540 (159 ± 30 nm shown in Figure 1g) and XLE-2540 (222 ± 37 nm shown in Figure 1h) are marginally close, indicating a similar flux. The selective layer is a hydrophilic ultrathin barrier layer with PA, making it possible to preferably permeate water. These commercial membranes consist of a selective layer, an interlayer, and structural support. The three different layers are clearly visible by combining with their wide view HR-SEM images in Figure 2. The structural support is a thick polyester web (solid arrows), and the microporous interlayer (dotted arrow) consists of polysulfone (PSf). The entire membrane thicknesses, ranging from 133 to 142 μm, are summarized in Figure 3. Figure 4 shows the Fourier transform infrared/attenuated total reflection (FT-IR/ATR) spectra of commercial membranes from 1450 to 1800 cm–1, related to the PA selective layer and the underlying PSf support layer. The FT-IR/ATR spectra of commercial membranes are almost identical, referring that the chemical structure of membranes is the same. The commercial membranes had peaks at 1668 cm–1 (C=O bond stretching), 1610 cm–1 (H-bonded C=O bond stretching), and 1542 cm–1 (N–H bond in-plane bending), which correspond to the characteristic peaks of PA.27,28 NF270 had relatively weak peaks, indicating a less dense PA layer compared to others. The PSf peaks at 1585 and 1487 cm–1 were detected in all commercial membranes, which clearly demonstrated that PSf was used as a support layer for these membranes. The PSf layer was detected because the penetration depth of the IR is larger than the thickness of the selective layer ranging from 70 to 250 nm. From the thermogravimetric analysis (TGA) results shown in Figure 5, two different weight losses were seen for commercial membranes at temperature ranging from 400 to 450 and 500 to 550 °C while displaying almost similar patterns of curve. Each weight loss is assigned to initial decomposition temperatures of the PSf and PA blends and decomposition of PSf.29 NF90 has 8 weight % larger loading of PSf and PA blends than NF270, which implies a dense selective layer. However, RO membranes have less than 3 weight % difference in the loading of PSf and PA blends, indicating comparable density of selective layer between TW-2540 and XLE-2540. The in-depth explanation for the difference in chemical structure, bonding, and loading of commercial membranes is limited because of proprietary compositions. Nevertheless, it was hypothesized to stem from their inherently different selective layer thickness, relative peak intensities of PA, and different loading of PSf and PA blends.

Figure 1.

Figure 1

Open in a new tab

(a–d) Surface and (e–h) cross-sectional HR-SEM images of (a,e) NF90, (b,f) NF270, (c,g) TW-2540, and (d,h) XLE-2540 membranes.

Figure 2.

Figure 2

Open in a new tab

Cross-sectional HR-SEM images of (a) NF90, (b) NF270, (c) TW-2540, and (d) XLE-2540 membranes with low magnification.

Figure 3.

Figure 3

Open in a new tab

Entire thickness of commercial membranes measured by digital caliper.

Figure 4.

Figure 4

Open in a new tab

FT-IR/ATR spectra of commercial membranes.

Figure 5.

Figure 5

Open in a new tab

TGA curves of (a) commercial membranes, (b) comparison in NF membranes, and (c) comparison in RO membranes.

2.2. Desalination Characteristics

Figure 6 shows the desalination results using commercial membranes (NF90, NF270, TW-2540, and XLE-2540) under 30 bar at 298 K. For the feed reservoir, seawater was simulated with 35,000 ppm (3.5 weight %) of sodium chloride (NaCl) aqueous solution. As shown in Figure 6a, the conductivity values coincided well with the designated value 54.7 mS cm–1, standing for 35,000 ppm of NaCl aqueous solution. After permeating the NF membranes (NF90 and NF270), the conductivity values dropped to 19.5 and 37.8 mS cm–1, corresponding to 12,400 ppm and 24,300 ppm. The rejections are 64 and 31%, which are well matched with the PA peak density and PA loading. However, these rejections are not the qualified values of the desalination membranes. Because of the usual trade-off property coming from the separation membranes between flux and rejection,30 NF90 shows a lower flux (160 L m–2 h–1) than NF270 (210 L m–2 h–1), coinciding with their thickness difference of selective layer. Even if NF270 has a higher flux, NF90, which shows a higher rejection, is a more appropriate membrane for nuclide separation.

Figure 6.

Figure 6

Open in a new tab

Desalination performances of NF and RO membranes under 30 bar at 298 K: (a) conductivity and (b) flux and rejection from NF90 and NF270, (c) conductivity, and (d) flux and rejection from TW-2540 and XLE-2540, respectively.

The RO membranes were also tested for the desalination of simulated seawater. After permeating the RO membranes (TW-2540 and XLE-2540), the conductivity values significantly dropped to 14 and 12 mS cm–1, which correspond to 8900 and 8000 ppm. The rejection values are 74 and 77%, which are consistent with the similar PA density and loading. The rejection is tailored by intrinsic properties of PA chemistry repelling NaCl while permeating water.31 Besides, the flux was significantly reduced to 13 L m–2 h–1 for TW-2540 and 22 L m–2 h–1 for XLE-2540, indicating that the RO membranes are almost 10 times less permeable compared to NF membranes. Considering the high rejection requirement for wastewater purification related to the NPPs, even if the flux is low, the RO membranes are more appropriate for nuclide separation than the NF membranes. Thus, the RO membranes (TW-2540 and XLE-2540) were further investigated to separate aqueous solutions containing nuclides such as Cs, Sr, and Co.

2.3. Nuclide Separation Characteristics

Figure 7 shows the nuclide separation results for aqueous nuclide solutions containing 10 ppm of Cs, Sr, and Co ions using RO membranes under 30 bar at 298 K. For TW-2540, the nuclides were rejected from around 10 to 0.4 ppm for Cs, 0.8 ppm for Sr, and 0.05 ppm for Co. In the case of permeation using XLE-2540, the nuclides were rejected from around 10 to 0.1 ppm for Cs, 0.2 ppm for Sr, and 0.5 ppm for Co. The concentration downgrade corresponds to the rejection of 95.1, 93.0, and 99.6% for TW-2540 and 98.8, 98.0, and 96.0% for XLE-2540, respectively. The significant downgrade of concentration is attributed to the nuclide rejection by PA chains and their high affinity to water. It was also demonstrated that the concentration downgrade is valid in different feed concentrations, such as 5 and 20 ppm, as shown in Figure S1 (Supporting Information). Because the RO membranes separate almost all of the target species via physicochemical mechanisms, there is no significant rejection difference among each nuclide solutions.

Figure 7.

Figure 7

Open in a new tab

Single-nuclide separation performances from RO membranes under 30 bar at 298 K: (a) concentration downgrade and (b) flux and rejection from TW-2540, (c) concentration downgrade, and (d) flux and rejection from XLE-2540.

For completeness, the NF membranes were also tested for nuclide separation under the same conditions, as shown in Figure S2 (Supporting Information). In comparison with the RO membranes, they had a higher flux with low rejection, which is not fully suited for nuclide separation. More specifically, the fluxes of NF90 for the three nuclides ranged from 237 to 279 L m–2 h–1, with rejections ranging from 63 to 83%. NF270 had the same rejection period, with higher fluxes ranging from 321 to 354 L m–2 h–1. The overall tendency of NF membranes coincides well with the desalination process and the traditional trade-off property of membrane separation. In contrast, the rejection order (RCs > RSr > RCo) of nuclides is clearly different from that of the RO membranes, repelling all nuclides with same intensity. This result indicates that the nuclide flux is more highly dependent on diffusivity, the size selective property, in polymer chains rather than solubility in the membrane when molecules permeate through NF membranes. Because the permeation of NF membranes displayed relatively poor performance on rejection in comparison to RO membranes, mixed nuclide separation was focused on RO membranes.

For application to the treatment of more realistic wastewater, mixed nuclide aqueous solution permeated to the RO membranes under 30 bar at 298 K (Figure 8). The mixed nuclide aqueous solution was a combination of the above-described nuclides, Cs, Sr, and Co. Both of the RO membranes showed slightly improved rejection with comparable flux compared to the single-component permeation. This is presumably due to the increased friction among the nuclides in the solution and membrane. In the molecular separations, the empirical fact is that mixed component permeations present similar or better rejections.32 Solutions containing around 10 ppm of each nuclide (Cs, Sr, and Co) were purified to 0.2, 0.1, and 0.1 ppm by TW-2540. Moreover, XLE-2540 downgraded the concentration to 0.17, 0.09, and 0.11 ppm. In terms of rejection, this is above 98%, which indicates potential option to purify wastewater from NPPs for safe discharge into the environment. More importantly, XLE-2540 had an extremely high rejection of 98.5% for Cs, 99.2% for Sr, and 99.0% for Co. These rejection values are comparable to the currently operating RO membranes in NPPs.33 However, they showed similar fluxes as single-component permeations, 76.8 L m–2 h–1 for TW-2540 and 155.4 L m–2 h–1 for XLE-2540.

Figure 8.

Figure 8

Open in a new tab

Mixed nuclide separation performances from RO membranes under 30 bar at 298 K: (a) concentration downgrade and (b) flux and rejection from TW-2540, (c) concentration downgrade, and (d) flux and rejection from XLE-2540.

If the flux through a membrane is higher, the time consumption for the wastewater treatment process decreases. In the treatment of wastewater coming from a nuclear facility, time consumption is a very important factor because time requirement directly correlates with the total cost of wastewater treatment.34 Also, wastewater generated during NPP operation is supposed to be rapidly recycled. To obtain a higher flux in the RO system, mixed nuclide separation was performed under 40 bar at 298 K (Figure 9). Higher flux, by the higher driving force, was obtained for both TW-2540 and XLE-2540 membranes, as expected. Because of the trade-off from increased flux, both membranes incur slightly lower rejection for all nuclides. In the case of permeation of TW-2540, stable rejections, with a maximum loss of 1.3%, were acquired for each nuclide, but XLE-2540 lost up to 8.2% in rejection. The same trend was observed when mixed nuclide separation was performed under 20 bar at 298 K as shown in Figure S3 (Supporting Information). This is probably due to the structural endurance difference between TW-2540 and XLE-2540. Based on the universal test, TW-2540 has a higher tensile strength value, 55.2 MPa, which indicates requiring more load until failure compared with that of XLE-2540, 50.5 MPa. It is speculated that high tensile strength alleviates the negative effect of the TW-2540 membrane caused by the increased feed pressure. An even higher rejection can be anticipated tuning the feed pressure and membrane surfaces.

Figure 9.

Figure 9

Open in a new tab

Mixed nuclide separation performances from RO membranes under 40 bar at 298 K: (a) concentration downgrade and (b) flux and rejection from TW-2540, and (c) concentration downgrade, and (d) flux and rejection from XLE-2540.

3. Conclusions

This work demonstrates that commercial NF and RO membranes can perform not only desalination but also the separation of aqueous nuclide solution. Commercial PA-based membranes exhibit decent rejection for simulated seawater and high rejection for aqueous nuclide solutions (Cs, Sr, and Co) at 298 K. This is because the PA group repels the NaCl and nuclide ions, reducing the effective permeation of them, which highly coincides with their physicochemical properties. RO membranes are more appropriate for nuclide separation because of their high rejection while NF membranes display high flux. This study on polymeric membranes for nuclide separation contributes to the understanding of the permeation of nuclide solutions through polymeric membranes. Furthermore, RO membranes are effective in the permeation of mixed nuclide solution, which indicates that they are applicable to more realistic conditions. Future work using radioactive nuclides is warranted.

4. Experimental Section

4.1. Materials

The following chemicals were used as received: NaCl (99.5%, Junsei), cesium chloride (99.9%, Sigma-Aldrich), cobalt dichloride hexahydrate (99%, Showa), strontium dichloride hexahydrate (98%, Junsei), and nitric acid (0.1 mol/L, SAMCHUN). The NF membranes (NF90 and NF270) and RO membranes (TW-2540 and XLE-2540) were purchased from DOW FILMTEC.

4.2. Experimental Equipment

The NF and RO membrane performance was evaluated with the lab-made cross-flow test unit shown in Figure 10. A feed tank with 20 L volume was included in the instrument for running a long-term separation process. Two different pumps were installed for the operation of a wide range of techniques depending on diverse feed pressure. Two cells were designed in parallel to obtain permeation data of the same membrane simultaneously. During operation, digital balances monitored the mass of permeated liquid with respect to the time of operation to acquire the mass flow rate. Temperature, volumetric flow rate, and pressure were monitored using a temperature indicator (TI), volumetric flow transmitter (FT), and pressure transmitter (PT), respectively. The liquid temperature was maintained as 298 K by circulating water through a water bath.

Figure 10.

Figure 10

Open in a new tab

Schematic diagram of the NF, RO system.

The permeation flux (F, L m–2 h–1) and rejection (R, %) of the membrane were calculated by the following equations28

4.2. 1
4.2. 2

where V is the volume of the solution permeated during the operation time, A is the effective membrane area, Δt is the operation time period (2 h in this study), and Cp and Cf are the concentrations of the target element in the permeate and the feed sides, respectively.

4.3. Membrane Permeation Experiments

The membrane module was prepared as follows. Membranes were immersed in deionized (DI) water just after purchase so as to protect the membrane pores from any pollutants. There was no pretreatment such as solvent activation or heat treatment to avoid additional effect on permeation performances of pristine commercial membranes.27,35 Then, a sample was cut into the same size as circular porous steel (4.7 cm diameter). Also, the membranes were mounted to be a selective layer facing the feed side. During the experiment, the active area of flat circle membrane was 12.57 cm2 because of the space loss by the o-ring. Simulated seawater, which has a concentration of 35,000 ppm, was prepared with NaCl and DI water, while nuclide solutions were prepared from the stock solution. The stock solution (1000 ppm of each nuclide) was prepared with cesium chloride, strontium dichloride hexahydrate, cobalt dichloride hexahydrate, and nitric acid. Then, the feed reservoir (10 ppm) was prepared by hundred times dilution from the stock solution. For the mixed nuclide solution, prepared 10 ppm of each solution was merged. The feed and permeate solutions were collected in glass vials separately. It was confirmed that the concentration variation of feed solution was negligible in terms of operation time because of the huge amount of feed reservoir. During the experiment, the whole system was maintained at 298 K. Each membrane test was conducted after 2 h of operation to reach the steady state before data collection. The membrane permeation test was performed three times to obtain reasonable reproducibility, and the reproducibility was expressed as error bars in Figures 6–9. Feed samples were collected several times to monitor the concentration difference during operation, and permeate samples were also collected from two different membrane cells.

4.4. Characterization

HR-SEM (SU8230, Hitachi) was used to examine the membrane surface and cross section. The membranes were immersed in liquid nitrogen to avoid rupture during cutting, and then, the samples were placed on a carbon tape and coated with osmium tetroxide (OsO4) to prevent surface discharging. Five different regions of each membrane sample were examined to determine the average thickness of the selective layer. The entire membrane thickness was measured using a digital caliper (547–401, Mitutoyo). Five different regions of each membrane sample were measured to display the average thickness. FT-IR/ATR (Nicolet iS5, Thermo Fisher Scientific) spectra were collected over a wavenumber range of 1800–1450 cm–1 to analyze the functional groups of the membranes. TGA (SDT Q600, TA Instruments) was used to obtain thermogravimetric curves. Prior to the TGA measurement, the sample was dried overnight in a vacuum oven at 40 °C to remove physically adsorbed moisture on the membrane surface. For the analysis, the sample was heated from room temperature to 800 °C with a heating rate of 10 °C min–1 under 99.999% purity nitrogen gas. The conductivity of aqueous NaCl solution was measured by a conductivity meter (Orion Star A215, Thermo Scientific). Inductively coupled plasma–optical emission spectroscopy (PQ9000 Elite, Analytikjena) was used to obtain the nuclide concentrations of feed and permeate solutions. For the calibration, standard solutions were prepared, ranging from 0 to 1 ppm. Also, the feed samples were diluted 10 times to normalize these into calibration range, while the permeate samples were analyzed as collected. The tensile property of the membranes was measured using a universal testing machine (QM100S, Qmesys). The rectangular test specimens (50 × 35 mm) with known thickness were mounted between the grips of the machine and were then moved to the opposite direction at a constant speed of 1 mm per minute. The load required to break the membrane was measured and calculated as tensile strength.

Acknowledgments

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (no. NRF-2017M2A8A5015147 and NRF-2017M2A8A5015148).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c02106.

  • Single-nuclide separation performances from RO membranes in different feed concentrations, single-nuclide separation performances from NF membranes, and mixed nuclide separation performances from RO membranes under 20 bar (PDF)

Author Contributions

S.-J.K., S.H., and H.H.K. contributed equally. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Supplementary Material

ao0c02106_si_001.pdf (626.9KB, pdf)

References

  1. Xue Z.; Wang S.; Lin L.; Chen L.; Liu M.; Feng L.; Jiang L. A Novel Superhydrophilic and Underwater Superoleophobic Hydrogel-Coated Mesh for Oil/Water Separation. Adv. Mater. 2011, 23, 4270–4273. 10.1002/adma.201102616. [DOI] [PubMed] [Google Scholar]
  2. Hillie T.; Hlophe M. Nanotechnology and the Challenge of Clean Water. Nat. Nanotechnol. 2007, 2, 663–664. 10.1038/nnano.2007.350. [DOI] [PubMed] [Google Scholar]
  3. Wada T.; Fujita T.; Nemoto Y.; Shimamura S.; Mizuno T.; Sohtome T.; Kamiyama K.; Narita K.; Watanabe M.; Hatta N.; Ogata Y.; Morita T.; Igarashi S. Effects of the Nuclear Disaster on Marine Products in Fukushima: An Update after Five Years. J. Environ. Radioact. 2016, 164, 312–324. 10.1016/j.jenvrad.2016.06.028. [DOI] [PubMed] [Google Scholar]
  4. Terada H.; Katata G.; Chino M.; Nagai H. Atmospheric Discharge and Dispersion of Radionuclides during the Fukushima Dai-ichi Nuclear Power Plant Accident. Part II: Verification of the Source Term and Analysis of Regional-Scale Atmospheric Dispersion. J. Environ. Radioact. 2012, 112, 141–154. 10.1016/j.jenvrad.2012.05.023. [DOI] [PubMed] [Google Scholar]
  5. Liu X.; Wu J.; Hou L.-a.; Wang J. Removal of Co, Sr and Cs Ions from Simulated Radioactive Wastewater by Forward Osmosis. Chemosphere 2019, 232, 87–95. 10.1016/j.chemosphere.2019.05.210. [DOI] [PubMed] [Google Scholar]
  6. Delacroix D.; Guerre J. P.; Leblanc P.; Hickman C.. Radionuclide And Radiation Protection Data Handbook 2002; Oxford University Press, 2002. [DOI] [PubMed] [Google Scholar]
  7. The International Atomic Energy Agency (IAEA) . Application of the Concepts of Exclusion, Exemption and Clearance; International Atomic Energy Agency, 2004; RS-G-1.7. [Google Scholar]
  8. Park Y.; Lee Y.-C.; Shin W. S.; Choi S.-J. Removal of Cobalt, Strontium and Cesium from Radioactive Laundry Wastewater by Ammonium Molybdophosphate–Polyacrylonitrile (AMP–PAN). Chem. Eng. J. 2010, 162, 685–695. 10.1016/j.cej.2010.06.026. [DOI] [Google Scholar]
  9. Chen D.; Zhao X.; Li F. Treatment of Low Level Radioactive Wastewater by Means of NF Process. Nucl. Eng. Des. 2014, 278, 249–254. 10.1016/j.nucengdes.2014.08.001. [DOI] [Google Scholar]
  10. Strathmann H. Membrane Separation Processes: Current Relevance and Future Opportunities. AIChE J. 2001, 47, 1077–1087. 10.1002/aic.690470514. [DOI] [Google Scholar]
  11. Bernardo P.; Drioli E.; Golemme G. Membrane Gas Separation: A Review/State of the Art. Ind. Eng. Chem. Res. 2009, 48, 4638–4663. 10.1021/ie8019032. [DOI] [Google Scholar]
  12. Robeson L. M. The Upper Bound Revisited. J. Membr. Sci. 2008, 320, 390–400. 10.1016/j.memsci.2008.04.030. [DOI] [Google Scholar]
  13. Werber J. R.; Osuji C. O.; Elimelech M. Materials for Next-Generation Desalination and Water Purification Membranes. Nat. Rev. Mater. 2016, 1, 16018–16034. 10.1038/natrevmats.2016.18. [DOI] [Google Scholar]
  14. Lee A.; Elam J. W.; Darling S. B. Membrane Materials for Water Purification: Design, Development, and Application. Environ. Sci.: Water Res. Technol. 2016, 2, 17–42. 10.1039/c5ew00159e. [DOI] [Google Scholar]
  15. Yong G.; Jun Z.; Guanghui Z.; Dong Z.; Weiwen C.; Guoqi Y.; Xuejun L.; Bangzhong M.; Junhui Z.; Ping G. Treatment of the Wastewater Containing Low-Level 241Am Using Flocculation-Microfiltration Process. Sep. Purif. Technol. 2004, 40, 183–189. 10.1016/j.seppur.2004.02.009. [DOI] [Google Scholar]
  16. Sancho M.; Arnal J. M.; Verdú G.; Lora J.; Villaescusa J. I. Ultrafiltration and Reverse Osmosis Performance in the Treatment of Radioimmunoassay Liquid Wastes. Desalination 2006, 201, 207–215. 10.1016/j.desal.2006.02.015. [DOI] [Google Scholar]
  17. Jia F.; Li J.; Wang J.; Sun Y. Removal of Cesium from Simulated Radioactive Wastewater Using a Novel Disc Tubular Reverse Osmosis System. Nucl. Technol. 2017, 197, 219–224. 10.13182/nt16-6. [DOI] [Google Scholar]
  18. Jia F.; Wang J. Separation of Cesium Ions from Aqueous Solution by Vacuum Membrane Distillation Process. Prog. Nucl. Energy 2017, 98, 293–300. 10.1016/j.pnucene.2017.04.008. [DOI] [Google Scholar]
  19. Jia F.; Li J.; Wang J.; Sun Y. Removal of Strontium Ions from Simulated Radioactive Wastewater by Vacuum Membrane Distillation. Ann. Nucl. Energy 2017, 103, 363–368. 10.1016/j.anucene.2017.02.003. [DOI] [Google Scholar]
  20. Khayet M. Treatment of Radioactive Wastewater Solutions by Direct Contact Membrane Distillation Using Surface Modified Membranes. Desalination 2013, 321, 60–66. 10.1016/j.desal.2013.02.023. [DOI] [Google Scholar]
  21. Liu H.; Wang J. Treatment of Radioactive Wastewater Using Direct Contact Membrane Distillation. J. Hazard. Mater. 2013, 261, 307–315. 10.1016/j.jhazmat.2013.07.045. [DOI] [PubMed] [Google Scholar]
  22. Jin X.; She Q.; Ang X.; Tang C. Y. Removal of Boron and Arsenic by Forward Osmosis Membrane: Influence of Membrane Orientation and Organic Fouling. J. Membr. Sci. 2012, 389, 182–187. 10.1016/j.memsci.2011.10.028. [DOI] [Google Scholar]
  23. Strathmann H. Membrane Separation Processes. J. Membr. Sci. 1981, 9, 121–189. 10.1016/s0376-7388(00)85121-2. [DOI] [Google Scholar]
  24. Kieu H. T.; Liu B.; Zhou K.; Law A. W.-K. Pressure-Driven Water Permeation through Multilayer Graphene Nanosheets. Phys. Status Solidi 2017, 254, 1700074–1700084. 10.1002/pssb.201700074. [DOI] [Google Scholar]
  25. Zhang H.; Wu M. S.; Zhou K.; Law A. W.-K. Molecular Insights into the Composition–Structure–Property Relationships of Polyamide Thin Films for Reverse Osmosis Desalination. Environ. Sci. Technol. 2019, 53, 6374–6382. 10.1021/acs.est.9b02214. [DOI] [PubMed] [Google Scholar]
  26. Werber J. R.; Bull S. K.; Elimelech M. Acyl-Chloride Quenching Following Interfacial Polymerization to Modulate the Water Permeability, Selectivity, and Surface Charge of Desalination Membranes. J. Membr. Sci. 2017, 535, 357–364. 10.1016/j.memsci.2017.04.041. [DOI] [Google Scholar]
  27. Shin M. G.; Kwon S. J.; Park H.; Park Y.-I.; Lee J.-H. High-Performance and Acid-Resistant Nanofiltration Membranes Prepared by Solvent Activation on Polyamide Reverse Osmosis Membranes. J. Membr. Sci. 2020, 595, 117590–117600. 10.1016/j.memsci.2019.117590. [DOI] [Google Scholar]
  28. Shin M. G.; Park S.-H.; Kwon S. J.; Kwon H.-E.; Park J. B.; Lee J.-H. Facile Performance Enhancement of Reverse Osmosis Membranes via Solvent Activation with Benzyl Alcohol. J. Membr. Sci. 2019, 578, 220–229. 10.1016/j.memsci.2019.02.027. [DOI] [Google Scholar]
  29. Mohan D. J.; Kullová L. A Study on the Relationship between Preparation Condition and Properties/Performance of Polyamide TFC Membrane by IR, DSC, TGA, and SEM Techniques. Desalin. Water Treat. 2013, 51, 586–596. 10.1080/19443994.2012.693655. [DOI] [Google Scholar]
  30. Park H. B.; Kamcev J.; Robeson L. M.; Elimelech M.; Freeman B. D. Maximizing the Right Stuff: The Trade-off between Membrane Permeability and Selectivity. Science 2017, 356, eaab0530 10.1126/science.aab0530. [DOI] [PubMed] [Google Scholar]
  31. Buch P. R.; Jagan Mohan D.; Reddy A. V. R. Preparation, Characterization and Chlorine Stability of Aromatic–Cycloaliphatic Polyamide Thin Film Composite Membranes. J. Membr. Sci. 2008, 309, 36–44. 10.1016/j.memsci.2007.10.004. [DOI] [Google Scholar]
  32. Bargeman G.; Vollenbroek J. M.; Straatsma J.; Schroën C. G. P. H.; Boom R. M. Nanofiltration of Multi-Component Feeds. Interactions between Neutral and Charged Components and Their Effect on Retention. J. Membr. Sci. 2005, 247, 11–20. 10.1016/j.memsci.2004.05.022. [DOI] [Google Scholar]
  33. Ambashta R. D.; Sillanpää M. E. T. Membrane Purification in Radioactive Waste Management: A Short Review. J. Environ. Radioact. 2012, 105, 76–84. 10.1016/j.jenvrad.2011.12.002. [DOI] [PubMed] [Google Scholar]
  34. Su-xia H.; Ji-jun L.; Bin H.; Ru-song L.; Tao S. The Treatment of Radioactive Wastewater by Ultrasonic Standing Wave Method. J. Hazard. Mater. 2014, 274, 41–45. 10.1016/j.jhazmat.2014.03.068. [DOI] [PubMed] [Google Scholar]
  35. Shi M.; Yan W.; Dong C.; Liu L.; Xie S.; Gao C. Solvent Activation before Heat-Treatment for Improving Reverse Osmosis Membrane Performance. J. Membr. Sci. 2020, 595, 117565–117574. 10.1016/j.memsci.2019.117565. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c02106_si_001.pdf (626.9KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES