Manish
Kumar
a,
Sumanta
Chowdhury
b and
Jaspreet Kaur
Randhawa
*a
aSchool of Mechanical & Materials Engineering, Indian Institute of Technology Mandi, India. E-mail: jaspreet@iitmandi.ac.in
bSchool of Chemical Sciences, Indian Institute of Technology Mandi, India
First published on 9th November 2023
The increasing global population and its reliance on water-based activities have led to freshwater scarcity and unequal distribution. Researchers have responded to the pressing need for effective wastewater treatment by developing new adsorbent materials. However, practical application of these nanomaterials is hindered by challenges in scalability, regeneration, and agglomeration. The integration of nano-adsorbents into polymeric membranes, particularly electrospun nanofibrous membranes, holds potential for improving their performance, reusability, and durability, while also addressing separation concerns. This comprehensive review examines the selection and application of innovative adsorbent materials in membrane technology for wastewater treatment. It explores synthetic techniques for membrane fabrication, emphasizes the impact of materials like 2D materials, MOFs, and COFs, and also addresses problems such as biofouling. Additionally, this study highlights the potential of zwitterionic materials in mitigating biofouling and discusses the concept of biomineralization for water remineralization. Altogether, this review provides valuable insights into the current progress and prospects of material selection in membrane technology for water remediation.
Water impactThe current status of technologies for wastewater remediation is reviewed. Membrane technologies and their functionalization strategies for organic dye, hazardous heavy metal, and salt rejection are examined critically. The prospects of membrane biofouling mitigation and purified water remineralization techniques are conferred in relation to the functionalization of cutting-edge membranes. The knowledge gap in advanced material selection for hybrid membrane design has been identified critically. |
A material's effectiveness as an adsorbent is predominantly determined by its surface area and anchoring sites, such as functional groups. In order to augment the surface area of adsorbents, two potential approaches can be considered: exterior and internal surface area augmentation. The enhancement of nanomaterials' exterior surface area and subsequent improvement in efficiency can be achieved through the synthesis of these materials using either bottom-up or top-down methodologies. A wide array of cost-effective and environmentally sustainable nanomaterials possessing unique properties have been proposed for application in the treatment of industrial wastewater, groundwater, surface water, and seawater.4 The scientific literature extensively acknowledges nanotechnology as an innovative and effective approach for the treatment of wastewater. Nano-adsorbents are well acknowledged for their substantial external surface area and their ability to undergo functionalization through various means. The exterior morphology, observable dimensions, and intrinsic and extrinsic composition, as well as the physical, chemical, and material characteristics of these entities exhibit notable differences. Nanomaterials have been employed in the field of wastewater treatment through many methods, such as adsorption,5–7 photocatalysis,8–10 membrane filtration,11–13 and chemical disinfection.14–16 The global nanotechnology market was valued at $1.76 billion in 2020 and is predicted to reach $33.63 billion by 2030, expanding at a CAGR of 36.4% from 2021 to 2030.17 Water and wastewater treatment enabled by nanotechnology has the ability to not only overcome fundamental obstacles confronting present treatment methods, but also to create unique treatment capabilities that could enable cost-effective, highly efficient, and adaptable wastewater treatment solutions.18–20 Yet, the application of nanomaterials as nanoadsorbents (NAs) is limited due to challenges such as regeneration, agglomeration/aggregation, and instability in the waste stream.21
The assessment of internal surface area in adsorbents holds vital importance and should not be overlooked, in conjunction with the control of external surface areas of nanomaterials. As a result, there has been a considerable amount of academic research conducted on the advancement of porous adsorbents, including activated carbon, charcoal, mesoporous silica, and zeolites. The primary objective of these investigations is to tackle the issue of water remediation needs.22 But the challenge of attaining the desired attributes, such as adjustable pore size, form, functionality, particle size, and morphology, in these porous materials has presented significant difficulties. In recent times, there has been significant progress in the advancement of reticular porous materials, particularly metal–organic frameworks (MOFs). MOFs,23 covalent organic frameworks (COFs),24 hydrogen bonded organic frameworks (HOFs),25etc. have seen an exponential growth of interest in a variety of science disciplines. In this regard, the science of linking molecular building blocks using bonds as glue to create extended crystalline structures is called reticular chemistry.26,27 Reticular chemistry has also had an impact on the development of advanced water purification technology, owing to its architectural, chemical, and thermal stability, all of which could be combined to preserve the porous nature of reticules as well as their hydrolytic stability. Starting from the discovery of an extended crystal structure of manganese germanium sulfide reported by O. M. Yaghi and co-workers in 1994,28 reticular chemistry has developed into one of the fastest growing fields of science that is being currently practiced over hundreds of countries worldwide.29 The synergistic combination of large surface area, adjustable pore size, and surface design capabilities can effectively address the prevalent issues encountered in water remediation using membrane technology.
Adsorbent materials are frequently utilised in the process of water purification to eliminate contaminants. Nonetheless, it is important to acknowledge that these materials do possess certain constraints. Powdered nano or micro materials occasionally depend on chemical coagulants or flocculants, hence resulting in increased operating expenses.30 Furthermore, extracting powdered components from treated water might be difficult, potentially resulting in recontamination.31 Membrane technology, on the other hand, provides significant advantages. Membranes act as selective barriers, enabling water molecules to pass but preventing contaminants from passing.32 This methodology ensures comprehensive and reliable processing, leading to the production of uncontaminated water without the need for further chemical substances. In addition, membrane technology possesses a reduced carbon footprint, exhibits a modular architecture, and offers enhanced control over the treatment process, rendering it a favored solution for desalination and wastewater treatment.33
The utilization of polymer-based membrane technology presents a promising solution for addressing diverse remediation challenges, including but not limited to air purification and wastewater treatment, as evidenced by recent research conducted by Goh et al.34 Tan and Rodrigue conducted a study on the production of polymeric membranes. These membranes are created using materials that can be handled in solution. Subsequently, they are shaped into several membrane morphologies, including fiber mats, hollow fibers, and flat sheet membranes.35 Accordingly, the research for materials that work well together as better adsorbents for membranes has led researchers to look for new ways to combine materials science and membrane technology.36 Hybrid nano composites have better qualities than their pristine counterparts alone. Functional nanomaterials attached to a membrane structure improve the membrane stability, functionality, tunable rejection, mechanical strength, fouling liability and beyond.37–40 Among the various manufacturing techniques, electrospinning produces high-quality fibres with a three-dimensional network, making it ideal for filtration. Electrospun nanofibrous membranes (ENMs) have outstanding mechanical properties, a high flux, up to 80% porosity, and configurable pore sizes in the sub-micron range. Because of their large external surface area combined with increased mass transfer aided by their internal surface areas, including porosity, ENMs provide a substantial advantage. This characteristic allows for more effective interactions with the environment, increasing mass transfer efficiency and aiding successful filtration operations. Furthermore, because of their tunable pore sizes and high surface-to-volume ratio, ENMs are well suited for microfiltration applications and constitute attractive candidates for pollutant exclusion. Core–shell, porous, and hollow nanofibers with high porosity and low tortuosity could be synthesized using the process of electrospinning.41,42
Nevertheless, the utilization of polymeric membranes in wastewater treatment poses several significant drawbacks that must be addressed. These include challenges such as unstable morphology, limited reusability, and the potential for fouling, along with inconsistent or non-uniform filtration.43 These issues demand careful attention and innovative solutions to optimize the performance and durability of polymeric membranes in wastewater treatment processes. Furthermore, fouling control is another difficulty to be addressed in most of the membrane-based separation applications.44 Feed composition and foulant characteristics, as well as hydrodynamic circumstances, draw solution composition, and membrane orientation, all impact fouling in membranes.45–47 Antifouling modifications include polymer brushes, zwitterions, ultra-hydrophilic nanomaterials, etc.48–50 The integration of nanomaterials into pre-existing membranes is expected to enhance the efficiency of wastewater treatment due to their distinctive attributes, including but not limited to their hydrophilicity, thermal stability, surface roughness, hydraulic stability, heightened permeability, fouling control, and superior selectivity.4
This comprehensive review article offers a detailed examination of the selection and utilization of new and efficient adsorbent materials in the realm of membrane technology within the current global context. The article commences by analyzing the extent of the market and the notable influence of membrane technology in the context of wastewater cleanup. Subsequently, the focus shifts towards the methodologies employed in the manufacturing of membranes, with particular emphasis on the application of synthetic approaches. Significantly, there exists a specific focus on the development of environmentally friendly and readily scalable engineered nanomaterials. The study also provides a comprehensive analysis of recent developments in material selection and their subsequent application in hybrid membranes. Moreover, this current research explores the incorporation of diverse materials, such as transition metal dichalcogenides, (TMDs), graphene, and graphitic carbon nitride (GCN), and offers an in-depth examination of their respective influences. Moreover, this research extensively examines the utilization of reticular porous materials, specifically metal–organic frameworks (MOFs) and covalent organic frameworks (COFs), as advanced additives in membranes. The primary focus is on evaluating their efficacy in the removal of dyes and heavy metals from wastewater, as well as their potential application in desalination processes. Furthermore, this scholarly review article explores the matter of biofouling in membrane-based water filtration systems. This statement emphasizes the potential of zwitterionic materials as a promising solution to address and alleviate this difficulty. The present discourse provides an elucidation of the idea of biomineralization, emphasizing its importance in the process of water remineralization subsequent to water purification. This paper provides a comprehensive and instructive analysis of the advancements achieved in the utilization of fillers as material selections and the prospective trajectory of membrane technology in the field of water treatment.
The use of traditional chemical substances in the process of membrane manufacturing poses a significant threat to the preservation of the environment, hence raising concerns regarding the classification of membrane technology as an environmentally friendly separation alternative. Consequently, there exists a necessity to transition membrane production techniques towards environmentally sustainable alternatives, characterised by reduced utilization of hazardous solvents and adherence to principles of sustainability.51–55 Furthermore, in order to optimize the effectiveness of the separation process, a range of modifications and manufacturing techniques have been employed with regard to the membrane. Significant enhancements in membrane performance have been achieved by modifying surface properties using various techniques, including plasma etching,56 thermal crosslinking,57 and oxidation.58 Furthermore, the direct integration of various nanoparticles, such as MOFs,59 SiO2,60 TiO2,61 graphene oxide (GO),62 zeolites,63 and MXenes,64 during the electrospinning process has exhibited promising outcomes in terms of improving membrane rejection and augmenting flux.
Two factors distinguish water-treatment membranes: action mechanism and pore size (Fig. 1). Membrane separation technologies are classified as pressure-driven microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), reverse osmosis (RO), etc.65,66 In this regard, the estimated market size of MF membranes was expected to reach 12.8 billion USD in 2022, while the market for UF membranes is projected to reach 5.5 billion USD by 2025. In the same timeframe, the market size for RO membranes is anticipated to reach 13.5 billion USD, and the market for NF membranes is predicted to be around 950 million USD.67,68
Fig. 1 Comparison of the filtration technique spectrum of RO, NF, UF, and MF, based on size exclusion. |
The MF method is used to remove turbidity, color, and pathogens. The MF technique can reduce the greatest amount of turbidity (more than 97%) while restricting just around 10% of natural organic matter. UF can remove practically all microorganisms and suspended particles with pore dimensions ranging from 0.01–0.1 μm. Due to the wide pore size of MF and UF membranes, the transmembrane pressure required for operation is minimal, resulting in low energy consumption. For the treatment of surface and groundwater, MF and UF have been widely used. NF membranes are highly flexible, cost-effective, and simple to manufacture. Polymeric membranes and ceramic membranes are the two most common types of NF membranes. Polymeric membranes have a short lifetime due to their poor chemical resistance and high fouling rate. NF is known as a more advanced filtration method than UF; it can be used to separate small colloids to small molecular ions with pressures ranging from 4–20 MPa. RO is a type of reverse osmosis that operates at pressures ranging from 7–100 bar.69 RO pressure separates smaller molecules with a semi-permeable membrane (pore size 0.5–1.5 nm). RO removed 95–99% inorganic salts and charged organics. FO membranes strike a balance between selectivity and penetrating water flux. Semi-permeable membranes are used for FO feed/draw. Draw solutions have a higher osmotic potential than feed solutions. Water is transported from the feed solution to the draw solution by osmotic pressure differentials, which retain rejected solutes on the feed side and purified water on the draw side. Energy-saving FO does not require hydraulic pressure. For wastewater treatment, FO is safe and easy to clean and has a low fouling rate. FO is plagued by draw solution re-concentration, membrane selection, and internal and external concentration polarization. The membrane distillation (MD) method uses a microporous hydrophobic membrane and a temperature difference between the wastewater stream and the permeate stream to evaporate and condense water across the membrane. Only vapor can pass through MD pores, while other molecules are blocked. The MD method has been shown to remove over 96% of Ca2+, Mg2+, Fe3+, and Fe2+, and more than 99% of As3+ and As5+.
Membranes differ primarily in the material used to make them and the type of technique involved. Water treatment membranes can be made of a wide range of inorganic and organic materials, such as polymers, composite materials, and natural materials.70,71 There are two modes of membrane filtration, dead-end and cross-flow membrane filtration.72 In the process of dead-end filtration, the complete feed flow is propelled through the membrane, resulting in the accumulation of filtered materials on the membrane's surface. Dead-end filtration is considered a batch technique due to the fact that the filtration capability of the filter diminishes as materials accumulate on its surface. The presence of a consistent turbulent flow adjacent to the membrane surface hinders the build-up of substances in the process of cross-flow filtration. The membranes employed in this methodology often consist of tubular structures with an inner tube wall coated with a membrane layer. The flow velocity of the feed passing through the membrane tube is maintained at a high level in order to induce turbulent conditions and generate a significant pressure differential, which in turn facilitates the filtration process. The process is referred to as cross flow filtration due to the perpendicular orientation of the feed flow and the filtration flow. Crossflow filtration is a very efficient technique utilized for the filtration of liquid substances that possess a substantial concentration of filterable constituents.
Fig. 2a depicts the existing technologies for global water purification in terms of their economic turnover contribution to the water purification industry. Some of the reported removal efficiencies for a variety of heavy metals for four well-known technologies are shown in Fig. 2b. The entire cost per million liters of treated water is shown in Fig. 2c. Despite substantial variation in the statistics, it becomes clear that techniques including adsorption and ion exchange tend to give the most cost-effective options for water treatment. To offer a comprehensive explanation, it is imperative to differentiate between investments and operational costs, which are detailed in Fig. 2d and e, respectively. Based on the provided data, it is clear that the two examined techniques require the least quantity of initial capital investment out of the five under consideration. In terms of operational expenses, the sorption process is also notable for its cost-effectiveness. Fig. 2f depicts the average quantities of energy required for the operation of these techniques, which are supported by research. Consequently, the quantitative evaluation of emerging effluent technologies and the diversity of their sources constitute a major global concern.
Fig. 2 Current water purification technologies: (a) by global economic turnover by sector; (b) by efficiency (up to the first decimal) for a selected number of technologies and a small representative selection of heavy metal ions; cost (upper and lower limits) of wastewater and drinking water treatment technologies: (c) total cost per volume of treated water, (d) installation costs, (e) operating expenses and (f) energy consumption [reproduced from ref. 73 with permission from The Royal Society of Chemistry, copyright 2023]. |
Natural colors or dyes originating from plants and animals are thought to be environmentally friendly. The market was then revolutionized by the discovery of the synthetic dye mauvine in 1856. The industrial sector is anticipated to use more than 10000 different dyes and pigments, with over 7 × 107 tons of synthetic dyes generated annually.74 Synthetic dyes are widely available, inexpensive, and colorfast. Table 1 also contributes to the basic characteristics of today's salt and dye removal membranes.
Entry | Membrane | Membrane materials | Producer | Permeability (L m−2 h−1 bar−1) | MWCO (Da) | Salt rejection | Dye removal efficiency | Ref. |
---|---|---|---|---|---|---|---|---|
1. | NF 90 | Polyamide (PA) thin film composite (TFC) | Dow Filmtec | 5.3 | 200–400 | NaCl 90–96% | 100% Remazol Turquoise Blue G, Remazol Yellow GR and Lanaset Blue 2R | 77 |
MgSO4 98% | ||||||||
2. | DK | PA thin film | Osmonics | 3.3 | 150–300 | MgSO4 98% | 100% | |
3. | Desal-5 DK | Polysulfone (PSF)/PA | Osmonics | 5.1 | 150–300 | NaCl 17.1–47.8% | Reactive black 5 (RB5) | 78 |
4. | NF 20 | TFC | Sepro | 4.1 | NaCl 10.7–34.8% | — | 79 | |
5. | NF 270 | PS | Dow Filmtec | 13.3 | 270 | NaCl 7–21.8% | — | |
6. | ESNA-1-LF | meta-Phenylene diamine-based PA | Nitto Denko Hydranautics | 4.25 | 100–300 | NaCl 29–94.8% | — | |
7. | TFCSR | PS/PA thin film | Koch | — | — | NaCl 58–64.3% | — | 80 |
8. | TR 60 | PA spiral wound | Toray | 3.9 | 400 | NaCl 23–86.3% | — | 81 |
9. | Sepro NF 6 | PA | Ultura | 13.7 | 862 | NaCl 2.6–17.9% | 99% direct red 80, direct red 23, and Congo red (CR) | 82 |
10. | Sepro NF 2A | PA | Ultura | 10.5 | 493 | NaCl 6.9–33.3% | 99.9% | |
11. | K-5 | PA | KEKI, Hungary | ∼5 | 500 | — | 76% | 83 |
12. | NP010 | PES | Microdyn Nadir | >5 | 1000 | MgSO4 35–75% | 64.7% cyanidin-3 | 84 |
13. | NP030 | PES | Microdyn Nadir | >1 | 400 | MgSO4 80–95% | 89.8% cyanidin-3 | |
14. | NTR-7450 | Sulfonated PES | Nitto-Denko | 14.6 | 600–800 | Sucrose retention 80% | 95% molasses | 85 |
15. | TS-40 | PA | MEY | 5.2 | 400 | Sucrose retention 80% | 96% | 86 |
16. | SW30HRLE-400 | PA TFC | Dow Filmtec, USA | — | — | NaCl 99.8% | — | 87 |
17. | HB10255 | CTA hollow fiber | Toyobo, Japan | — | — | NaCl >99.4% | ||
18. | NFX | Proprietary PA TFC | Synder, USA | NaCl >40% | ||||
MgSO4>99 | ||||||||
Lactose >99% |
The use of artificial dyes is crucial in various sectors, such as textile manufacturing, dyeing processes, paper production, pulp industry, tannery operations, and paint applications. Dyes, which are organic substances, exhibit significant water solubility, hence presenting difficulties in their removal and control. Polar dyes, in particular, demonstrate a notable inclination to cling to polar surfaces of fabrics with a considerable level of easiness. According to Katheresan et al.,75 the textile industry accounts for approximately 55% of the dye effluents that are released. During the normal process of dyeing textiles, approximately 80% of the dye molecules are absorbed by the fabric while the remaining 20% are released into the surrounding water. The presence of pigments in water impedes the penetration of light, resulting in a decrease in the rate of photosynthesis and the concentration of dissolved oxygen, which impacts the entire aquatic ecosystem. Furthermore, dyes are carcinogenic and toxic substances that can spread throughout the food chain. Therefore, expedient measures and innovative technologies are required for the effective removal of dyes from effluent. It has been determined that adsorption is the most effective method for dye removal due to its high removal efficiency and low manufacturing cost. The adsorption process is dependent on the constitution of the dyes, which can be categorized as either ionic or nonionic. The efficacy of ion exchange depends on the adsorption and subsequent release of dyes.76
Another key socioeconomic challenge that developing countries must address is a sizable rural or distant population. One of the most significant impediments to the nation's adoption of such technology is consumers' capacity to acquire more advanced membrane-based filtering systems. Installation of regulated decentralized water treatment plants to serve a local population is one technique for addressing clean water shortages in rural areas. Furthermore, a pre-treatment system utilized before membrane filtration improves the efficacy and lifespan of the membrane by reducing fouling. Table 2 lists the various membrane systems used in developing countries. According to the literature, a range of over 200 synthetic and natural polymers have been utilized in the creation of nanofiber membranes through the electrospinning process.88 The aforementioned membranes have been employed in diverse separation and filtration applications, including but not limited to the elimination of dyes and heavy metals, separation of oil and water, capture of toxic gases, and removal of bacteria.89–91
Type of technology | Pre-treatment (capacity) | Nations | Source of water | Characteristics | Ref. |
---|---|---|---|---|---|
UF | NA (not applicable) (15536 ML d−1) | Malaysia | Surface water and groundwater | Cost-effective, 92–100% efficient heavy metal removal | 92 |
Gravity driven UF | NA (5000 L d−1) | South Africa | Groundwater | Elimination of all coliform bacteria (from 2419.2 to 7 cfu/100 mL), E. coli, and enterococci membrane demonstrates prolonged durability | 93 |
RO and desalination | Double media sand using 2.5 μm cassette filter | South Africa | Borehole water | RO TDS was reduced from 1292 ppm to 24 ppm, and nitrate-nitrogen from 42.5 to 0.9 ppm | 94 |
Low pressure UF | NA | South Africa | Surface water | Clean water requires 100–150 kPa pressure in UF systems. Effective turbidity and coliform removal | 95 |
AGMD (air gap MD) | MF (552 L per day) | Vietnam | Seawater | Using 87 kW h m−3 of water, produce 46 litres per hour of distillate without fouling or moisture issues in membranes | 96 |
MF, UF, NF, RO, FO and MD | Purification | Vietnam | Wastewater and seawater | A membrane is compact, modular, and efficient | 97 |
NA | |||||
NF and ED hybrid process | Electrodialysis | Vietnam | Surface water | Small water plants in rural Vietnam can use ED-NF. Water quality exceeds VN norms | 98 |
NA | |||||
Decentralized membrane filtration | Purification | Southern India | Household wastewater and membrane-filtered | Fecal coliform bacteria were reduced by membrane filters. The setup and maintenance costs are affordable | 99 |
NA | |||||
NF and RO | Adsorption and coagulation | India | Pesticide contaminated surface water | Hardness, chemical oxygen demand, total organic carbon, and microbes decreased with NF | 100 |
NF | NA | India | Arsenic contaminated water | NF eliminates arsenic by 99.80%, meeting WHO standards | 101 |
UF | Sand mesh of 25 μm and 150 μm | Mozambique | Freshwater | To meet microbiological regulations, the permeate tank must be post-chlorinated before distribution. Permeate flux has not changed | 102 |
Ozonation submerged ceramic MD and UF | Mesh with 50 μm (5 m3 h−1) | Thailand | Freshwater | Multi-step process ensures virus- and pathogen-free water. Compact treatment ideal for decentralized sites | 103 |
RO-desalination | Coagulation, degasification, and a dual-media filter | Indonesia | Brackish water | Solar or reverse osmosis can treat groundwater. Modular methods that are cost-effective and easy to modify and maintain | 104 |
Hollow fiber UF | Purification (100000 m3 d−1) | China | Reservoir | No organic leakage in 7 year operation, confirming UF membrane effectiveness | 105 |
UF | Coagulation | China | Raw water | Effectively eliminates metals, coliforms, and turbidity | 106 |
NA | |||||
RO and NF-desalination | NF | Turkey | Seawater | NF as pre-treatment for seawater RO solves scale, cuts costs, boosts yield | 107 |
NF | Precision filter (20 m3 d−1) cation exchange resin, sand filters | Sri Lanka | Ground water | In permeation, NF plant reduces hardness, fluoride, and organic carbon | 108 |
RO-desalination | NA | Brazil | Brackish water | The desalinated water had 94% rejection rates for SO42−, 97% for TDS, and 100% for F− | 109 |
The fundamental driving force behind the creation of high-performing and durable membrane materials is the fine-tuning of surface shape and/or chemistry through in situ or ex situ alteration processes. The alteration of membrane materials takes place simultaneously with the production of the membrane in the process of direct membrane material development. Following the formation of the membrane, the subsequent step involves the alteration of the said membrane. Enhancing the wettability of a membrane, as quantified by the contact angle exhibited by a liquid droplet (such as water or oil) placed on the membrane's surface, is the principal strategy employed to alter its surface chemistry. The significance of membrane porosity and pore size distribution holds considerable importance within the realm of efficient wastewater treatment. The performance of the membrane is affected by a range of factors associated with the chemical properties and composition of the feed, including pH, ionic strength, and the presence of foulants. Furthermore, there are concerns regarding the wettability and antifouling properties of the membrane.
The total membrane resistance (Rtot) is used to describe fouling in membranes.
Rtot = Rm + Rc + Rp + Ra + Rcp |
Fig. 3 Schematic representation of (a) electrospun nanofibers with varying (hollow, functionalized and core–shell) structures, (b) PI process, (c) track etching process and (d) IP process. |
In 1963, the invention of electric current and the formation of fibres from viscous liquid using an air stream assisted route were reported.111 One of the factors considered in the electrospinning technique is electric current. The electric current is determined by the movement of the charge carrier through the electrospinning distance. Furthermore, most charge carriers are obtained by ionising the air and exposing it to the sharp needle and polymer jet solution. The concentration of charge carriers is measured using the amount of salt in the polymer solution. Some independent parameters influence the electric current in the jet, including solution feed, relative humidity, applied voltage, solution conductivity, needle (hollow) diameter, and a few geometrical properties. The current in a jet can be calculated by using the theoretical equation mentioned below:
Itotal ≈ EQ0.5K0.4 |
Recent developments in the field have demonstrated that both pristine and functionalized electrospun polymeric nanofibrous membranes have enormous potential in water treatment applications. These membranes may consist of natural and synthetic polymers, functionalized polymers, and (inorganic/organic) additives.
In light of the current global environmental crisis and the escalation of water, it is essential to develop innovative ENMs with enhanced efficacy. Electrospinning has a number of limitations that must be addressed in order to address environmental concerns. These limitations include bulk productivity, tip obstruction during fabrication, unwanted repulsive forces leading to waste in the case of multiple syringes, electrostatic charge disturbance, and the requirement for real-time active system monitoring of parameters to ensure reproducibility. This matter has been highlighted in recent studies by Li et al.112
Fuat Topuz et al. fabricated a very effective adsorbent platform composed of hyper crosslinked cyclodextrin networks (HCNs) based on ENMs. After being treated with acidic methanol, the HCN membrane was employed to scavenge textile dyes and polycyclic aromatic hydrocarbons from polluted water, and it displayed great sorption performance (Qmax = 692 mg g −1 dye) and remarkable reusability.121
Additionally, polymeric ENMs are also found to be excellent adsorbents for toxic heavy metal adsorption. For example, Li et al. created a chitosan-based poly(methyl acrylate) electrospun membrane for the removal of chromium Cr6+ from polluted water in another research study.122 Aliabadi et al. developed a hybrid mesoporous membrane of polyethylene oxide (PEO) and chitosan nanofibers using the electrospinning technique.123 The study showed that the high surface area (312.2 m2 g−1) of the PEO/chitosan-based ENMs was responsible for the selective adsorption of heavy metal ions. The measured removal efficiency was in the order 89%, 82%, 72%, and 68% for Pb2+ < Cd2+ < Cu2+ < Ni2+, respectively. The charge density of metal ions decreases as the ionic radius increases. As a result, the number of active sites on the membrane available for adsorption decreases, leading to a decrease in the efficiency of removal. Yang et al. prepared a functionalized chitosan electrospun membrane using poly(glycidyl methacrylate) (PGMA) and polyethyleneimine (PEI) and utilized it for heavy metal removal.124 The study showed maximum adsorption capacities of 138.96, 69.27 and 68.31 mg g−1 for Cr6+, Cu2+, and Co2+, respectively. Zhao et al. developed a branched PEI embedded polyacrylonitrile (PAN) nanofibrous membrane using electrospinning.125 The study showed potential for Cr6+ removal from water with improved hydrophilicity of the membrane. An adsorption capacity of Cr6+ as high as 637.46 mg g−1 could be obtained. The adsorption capacity increased along with decreasing fiber-diameter. In another study, Zhao et al. prepared a phosphorylated PAN membrane using electrospinning and further modified it using chemical grafting.126 The membrane exhibits high adsorption affinity towards selected heavy metal ions. The adsorption amount was calculated to be 98.06, 78.03, 102.40 and 18.89 mg g−1 for Pb2+, Cu2+, Ag+ and Cd2+, respectively. Karim et al. used the electrospinning method to create ENMs from PVA/chitosan (PVA/Chi) for selective and high adsorption of Pb2+ and cadmium Cd2+ ions depending on the acidity of the solution.127 The manufactured ENMs have proven to be a superior material due to their high functionality and numerous active sites for interaction with Pb2+ and Cd2+ ions that provided a high adsorption capacity of 266.0 and 148.0 mg g−1, respectively.
Hence, along this line of discussion, it could be well understood that ENMs can act as superior adsorbents owing to their capillary action. Furthermore, the tunability of their fiber diameter along with the wide choice of polymeric backbone makes them highly desirable in the field of water purification. Additionally, it is simple to functionalize the pristine polymeric ENMs to create hybrid membranes that can target a particular pollutant to be adsorbed from wastewater sources in addition to amplifying their physicochemical qualities.
For example, novel hollow α-Fe2O3 nanofibers of rice-like nanorods were successfully fabricated by Gao et al. on an electrospun PVA nanofiber template using a simple hydrothermal process followed by calcination.135 Furthermore, hollow α-Fe2O3 fiber arrangements exhibited a magnetic response as well as effective methyl orange (MO) dye adsorption in water. Du et al. created a conductive CNT nanofiber hybrid hollow fiber membrane (CNT-HFM) by coating and crosslinking CNTs on an electrospun PAN hollow nanofiber support layer with considerable permeability and mechanical strength enhancements.136 The CNC-HFMs had 3.06 and 12.7 times the tensile stress and Young's modulus of commercial phase inverted HFMs. The CNC-HFMs also had 7.3 times the permeability of commercial materials. The membrane removed turbidity and total organic carbon (TOC) efficiently. According to the study, the synergy of positive electro-assistance and single membrane separation reduced fouling on the CNC-HFMs and enhanced turbidity and TOC removal efficiency. Another study by Xizi Xu fabricated MgO/PAN functional core–shell nanofibers for the removal of copper with a maximum adsorption capacity of 354 mg g−1 at pH 5.137
Mousa et al. used the electrospinning process to create a coaxial nanofibrous membrane.138 In their study, they employed hydrophilic cellulose acetate (CA) as the core and hydrophobic polysulfone (PSf) polymers as the shell for creating a core–shell membrane, supplemented with 0.1 wt% ZnO nanoparticles. The membranes were also treated with 2 M NaOH to increase hydrophilicity and hence water separation flow. The mechanical parameters of the polymeric membranes in terms of Young's modulus, tensile strength, and toughness are higher than those of the NaOH-treated membranes. Furthermore, the membranes have strong antimicrobial properties and have successfully isolated water from oil–water effluent. Moreover, when compared to untreated membranes, the water flux for the modified membranes was increased by 1.6 times. Tijing et al. designed and prepared a superhydrophobic PVDF-co-hexafluoropropylene nanofiber membrane doped with CNTs by the electrospinning process.139 The mechanical strength and hydrophobicity of the membrane depended on the concentration of CNTs. These properties were controlled by different concentrations of the incorporated CNTs. The 5% CNT concentration showed the highest water flux (24–29.5 L m−2) and high salt rejection under an external pressure of 99 kPa. Deng et al. synthesized a novel nanocomposite MWCNT–PEI embedded with PAN nanofibers.140 Their study showed higher mechanical strength and improved hydrophilicity of the membrane with 145.8 L m−2 h−1 water flux. The membrane exhibited more than 93% removal efficiency of Cu2+ and Pb2+ heavy metals.
Regarding the fact that electrospinning technology offers a fascinating perspective on the creation of novel nanofiber-based materials, a substantial obstacle remains. The current limitations of electrospinning in terms of providing reliable processing of nanofibers have resulted in instability and inconsistency. As a result, there are significant limits on existing developments and the range of spinnable fluids. Due to the presence of these impediments, there exists an important reason to enhance manufacturing efficiency with the aim of mitigating the interconnected issues posed by macro-level production limitations.
With the PI method, Sharma et al. fabricated a highly efficient graphene-oxide doped CA-based IEM.143 Cation exchange membranes' ion exchange capacity (IEC), fixed charged density, water uptake, swelling degree, and proton conductivity were optimized by adjusting the compositional ratio of GO/CA/polyethylene glycol. The maximum GO content led to a membrane with a proton conductivity of 0.273 S cm−1 and an ionic exchange capacity of 1.08 m mol g−1 at 30 °C (CAG3). The WCA of the modified CAG3 membrane was only 53 degrees, making it more hydrophilic. The rejection rate was low for cationic dye (20%) but high for anionic dye (92.5%). Improvements in acid recovery and selectivity for anionic dye indicate the development of a more efficient IEM with increased proton conductivity. PI was used by Bhran et al.144 to create a membrane using PVC and PVP. Increased membrane wettability and decreased filter pressure are what the hydrophobicity results show. This may be because the membrane in question has a lower WCA, a factor known to significantly impact the drop in capillary pressure experienced by filter media. Salt rejection (NaCl) by the membranes was measured at 98% at a very high flux.
Using the PI method, Aji et al. created an UF membrane from recycled polyvinyl chloride.145 In addition to a BSA rejection rate of over 90% for the neat PVC membrane, the results showed that the membrane had a PWF of 85 L m−2 h−1. The WCA was reduced by about 25%, the hydrophilicity was improved, and the equilibrium water content was increased by about 19%. Han et al. created Mg(OH)2/PES (magnesium hydroxide/polyethersulfone) hybrid membranes using synthetic PI.146 The hybrid Mg(OH)2/PES membrane showed the best performance, the pure water flux increased from 430 to 720 L m−2 h−1 MPa, the egg albumin rejection rate was 94.58%, and the recovery flux ratio was around 70%. Another study from Masheane et al. showcased the fabrication of Fe–Ag/functionalized-multiwalled carbon nanotube (Fe–Ag/fMWCNT/PES) nanostructured hybrid membranes via modified PI.147 The addition of Fe–Ag/f-MWCNTs to the PES polymer boosted the membrane hydrophilicity, thermal stability, crystallinity, and fouling resistance. Nevertheless, the Fe–Ag/f-MWCNTs did not leach from the PES membranes. In crossflow systems, adding Fe–Ag/f-MWCNTs to the PES polymer matrix increased water flux from 26.5 to 36.9 L m−2 h−1 and improved Cr6+ ion rejection up to 94%.
Korolkov et al. demonstrated that hydrophobic TEMs with pore sizes of 180 nm and thicknesses of 12 mm based on polyethylene terephthalate (PET) were prepared using photo-induced graft polymerization of silicon monomers such as triethoxyvinylsilane (TEVS), resulting in a WCA of up to 104°.154 Korolkov et al. performed additional hydrophobization of PET TEMs by polycondensation reaction on the membrane surface via soaking in trichloro(octyl)silane solution for separation of oil–water emulsion.155 The best pressure for maximal flux and separation efficiency was 700 mbar. Membranes with pore diameters of 350 nm and vacuum pressures of 700 mbar demonstrated fluxes of 305 mL m−2 h−1 for chloroform–water emulsions and 75 mL m−2 h−1 for cetane–water emulsions. Choi et al. fabricated track etched porous MF membranes for filtration which overcome the fouling issue in a submerged MBR system.156 They fabricated three different membranes PETE, PCTE and PTFE. Their result shows that the filtration resistance increased the fastest for the PETE membrane with the lowest PWP value, and the slowest for the PCTE membrane with the intermediate PWP value. With the greatest PWP and a microstructure resembling a sponge, the PTFE membrane demonstrated a faster increase in filtration resistance than the PCTE. This was probably because of the PTFE membrane's rougher surface, which facilitated the foulant's quicker deposition. Zdorovets et al. prepared a hydrophobized PET track etched membrane for low level radioactive waste treatment and salt rejection.157 UV-induced grafting was used to modify PET track etched with styrene and TEVS. The alteration led to an increase in the wetting contact angle (WCA) of polyethylene terephthalate (PET) track-etched membranes to 99° when the pore size was increased from 150 to 300 nm. The results indicate that the rejection rates for Cs, Sb, Mo, Al, Ca, Sr, Mg, Na, K, and Fe ions were consistently high, approaching 100% in the majority of cases.
Based on the aforementioned instances, it is evident that the utilization of diverse fabrication techniques in the creation of unspoiled membranes shows potential for their application in water restoration activities. Nevertheless, via meticulous selection of tailor-made materials, there is potential for substantial enhancement in both performance and fouling characteristics. Therefore, the integration of high-performance materials with self-supporting membranes may serve as a crucial factor in the development of cost-effective and exceptionally efficient water treatment systems.
In this regard, the incorporation of organic/inorganic NAs into the polymer matrix has led to the development of a new generation of membranes.159 The nanoparticles (NAs) possess distinct features that confer selectivity towards contaminants, while the polymer matrices serve as a supportive medium for the NAs. The behaviour of numerous interfacial atoms distributed as nanoparticles (0D, 1D, 2D) at the interface effects the behaviour of the bulk material in a system. In nanoparticles, this behaviour manifests as changes in optical, magnetic, thermodynamic, thermomechanical, electrical, and structural properties. As a result, the intended electrical, thermal, mechanical, and rheological properties of the targeted nanocomposite may be increased based on the size, shape, and composition of these nanomaterials, as well as their interactions with the host polymeric matrix. In total, adding nanoparticles to the membrane may help lower the amount of energy required, the quantity of chemicals necessary to clean the membrane, and the total operating cost. Nonetheless, they face a number of challenges, including fouling, limited flux, and the trade-off between flow and selectivity.
TiO2 and silver (Ag) with their 0D nanostructures are among the most frequently investigated metals and metal-oxides in order to address energy and environmental challenges.160–162 Peydayesh et al. constructed a composite NF membrane from polyethersulfone (PES) and ethylenediamine (ED) grafted MWCNTs.163 The membrane surface roughness, hydrophilicity, isoelectric point, and thermal and mechanical properties were optimized using different loading concentrations of the CNTs. The hybrid membrane reveals excellent heavy metal removal in the order Zn2+ (96.7%) > Mg2+ (95.01%) > Cd2+ (92.4%) > Cu2+ (91.9%) > Ca2+ (91.3%) > Ni2+ (90.7%) > Pb2+ (90.5%). Additionally, the water flux of the hybrid membrane was enhanced by 122% compared to the pure PES membrane and achieved around 80.5 L m−2 h−1. In another study, Orudzhev et al. produced a novel piezophotocatalytically active membrane based on PVDF nanofibers with α-Fe2O3 nanoparticles integrated into them via electrospinning.164 Immobilization of α-Fe2O3 nanoparticles in PVDF nanofibers improves crystallization and promotes increased self-polarization of PVDF nanofibers into the electroactive phase due to ion–dipole interphase interaction. The rate of MB breakdown during piezophotocatalysis employing α-Fe2O3/PVDF increased 4.7 times when compared to piezocatalysis due to the piezophototronic effect. Xiao-Qiong Wu et al. established an effective approach for fabricating a high-performance triple-layer (PVDF-HFP and SiNPs) superhydrophobic/hydrophobic/hydrophilic membrane for potential practical MD applications for industrial wastewater treatment.165 The triple-layer membrane exhibited stable MD performances when using real seawater and industrial flue gas desulfurization wastewater as the feed solutions, while no obvious fouling and wetting were observed even at 60% water recovery. In contrast, the TL membrane maintained a stable water flux of ∼15.1 kg m−2 h−1, indicating a better fouling resistance towards inorganic scalants of the TL membrane than that of the DL membrane. The WCAs of the TL membrane remained above 150° even after 1 h ultrasonication or 40 abrasion cycles. Similarly, Chenyi Fang et al. prepared a PVDF membrane where the surface chemistry was altered by Teflon coating that showed high WCA (142.5 ± 2.3°), low water sliding angle (25.8 ± 3.8°) and fouling/scaling resistance in the presence of salt and surfactants.166 In another study, the interlayer of the TFC membrane is improved by incorporating doping sulfonated polyaniline (SPANI) nanofibers. These nanofibers exhibit exceptional hydrophilicity and distinctive chemical characteristics. The membrane exhibits a commendable permeate flux of 35.35 L m−2 h−1 bar−1, along with satisfactory retention capacities for Na2SO4 (98%) and MgSO4 (95%).167 Cong Yang et al. fabricated a solely green TFC system that contains a biodegradable green electrospun support layer (PLA), green solvents, and green monomers (genipin–priamine) using a scalable electrospinning technique. A gelatin interlayer between the PLA support and the selective layer (genipin–priamine) was utilized to improve the compatibility of these two layers.168 Further, the robustness of the fabricated PLA support was enhanced by hot annealing, as the tensile strength and Young's modulus of the support increased from 1.62 to 4.55 MPa and from 25.36 to 60.12 MPa, respectively. The prepared green TFC membrane also exhibited excellent oil removal efficiency (99.6%) with a water permeance of 5.6 L m−2 h−1 bar−1. The incorporation of zero-dimensional (0D) and one-dimensional (1D) nanoparticles into the polymeric membrane matrix has been attempted to improve its performance. However, the lack of appropriate interfacial compatibility between the nanofillers and polymers can lead to the formation of defects, which in turn create non-selective channels. As a result, achieving a high level of selectivity becomes challenging. A noteworthy attribute of these entities is their proclivity to aggregate upon dissolution, leading to a substantial reduction in surface area. However, the objective of attaining accurate separation of molecules with similar sizes from a wide variety of solutes is a considerable obstacle when utilising zero-dimensional (0D) and one-dimensional (1D) nanomaterials. Therefore, there exists a significant need for the development of novel and advanced membrane materials that demonstrate both rapid and accurate ion/molecular transport. In the context of this paradigm, 2D nanomaterials possess structures that are atomically thin and demonstrate potential in terms of low material utilisation, quick manipulation, and extensive treatment adaptability.
In TMDs, two chalcogen layers are divided by a transition metal atomic layer, resulting in a sandwich-like structure.169 Foreign molecules can easily intercalate between the layers due to their weak interlayer van der Waals contact. As a result of the simple permeability of guest molecules across TMD interlayers, efforts have been made to develop TMD-based hybrid membranes.170 In this regard, TMD materials based on MoS2 and WS2 have shown tremendous promise in hybrid membrane-based water treatment systems. For example, Liang et al. used a reverse atom transfer radical polymerization method to prepare a MoS2-PSBMA/PES composite membrane using the PI method.171 The zwitterionic PSBMA compound modified the hydrophilicity of the MoS2 nanosheets as fillers. This approach resulted in a water flux of 108.3 L m−2 h−1 at 0.6 MPa pressure. Furthermore, dyes like reactive black 5% and reactive green 19 could be efficiently rejected from water with an efficiency of 98.2% and 99.3%, respectively. However, the fabricated membrane showed insufficient rejection of salts. In another study, Yang et al. fabricated an oxidized MoS2 (O-MoS2) based TFN membrane using a PA layer for salt (Na2SO4, MgSO4, MgCl2 and NaCl) rejection from water.172 The few-layered O-MoS2 nanosheets with strong hydrophilicity and negative charge provided a theoretical foundation for developing NF membranes with good selectivity, permeability, and antifouling performance. TMD-based 2D hybrid membranes could also be beneficial for the adsorption of heavy metal ions from water owing to their chalcogen rich atomic layers. For instance, Zhao et al. prepared a MoS2 nanosheet decorated PVDF membrane to remove toxic Hg2+ ions from water with a maximum adsorption capacity of 578 mg g−1.173 In addition, the study revealed the role of sulfur in such a high adsorption capacity.
Apart from TMDs, carbon based 2D materials like pristine graphene and its analogues GO and reduced graphene oxide (rGO) are extensively studied for their use in hybrid membranes to mitigate water contaminants.174 Their excellent robustness, tunable hydrophilicity, interlayer transport channels and facile surface functionalization make them appealing as fillers in hybrid membrane technology. However, tunability in surface functionalization, hydrophilicity and surface charge is challenging for pristine graphene as compared to oxidized graphene. Furthermore, large-scale graphene production is complex, whereas industrial-scale GO production is feasible. Hence, research related to GO based hybrid membranes is more explored than that related to pristine graphene-based membranes. For example, Liu et al. fabricated about 100 nm thick layers of GO based nanocellulose hybrid membrane to reject more than 90% positively and negatively charged dye molecules (Victoria Blue 2B, methyl violet 2B and Rhodamine 6G) with a water flux as high as 18123 ± 574 L m−2 h−1 bar−1.175 Additionally, the membranes exhibited significantly improved mechanical stability in both dry and wet states and exceptionally high water flux. Another study from Tan et al. demonstrated the capability of GO based membranes in heavy metal removal applications.176 According to the study, PVA/GO@PTFE hybrid membranes could effectively separate toxic metal ions like Cu2+ and Cd2+ from water with 72.6 and 83.8 mg g−1 adsorption capacities with six regeneration cycles. Moreover, GO based hybrid membranes can efficiently reject salts from water, featuring them as promising candidates for desalination.177
Another type of 2D material is GCN, which has a sp2 layered structure with weak van der Waals forces.178 GCN is made up of two essential structural (triazine and heptazine) units made up of N-heterocyclic rings. The principal components are carbon and nitrogen, and at least certain portions of the structure can be compared to elongated planes of graphite. When compared to other carbon-based materials, GCN delivers outstanding properties in terms of tunable band structure, nitrogen rich sites, tunable defects and tailor-made functionalization.179 Furthermore, the evenly distributed tiny triangular nanopores (3.11 Å) in GCN allow for the preferential transport of small molecules such as H2O (2.65 Å kinetic diameter). As a result, GCN has been regarded as an advanced material with potential applications in membrane based water remediation.180 Fabrication techniques such as vacuum filtration, dip coating, blending, IP, 3D printing, and electrospinning could all be used to fabricate GCN-based membranes. Nevertheless, the relationship between the GCN morphology and membrane performance was not established in prior research. In this regard, Ge et al. constructed three different g-C3N4 nanostructures including 1D nanorods, 2D nanosheets, and a 1D/2D nanohybrid, to fill a PA layer and create hybrid TFN membranes. The 1D/2D nanohybrid like structure aided in creating more open channels. Moreover, g-C3N4 nanohybrid membranes exhibited good antifouling properties in response to BSA and silica foulants.181 In a study conducted by Shahabi et al., similar outcomes such as enhanced surface roughness and hydrophilicity improved the salt rejection efficiency of GCN-based hybrid membranes. With only 0.015 wt% of GCN nanosheets as fillers in the PA membrane, it was possible to achieve a NaCl rejection performance of up to 99.7%.182 On the other hand, owing to its rich nitrogen-rich structure GCN can improve membrane hydrophilicity, surface charge density and adsorption sites. In this regard, the adsorptive capacity of pristine membranes could be enhanced using fillers like GCN.183 Very recently, Nadig et al. fabricated a PSF based MMM using 2D nanosheets of GCN to separate heavy metal ions like Pb (95%), Cd (80%) and As (70%) from water.183
Apart from these examples, 2D materials like hexagonal boron nitrides, black phosphorous, layered double hydroxides, and MXenes have also gained a lot of attention due to their outstanding physicochemical properties.184–186 In general, recent developments in hybrid membranes based on their designs, morphologies, physical properties, and chemical characteristics display a range of performance enhancements, including enhanced water permeability, rejection of pollutants and salts, antichlorine behavior and antifouling activity. Unfortunately, despite their intriguing membrane separation properties, their ability to achieve a higher flux is severely limited by their inherent nonporousness and the ensuing excessively long permeation pathways along the nanosheet interlayers. This is primarily due to lack of vertical transport pathways which limits their water permeability performance. Although drilling with chemical etching or electron beams has created pores in these 2D materials, the pore-forming conditions are too harsh for practical use. Porous materials with tunable nanopores facilitate mass transport of smaller molecules such as water, whereas they act as molecular sieves to reject pollutants. Thus, it is anticipated that adding reticular porous materials to the polymer matrix will significantly increase the water flux and selectivity of traditional polymeric membranes. In 2019, Jiang and Wu's group developed mixed nanosheet membranes using GO and covalent triazine frameworks (CTFs) to reject organic dyes (>90%) from water.187 According to their report, exfoliated CTFs were chemically grafted with GO by forming amide bonds between amidine terminated CTFs and COOH terminated GO to form GO-CTF membranes. The GO-CTF membranes showed a thickness of 32 nm in the layered configuration that exhibited a 12-fold increment in water flux compared to pure GO membranes. As a result, novel and ground-breaking materials with intrinsically porous structures could further enhance the water flux and may provide better permeation pathways.
Fig. 4 Schematic representation of the reticular-porous material-based synthetic advantages, as well as the molecular sieving mechanism in the pore channels. |
MOFs are a fascinating class of porous materials with numerous uses in a variety of fields.23 The synthesis of MOFs typically involves mixing metal salts and organic ligands in a suitable solvent under controlled conditions, such as temperature, pH, and time, resulting in a three-dimensional framework with high surface areas and tunable pore sizes.203 A number of techniques, including X-ray diffraction and gas sorption studies, can be used to better characterize MOF crystals. MOFs have shown potential as heterogeneous catalysts, gas storage materials, drug delivery vehicles, and sensing and imaging agents, among other applications. MOFs also represent a significant breakthrough in realizing the prospects of harvesting water from air. Notably, Yaghi et al. demonstrated that certain MOF materials, when utilized in an electrified device with frequent operation, can achieve a tenfold enhancement in water productivity.204 This research finding highlights the potential of MOFs as a promising technology for addressing global water scarcity issues. Ongoing research in MOF synthesis and applications is expected to lead to further advances and innovations in these fields.
Apart from inorganic–organic porous hybrids like MOFs, a class of highly organized, porous, and crystalline materials known as COFs are made of organic building units that are connected by covalent bonds.24 As the glue between the knots and linkers is made up of covalent bonds, the kinetics of the bond formation is faster, resulting in mostly amorphous polymeric networks. Nevertheless, the choice of linking covalent bonds is dynamic in nature and thus brings reversibility in the reaction media via dynamic covalent chemistry,205 yielding crystalline porous solids via the error–correction mechanism.206 COFs have gained increasing attention in recent years due to their unique properties, including organic skeletons, high surface area, tunable pore size and shape and pore-interface, along with excellent stability under various conditions.207 These frameworks are identified by their precise spatial structures and distribution of building blocks, forming either extended two-dimensional (2D) or three-dimensional (3D) frameworks. COFs can be synthesized through different methods, including Schiff base condensation, boronate ester formation, and beyond.208 In addition to these common linkages, other types of linkages have also been used for COF synthesis, including triazine formation, acetylene linkage formation, etc.208 The choice of linkages used for COF synthesis can influence the properties and performance of the resulting COF, making it an important consideration in COF design. These methods offer flexibility in controlling the structure and properties of COFs, enabling the design of materials for specific applications.
The formation of pores in COFs is intricately linked to the stacking of 2D layers in the case of 2D COFs or the folding and interpenetration of 3D networks in 3D COFs. These architectural features are primarily governed by non-covalent interactions, including π–π and hydrogen bonding interactions. Such interactions play a crucial role in tailoring the properties of COFs for various applications. In the realm of 2D COFs, the face-to-face stacking of sheets (AA, AB, or ABC mode) offers an interlayer distance ranging from 3.2 to 5 Å.207 This stacking arrangement allows for precise control over the interlayer spacing and thus influences the resulting porosity. Meanwhile, in 3D COFs, the intricate folding and interpenetration of networks contribute to their unique porous structures. However, the design and synthesis of 3D COFs pose challenges due to the high cost of knots/linkers and the complexity of achieving desired structural arrangements.
Researchers have investigated the functionalization of COF backbones or skeletons with various functional groups or heteroatoms to improve the function of COFs.207 This approach not only enables the specific adsorption of guest molecules or ions but also allows for the development of biomimetic systems, reminiscent of ion-pump channels such as Na+, K+, and Ca2+. This tailored functionalization opens up exciting opportunities for applications ranging from water remediation to catalysis and gas storage. Given their selective adsorption of gases like carbon dioxide, methane, and hydrogen, COFs have been proven to exhibit remarkable gas storage and separation capabilities.209 Additionally, COFs have been explored as heterogeneous catalysts, where their high surface area and ordered structure allow for efficient mass transport and high catalytic activity.210The chemical tunability, highly π-conjugated electronic structure and high surface area of COFs also enable their use in selective sensing applications for the detection of various analytes.211 Finally, because of their improved charge transport and accessible surface area for electrochemical reactions, COFs have been investigated as electrode materials for energy conversion as well as storage devices, such as batteries, supercapacitors, and fuel cells.212 However, the scope of this review article is limited to water purification. Hence, detailed discussions pertaining to other applications are beyond the scope of this article.
While both 2D and 3D COFs hold promise for a wide range of applications, the extensive exploration of 2D COFs in water remediation applications is primarily driven by practical considerations. The cost of knots/linkers used in constructing 3D COFs and the challenges associated with tailor-made design hinder their widespread application. However, ongoing research efforts continue to tackle these obstacles, paving the way for future advancements in 3D COFs.
In the future, MOFs and COFs have the potential to revolutionize the field of water remediation by providing new and innovative solutions to address pressing environmental challenges. The unique structural properties of MOFs and COFs offer several advantages for water treatment applications, including high surface areas, tailored pore sizes, and selective adsorption properties. These materials can selectively adsorb organic pollutants, including dyes, pesticides, and pharmaceuticals, due to their high surface area and functionalized pore structures. Additionally, MOFs and COFs can be modified to enhance their adsorption properties, providing new opportunities for the development of more efficient and selective adsorbents. Moreover, MOFs and COFs can be designed to have high stability and reusability, which can reduce the cost and environmental impact of water treatment processes. Looking ahead, further research is needed to fully explore the potential of MOFs and COFs in water remediation applications. This includes the development of new MOFs and COFs with tailored properties for specific water treatment applications, as well as the optimization of existing materials to enhance their performance and reduce their cost.
Fig. 5 Schematic illustration of design strategies for MOF-based membranes in liquid separation: (a) in situ preparative route; (b) blending method; (c) IP process [reproduced from ref. 226 with permission from The Royal Society of Chemistry, copyright, 2023]. MOF membrane fabrication methods. (d) One-step growth of ZIF-8 membranes on an AAO support using PDA.11. (e) Secondary development of ZIF-8 membranes (in situ). (f) ZIF-8 membrane counter-diffusion growth. (g) Asynchronous development of ZIF-8 membranes with a sol–gel method. (h) ZIF-8 membrane fabrication using the atomic layer deposition (ALD) technique. (i) Gel-vapor deposition for the manufacture of ultrathin ZIF-8 membranes. (j) An electric current drives the production of ZIF-8 membranes. (k) Microfluidic interfacial processing of ZIF-8 fibers [reproduced from ref. 227 with permission from The Royal Society of Chemistry, copyright, 2023]. |
Another route to obtain continuous MOF films is the liquid phase epitaxy method (e.g., layer-by-layer method) resulting in surface-attached MOFs (SURMOFs).216,217 In this method, a substrate is repeatedly exposed to the metal precursor solution and ligand precursor solution (Fig. 5c) alternately with intermedial washing. Although a higher degree of control and a smaller number of defects could be achieved by this method, SURMOF films suffer from scalability issues. Additionally, the interfacial diffusion technique has also been employed for the selective growth of continuous MOF films.218,219 In this technique, precursor solutions of both the metal and organic linker are introduced to the two opposite sides of a porous substrate (Fig. 5). Hence, the precursors meet each other by diffusion through the voids of the substrate and react, resulting in a MOF film at the interface. Furthermore, the MOF nucleation and growth only takes place at the interface of the two precursor solutions forming continuous films with improved uniformity. So far, this technique has been mostly applied to MOFs (ZIF-8, ZIF-71 and HKUST-1) that are formed under ambient conditions.
Aggregation of MOF crystallites takes place when the loading% of MOFs is higher in the MOF based MMMs, which leads to low water flux and solute rejection performance, swelling and poor long term operational capability. To address this problem, there should be a compatibility between the polymer and MOF crystallites present in the MMMs. Further, the hydrolytic stability of the MOFs is another factor to consider. Originating from the labile nature of metal–ligand coordinative bonds, the sensitivity of most of the MOFs to water or moisture is a well-known problem to reticular chemists. In order to address water sensitivity issues, researchers have designed MOFs with hard–hard or soft–soft interactions between SBUs and ligand donor sites. For example, Zr4+ with oxygen donor sites of carboxylates (UiO series, CAU series, etc.)228–230 and Zn2+ with azole-based (imidazole, triazole, tetrazole) nitrogen donor sites [zeolitic imidazolate frameworks (ZIFs)] are evident in the literature.231 MOFs with these attributes are more desired for water treatment-based applications. Nevertheless, post synthetic modification (PSM) has also been used to modulate the key properties of MOF based membranes in water treatment applications.232–236 To summarize, water stable MOFs that could be used as continuous membranes or fillers in MMMs are generally applied for water treatment applications such as dye removal, heavy metal removal and desalination of saline water sources based on their NF, UF and RO-based mechanisms, respectively. Selective and careful design of monomers results in variation of pore windows and diameter of MOFs. So far, pore windows of MOFs may vary from 0.3 nm to 10 nm. Hence, MOF membranes can act in the nano/UF regime and could act as RO membranes.
Fig. 6 SEM images of (a) ZIF-8, (b) ZIF-L (no CTAB), (c) ZIF-L (CTAB, 1×), (d) ZIF-L (CTAB, 2×), and (e) ZIF-L (CTAB, 8×); (f) molecular structures and dimensions of the adsorbents and adsorbate; (g) water permeability during the dynamic adsorption of RB; (h) suggested microstructure of the membrane (ZIF-8 at the top, ZIF-L without CTAB in the middle, and ZIF-L synthesized with CTAB at the bottom) [reproduced from ref. 240 with permission from The Royal Society of Chemistry, copyright, 2023]. |
In 2018, Chi et al. identified the key factors affecting the dye adsorption performance of ZIF-L.242 The authors found that, during sorption free imidazole molecules from ZIF-L leached into water and exhibited the highest adsorption capacity for RB dye. Zhu et al. constructed a PSS modified ZIF-8 based TFN membrane in a PA layer using the IP method.243 The as-synthesized membrane showed ultrahigh retention (>99%) of reactive black 5 and reactive black 2 dyes. Li et al. constructed a robust ZIF based JUC-160 membrane (approx. 2.5 μm thick) as a molecular sieving membrane for dye molecules from water.244 The membrane possessed uniform pores which were larger than water molecules and smaller than dye molecules at the same time. Owing to the pore structure, the JUC-160 membrane exhibited higher rejection rates for different dye molecules (>99.8%) with a water flux of 100 L m−2 h−1. The as-constructed membrane was assessed for a total duration of 150 hours with uncompromised rejection rates and water flux. This indicated the superior recyclability of the membrane.
The utilization of ZIFs as fillers or adsorbents in membranes for water remediation presents certain inherent pitfalls. ZIFs typically exhibit inadequate chemical stability especially under slightly acidic conditions, thereby compromising their long-term performance and durability in water treatment applications. Excessive ZIF growth is another concern, leading to diminished mass transfer rates and consequent compromised water permeance. Additionally, fouling issues may arise, impeding the effective removal of contaminants from water. These scientific challenges underscore the imperative for continued research and development to surmount these limitations and fully unlock the potential of ZIFs in membrane-based water remediation systems.
Apart from ZIFs, UiO-66 based isoreticular MOFs have also been widely explored in the field of dye removal from wastewater owing to their high surface area, small pore aperture, high hydrolytic stability and hydrophilicity. Most of the reports related to MOF based membranes for dye removal were focused on utilizing a variety of MOFs. However, very few were focused on the loading% of MOFs in the membranes. Denny et al. achieved the introduction of MOF particles in a MMM with a high MOF loading (approx. 67%) using a wide range of MOFs [UiO-66, UiO-66-NH2, MIL-101-(Cr), MIL-101-(Fe), HKUST-1, MIL-53(Fe), ZIF-8] to prepare MMMs that could be fabricated on a large scale using PVDF as a support material.245 All of the as-prepared MMMs retained the high surface areas of the parent MOFs and were also subjected to PSM with promising results. The as-prepared UiO-66-MMMs showed >99% removal of Coomassie brilliant blue (CBB) dye from water, whereas the PVDF support didn't allow analyte solution to pass through. In another follow up study, Moreton et al. utilized styrene/butadiene polymer based MMMs with up to 90% UiO-66 loading.246 In contrast to the previous work, the SBS MMMs achieved higher MOF loadings than PVDF and also provided better mechanical tunability than PVDF (Fig. 7). Nevertheless, the MOF loading of <70% was devoid of any cracks and showed homogeneous particle size distribution over the MMMs. In addition, the authors demonstrated selective retention of CBB (60%) over MO (22%) from aqueous solutions using the 80 wt% UiO-66/SBS MMM. In 2016, Yao et al. developed UiO-66-MOF-urea based MMMs by post synthetic polymerization of UiO-66-NH2 and polyurethane oligomer to yield 50, 60 and 70 wt% loading of the UiO-66-NH2 MOF in the MMMs.247 The 70% UiO-66-urea based MMM exhibited different adsorption affinities to hydrophilic dyes (EY, RB, MG and MB) in aqueous solutions. The authors showed that the as-prepared MMMs acted as superior adsorbents to totally separate EY/MB, RB/MG and RB/MB mixtures owing to their preferential adsorption of EY and RB over MG and MB in aqueous solutions.
Fig. 7 UiO-66 based MMM membranes made up of polystyrene-block-polybutadiene (SBS) polymers: (a) schematic showing fabrication techniques, (b) PXRD patterns and (c) N2 sorption studies of the UiO-66 based MMMs [reproduced from ref. 246 with permission from The Royal Society of Chemistry, copyright, 2023]. |
Nonsolvent-induced phase separation (NIPS) and thermally induced phase separation (TIPS) techniques are widely recognized in the realm of large-scale membrane manufacturing processes. Nonetheless, membranes produced through the NIPS or TIPS technique commonly exhibit a tendency to exclude sizable macromolecules and particulate substances such as proteins, suspended solids, bacteria, viruses, and colloids. However, they struggle to effectively segregate small organic molecules with precision. The addition of MOFs with customized functionalities and adsorption sites to NIPS or TIPS membranes may enhance the current MF and UF membrane technology. In this regard, Wang et al. reported a thermally induced phase separation-hot pressing (TIPS-HoP) strategy to produce 10 distinct MOF-membranes (e.g., UiO-66-NH2, ZIF-8, MOF-5, MOF-801, MOF-808, HKUST-1, MIL-100(Cr), BIT-72, Mg-MOF-74, and Zn-BLD) with up to 86 wt% loading of MOFs.248 The authors used ultrahigh-molecular weight polyethylene for the interweaving of MOF particles which in turn increased their mechanical strength. A CR rejection of >99.0% with 126.9 L m−2 h−1 bar−1 water permeance was achieved using UiO-66-NH2-PE MMM-92% in a cross-flow system. According to the study, even after ten CR removal cycles (5 h for each cycle), the membrane did not show any obvious compromise in the water flux and CR rejection performance. This work has shown great promise for the large-scale roll-to-roll production of MOF based membranes with easy retrofitting possibility in existing industrial processes to exclude small molecules such as organic dyes.
In addition to employing ZIFs and UiO-66, several other MOF structures including two-dimensional MOFs have been documented for the purpose of dye removal. For example, Ang et al. utilized polycationic polymers like PEI or PDDA for regulating the assembly of 2D Zn-TCP(Fe) nanosheets by crosslinking of polycationic polymers and the surface terminated carboxylic groups of the 2D MOF nanosheets.249 The lamellar NF membrane (thickness 48 nm) exhibited superior BB dye separation (>90%) with an ultrahigh water permeance of 4243 L m−2 h−1 bar−1 (2-fold higher than the control adsorbent). Moreover, the membrane prepared was exposed to BB dye separation to assess its long-term stability in the presence of the fouling agent BSA. Results revealed that the membrane exhibited exceptional stability for a duration of 8 hours, experiencing a marginal decrease of about 1% in rejection rate. Furthermore, the permeance rate reached a plateau after 1.5 hours. El-Mehalmey described an efficient pathway to fabricate a UiO-66-NH2 based MMM comprising CA as a polymer matrix for superior dye (both cationic and anionic) rejection performance.250 According to the report, CaCO3 was used as a sacrificial porogen which can be removed post synthetically to allow better water permeation by generating a network of channels inside the membrane. The formulated MMM showed superior mechanical durability, high MB and MO uptake capacity and recyclability. Zhao et al. fabricated Zr-porphyrin MOF-based (2D and 3D) photocatalytic self-cleaning TFN membranes using PAN as a substrate.251 The formulated membrane exhibited self-cleaning properties owing to its photocatalytic activity. Moreover, the optimum membrane (TFC-3D-2) showed a high water permeate flux of 110.4 L m−2 h−1 with very high dye rejection (approx. 100%) for five distinct dye molecules viz. MB, RhB, CR, DR 23, and RB 5. The Zr-porphyrin MOF-based TFN membrane consisting of 1D triangular channels with improved hydrophilicity and surface charge resulted in the superior separation and self-cleaning ability of the membrane. Nevertheless, fabrication of membranes utilizing these processes possesses intrinsic limitations, such as substrate stability requirements during MOF growth, limited selection of mild-condition-grown MOFs, and the resulting membranes' rigidity and brittleness. Table 3 provides a comprehensive summary of recent advancements in the performance of MOF-based membranes for dye removal.
MOF | Metal | Linker | Polymer support | Membrane | Fabrication method | Pore aperture (nm) & size (nm) | Stability | Dye and removal % | Water permeance (L m−2 h−1 bar−1) | Water flux (L m−2 h−1) | Recyclability | Ref. |
---|---|---|---|---|---|---|---|---|---|---|---|---|
ZIF-8 | Zn | 2-Methylimidazole (Hmim) | NaOH treated PAN and poly(sodium 4-styrenesulfonate) [PSS] | ZIF-8/PSS | Co-ordination driven in situ self-assembly | — | — | MB 98.6% | 26.5 | — | — | 237 |
ZIF-8 | Zn | Hmim | PSF as a support and PA as a coating layer | PA/ZIF-8 (LBL#3) | Layer-by-layer (LBL) | — | CR 99.8% | — | 20.1 | 238 | ||
PA/ZIF-8 (LBL#4) | CR 99.2% | — | 27.1 | 238 | ||||||||
ZIF-8 | Zn | Hmim | NaOH treated PAN as a support and PEI as a coating layer | ZIF-8/PEI-HPAN | Chelation assisted interfacial reaction (CAIR) | — | 60 h operation | CR 99.8% | 78 | — | — | 239 |
MB 98.9% | 75 | |||||||||||
AF 87.2% | 97 | |||||||||||
CV 45.8% | 115 | |||||||||||
JUC-160 | Zn | 2-Methylbenzimidazole and benzimidazole | α-Alumina support | JUC-160 membrane | Seed mediated secondary growth method | 0.33–0.36 nm | Soaked in HCl, NaOH and boiling water, for 7 days | CR 99.8% | — | 102.3 | 50 h of continuous operation with negligible change in water flux and dye rejection | 244 |
CBB R250 99.8% | — | 98.6 | ||||||||||
RhB 99.8% | — | 103.1 | ||||||||||
UiO-66 | Zr | Terephthalate acid | PVDF | UiO-66 MMM | Drawdown coating | 0.6 | Organic solvents, mechanically stable | CBB 99% | — | — | Two cycles | 246 |
Zr | 2-Amino terephthalate acid | High density polyethylene (HDPE) and paraffin | NH2-UiO-66 PE MMM-86% | TIPS-HoP strategy | — | — | CR 99% | 126.9 | — | 5 h cycle for 10 times | 248 | |
FA 99.1% | 111.4 | |||||||||||
OG 99% | 115.9 | |||||||||||
Cr | Benzene-tricarboxylic acid | High density polyethylene (HDPE) and paraffin | MIL-100(Cr) PE MMM-86% | CV 99% | 112.5 | |||||||
RB 99.2% | 108 | |||||||||||
MB 99.2% | 120 | |||||||||||
UiO-66-NH2, MIL-100 (Cr) | Zr, Cr | 2-Amino terephthalate acid, benzene-tricarboxylic acid | High density polyethylene (HDPE) and paraffin | NH2-UiO-66-MIL-100(Cr) PE MMM-86% (1:1) | CR, FA, OG, CV, RB, MB >99% | >100 |
The most frequently utilized approaches involve increasing the number of adsorptive sites within the frameworks to enhance the capacity for adsorbing various toxic heavy metal ions. This can be achieved through several means, including electrostatic interactions, acid–base interactions, hydrogen-bonding interactions, and the presence of coordinatively unsaturated adsorptive sites or defect sites. For example, Chen et al. prepared formic acid modulated UiO-66-NH2 to adsorb Cr6+ ions from aqueous solution. The pH dependent surface charge modulation resulted in the enhanced adsorption of Cr6+ ions. At pH 2, the pendant amino groups were protonated resulting in a zeta-potential of >30 mV. Hence, under the acidic pH conditions the Form-UiO-66-NH2 adsorbed 169.4 mg g−1 Cr6+ ions [HCrO4−, CrO42−], whereas under neutral pH conditions the adsorptive capacity was limited to only 28.1 mg g−1.258 Hence, this work showcases the role of electrostatic interactions in enhancing the adsorptive properties of MOFs as heavy metal adsorbents. Acid–base interactions also play a key role in the adsorption of heavy metal ions. Metal ions that are chemically hard (Fe3+, Mn2+, Co2+, etc.), specifically acids, have small sizes and are not easily polarized. On the other hand, soft metal ions (Hg2+, Cd2+, Cu+, Hg+, Ag+, etc.) are larger in size and are readily polarized. Meanwhile, Pb2+, Cu2+, Ni2+, Zn2+, etc. are considered as borderline acids. Ligands containing highly electronegative donor atoms like oxygen or nitrogen are classified as hard bases, while ligands with sulfur donors are considered soft bases due to their higher polarizability. On this basis, numerous functionalities have been designed in order to selectively capture heavy metal ions in functionalized MOF pores. Different types of functional groups such as thiols, amines, sulfamine, and others can serve as chelating agents to increase the adsorption capacity of MOF-based adsorbent materials for Hg2+. For example, Zhao et al. demonstrated that introducing cysteine moieties (pendant –NH2, –SH and –COO− functional groups) onto UiO-66 surfaces can enhance the removal efficiency of Hg2+ ions from water.259 To highlight the role of coordinative interactions between the functional groups and metal ions, Ahmadijokani et al. used an ethylene diamine functionalized UiO-66 MOF for the removal of Pb2+, Cd2+and Cu2+ ions from aqueous solutions.260 The UiO-66-EDA membrane exhibited maximum adsorption capacities of 243.90, 217.39 and 208.33 mg g−1 for Pb2+, Cd2+and Cu2+ ions, respectively. According to the study, electron exchange, electron sharing, and electrostatic and coordinative interactions between the metal ions and the abundant amine groups were key to the adsorption of the metal cations on the UiO-66-EDA surface. In addition to functional groups, metal oxo-clusters are also proven to be beneficial for the adsorption of heavy metal ions.261 Often open metal sites (OMSs) as defect sites in MOFs play a crucial role in adsorbing toxic heavy metal ions from wastewater sources. To highlight this, Xu et al. created oxygen vacancies in a UiO-66 MOF to create open metal sites that acted as active sites to adsorb As5+ with an ultrahigh adsorption capacity of 248.75 mg g−1 under neutral pH conditions.262 Also, the external surfaces of MOF particles could enhance the adsorption efficiency towards heavy metal removal from water. To highlight this, Huang et al. synthesized a ZIF-8/hyphae membrane utilizing hyphae fungus (Mucor) as a robust scaffold to immobilize ZIF-8 by an in situ growth method followed by vacuum filtration.263 According to the study, the adsorption capacities of ZIF8@Mucor-0, ZIF8@Mucor-3 and ZIF8@Mucor-4 were found to be 973.53, 1443.29 and 1013.47, respectively. Owing to its smaller size and increased active sites, ZIF-8@Mucor-3 exhibited the maximum adsorption capacity.
However, when it comes to large-scale applications, neither pristine nor modified MOFs alone are capable of effectively driving heavy metal ion-based wastewater remediation, unlike their membrane counterparts. Nonetheless, the utilization of electrospun MOF-based fiber composites offers a promising solution for wastewater treatment, particularly in the removal of heavy metal ions. Therefore, this section explores the key findings regarding the adsorptive removal of heavy metals using MOF-based electrospun nanofibers. This is due to the fact that these nanofibers possess appropriate functionalities that can work in harmony to both anchor MOFs and adsorb toxic metal ions, potentially overcoming the traditional challenges associated with MOF-based membranes. As already discussed in the Synthetic strategies to form membranes section, ENMs that are produced by electrospinning have been widely explored for heavy metal ion removal from aqueous solutions.264–266 In addition to their porous nature, easy separation from water, high flux, and cost-effectiveness are some key benefits being offered by ENMs. Thus, ENMs could act as supports for the fabrication of MOF based nanofiber composites for the removal of heavy metal ions from water (Fig. 8). MOFs are generally fabricated into ENMs by direct electrospinning in the presence of MOF nanoparticles or in situ growth of MOFs on ENM substrates.267,268
Recent advancements in this research area have been focused on introducing various kinds of MOFs onto different polymeric nanofibers using the aforementioned strategies.268 Efome et al. fabricated a zirconium-based MOF-808/PAN composite membrane using electrospinning and studied its removal efficiencies for the filtration of Cd2+ and Zn2+ ions from water.269 A MOF loading of 20% could be achieved in the composite membrane which could remove almost 60–70% of the Cd2+ and Zn2+ ions. In this study, the authors achieved adsorption capacities as high as 225.1 and 287.1 mg g−1 for Cd2+ and Zn2+, respectively. Meanwhile, the pristine PAN membrane without MOF-808 as a filler adsorbed 21.6 and 4.9 mg g−1 for Cd2+ and Zn2+, respectively. In their study, the authors demonstrated that as compared to conventional thermal activation, the ‘hydractivation’ method regenerated an expanded MOF without further crystal downsizing and pore shrinkage. From the same research group, another study was conducted where process parameters like feed concentration, transmembrane pressure and membrane thickness could play a vital role in determining the efficiency of heavy metal ion removal from water.270 The authors also investigated the effect of competitive co-existing ions like Na+, Mg2+, and Ca2+ on the removal efficiencies of Zn2+, Cd2+, Pb2+ and Hg2+ ions. In another study, Jamshidifard et al. synthesized a zirconium based UiO-66-NH2 MOF using microwave heating and introduced it onto electrospun PAN/chitosan nanofibrous membranes.271 The as-prepared membranes were employed for the removal of Pb2+, Cd2+ and Cr6+ ions via both adsorption and membrane filtration. In their study, under optimized conditions (pH, temperature, MOF content and equilibrium time) the maximum adsorption capacity of the PAN/chitosan/UiO-66-NH2 membrane was 441.2, 415.6 and 372.6 mg g−1 for Pb2+, Cd2+ and Cr6+ ions, respectively, and the membrane could be recycled up to 5 cycles without significant loss in water flux and heavy metal removal capacities.
Hashem et al. used filter paper as a source of cellulose fibers to graft UiO-66-NH2 nanoparticles using a one-pot strategy.272 The grafting process became facile when chloroacetic acid was used for creating carboxylate anchoring sites on the cellulose nanofibers. Together with the UiO-66-NH2 nanoparticles, the porous support rejected 78.2% of Cr6+ ions from water. Li et al. proposed biomineralization-mimetic fabrication of ZIF-8 and -67 based electrospun-silk-nanofibers (ESF@MOFs) for As5+ and Cr6+ removal from water with ion rejection >99%.273 In 2018, Hou et al. fabricated ZIF-67 based “pearl-necklace-like” composite membranes by in situ growth of ZIF-67 on a 2-methylimidazole/CA (MIM/CA) ENM.274 The BET surface area of the control membrane increased after the grafting of ZIF-67 from 6.9 to 463.1 m2 g−1. The as-synthesized composites demonstrated exclusion of 18.9 and 14.5 mg g−1 for Cu2+ and Cr6+ ions, respectively. According to the study, electrostatic adsorption and ion-exchange were the important factors to adsorb such quantities of metal ions. Nevertheless, the recyclability studies revealed that the removal efficiency decreased to 30% after three cycles, showing incompatibility in their long-term performance. Another study from Efome et al. was done by selecting two water-stable MOFs (MIL-100-Fe and MOF-808) for fabricating electrospun PAN and PVDF based nanofibrous MOF membranes (NMOMs).275 The MIL-100-Fe/PAN based NMOM showed water flux as high as 348 L m−2 h−1 with a permeance of 870 L m−1 h−1 bar−1. Owing to the high affinity between the MOFs and electrospun PVDF nanofibers, the NMOM retained its adsorption capability to 90% even after four cycles of sorption experiments. According to the study, removal of Pb2+ ions from water as high as 248.6 L m−2 h−1 bar−1 could be achieved with a membrane thickness of 560 μm. Chen et al. developed an EDTA modified MOF-808@PAN based PME membrane via the electrospinning method to remove heavy metal ions from water.276 A MOF loading of 60 wt% (PME-60) was found to be the optimum loading% for the removal of Cu2+ and Cd2+ ions with adsorption efficiencies of 81.9% and 85.5%, respectively. The as-prepared PME membranes can perform in an acidic environment with good corrosion resistance. Furthermore, the PME membrane could be easily reused by treating with EDTA-2Na solution.
As shown in Table 4, MOFs are a promising class of materials for removing heavy metal ions from wastewater. However, there are several challenges that need to be addressed before MOFs can be widely adopted for this application. One challenge is the stability of MOFs in water and under harsh chemical conditions. MOFs are typically made up of metal ions and organic linkers, and these components can be susceptible to degradation in the presence of water and other chemicals. This can lead to the loss of the MOF-structure and its ability to adsorb heavy metal ions. Another challenge is the synthesis of MOFs. The synthesis of MOFs can be complex and time-consuming, and it is often difficult to scale up the production of MOFs to meet the needs of industrial applications. MOFs must also be selective in their ability to adsorb heavy metal ions. This means that they should preferentially adsorb heavy metal ions over other ions that may be present in wastewater, such as sodium or chloride ions. MOFs should also have a high adsorption capacity, meaning that they should be able to remove a large number of heavy metal ions from a given volume of wastewater. The kinetics and mass transfer properties of MOFs also need to be improved. The kinetics of adsorption refers to the rate at which heavy metal ions are adsorbed onto the MOF surface. The mass transfer properties of MOFs refer to the rate at which heavy metal ions diffuse through the MOF's structure. Both of these properties need to be improved in order to achieve faster adsorption rates. Finally, regeneration methods that effectively remove adsorbed metal ions without damaging the MOF structure are necessary for reuse. MOFs can be regenerated by exposing them to heat, light, or other stimuli. However, it is important to ensure that these regeneration methods do not damage the MOF structure, as this could reduce its ability to adsorb heavy metal ions in the future. The cost-effectiveness of MOFs, including their synthesis and raw materials, also needs to be addressed for large-scale implementation. MOFs are typically made from expensive metal ions and organic linkers. This can make them prohibitively expensive for large-scale use. Real-world wastewater is often complex, containing a mixture of heavy metal ions and other contaminants. It is important to understand how MOFs perform in such environments. For example, some MOFs may be more effective at adsorbing certain heavy metal ions than others. It is also important to consider the effects of other contaminants in wastewater on the adsorption of heavy metal ions by MOFs.
MOF with loading% | Metal & linker | Support membrane | Functionality in MOF | Pore size (MOF) (nm) | Heavy metals | Maximum removal capacity (mg g−1) | Equilibrium time (min) | Recyclability | Water flux (L m−2 h−1) | Water permeance (L m−2 h−1 bar−1) | Ref. |
---|---|---|---|---|---|---|---|---|---|---|---|
MOF-808, 20 wt% | Zr4+ and 1,3,5-benzene tricarboxylic acid | PAN ENM on PVDF ENM | –COOH, –OH | — | Cd2+ | 225.05 | 180 | 4 cycles | — | 139.2 ± 22 | 269 |
Zn2+ | 287.06 | ||||||||||
UiO-66-NH2, 10 wt% | Zr4+ and 2-aminoterephthalic acid | PAN/chitosan ENM on PVDF | –NH2, –OH | 0.9–1.3 | Pb2+ | 441.2 | 60 | 5 cycles | 452 | — | 271 |
Cd2+ | 415.6 | 463 | — | ||||||||
Cr6+ | 372.6 | 479 | — | ||||||||
ZIF-8, 36 wt% | Zn2+, 2-methylimidazole | ESF | –Imidazolyl nitrogen, –CH3 | — | As5+ | 50 | — | — | — | — | 273 |
ZIF-67, 34 wt% | Co2+, 2-methylimidazole | ESF | –Imidazolyl nitrogen, –CH3 | — | Cr6+ | 15.4 | 1400 | — | — | — | |
F300, 20 wt% | Fe3+, BTC | PAN | –COO− | 2.1 | Hg2+ | 229.66 | 120 | 4 cycles | 348 ± 25.8 | 870 | 275 |
PVDF | Pb2+ | 148.13 | — | — | |||||||
MOF-808, 20 wt% | Zr4+, BTC | PAN | –COO− | 0.5–1.8 | Hg2+ | 276.96 | — | — | |||
PVDF | Pb2+ | 170.74 | — | — |
The challenges described above can be addressed through interdisciplinary collaboration between chemists, engineers, and materials scientists. By working together, new MOFs can be developed that are more stable, selective, and cost-effective. By overcoming these challenges, MOFs could become valuable tools in tackling water pollution and ensuring cleaner water resources.
In the last two decades, the rapid development and advancements in the field of MOFs have led researchers to utilize MOFs in various water treatment processes. Despite synthesizing countless MOFs, their application to wastewater treatment is limited to only water stable MOFs among which very few could be used for removing salts from water. Until now, the selection of MOFs to be applied in water treatment has mostly relied upon water stability, pore size and aperture along with hydrophilicity/hydrophobicity. According to the discoveries made so far, UiO-66 and ZIF-8 are the most effective MOFs for separating monovalent ions from water. Their robust nature and high-water stability allow them to be operational for longer times. On the other hand, their smaller pore apertures allow them to retain the dissolved ions in the voids.
MOF based membranes have shown potential for desalination processes and their water desalination capability was investigated by Hu et al. by a molecular simulation study.277 ZIF-8 was investigated in this study because of their thermal/chemical stability and water/moisture tolerance. The computational study was performed using NaCl solution having concentration equivalent to seawater. The very small pore apertures of ZIF-8 membranes (3.4 Å) were successful to sieve Na+ or Cl− and allowed transport of water through the membrane. However, it has to be reminded that in contrast to experimental conditions, the simulated study was done without taking the defects of the membrane into consideration. In 2015, another simulation study was conducted by the same research group, using a series of ZIFs as potential RO membranes for seawater desalination.278 The ZIFs were ZIF-25, -71, -93, -96, and -97 with similar topology but varying in imidazolate organic linkers with different functional groups. The presence of different functional groups resulted in variation of aperture size and polarity difference in the ZIFs which in turn affected the desalination performance. For instance, the smaller aperture size of ZIF-93 and -97 (3.5–3.7 Å) demonstrated decreased water flux compared to ZIF-25, -71 and -96 (5.1–5.5 Å). Also, the authors have shown by analyzing the radial distribution functions that the framework affinity for water follows the trend ZIF-25 < -71 < -96 and the trend for water flux follows the order ZIF-25 > -71 > -96. In conclusion, the most hydrophobic ZIF-25 showed the highest water flux. This simulation study from Gupta et al. showed that ZIF-25 could be a potential RO membrane for water desalination.
The group of Liu et al. was one of the first groups to develop an alumina based UiO-66 membrane in 2015.279 The authors fabricated UiO-66 MOFs on hollow alumina fibers using an in situ solvothermal approach. Outstanding multivalent ion rejection performance (86.3% for Ca2+, 98.0% for Mg2+, and 99.3% for Al3+) with moderate permeance (0.14 L m−2 h−1 bar−1) and good permeability (0.28 L m−2 h−1 bar−1 μm) was exhibited by the as-prepared membrane. In particular, originating from the better water stability of UiO-66 MOF particles, the membrane showed long term operational stability even after 170 h of operation. In another study, Duan et al. employed ZIF-8 nanoparticles on a PA layer to investigate the desalination performance of the membrane. The increased interaction between PA and ZIF-8 could eliminate non-selective voids and thus maintained good rejection of NaCl (98%) with better water permeance (162% better than only PA membrane) under brackish water RO conditions. Recently, Bonnett et al. synthesized zirconium based PCN 222 MOF nanorods (300 nm) and incorporated them into an RO based TFN membrane.280 According to this work, MOFs containing relatively larger pores increased the water flux whereas at the same time decreased salt rejection. Nevertheless, PSM could achieve better tunability of the pore sizes of MOFs and hence contribute to better salt rejection with reduced water flux. In their seminal work, myristic acid was employed for the modification of PCN-222 MOF particles which could achieve >95.5% salt rejection with almost 100% water flux. Park et al. introduced a H2SO4 treated HKUST MOF into a PSF substrate to fabricate RO membranes.281 The agglomeration tendency of the MOF particles was reduced after acid treatment and finally the surface roughness of the substrate was lowered. The acid modified membrane has shown a 33% increment in water flux with uncompromised salt rejection performance. The membrane fouling resistance was also improved displaying stable permeance under longer operational conditions.
While most of the studies were focused on substrate-MOF compatibility or pore tunability of MOF particles to increase water flux and salt rejection, studies on the effect of MOF particle size were infrequent. Lee et al. demonstrated the effect of ZIF-8 particle size on the desalination performance of the ZIF-8/PSF membrane.282 The authors synthesized various sizes of ZIF-8 nanoparticles (60, 150 and 250) on the microporous PSF substrate. Employing different ZIF-8 sized particles, the interfacial area of the PA matrix and ZIF-8 particles affected the membrane performance. Analysis of AFM studies showed that the 150 nm sized ZIF-8 possessed the highest ZIF-8/PA interfacial area. This result can be attributed to the balance between the outside surface area of the nanoparticles and surface coverage by ZIF-8 based on the size of the MOF. Also, the size of the ZIF-8 nanofillers could affect the IP process which in turn influences the RO performance. The ZIF-8 (150 nm)/PA TFN membrane showed the highest water permeance of 3.95 L m−2 h−1 bar−1 with the highest NaCl rejection of 99.2% when compared to the pristine TFC and other sizes (60, 250 nm) of ZIF-8/PA TFN membranes.
MOF based membranes could also be utilized to sieve divalent cations of saline water like MgCl2, Na2SO4, MgSO4, etc. by the NF technique. For instance, Liu et al. post synthetically modified UiO-66-NH2 by palmitoyl chloride (UiO-66-NH2-PC) to increase the dispersibility of nanofillers within the PA layer in cyclohexane medium (Fig. 9).283 The aggregation tendency was drastically reduced due to the polarity enhancement with the organic solvent after modification by palmitoyl chloride. In their study, the pure water permeance was found to increase from 8.1 to 12.4 L m−2 h−1 bar−1 with 95% Na2SO4 rejection performance. All the as-prepared membranes showed negative zeta potential and exhibited rejection properties in the order MgSO4 > Na2SO4 > NaCl following the mechanism of NF membranes. The modified UiO-66-NH2 based membrane offered better membrane performance with outstanding long-term operation for continuous filtration of 80 h.
Fig. 9 FESEM images of UiO-66-NH2 (a) and UiO-66-NH2-PC (b) along with the schematic representation of the preparation process of UiO-66-NH2 based TFN membranes [reproduced from ref. 284 with permission from The Royal Society of Chemistry, copyright, 2023]. |
In most of the TFC NF membranes, the permeance and selectivity are primarily based upon the active PA layer. Efforts have been directed towards the enhancement of permeance of the PA active layer without compromising the selectivity. To address this, tailoring of intrinsic properties of the active layer is necessary. This could be achieved by the judicious design of the polymer structure itself or with the introduction of hydrophilic nanomaterials into the layer. In addition to that, an alternative strategy could be to make ultrathin active layers to decrease the water transport resistance. Following these strategies, Wang et al. reported a novel strategy on the nanostructure mediated IP process by using preloaded PDA/ZIF-8 nanoparticles as sacrificial templating materials supported on a porous SWCNT/PES membrane.285 IP was used to fabricate the PA layer on supported PD/ZIF-8 nanoparticles. After the completion of the IP process, the synthesized NF composite membrane was exposed to water to remove PD/ZIF-8 nanoparticles by dissolution to form a thin PA layer consisting crumpled nanoscale structures. Additionally, fine tuning of the crumpled surface morphology and the surface area could be achieved by controlling the loading% and particle size of sacrificial nanoparticles. By this strategy, the authors could achieve significant water permeation as high as 53.2 L m−2 h−1 bar−1 with 95.2% Na2SO4 rejection. Furthermore, owing to the geometric structure of ZIF-67, the use of it as a sacrificial templating material has contributed towards a larger effective surface area. This in turn affected the water permeance and a water flux of 1831 L m−2 h−1 was achieved with 97.2% Na2SO4 rejection. In another study, Zhao et al. used three different water stable MOFs (MIL-53(Al), UiO-66-NH2 and ZIF-8) to fabricate TFN NF membranes by both blending and preloading IP methods.286 The introduction of MOFs successfully decreased the degree of cross-linking and increased the membrane thickness, surface negative charge and roughness of the PA active layer. MIL-53 was found to interact more strongly to the PA layer than UiO-66-NH2 and ZIF-8. Stronger interaction along with MOF hydrophilicity could explain the variation in permselectivity of the as-prepared TFN NF membranes. In comparison to the control TFC membrane, the MOF incorporated TFN membranes showed better water permeability without compromising their permselectivity to reject NaCl (>40%). Among the three membranes, UiO-66-NH2 showed the highest permeability, almost 1.3 times higher than the other two MOF (MIL-53, ZIF-8) based TFN membranes.
Following this, the applicability and practicality of MOF based membranes are found to be more promising than those of commercial membranes. However, the large-scale synthesis of defect free MOF-based membranes still remains a challenging task. As the literature suggests, MOFs as fillers are more widely applied for water remediation than continuous MOF membranes. Hence in order to enable the purification of significant quantities of waste or saline water, it is necessary to synthesize MOFs on a large scale which is a challenging task to accomplish. This is due to their high production cost, which is mainly associated with the high cost of starting materials, solvents, and energy consumption required for their synthesis.287 The high costs of MOF production are also attributed to the absence of efficient production methods for large-scale manufacturing and the difficulties encountered in reproducing the desired MOF properties at such a scale.287
Although various synthesis methods have been proposed to address these issues, such as continuous flow synthesis, microwave-assisted synthesis, mechanochemistry and the spray drying technique, the feasibility of these methods for industrial-scale production is yet to be demonstrated. Moreover, the lack of a standardized synthesis protocol and the need for skilled personnel to carry out the synthesis further increase the cost of production. Despite these challenges, there is growing interest in developing low-cost synthesis methods for MOFs and COFs to enhance their economic viability. For instance, most of the MOFs reported to date are synthesized in DMF solvents at elevated temperatures. Some of them even require prolonged time duration to ensure crystallinity. In this regard, attempts have been made to use low-cost starting materials and solvents, such as biomass and water, waste plastics, etc. for the synthesis of MOFs. The market size of metal–organic frameworks (MOFs) is experiencing substantial growth, driven by the contributions of key industry players such as BASF, MOFapps, MOF Technologies,287 Framergy Inc., Lumtec, etc. The collective efforts of these companies have propelled the commercialization and utilization of MOFs and COFs across various sectors.
Recently, COFs have emerged as one of the best alternatives to overcome these existing problems.190 The inception of COFs and COF based thin films and membranes can solve the problem of attaining high permeability with good selectivity. COFs, which were discovered by Omar M. Yaghi and co-workers in 2005, have shown tremendous prospects in water treatment and separation processes so far.288 Considering the mighty possibility of structural designability, easy tunability of pore dimensions, excellent stability under harsh conditions, high surface areas, and tailored functionalities at external surfaces and pores, low density COFs could be a breakthrough material of this century (Fig. 10).289–291 So far, a wide diversity of COFs has been explored involving three broad classes: imine-based, triazine based, and boron-based COFs.208
Fig. 10 Schematic representation of water filtration of polluted water using the molecular sieving mechanism by a covalent-organic framework based membrane. |
Separation processes involving membranes contribute to small carbon footprint and consume less energy than adsorption-based technologies. Considering the advantages that membrane-based technology could provide, a surge in reports regarding COF based membranes can be found in the literature starting from 2009.200 Mostly, COF based membrane fabrication involves IP, layer-by-layer stacking and in situ growth (Fig. 11). The detailed methodology for the fabrication of these aforementioned methods is summarized elsewhere.200–202,292–294 So far, a plethora of reports on COF based membranes are found in the literature; however, shortcomings like the chemical instability, complexity in synthesis and scalability still remain to be addressed.
Fig. 11 Schematic diagrams of the various approaches for the fabrication of ultrathin COF membranes: (a) creating tubular COF-LZU1 membranes through an in situ solvothermal synthesis approach. (b) Crafting COF membranes using the blade-casting technique. (c) Immobilizing COF membranes onto polymeric substrates through interfacial polymerization. (d) Self-standing COF membrane production through liquid–liquid-phase interface-assisted polymerization. (e) Producing COF membranes via unidirectional diffusion synthesis. (f) Employing vacuum assistance to assemble single-layered CTF nanosheets into membranes. (g) Schematic representation of in situ grown COF based membrane. [reproduced from ref. 227 with permission from The Royal Society of Chemistry, copyright, 2023]. |
Additionally, most of the reports related to COF based membranes as adsorbents are based on the discovery of new structures, PSM, molecular sieving and the charge separation mechanism.187–189,295,296 Very less is explored on their mechanical behaviour or film-to substrate adhesion strength. To address this, Banerjee's group has recently reported the underlying structure–mechanical property relationships of porous and crystalline frameworks for the construction of COF based thin films and membranes using self-assembly of crystalline COF particles.297 The authors have also shown that the mechanical characteristics of COF based thin films are majorly regulated by packing efficiency, i.e., the quality and quantity of the defects.
The challenges associated with the large-scale fabrication of COFs are even more different than MOFs. Dynamic covalent chemistry plays a crucial role in the crystallization of COFs, as it facilitates the formation of bonds and the transition from amorphous to crystalline phases, leading to the repair of structural defects. However, the solvothermal synthetic strategy to yield high-quality COFs using dynamic covalent chemistry and error–correction mechanisms can be difficult without prolonged reaction times, organic solvents, and complex freeze–pump–thaw cycles. In this regard, the Banerjee research group has addressed this challenge by demonstrating a scalable, time-saving method for synthesizing highly crystalline COFs using a mechanochemical approach.298,299 Apart from this, the choice of solvents greatly affects the quality of the produced COFs. In this aspect, water is not commonly used as a solvent for COF synthesis because organic monomers have very low solubility in aqueous solutions. Nevertheless, the idea of synthesizing water-based COFs is ground breaking because the production of water as an end-product has the potential to facilitate the dynamic and reversible formation of imine (–CN) bonds in an aqueous environment.300 By using an inexpensive catalyst like p-toluene sulfonic acid and solvent like water, crystalline COFs with high surface area can be formed in just 60 seconds.299 The extruder-based “terracotta process” shows promise for scaling up COF synthesis; however, it may not be suitable for synthesizing ionic COFs (iCOFs) and ionic covalent organic nanosheets (iCONs).
A simple methodology using the liquid–liquid interfacial technique was demonstrated by Banerjee's group for the synthesis of COF thin films with thickness control.302 The authors have synthesized Tp-Bpy, Tp-Azo, Tp-Ttba, and Tp-Tta COF thin films using triformylphloroglucinol (Tp) as the trialdehyde with di- and tritopic amines (Fig. 12). Moreover, both the Tp-Bpy and Tp-Azo COF thin films show excellent rejection values as high as 94% (brilliant blue-G), 80% (CR), 97% (AF), and 98% (RhB) and 90% (BB), 79% (CR), 99% (AF), and 99% (RhB), respectively. The long-term applicability and recyclability of the COF thin films were well established by the negligible changes in rejection performance over five cycles.
Fig. 12 COF thin-films and their mechanism of formation from fibers to an assembled sheet like structure: (a) illustrated is the suggested mechanism for creating COF thin films (Tp-Bpy). On the left, SEM images are presented, while on the right, digital images show different stages of thin-film development, encompassing fiber formation, assembly of fibers, and sheet formation. (b) In the subsequent section, transparent COF thin films are characterized, featuring digital images on the left, SEM images in the middle, and AFM images on the right. [reproduced from ref. 302 with permission from American Chemical Society, copyright, 2023]. |
In their follow up work, Banerjee's group developed a scalable, simple strategy to obtain self-standing covalent organic membranes (COMs) like M-TpBD-OMe2, M-TpBD-Me2, M-TpAQ and M-TpAD by knife casting of COF precursors with a PTSA catalyst on a glass plate and then baking for 12–72 hours.303 The ordered porous framework structure of the as-obtained COMs led to high permeance of organic solvents like acetone and acetonitrile. The acetonitrile permeance of M-TpTD was almost 2.5 times higher than the pre-existing PA-based NF membranes with comparable solute rejection performance. Additionally, the ultrahigh stability of the as-prepared COMs allowed them to function in water, organic solvents and also under acidic conditions (3N HCl). More significantly, the as-prepared M-TpBD COM rejected RB dye up to 99%, whereas 94% and 96% for MB and CR, respectively, with an excellent water permeance of 120 L m−2 h−1 bar−1. Further, the COFs showed excellent recyclability of up to five cycles without a considerable drop in their performance.
In 2016, Yu and coworkers304 developed a polycationic 2D COF, with a pore size of 5.8 nm. The polycationic COF consisted of bipyridinium (BIPY) based aldehyde monomers which demonstrated high uptake (>97%) of anionic dye pollutants from water even at very low concentrations like 3.2 × 10−5 M. The uptake of the anionic dyes by the COF matrix was possible due to the presence of bipyridinium motifs in the framework where the exchange of the Cl− ions with the anionic dyes took place.
COF membranes could also be developed using a support material. For example, Caro's group synthesized a COF-LZU1 membrane of only 400 nm thickness using amine modified ceramic tubes, which not only ensured mechanical stability but also allowed better water permeance.305 The dye rejection performance of the COF-LZU1 membrane (pore size = 1.8 nm) was evaluated using different dye solutions (chrome black T (CB-T), MB, AF, CR, and RB). Rejection rates greater than 90% were observed for all dye molecules except for molecular dimensions less than 1.2 nm. Among other dyes, the lowest rejection rate of 91.4% for fuchsin was observed due to the spherical shape (1.13 nm × 1.17 nm × 1.17 nm) of the dye molecule. No notable decrease in permeation flux and rejection rate was observed over 80 h. According to their work, the COF-LZU1 membrane had superior permeation to other commercial NF membranes. Wang's group synthesized a composite COF-based membrane on PSF substrates using Tp and Pa as monomers.306 The proportion of COF crystallites was shown to be the critical aspect for developing permselective composite membranes. The resulting COF/PSF composite membrane demonstrates extraordinary rejection of CR up to 99.5%. The optimized TpPa/PSF membranes can reject MB (799.80 Da), CR (696.66 Da) AF (585.54 Da), CB-T (461.38 Da) and AO7 (350.32 Da) with a rate of 94.4%, 99.5%, 52.6%, 96.3% and 14.7%, respectively. Despite the good recyclability and stability under acidic/basic conditions, a low water permeability of 50 L m−2 h−1 bar−1 was obtained. Lai's group developed a 2D covalent organic thin film with a pore size of 1.5 nm by utilizing an amphiphilic truxene based amine and a simple dialdehyde as building blocks at the air/water interface by the Langmuir–Blodgett method.307 The 2D membrane supported on a porous PSF substrate separates organic dye molecules on the basis of their smallest projection size. The cut off molecular size is around 1.3 nm related to the membrane pore size. Thus, a molecular sieving phenomenon could be performed using the 2D COF membrane where smaller MO and RB molecules pass through whereas larger molecules like DR80 and PEG (MW 4000) could be absolutely rejected.
Charged (cationic and anionic) COFs are also desired in the field of water remediation. The intrinsic charge residing inside the pores and surfaces could increase the adsorption of oppositely charged pollutants leading to an enhanced dye rejection performance. In this regard, Ma & Li groups synthesized the first 2D cationic EB-COF:Br based membrane using ethidium bromide (EB) and Tp as the monomers for interfacial crystallization.308 The covalent organic nanosheets were restacked using a layer-by-layer assembly method to assemble a continuous and dense membrane on a nylon 66 support. The accurately prepared 2D cationic EB-COF:Br membrane displayed superior selectivity to sieve ionic pollutants of varying molecular dimensions and charges assisted with high solvent permeability. The imine linked cationic EB-COF:Br membrane demonstrated better solvent permeability performance than GO-membranes and other commercial NF membranes due to its intrinsic porous nature and hence better mass transport in the vertical direction. The cationic 1D pore channels in the EB-COF:Br based membrane successfully rejected dyes in the following order: anionic dyes > cationic dyes > neutral dyes. The recyclability studies exhibited outstanding rejection efficiency (>99%) even after six cycles of adsorption and after 10 hours of continuous processing. Moreover, the EB-COF:Br cationic nanochannels can be easily restored using 2 mol L−1 NaBr solution.
In addition, the role of functionality may often cause the performance of isostructural COFs to vary. For instance, He et al. synthesized two imine-based isostructural COF films by the introduction of H-bonding between COF crystallites, namely COF-TBDH and COF-TBDM.309 Both the COFs were synthesized using the same amine but with different functionalities present in the dialdehyde monomers namely –OH and –OCH3. Implementation of both the COF films in dye adsorption revealed the underlying role of H-bonding and indeed the COF-TBDH/nylon membrane consisting –OH groups showed outstanding performance in CR rejection (100%) and exhibited a water permeation of 4394 L m−2 h−1 MPa−1. According to the authors, the –OH groups in the COF-TBDH had an impact on both the quality of the COF film and the enhancement of rejection performance.
Creation of defects in microporous materials often leads to high adsorption capacities compared to their pristine counterparts. This enhancement may be linked to unsaturated adsorption sites, increased pore sizes and highly energetic surfaces. In this context, Siril and colleagues have devised a continuous flow-based synthetic approach for an imine-linked DAB-TFP COF that includes defects in the COF matrix (Table 5).310 These defects were discovered to be quite useful for the separation of textile industry-based MB dye. The authors demonstrated a recyclability of ten cycles without a noticeable reduction in dye removal performance.
COF | Functionality | Pore size (nm) | Surface charge | pH stability | Removal percentage | Recyclability | Ref. |
---|---|---|---|---|---|---|---|
SA-COF | Imine/enamine, enol/keto | 1.43 | −ve at pH 12 and +ve at pH 1 | 1–12 | MB (98.3%, 10 min) > RB (94.1%, 60 min) ≫ CA (none) | — | 301 |
MB (99.8%) > AA (none) at pH 12 | — | ||||||
Tp-Bpy (thin film) | Enamine, keto & pyridinic nitrogen | 2.5 | — | pH 1 | 94% (BB), 80% (CR), 97% (AF), 98% (RH), and 96% (TB) | 5 cycles | 302 |
Tp-Azo (thin film) | Enamine, keto & azo | 2.6 | — | pH 1 | 90% (BB), 79% (CR), 99% (AF), 99% (RH) and 96% (TB) | 5 cycles | |
Tp-BD (thin film) | Enamine, keto | 2.2 | — | 3 M HCl | 99% (RB), 96% (MB), 94% (CR) | 5 cycles | 303 |
Tp-TD (thin film) | Enamine, keto | 3.0 | — | 3 M HCl | 84% (RB), 80% (MB), 83% (CR) | 5 cycles | |
PC-COF | Imine, bipyridinium | 5.8 | +ve | — | 96.9% (MO), 99.6% (AG-25), 60.3% (DFBM), 99.7% (IC), 97.8% (AR-27) | — | 311 |
COF-LZU-1 (membrane) | Imine | 1.8 | — | pH 7, 30 days | 99.2% (MB), 98.6% (CR), 98.2% (CB-T), 99.1% (RB), 91.4% (AF) | — | 305 |
84.5% (RhB), <30% (MB, MO) | |||||||
EB-COF: Br@GO (membrane) | Enamine, keto, phenanthridinium-bromide | 1.68 | +ve | — | 99.6% (MO), 99.2% (FS) and 98.1% (PP) | 6 cycles (MO) | 308 |
91.2% (RB), 87.2% (MB) and 84.9% (DMPD) | |||||||
74.4% (CAc), 22.3% (NR) and 15.7% (NA) | |||||||
Tru(NH2)3-TPA@PSF (thin film) | Truxene | 1.5 | — | 2.4% (MO, RhB), 14.5% (BB), 46.1% (RB5), 97.8% (DR80) | — | 307 | |
GO@CTF membrane | Triazine | 1.2 | — | pH 7–11, unstable under acidic conditions | 98.8% (EBT), 92.6% (CR), 93.3% (MB), and 93.1% (AB) | 30 cycles | 187 |
COF-TBDH/nylon | Imine, hydroxy | 2.0 | −ve (−15.4 eV) | — | 100% (CR), 93.1% (CBB) | 5 cycles | 309 |
COF-TBDM/nylon | Imine, methoxy | 2.0 | −ve (−13 mV) | — | 86.1% (CR), 85% (CBB) | 5 cycles | |
DAB-TFP-flow | Enamine, keto | 0.5–1.0 | −31 mV | — | 99.3% (MB) | 10 cycles | 310 |
DAB-TFP-batch | Enamine, keto | 0.5–1.0 | −27 mV | — | 39.4% (MB) | — |
Hybrid reticular membranes made up of both MOFs and COFs are a new area of research that has the potential to revolutionize the field of water remediation. The best of both worlds, i.e., hierarchical porous networks and hydrolytic stability, could aid their water treatment performance. Rafiee's group developed a MOF/COF composite (M5C) using MOF-5 and a melamine-terephthaldehyde based COF for the rapid removal of auramine O and RB cationic dyes from water by electrostatic and other weak interplay like H-bonding, Lewis acid–base and π–π stacking interactions.312 According to the authors, at pH 7 the adsorption capacity of M5C for AO and RB dyes was found to be 17.95 and 16.18 mg g−1, respectively.
Thiols and thioether functionalities are known to specifically adsorb mercury ions from water. Hence, porous COFs being functionalized with these sulfur functionalities are expected to capture Hg2+ ions from water. In this context, Wang's group was one of the first groups to report the application of fluorescent COFs to sense and remove Hg2+ ions and other metals.316 They have designed and constructed thioether based hydrazone linked COF-LZU8 where continuous π-conjugation present in the COF matrix behaves as an outstanding signal transducer. This was made possible by emphasizing the alteration in fluorescence signals and also the soft thioether nucleophilic centres as side arms to adsorb Hg2+. Hence, by applying both the bottom-up approach and PSM, the COF scaffold could provide fluorescence sensing, soft ionophores and high surface area for its utilization in heavy metal sensing and removal from water. Quenching of fluorescence emission, the resultant colour change detectable in a UV-lamp (λ = 365 nm) was found to be selective for Hg2+ among other metal cations. Thus, the easy synthesis, low cost, excellent robustness, pH independent performance and long 1D channels of COF-LZU8 with a pore diameter of 1.3 nm work in unison to outperform most of the reported materials for the related detection and removal of Hg2+ ions from aqueous solutions. Jiang's group also developed a thioether based imine linked highly robust 2D COF to remove Hg2+ ions from water.317 The structural design enhances the adsorption on the basis of large mesopores and soft-nucleophilic thioether centres that can efficiently adsorb Hg2+ ions with better selectivity and recyclability. The TAPB-BMTTPA-COF possesses benchmark mercury adsorption capacity (734 mg g−1) outperforming most of the reported adsorbents in the literature.318–320 The methyl groups attached to the thioether moieties presented in the 13C CP/MAS NMR spectra of the adsorbent COF shifted downfield after adsorption, indicating an interaction between Hg2+ and the thioether group. Additionally, the distribution co-efficient (Kd) was found to be comparable (7.82 × 105 mL g−1) to those of other state-of-the-art materials such as UiO-66-(SH)2 (ref. 318) and sulphur functionalised porous carbon.319 Moreover, pH dependent stability and adsorption are one of the key factors to be addressed while working with porous adsorbents. Nevertheless, the TAPB-BMTTPA-COF stability and adsorption capability were found to have a negligible effect over a broad pH range. The TAPB-BMTTPA-COF could be easily regenerated using 6 M HCl solution and recyclable up to six cycles with 92% capacity. Ma's group constructed COF-S-SH using simple thiol–ene click reactions by post synthetically modifying vinyl functionalised 2D imine linked COF-V with 1,2-ethanedithiol (Fig. 13a).321 COF-S-SH effectively removed both Hg2+ (1350 mg g−1) and Hg0 (863 mg g−1) from water and air, respectively, twice as much as sulphur functionalized mesoporous carbon. The superior performance of the COF is because of the intramolecular cooperativity between two sulphur atoms and one mercury atom as revealed by the XAFS study. Moreover, COF-S-SH displayed a high distribution co-efficient value (Kd) as high as 2.3 × 109 mL g−1 such that the Hg2+ concentration goes down from 5 ppm to well below the acceptable limit of drinking water (2 ppb) after just 30 min. The presence of free sulphur atoms could eliminate Hg2+, Pb2+ and Cu2+ below the U.S. Environment Protection Agency elemental limits for drinking water, within just 10 min. More significantly, COF-S-SH excluded most of the non-toxic ions like Na+, Ca2+, Mg2+, and Zn2+ creating a possibility to eliminate toxic ions in the presence of competitive non-toxic ions. According to the authors, the flexibility of the 1,2-dithiol moieties along with the appended thiol functionalities inside the porous matrix was key to this ultrahigh adsorption capacities of COF-S-SH as compared to COF-S-Ph-SH and COF-S-Et (Fig. 13b) adsorbents. The COF-S-SH adsorbent was regenerated using acetone and it was further recycled for four cycles without significant loss of Hg uptake capacity.
Fig. 13 (a) Post synthetically modified vinyl functionalized COF (COF-V) for mercury removal. (b) Schematic depicting the trade-off between flexibility and rigidity of thiol linked COFs for the capture of mercury from water [reproduced from ref. 321 with permission from American Chemical Society, copyright, 2023]. |
In another seminal study from Mancheño's group, thiol-based TPB-DMTP-COF-SH was developed by PSM of ethynyl functionalised [HCC–]0.5-TPB-DMTP-COF using simple click-chemistry.322 The post-synthetically modified COF was very efficient and selective for the removal of mercury from water. Even the Hg2+ concentration as high as 10 mg L−1 (in water) has been reduced to <2 μg L−1 in just a few minutes (Kd = 3.23 × 109) in the presence of the thiol modified COF. The TPB-DMTP-COF-SH has shown a record uptake capacity of 4395 mg g−1 with a Hg2+ retention value of 99.98% in just 2 minutes. The thiol functionalised COF also showed good selectivity while capturing toxic heavy metal ions such as Sn2+ and Pb2+ in comparison with Cd2+ or As3+ ions. More significantly, TPB-DMTP-COF-SH could remove mercury ions even in the presence of non-toxic metal ions and also from seawater. Recyclability studies (using 6 M HCl) further showed that the TPB-DMTP-COF-SH COF could be reused up to four consecutive cycles without significant loss in the adsorption performance.
Imine-linked COFs are one of the most explored COFs in the water remediation field. They could be broadly subclassified as β-ketoenamine linked and imine-linked COFs. Imine COFs are vulnerable towards acidic water environments whereas β-ketoenamine linked COFs are robust under acidic as well as alkaline aqueous conditions. In this regard, Li et al. has developed hydrazone linked COFs, namely TpODH and TFBODH with two different aldehydes as building blocks (triformylphloroglucinol and 1,3,5-triformyl benzene respectively).323 The TpODH COF had better crystallinity and hence greater surface area than the TFBODH COF due to extensive irreversible keto–enol/imine–enamine tautomerism and intramolecular hydrogen bonding. The presence of functional groups and the higher surface area (835 m2 g−1) of the TpODH COF were found to be the key factors for its enhanced adsorption capacities of Hg2+ and Cu2+ cationic adsorbates. The TpODH COF had high adsorption capacities of 324 and 1692 mg g−1 for Cu2+ and Hg2+ respectively, exceeding most of the earlier reported COFs. In contrast, the adsorption capacity of the TFBODH COF was found to be only 23 mg g−1 for Cu2+. TpODH was regenerated using 6 M acid solutions (HCl/HNO3) and was further subjected to Hg2+ and Cu2+ adsorption to achieve only 51 and 67% of the original adsorption capacities. This demonstrates their poor recyclability when compared to other state-of-the-art adsorbents in the literature.
Apart from the studies related to appended functionalities and different covalent linkages, the nature of the organic linkers (aliphatic/aromatic) also plays a key role in adsorbing heavy metal ions. In light of this, Liu's group developed amide-linked COFs viz. COF-TP and COF-TE from diamines and acyl chloride by mechanical ball milling at room temperature for Pb2+ removal.324 Two different kinds of diamine monomers were selected to generate both the COFs i.e., COF-TP was constructed from aromatic diamines whereas COF-TE was prepared using aliphatic diamines. This approach resulted in different adsorption capacities of Pb2+ in the COFs. COF-TE being constructed from aliphatic diamines resulted in higher adsorption capacities of Pb2+ than COF-TP. This is due to the missing π–π stacking interaction in the adsorbent, resulting in better diffusion of Pb2+ ions. Moreover, the authors showcased recyclability experiments where 1% HCl solutions were employed to regenerate the adsorbents. Remarkably, even after undergoing 10 cycles of adsorption and desorption, nearly 95% of the initial adsorption capacity could be preserved.
Most recently, Khojastehnezhad et al. demonstrated the significance of PSM of a COF (covalent organic framework) in enhancing the heavy metal removal performance.325 Initially, they synthesized an imine-linked TFPOT-PDA COF on the surface of spherical Fe3O4 nanoparticles with a particle size of 200 nm. Subsequently, a three-step PSM was carried out to incorporate thiourea units into the COF structure. These thiourea units exhibited exceptional activity as selective sites for adsorbing Hg2+ ions from aqueous solutions, achieving an extraordinarily high uptake capacity of 1400.9 mg g−1. The study revealed that other functional groups such as imine and secondary amine groups played a minor role in capturing Hg2+, emphasizing the crucial role of the thiocarbonyl group in the adsorption process. Furthermore, the adsorbent demonstrated excellent recyclability, maintaining its effectiveness for up to 5 cycles without significant loss of activity. Furthermore, the incorporation of a Fe3O4 core in the core–shell adsorbents facilitated the convenient retrieval of the adsorbents following each utilization (Table 6).
Adsorbent | Linkage | Pore size | Functional group | Heavy metals | Maximum removal capacity | Equilibrium time of adsorption | Optimum pH | Competitive ions | Recyclability | Ref. |
---|---|---|---|---|---|---|---|---|---|---|
TAPB-BMTTPA-COF | Imine | 3.2 | Imine, thioether | Hg(II) and Pb(II) | 734 and 726 mg g−1 | 10 min | 3–4 | Zn(II), Fe(III), Mg(II), Ca(II), and K(I) | 6 cycles | 317 |
COF-S-SH | Imine | <2.8 | Imine, thiol and thioether | Hg(II) and Hg(0) | 1350 and 863 mg g−1 | 30 min | 3–10 | Hg(II), Pb(II), Cu(II), Ca(II), Mg(II), Zn(II), Hg(0), and Na(I) | 321 | |
TPB-DMTP-COF-SH | Imine | 2.14 | Imine, methoxy, triazole, thiol | Hg(II), Sn(II) | 4395 and 4350 mg g−1 | 2 min | 5.6 | Sn(II), Pb(II), Cd(II), As(III), Hg(II) | 322 | |
TpODH | Ketoenamine | 0.5–1.2 | Enamine, keto, urea | Cu(II) and Hg(II) | 324 and 1692 mg g−1 | 250 min | 4 | Hg(II), Cu(II), Pb(II), Cr(II), Cd(II) | 4 cycles | 326 |
COF-TP | Amide | — | Aromatic amide | Pb(II) | 140 mg g−1 | 25 min | 6.5 | — | 10 cycles | 324 |
COF-TE | Amide | — | Aliphatic amide | Pb(II) | 185.7 mg g−1 | 25 min | 6.5 | — | 10 cycles | |
Fe3O4@RCOF-EBH-TSC | Imine | 4.01 | Hydroxyl, amine and thiocarbonyl | Hg2+ | 1400.9 | 5 min | 6 | Pb(II), Zn(II), Fe(III), Cd(II), Ni(II), Mn(II), Ag(I) | 5 cycles | 325 |
Altogether, COFs as a subclass of organic polymers have shown tremendous potential in heavy metal removal from water. As evident from the above discussions, there are very few reports summarizing the benefits of COF based membranes for the separation of heavy metal ions. Nevertheless, a thorough examination of the literature reveals that the effectiveness of heavy metal adsorption is significantly influenced by various factors, namely linking units, interlayer stacking, functional groups, hydrophilicity, and surface area. Therefore, when designing a membrane based on COFs for heavy metal adsorption, it is crucial to take into account these aforementioned factors, as well as their structural stability under different pH conditions. Additionally, another important aspect that requires careful consideration is the regeneration of the adsorbents. In many large-scale processes, the use of corrosive acidic solutions or sodium sulphide solutions can introduce complexities and increase the cost of the process. In addition, it is worth mentioning that a majority of the existing literature focuses on batch adsorption processes, which not only result in compromised removal of heavy metal ions but also lead to a more cumbersome and space-intensive adsorption procedure. However, it should be noted that these issues can be readily addressed by employing continuous flow techniques, which offer a viable solution to enhance the efficiency and streamline the adsorption process. Considering the progress made in COF-particle based adsorbents so far, a surge in the development of COF based membranes for the removal of heavy metal ions could be expected in the near future.
In 2019, Wang and Wei groups came up with an idea to investigate the conduct of transportation of water and salt ions based on multilayer stacking.328 The authors showed that with an increase in the number of layers, the water permeation decreased whereas the salt ion rejection performance increased. The number of water-hydrogen bonding interactions increases with the increase in the number of layers, which results in decreased water permeance. Generally, COF monolayers tend to stack for the formation of laminated multi-layers to minimise the surface energy. Hence, there is a possibility that the way of stacking might affect the water transport and salt rejection performance. To gain further insight, the authors have modelled 25 monolayers of TpPa-1 COF, with a pore size of 1.58 nm. They have found that when the monolayers were stacked in a fully eclipsed fashion, the water permeation was 3201 L m−2 h−1 bar−1 whereas the MgCl2 rejection was restricted to 42%. In contrast, when the COF monolayers were stacked in an offset-eclipsed fashion, with an effective pore diameter of 0.89 nm, the water permeation decreased to 1118 L m−2 h−1 bar−1 while the MgCl2 rejection increased to 100%. Hence, a clear idea on the impact of layer thickness and subsequent arrangement of COF layers can be drawn from this work.
Implementation of MMMs for desalination by the introduction of COFs into TFC or TFN membranes is another research direction explored in the literature.329–332 However, poor pore connectivity, the presence of irregular pore channels and poor control over active COF layer thickness restrict the water permeance. For instance, in 2016, the Wu group developed TFN NF membranes using COFs as additives for the first time.331 The TFN membranes exhibited outstanding rejection performance (>90%) of NaSO4 from water. Interfacial assembly of a TpPa-COF interlayer on a PSF substrate followed by PA skin layer grafting by the interfacial polymerisation process led Gao’s group to achieve a sandwich-like TFC RO membrane which could reject NaCl as high as 99.2% with compromised (16.78 L m−2 h−1 MPa−1) water permeance.332 The introduction of the COF active layer on the substrate system resulted in 33.8% better water permeance and marginally increased NaCl rejection performance. However, COF-based membranes have been used as adsorbents to remove bivalent ions from water.
In another study, to overcome the problem of larger pore aperture in the COF matrix, Huang and Jiang groups developed a PSM strategy to narrow the pore apertures of IISERP-COF1 (Fig. 14a), resulting in restricted accessibility of ions inside the pores.333 The simple PSM approach was done using a ring opening reaction with a cyclic anhydride molecule converting the pore size from 12.7 to 6.5 Å (Fig. 14b and c). The post functionalization strategy not only constricted the pore size, but also reduced the transport through intracrystalline defects. As evident from the studies, the ion rejection rates for the desalination of Na2SO4, MgSO4, FeCl3, MgCl2 and NaCl solutions increased from 73.1%, 76.5%, 89.8%, 64.6%, and 56.4% to 96.3%, 97.2%, 99.6%, 90.6%, and 82.9%, respectively (Fig. 14d).
Fig. 14 Salt rejection performance of IISERP-COF1 and IISERP-COOH-COF1: (a) schematic showing the PSM and membrane fabrication process; (b) N2 sorption isotherms and (c) pore-size distribution of unmodified and modified IISERP-COF1along with (d) salt rejection performance [reproduced from ref. 333 with permission from The Royal Society of Chemistry, copyright, 2023]. |
So far, careful investigation of the literature has indicated two key aspects to look for when designing COF-based membranes for improved desalination performance, which are smaller pore apertures and defect-free ultrathin COF layers. By considering both factors, Wang and Wei groups obtained an imine-based 2D COF (TpHz) on a PEI modified PES substrate by using a secondary growth approach.334 The authors grew TpHz on top of the PES substrate by a unidirectionally diffused growth strategy such that the overall thickness of the COF layer was adequate for the fast permeation of water. The as-prepared TpHz membrane, with a pore aperture of 0.8 nm, exhibited a Na2SO4 rejection of 58.3% and a water permeance of 40.5 L m−1 h−1 MPa−1, being one order of magnitude better than other reported COF membranes with similar salt rejection performance.
COFs being made of solely organic building blocks are capable of better dispersibility and compatibility in polymer solutions, making them a good candidate for MMMs. More significantly, COFs can be loaded as high as 50% in the MMMs being difficult for most of the other 2D materials and porous materials.335 In addition, COFs are better suitable as molecular sieves in water treatment owing to their larger pore sizes (1–4 nm). However, COFs with an ultra-microporous nature are infrequent in the literature, making COFs even harder to utilize as RO membranes for desalination. Currently it is challenging to fabricate COF-based membranes at the large scale due to the complex and sluggish fabrication methods. Hence, advancements in cost friendly fabrication methods need to be evolved to use COFs as membranes at the industrial scale.
A comparison of RO membranes before and after ordinary alkaline and acid washing was conducted by Yu et al. to examine the impact of chemical cleaning on the elimination of membrane foulants.339 Due to its highest content in the feed water, calcium was the main inorganic component of the foulants. Due to their high deposition ratios and low removal efficiency, aluminum and iron were also prevalent components on the membranes. Before cleaning, the two greatest dissolved organic matter (DOM) fractions on the membranes were hydrophilic neutrals (HIN) and hydrophobic neutrals (HON). Hydrophilic acids (HIA) and HIN could not be sufficiently eliminated. 90% and 94% of all bacteria on the lead and tail membranes, respectively, were eliminated by chemical cleaning, which also significantly altered the composition of the microbial populations.
A UF hollow-fiber membrane based on PAN was developed by Kim et al. for use in water filtration. In comparison to a virgin UF hollow-fiber membrane, the modified membrane displayed increased negative charge and hydrophilicity and decreased surface roughness.340 In a lab setting, humic acid and sodium alginate were used to mimic the effects of fouling and performance in a cross-flow system operating in in-out mode. Humic acid and sodium alginate were removed with 65% and 73% efficiency, respectively, in the unaltered hollow-fiber membrane, while the proposed membrane achieved 93% and 95%. They used sodium NaOCl as a chemically improved backwashing agent to reduce organic fouling on the suggested membrane. The membrane flux recovery was maximized at 92.1% of the initial permeability of the PAN-NF hollow-fiber membrane at a NaOCl concentration of 1 mg L−1 with a backwashing period of 30 s, with minimal membrane degradation.
Nilusha et al. developed and optimized a unique backwashing, filtration, and relaxation approach for steady operation of a side stream tubular anaerobic ceramic membrane bioreactor handling domestic wastewater at room temperature.341 Two in situ backwashing schedules were examined, one that backwashed (once a day at 60 s per day, and twice a day at 60 s × 2 per day) rest times at 55 minutes and 5 minutes, respectively. Permeate backwashing in situ stabilized the flow at a level greater than 70% of the original flux across the membrane. After 60 days of operation, ex situ chemical cleaning was performed utilizing a sequence of pure water, NaOCl, and citric acid. In situ backwashing was able to successfully thin the dominating cake layer, and fulvic acid analogues and humic acid analogues were found to be the most pervasive organic foulants. The most common microbes adhered to the ceramic membrane fouling layer were proteobacteria, firmicutes, epsilon bacteria, and bacteroides, all of which were easily washed away by NaOCl.
Membrane fouling has a negative effect on process efficacy, resulting in increased use of energy and a shorter membrane lifetime. Variations in light/dark photoperiods have been introduced to optimize treatment performance and improve biomass properties, with the goal of minimizing membrane contamination, in order to mitigate this issue. E. Segredo-Morales et al. observed the effect of photoperiod on the efficacy of a membrane bioreactor (MBR), specifically membrane fouling.342 Excellent amounts of biomass concentration (3.21 ± 0.45 g L−1) and nutrient elimination rates (4.71 mg N L−1 d−1 and 0.67 mg P L−1 d−1, respectively) were attained under a moderate photoperiod of 16/8 h.
There are significant technical concerns regarding the potential for secondary contamination due to nanoparticles leaking from polymeric membranes (such as polyethersulfone (PES) UF membranes) during filtering. Studies have demonstrated that nanoparticles have the capacity to leach from the membrane polymer matrix, despite the fact that this has not been widely documented. The antibacterial and antibiofouling capabilities of PES membranes treated with biogenic silver nanoparticles, for instance, were studied by Zhang et al.343 Silver nanoparticles enhanced the hydrophilicity and permeated flux of membranes. The maximal leaching rate of the nanoparticles from the polymer matrix was reported to be 92.0 g m−2 h−1. In a different study, Liu et al.344 examined silver leaching from nanocomposite membranes using a quartz crystal microbalance with dissipation. The initial leaching rate from the nanocomposite membrane reached 800 ng cm−2 h−1, but it gradually dropped over time, according to the authors. Given that it is unknown how poisonous nanomaterials are to humans and other living things, leaching of nanoparticles could be hazardous to health. According to several investigations, the membrane polymer matrix does not allow nanoparticles anchored in GO sheets to leak.345 The majority of the time, however, nanoparticles are integrated directly into the membrane matrix rather than being anchored in GO.
The integration of nanoparticles on membranes is still the subject of several investigations; however, most studies are primarily concerned with enhancing separation qualities rather than long-term adherence of the nanoparticles to the membrane polymer. In addition to long-term use, when the retentate is washed on the membrane surface during filtering in cross-flow mode, leaching could arise during membrane cleaning, where the cleaning solution may impair connections between the membrane polymer and nanoparticles. The effluent (permeate) from the membrane filtration system enters the water distribution system directly during membrane filtration. If nanoparticles are present in the effluent and manage to pass through the membrane during the filtration process, they could cause secondary contamination.346 The nanoparticles may enter the retained water and cause further pollution even if they do not pass through the membrane. It is thought that changes in the membrane surface characteristics and membrane filtration efficiency would indicate the leaching of nanoparticles from the membrane surface. For instance, when the nanoparticles have been washed from the membrane matrix, a change in the membrane surface's shape, hydrophobicity, roughness, and permeability to pure water is anticipated. Previous research has demonstrated that after being exposed to pH = 8 for 144 h, the polymeric membrane permeability rose to almost double its starting value.347 After washing with NaOCl and NaOH, the membrane pore size increased from around 200 to 640 nm, demonstrating the influence of membrane wear out due to cleaning.348 Nanoplastics and microplastics may leach from membrane systems due to mechanical stress, chemical agents, physical flushing, ageing, wear, and other reasons, according to a recent paper by Ding et al.349 This is a dilemma because poisonous organic molecules might adhere to the hydrophobic plastics. The disposal of old membranes is seriously hampered by the presence of nanoparticles in the polymer matrix. Only by minimizing the wastewater produced during the membrane manufacturing process will membrane technology become green and sustainable. For the first time, Mayamin Razali et al. demonstrated and confirmed a realistic and successful method for continuously recycling membrane effluent, lowering waste creation by 99%. Using membrane production wastewater contaminated with DMF and NMP, seven different classes of adsorbent materials charcoal, MIPs, zeolites, MOFs, graphene, PIMs, and resins were fully screened, with the lowest consumption of energy of about 1200 kJ per m2 of the membrane manufactured.68
As an alternative to zwitterions, mixed-charge materials have also been investigated. Chen et al. demonstrated that when the surface composition of counter charges is 1:1, mixed self-assembled monolayers (SAMs) containing tetraalkylammonium groups and monovalent acid thiols resist fibrinogen adsorption.354 Bernards et al. demonstrated that when the monomer ratio of positively charged [2-(meth-acryloyloxy)ethyl]trimethylammonium chloride and negatively charged 3-sulfopropyl methacrylate potassium salt in the feed solution is 1:1, mixed-charge polymer brushes grafted from surfaces via surface-initiated atom transfer polymerization (SI-ATRP) can eliminate protein adsorption.355 Furthermore, Chen et al. demonstrated that hydrogels made from valence-balanced polyampholyte aminoethyl methacrylate hydrochloride and 2-carboxyethyl acrylate are highly resistant to protein adsorption.356 These findings suggested that zero charge valence from either zwitterion or mixed-charge groups is key to these surfaces' antifouling properties, as electrostatic interaction is presumably minimized in this way. Several studies reported adjustment of the isoelectric point (IsP) of surfaces using mixed-charge groups via SAMs. Lin et al. co-utilized carboxylic acid (from 16-mercaptohexa-decanoic acid) and amine groups (from 8-amino-1-octanethiol) in solution and achieved PIs from 3.2 to7.3.357 Kuo et al. used mixed thiol (from 3-mercaptopropyltrimethoxysilane) and amine (from 3-aminopropyltrimethoxysiliane) groups to obtain IsPs from <2 to 7.358 SAMs are useful for molecular-level studies but they have limited stability to resist protein adsorption and bacterial adhesion. Alternatively, polymer brushes formed using SI-ATRP have shown to be more chemically stable and more densely packed. They also exhibit stronger resistance to protein adsorption and bacterial adhesion than their SAM counterparts. Thus, SI-ATRP is an attractive polymerization method to introduce charged groups to surfaces. The difficulty to control a single surface property while maintaining other essential physicochemical properties unchanged has been pointed out by several reports.48 This interdependency of surface properties makes it hard to isolate the effect of a single property in antifouling studies. When controlling surface IsPs is the objective, this problem proves to be difficult to resolve as well. For example, recently Peng et al. used photoinitiated graft polymerization to introduce mixed-charge SO3− and N(CH3)3+ groups onto cyclic olefin copolymer microchips.359 They demonstrated that the PIs of the surfaces can be tuned by varying ratios of these two monomers and obtained three PIs of <3, 5 and 7. However, the variation in contact angles of these surfaces can reach around 15, thus the effect of surface charge on fouling cannot be disentangled from the effect of surface energy. Nevertheless, the ideal protocol for controlling surface PIs should be capable of isolating this material parameter from other properties.
Fig. 15 A conclusive schematic of the current trends in membrane modifications towards sustainable and highly efficient wastewater remediation. |
MB | Methylene blue |
MO | Methyl orange |
CR | Congo red |
RB | Rose bengal |
AF | Acid fuchsin |
CV | Crystal violet |
CBB | Coomassie brilliant blue R250 |
EY | Eosin yellow |
MG | Malachite green |
RhB | Rhodamine B |
RB5 | Reactive black 5 |
OG | Orange G |
MR | Methyl red |
CB-T | Chrome black T |
CA | Cellulose acetate |
CA-dye | Chrome azurol |
AA | Anthraflavic acid |
BB | Brilliant blue G |
TB | Thymolphthalein blue |
AG-25 | Acid green 25 |
DFBM | Direct fast brown M |
IC | Indigo carmine |
AR-27 | Acid blue 27 |
FS | Fluorescein sodium salt |
PP | Potassium permanganate |
DFPB | N,N-Dimethyl-p-phenylene-diamine dihydrochloride |
NR | Nile red |
NA | p-Nitroaniline |
CAc | Calcein |
DR 80 | Direct red 80 |
EBT | Eriochrome black T |
AB | Alcian blue |
This journal is © The Royal Society of Chemistry 2024 |