*3.3. Membrane Filtration*/*Adsorption*

Membrane separation is another emerging technology that has shown grea<sup>t</sup> promise in HM separation from polluted waters. The focus for development of a membrane-based solution for HM remediation intensified when researchers discovered problems in other common remediation, which include the poor reusability of adsorbents, high cost of material development, and di fficulty in separating nanomaterials from the water after remediation [50]. The employment of membrane-based remediation method can address these problems, as membrane that is incorporated with novel nanomaterials is a one-step method that can be reused while maintaining excellent rejection of HM in aqueous solution without the need of chemicals or pre-treatments. In the search for the development of membranes that are low in cost, high reusability, greater selectivity, better water transport, and high HM ions rejection, researchers are exploring the possibility of incorporation of various types of nanomaterials in a bid to impart the unique characteristics of nanomaterials into membranes. The addition of high surface area nanomaterials has developed highly adsorptive membranes [51], while the incorporation of highly hydrophilic nanomaterials on the membrane surface has significantly improved the water permeability of membranes [52]. Photocatalytic hybrid nanomaterials, such as Graphitic carbon nitride (g-C3N4) quantum dots (QD) [53] and TNT array [54], are also developed to allow for simultaneous photocatalysis and membrane filtration in efficient wastewater treatment [55]. Figure 4 shows the utilization of several membrane-based technologies in the remediation of water bodies laden with HM ions [56–58].

**Figure 4.** HM removal via (**a**) adsorptive membrane technique. Adapted from [56], with permission from Elsevier, 2017. (**b**) surface-charged modified membrane repellent. Adapted from [57], with permission from Elsevier, 2019. and (**c**) size exclusion of HM ions. Adapted from [58], with permission from Elsevier, 2019.

Even though there are many ways that membranes are employed to remove HM, including surface charged membranes for HM ion repellent, membrane distillation [59], adsorptive membrane [60], size exclusion removal [57], and more, there are two popular ways where HM can be removed from water bodies using membrane-based remediation, which is size exclusion removal, or using adsorptive membranes [61]. Commonly, membrane works by sieving the molecules according to size. Only particles that are larger than the pore size are retained. In this case, the removal of HM requires the employment of nanofiltration (NF) membranes, as the pore size of common ultrafiltration (UF) membranes are large and would allow HM ions to pass through. However, NF membranes face one prominent problem. They will restrict the movement of water through the membrane since the pore size is very small, which drastically reduces membrane flux [62]. To overcome this, researchers have explored the idea of adding novel nanomaterials into these membranes to improve the permeation, whilst maintaining or enhancing the HM rejection rate. Nanomaterials, such as halloysite nanotubes [63], CNT [64], metal oxide nanoparticles [65], and many more are blended into a polymeric matrix or are deposited on the membrane surface to improve selectivity and permeation, whilst maintaining or enhancing rejection capabilities. The improved performance of membranes is owed to the modification of characteristics due to the presence of nanomaterials. Materials, such as CNT's, create a pathway for water transport through the membrane matrix, increasing water permeation [66]. Other nanomaterials, such as amine grafted SiO2, are able to impart superhydrophilicity into a polymeric membrane, which is generally hydrophobic in nature [67]. Adsorptive filtration is another way where membranes can be employed for HM removal. Albeit, adsorption itself is a promising way to treat HM waters, it is bugged by a few problems. Sorbents always tend to float on the water surface, and this would render proper mixing with water to achieve maximum contact between adsorbent and HM ions untenable [68]. If the adsorbents are properly mixed, then their form as a particle would require a form of method to separate the used adsorbent from the water source [69]. This would incur additional time and resource. Hence, adsorptive membranes have been developed to overcome this, where potent adsorbents are immobilized into the membrane matrix [70]. Adsorptive membranes exhibit the ability to trap the HM ions and at the same time, allow the filterability of water, and produce clean permeates with the metal trapped in the membrane matrix. Commonly, UF membranes are used as an adsorptive membrane, as it has a larger pore size as compared to the NF membranes. To overcome the trade-o ff for the low rejection of HM ions, nano-adsorbent is incorporated into UF membranes to maintain both high membrane flux due to the larger pore size and excellent HM removal.

#### **4. Nanomaterials for Heavy Metal Removal**

Adsorption has shown the greatest potential in terms of cost and e ffectiveness, even though all of the techniques are able to perform admirably in removing HM from water sources, while membrane filtration techniques have also shown good performance with long term stability and usage [62]. In addition to this, photocatalysis, have shown significant promise, owing to its non-selective degradation, excellent ability to mineralise pollutants, and good reusability [71]. Nanomaterials have played very significant roles in advancing the HM removal technologies in all of the methods that were mentioned above. Nanomaterials are described as a material with a length of 1 to 100 nm in at least one dimension [14]. Their small size allows for them to exhibit unique properties that are not shown in bulk, with some examples of properties including increased surface pore and surface area [72], improved electrical properties, increased material strength and conductivity, and self-cleaning properties [73]. Nanomaterials can be classified into metallic, non-metallic, and their composites. Metal oxides and semiconductors are common sources of metallic nanomaterials, whereas non-metallic nanomaterials include carbonous nanomaterials, such as (CNT) [74] and graphene. Nanocomposites can be multiphase nanomaterials, where one part is defined as a nanomaterial in terms of size, whilst the other may also be a nanomaterial or they can be materials that are larger with bulk-type property. In addition to this, nanomaterials are also known according to the shape that they exhibit. Some of the nanomaterial shapes include nanoparticles [75] (spherical/globular), nanosheets [76], nanowires, nanoflowers [77], nanotubes [78], and nanorods [79].

#### *4.1. Motivation of Using Nanomaterial for Heavy Metal Removal*

The utilization of nanoparticles for the remediation of environmental problems has shown remarkable potential in line with the rapid development of nanoscience and nanotechnology. The modification in atomic level to produce particles that are independent in nanoscale has provided a myriad of novel characteristics that cannot be found in bulk materials. Nanomaterials can be synthesized using bottom-up approach to carefully tailor the desired properties, such as surface charges and functionalities to interact with HM. Nanostructured materials have shown exceptionally high surface area and porosity, higher e fficiency as an absorbent due to their superior surface to volume ratio, improved solubility, abundant reaction sides, photocatalytic properties, grea<sup>t</sup> surface charge, and lighter in weight or mass. These nanomaterials can also be modified via various techniques, such as surface grafting and gamma irradiation, to increase its surface reactivity [80]. CNT has also been used as a prominent adsorbent for HM removal due to their high surface area to volume ratio and their highly tunable characteristics. Various functional groups, such as hydroxyl and carbonyl groups, which can provide new adsorption sites, are easily tuned on the surface of CNT [81]. Clay can also be utilized as an e fficient absorbent of HM via nanoscience. Studies have shown that common clay that is combined with activated carbon, another low cost and a common material, are able to absorb HM ions, such as Cd (II), Ba (II), and Cu (II) from pulp wastewater [82]. The synthesized absorbent was able to exhibit a surface area of close to 800 m<sup>2</sup>/g, which is ten-fold larger when compared to individual nanomaterial. Another research also showed that the creation of nanomaterial via a nanocasting process using mesoporous hybrid material with ZnO and TiO2 exhibited a surface area between 120–332 m<sup>2</sup>/<sup>L</sup> [83]. The absorbent also showed that it could be reused up to three times due to the micrometer-sized structure with high surface area, which has the benefit of reducing the overall cost in the adsorption process. In addition to this, semiconducting nanomaterials, which are also known as photocatalyst, can be employed in HM reduction. This is based on the fact that it has good optical properties and the energy band can be easily modified through facile modification or hybrid to render improved properties, such as lower bang gap energy, lower recombination rates, and larger active sites for photocatalysis. Nanomaterials of unique features can also be incorporated with other technologies to create a synergistic improvement in HM removal. For instance, the incorporation of novel nanomaterials into polymeric membrane matrices, such as CNT, TiO2, and hydrous manganese dioxide (HMO) can increase the pathways for water transport, impart photocatalytic activity, and increase membrane hydrophilicity respectively. These nanomaterials can also deposit onto the membrane surface that can greatly govern the selectivity of membranes, something that is not possible with polymers alone. The utilization of nanoscience also enabled the creation of functional nanomaterials from waste source, such as the formation of a biogenic iron (Fe) compound at a size of 500 nm, using a natural microbial consortium that was sourced from an abandoned mine containing iron oxides (Fe2O3) and siderite by bioreduction of ferric citrate [84].

#### *4.2. Classification of Nanomaterials*

As researchers further pushed the boundary of nanoscience in the development of novel, functional materials, they discovered that the shape of nanomaterials could be manipulated. The di ffering shapes opened a vast array of new and unique characteristics that were not possible in its benign shape. Carbonous materials led the way, where carbon was used as the building block in building two di fferent variants, namely CNT and graphene [76]. CNT follows the shape of tubes, while graphene took the shape of sheets, in multiple layers. Other forms also emerged, which includes the formation of nanorods and nanoflowers of di fferent, as shown in the micrographs in Figure 5 [85–88].

**Figure 5.** Examples of nanomaterial structures (**a**) nanoflowers. Adapted from [85]. (**b**) nanotubes. Adapted from [86]. (**c**) nanosheets. Adapted from [87], and (**d**) nanorods. Adapted from [88].

The change in structure can bring about new and unique features that otherwise are not exhibited in the bulk phase. CNT's are produced when sheets of carbon, which are called graphene, are rolled up to produce single-walled CNT's or multi-walled CNT's [89]. The formation of tubular structure allows graphene to be 400 times stronger than steel, allowing for conducting electricity and even working as a semiconductor, something that is not possible in bulk carbon [90]. This allows CNT's to be cheaper and more environmentally stable materials in the development of electrical and electronic products that rely heavily on rare earth metals. TiO2 is a metal oxide that can be used in paints and personal cosmetic products as pigments due to its stable and environmentally safe nature. TNT have been developed and it has shown grea<sup>t</sup> promise in gas sensing and increased photocatalytic activity [91]. The development of tubular and porous structure, such as nanotubes and hydrated manganese oxide nanoparticles, have shown grea<sup>t</sup> improvement in terms of e ffective surface area that can reach the region of 400 m<sup>2</sup>/g, which is a ten-fold increase [92]. Higher surface area leads to increased reaction sites, which is valuable in the catalytic industry. Nanoflowers have been developed from materials, such as TiO2 and Fe2O3, exhibiting a superb volume to area ratio, better charge transfer, carrier immobility, and an enhanced number of adsorption sites [93]. All of these characteristics can significantly contribute to the field, such as drug delivery, catalytic process, chelation, and adsorption of HM ions [85]. In addition to these structures, nanorods/nanowires have also attracted the attention of researchers. Particularly, gold nanorods have received extensive attention, owing to their extremely attractive applications in biomedical technologies, plasmon-enhanced spectroscopies, pollutant remediation, and optical and optoelectronic devices. Copper nanoflowers have been shown to exhibit impressive adsorption of Pb (II) in aqueous solution, owing to its porous, high surface area structure, which significantly increases the availability of active sides and the presence of carboxylic (COOH) functional groups [94]. The flower structure is useful in drug delivery, as e fficient drug deliveries require a carrier that is non-hydrolysable, controlled release pattern, and reducing drug toxicity. Nanoflowers tick all of these boxes, as displayed by the e fficiency of sodium alginate/chitosan nanoflowers in drug delivery. Table 2 shows the types of nanomaterials and the important features that they exhibit as compared to the bulk material.


There are two ways where nanomaterials are structured, which can use a top-down or and bottom-up approach [107]. The top-down approach uses synthesis methods, such as lithography etching and exfoliating. This method is sparsely used due to the lack of versatility. Commonly, the bottom-up method is used, as it allows for control of the structure of nanomaterials at an atomic level, as different parameters can be manipulated to design the nanomaterial structure according to desire. Some of the common bottom-up methods in the design of nanomaterials include wet chemical precipitation, sol-gel, chemical vapor deposition, hydrothermal, sputtering, template growth, electrospinning, and atomic layer deposition [108]. In general, all of the techniques mentioned exhibit similar characteristics, which include high control towards to growth of structure, an abundant selection of precursors/starting material, control towards heat and temperature, high purity, and uniformity [109]. Section 4.3 greatly discusses the synthesis of nanomaterials.

#### 4.2.1. Metallic Nanomaterials

Metallic nanomaterials are formed from metal sources, such as titanium, iron, silver, gold, manganese, copper, and many more [110]. These nanomaterials can be produced in pure metallic form or in the metal oxide forms. Metal oxides are more favourable, as they exhibit increased stability over pure metal nanomaterials. Research that was conducted on the synthesis of nanomaterials using metals is aplenty based on current literature [111–114]. Fe2O3 is one of the commonly used metal oxide nanoparticles. Iron can work as an adsorbent for wastewater laden with HM ions. Castro et al. produced a biogenic iron compound using metal compounds that were sourced from mining wastewater via the bioreduction of ferric citrate [84]. The biogenic nature of iron compounds has high specific surface areas and high binding energies hence act as efficient adsorbents for HMs. In addition, the bacterial matrix surrounding the iron nanoprecipitates can bind harmful metals. The metal exhibited a surface area of 56.978 m<sup>2</sup>/g and a pore size of 8.304 nm. The large pore size allows access towards the more reactive sides to improve binding and capture the HM ions. Gold nanorods have been extensively studied due to the fact that they exhibit excellent plasmon-enhanced spectroscopies and optical and optoelectronic applications, which are hugely beneficial in the detection of HM in water samples via the colorimetric detection technique [63,77,115]. Many researchers have continuously studied upon its characteristics and its application for environmental remediation ever since the discovery of TiO2s ability to split water [116]. TiO2 in the form of anatase crystallinity exhibits grea<sup>t</sup> photocatalytic activity under ultraviolet (UV) light irradiation. The good crystallinity, together with its low band gap value (3.2 eV) and stable recombination rate, makes it an excellent candidate to reduce HMs into less harmful configurations (Cr (VI) to Cr (III)) and degrade various organic pollutants [117].

## 4.2.2. Non-Metallic Nanomaterials

The development of quantum dots is currently on the rise due to their superior features, such as edge morphology and increased in surface functional groups. Research that was done by Abdelsalam on the development of graphene quantum dots (GQDs) showed improved surface chemical functionality and well-defined edges, which is better when compared to the sheet structure that was exhibited by graphene [118]. GQD's adsorption capability of hydrated Cd (II) and Pb (II) calculated via density functional theory (DFT) showed that it was able to absorb the hydrated HMs through different positions and interactions, including edge and surface adsorption, interaction with unsaturated carbon atoms, and adsorption on the edge of the functionalized group. The edge of surface adsorption interaction is non-existent with other nanomaterials, which shows the uniqueness of GQDs [118]. Another example of efficient non-metallic adsorbent produced from waste is the development of CNT in the adsorption of Cd (II), Cu (II), Pb (II), and Hg (II) via chemical vapor deposition (CVD). CNT developed exhibited increased surface area and porosity, with evidence of abundant functional groups, such as O-H, C-H, C=O, C-N, C=N, phenols, aromatic rings, and aromatic groups, which provides high potential for the adsorption of HMs [47]. The developed CNT was able to act as an absorbent also shows promise, as the presence of a various functional reactive group, together with the porous structure and large

network of functional group, allows it to perform well as an adsorbent for HMs, such as Cu (II), Cd (II), and Pb (II) [119].
