1. Introduction

Purification processes such as distillation, evaporation, concentration, and crystallization are important in basic research as well as playing an important role in industries. These processes, however, operate at the expense of environmental deterioration by consuming an enormous amount of energy, further promoting global warming. Industrial effluents also contain a large amount of chemical and bio-chemical hazardous ingredients polluting the already scarce freshwater resources. Moreover, many industries waste a large amount of precious chemical compounds and organic solvents due to the lack of economical separation/purification materials. Therefore, purification processes with low energy requirements may benefit the environment by saving energy on one hand and saving important capital costs on the other. In recent years, adsorption- (entrapment) and membrane (size exclusion)-based purification have attracted immense research and industrial interest due to their low energy consumption as well as simple and environmentally friendly operation. Various amorphous materials such as hyper-cross-linked polymers (HCPs) [1,2], porous organic polymers (POPs) [3,4], conjugated microporous polymers (CMPs) [5,6,7], and activated carbon [8,9,10]; and crystalline materials such as metal-organic frameworks (MOFs) [11,12,13,14] and zeolites [15,16,17,18,19], have shown excellent preliminary separation performance. Covalent organic frameworks (COFs) are a class of crystalline framework material synthesized from purely organic building blocks. Their synthesis involves reticular chemistry, which gives immense freedom of pre-design. Their pore geometry, size, and functionalities can be pre-determined by choosing building units from a large bank. Yaghi and co-workers first reported COFs based on boroxine and boronate ester rings [20]. These COFs, however, are prone to deformation in the presence of even trace amounts of humidity rendering them unsuitable in aqueous conditions. Later, imine-based COFs were reported to have superior chemical and solvent stability. Banerjee and co-workers prepared β-ketoenamine COFs with exceptional stability at high temperatures and in extreme acidic and basic conditions [21]. Similarly, Thomas and co-workers reported triazine-based COFs prepared at 400 °C, exhibiting excellent thermo-chemical stabilities [22]. Some of the important applications and various types of COFs are shown in Scheme 1 and Figure 1, respectively. Among many important aspects, COFs offer pore post-functionalization due to their organic nature, for specific desirable applications such as gas storage [23,24,25], catalysis [26,27], electronic devices [11,28], electrode material for batteries [29,30,31,32], etc.

The most attractive feature of COFs is their framework structure with uniform and extended porous channels, attracting the interest of scientists in the purification/separation field. Earth’s environment is under threat from the increasing disposal of hazardous ingredients such as heavy metals and poisonous organic and bio-organic chemicals [33,34,35,36,37,38,39]. Due to their distinct features, COFs represent themselves as viable alternative materials to address these issues [40,41]. Suitable pore designs and functionalities can render COFs as adsorbents for trapping hazardous metals, organic and bio-pollutants, and greenhouse gases [39,42,43,44,45,46,47]. As adsorbents, COF pores attract and trap pollutants based on their affinity towards functional COFs. COFs are also touted as excellent robust materials to fabricate separation membranes [48,49]. In the membrane form, pollutants are separated based on diffusion rates relying on the size, geometry, or charge of permeate and retentate. COFs have been reported to remove toxic components from both gases and liquids. Moreover, functionally decorated COFs can facilitate catalytic degradation of pollutants and convert them to clean energy. Our group has extensively worked on metal oxides and their composites for the application of photo- and electro-catalytic process [50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67], but recently we have shifted our focus towards COFs due to their excellent characteristic properties. Although COFs were first reported in 2005, their environmental applications have only attracted interest very recently [68]. Since then, COFs have been extensively explored for various environmentally related applications. We believe that an up-to-date account is urgently required for scientists in this field.

2. Evolution of COF Synthesis through Time

A brief evolution overview of various COF synthetic methods since their first report will be discussed here. The first COF was based on boroxine and boronate-ester linkages [20]. Although highly crystalline, these COFs are not suitable for an aqueous environment as the crystallinity and framework nature is destroyed even in the presence of a trace amount of water due to the electron-deficient boron. Later on, various other COFs based on imine [27,70], triazine [71], hydrazine [72], and keto-enol [29,73] linkages were reported, exhibiting excellent stability in organic, aqueous environments as well as harsh acidic and basic conditions. The synthesis of COFs intended for environmental applications generally follows bottom-up and top-down approaches. The former involves solvothermal [20], interfacial polymerization [74,75], in-situ growth [76], micro-wave assisted [77], and on-surface crystallization [78,79] methods, and the latter involves delamination of COF powders into mono/few-layer sheets for further applications [80]. COF pore geometry can be pre-determined by choosing suitable linker symmetry (Figure 1) or modified through post-functionalization. Further elaboration of these synthetic and post-functionalization methods is beyond the scope of this review. Many other reviews have covered various aspects of COF reticular synthesis and properties in detail [81,82,83,84,85,86,87,88,89,90].

3. Important Aspects of COF Materials towards Cleaner Environment

Industries are polluting our environment in two ways: (i) disposal of hazardous chemicals in water reservoirs; (ii) and greenhouse gases into the air. Materials with advanced functionalities to trap these hazardous chemicals and gases are urgently required to address these issues [91]. COFs are materials that have intrinsic, uniform, ordered and tailorable pores along with high surface area, making them very attractive for trapping and separating these hazardous molecules. COFs with multi-functional pores have extended their applicability in environmental cleansing. Due to their high surface area and ordered porous channels, they can not only be used to trap/store gas molecules but can also function as separating media to remove unwanted chemicals from the waste solution. This application can reduce environmental pollution as well as help in the recovery of precious solvents to be re-used in industries, rendering the whole operation environmentally and economically friendly. Recent development has yielded COFs with extraordinary stability in harsh acidic and basic conditions, rendering them highly desirable in industrial purification and trapping applications [21,92]. The rational design of COFs needs careful consideration of many aspects for intended purification applications. In this section, we will discuss some important aspects of COFs such as structure, morphology, and charge of their pores, as well as their stability, which makes them ideal alternative materials for environmental applications.

3.1. Pore Structure

Trapping or separation processes involve distinct characters of pores, the most important being their size. The pore geometry and structure can be pre-designed by choosing monomers with appropriate symmetry, as shown in Figure 1. The size of environmental pollutants ranges from sub-nanometer (metals) to several nanometers (dyes) and even up to micrometer (bio-pollutants such as bacteria, etc.) [93,94]. Therefore, careful consideration should be given to designing COFs for specific purification applications. COFs with pore sizes ranging from 0.5–5 nm have been reported so far, making them highly desirable in size-dependent separation processes. Banerjee et al. exhibited control over pore size by cross-linking precursors of different lengths and obtained COFs with pores in the range of 1.4–2.6 nm. 1,3,5-trifromylphloroglucinol (TFP) was chosen as the aldehyde-bearing monomer, whereas four amine-bearing monomers with different lengths were chosen as linkers (Figure 2) [74].

Figure 2. Control of pore size through linker selection (top) and post-functionalization (bottom). Adapted with permission from references [74,95], copyright 2017 and 2019 American Chemical Society.

Similarly, Ditchel et al. efficiently reduced the pore size of COFs by choosing linkers with six methyl or ethyl groups directing into the pores of the framework [95]. Pore surface engineering or pore post-functionalization is another strategy to tune the pore structure of COFs. Jiang et al., for the first time, reported an interesting strategy to first synthesize COFs with azide functionalities [96]. The COFs were synthesized through the condensation reaction between azide-appended benzene diboronic acid (N3-BDBA) and benzene diboronic acid (BDBA) with hexahydroxytriphenylene (HHTP). The azide could be easily cross-linked with many moieties such as propyl acetate, –COOH, –NH₂, –COOMe, –OH, and –C≡C through click chemistry. The pore size was controlled between 1.2 and 3 nm by employing this strategy. Moreover, introducing these functional groups through post-functionalization also rendered COFs with desirable wettability and charge surfaces.

3.2. Hydrophobicity/Hydrophilicity of COFs

The morphology of COFs, such as their surface area and hydrophilic/hydrophobic nature, is very important for purification processes. COFs with a surface area of >2000 m²·g⁻¹ have already been reported [73,97]. Ordered porous channels along with such high surface area are highly desired for purification/trapping applications. Hydrophilic/hydrophobic properties of COFs are another important controllable aspect, which are exploited by merely choosing desired linkers or through post-functionalization. COFs with hydrophobic nature [98] will enhance their applications in organic media, whereas hydrophilic COFs [99] will work better in aqueous media. Zhang et al. prepared a superhydrophobic COF through pore surface functionalization and evaluated their application in harsh conditions [100]. They first synthesized the COF and the pores were grafted with fluoride. The contact angle was increased from 0° to 150° by varying fluoride grafting. The modified COF retained its crystallinity and hydrophobicity under extremely harsh conditions such as in boiling water and in solutions with pH ranging from 1 to 14. Similarly, Hu et al. synthesized a hydrophilic triazine-based COF (Figure 3) and used it as a sensor for the detection of gallic acid (GA) and uric acid (UA) [101].

Figure 3. Examples of hydrophobic and hydrophilic COFs. Adapted with permission from references [100], copyright 2020 American Chemical Society and [101], copyright 2021 Elsevier.

3.3. Structural Stability

The first reported COFs based on boroxine and boronate ester linkages had poor stability in even a small amount of water. This phenomenon arises from the fact that the boron sites are electron deficient and can undergo nucleophilic reaction, resulting in structural degradation through hydrolysis. The effect is more severe as water is produced as a by-product during COF synthesis, which can facilitate the backward reaction and therefore severely affect their industrial applications. For environmental applications, COFs need to retain their ordered structure in practically harsh conditions. Linkers with strong covalent bonding to extend the framework, along with hydrogen bonding between interlayers, can overcome this shortcoming to some extent. Imine, azines, hydrazine, imides, and triazine-based COFs exhibit exceptional stability in harsh conditions because they are synthesized through acidic catalyst-based reversible reactions. Therefore, the backward reaction will be facilitated in an acidic environment but the COF should be stable in water as well as organic solvents [102]. Banerjee et al. explored a reversible/irreversible approach to improve COF stability one step further in solvents with pH ranging from 1 to 14 [21]. The same group also exhibited that increasing hydrogen bonding in the framework can also improve the stability of COFs [103]. They incorporated –OH functionalities adjacent to the –C=N bonds to introduce –OH–N=C hydrogen bonds, which ultimately safeguarded the imine nitrogen from hydrolysis in the presence of both acids and water. Similarly, other groups have improved COF stability through increasing intra-molecular hydrogen bonding to improve the COFs overall thermo-chemical stability [104,105].

3.4. Pore Charge

Pore charge is a crucial factor in the separation or trapping of hazardous chemicals through electrostatic interactions. According to Donnan’s theory, a negative charge-separating barrier will repulse divalent anions whereas divalent cations will be attracted. Therefore, if the negatively charged separating barrier is an adsorption agent then the cations will be trapped inside the matrix, but if the barrier is a membrane then the positively charged ions will transport more efficiently leaving anions in the feed. The same phenomenon is applied to the oppositely charged separating barrier. So far, very little attention has been given to design-charged COFs. Ma and co-workers introduced cationic sites in the COF (EB-COF:Br) by reacting 1,3,5-triformylphloroglucnol and ethidium bromide (EB) (3,8-diamino-5-ethyl-6-phenylphenanthridinium bromide) [106]. Similarly, Oakey et al. prepared a negatively charged COF by introducing carboxyl functional groups in the COF’s backbone and introduced them as fillers in the preparation of mixed matrix membranes [107]. In addition to the narrow size distribution of COF pores, the deprotonated -COOH⁻ enhanced the rejection of bovine serum albumin (negatively charged protein) to as high as 81% at a COF loading of 0.8%. A similar approach was adopted to prepare a COF with modifiable carboxyl functional groups to prepare 12 COFs with variable aperture size and was self-assembled as continuous membranes [108]. It is expected that the synthesis of COFs based on charged pores for environmental application will attract more attention in coming years by incorporating functional groups such as hydroxyl, sulfonic acid, amine, etc.

This entry is adapted from the peer-reviewed paper 10.3390/en14082267