Review on some metal oxide nanoparticles as effective adsorbent in wastewater treatment

Between 1990 and 2015, the world's population increased from 5.3 billion in 1990 to 7.3 billion in 2015, and it is currently growing at an annual rate of 81 million people. Due to this huge growth rate of the population, the demand for fresh and clean drinking water increases excessively (Information Resources Management Association 2020). Water and surrounding soils are becoming severely contaminated due to industry growth, unplanned urbanisation, and population growth. The use of contaminated water is responsible for 70–80% of all diseases (Boretti & Rosa 2019). Meanwhile, heavy metals and toxic azo dyes are the main cause of water pollutants, which has become a global environmental concern.

The predominant uses of dyes are in the textile industry as colouring agents. The textile industry is a global phenomenon and one of the largest industries on the planet, generating more than $1 trillion in revenue and accounting for 7% of global export. Advancement of textile and dye industry results in formation of harmful organic wastes and thus increases environmental pollution. This is due to the discharge of untreated effluents containing a large amount of dyes into the water bodies. Dyes and heavy metals are discharged mainly into the environment through mining, chemical plants, electroplating, agriculture, industrial effluents, and domestic wastewater, among other sources. Excessive exposure to heavy metals like Cu, Zn, Hg, Cr, Pb and others can result in a variety of health problems such as diarrhoea, dysentery, typhoid, respiratory infection, liver or kidney damage, cancer, and so on (Ernhart 1992; Goyal 2005; Khlifi & Hamza-Chaffai 2010). As a result, advanced wastewater purification technologies must be developed and used to expand, build the economy, and sustain health. There are numerous traditional methods, including electro coagulation (Ernhart 1992), potentiometric (Khani et al. 2010), visible light degradation (Saravanan et al. 2013, 2014), adsorption (Dahri et al. 2014), and bio degradation (Ayed et al. 2009; Gupta et al. 2015). However, traditional techniques have limitations, such as lower efficiency, more delicate operating conditions, and a costlier disposal process. The adsorption process, on the other hand, has become increasingly popular because of its numerous advantages. Liquid phase adsorption exhibits a momentous advantage over other techniques because of its great effectiveness in removing hazardous contaminants, ease of operation and cheaper cost. Depending on their size, shape, and chemical and physical properties, nanoparticles can be categorised into different groups. Some of them are considered semiconductor nanoparticles, ceramic nanoparticles, polymeric nanoparticles, carbon-based nanoparticles, lipid-based nanoparticles, and metal nanoparticles. Because of their atomic and molecular origins, nanoparticles have complicated chemical and physical properties. Small clusters’ electrical and optical properties, as well as their reactivity, are fundamentally different from the more familiar properties of each component on a wide or extended surface. Van der Waals forces, greater polar and electrostatic contacts, and covalent interactions can all affect particle interactions at the nanometer scale. Particle aggregation is influenced by the viscosity and polarity of the fluid. The propensity of a colloid to coagulate can be increased or decreased by altering the surface. Nanoscience is garnering a lot of research funds, some of which should be allocated to raise public awareness regarding proper nanoparticle disposal. Metal oxide nanoparticles (MONPs) are important in a variety of fields including physics, chemistry, and materials research. Thermal elements can produce a wide variety of oxide compounds. These can have a wide range of structural geometries, as well as electrical properties, including insulator, semiconductor, or metallic properties. Metal oxide nanoparticles such as iron oxide (Van Benschoten et al. 1994; Raven et al. 1998; Dixit & Hering 2003), aluminium oxide (Coston et al. 1995; Lin & Wu 2001; Patra et al. 2012), titanium oxide (Pena et al. 2005; Jegadeesan et al. 2010; Xu et al. 2010), manganese oxide (Lenoble et al. 2004; Agrawal & Sahu 2006; Lafferty et al. 2010), zirconium oxide (Hristovski et al. 2008; Hang et al. 2012), and other metal oxide nanoparticles, as sorbents have recently garnered a great deal of interest. Nanosized metal oxides play an important role because they can easily remove heavy metals and dyes from aqueous surroundings (Hua et al. 2012) due to the size quantisation effect, resulting in large surface areas and high activity (Henglein 1989; El-Sayed 2001). Metal oxides have a wide range of functional characteristics, which are determined by their crystal structure, composition, morphology, intrinsic defects, doping, and other factors. Metal oxide nanoparticles are found in a variety of shapes, including tubes, particles, and sheets (Gupta et al. 2016). They are low-cost, environmentally friendly adsorbents for adsorbing heavy metals, azo dyes, and other wastewater-derived substrates (Andersen et al. 2003; Carballa et al. 2004; Miao et al. 2004; Joss et al. 2006; Lissemore et al. 2006; Zwiener et al. 2007). This paper focuses on some recent advances and applications in wastewater treatment for metal oxide nanoparticles, as well as the potential uses of such techniques to address various challenges confronting existing wastewater treatment technologies, in order to fill this gap and provide a comprehensive review. Titanium oxide nanoparticles, iron oxide nanoparticles, copper oxide nanoparticles, zinc oxide nanoparticles and manganese dioxide oxide nanoparticles, and their utilisation as adsorbents for the removal of dyes and toxic metal ions from water/wastewater, are discussed in this review paper. The goal of this research is to find ecologically safe, low-cost, and effective nano-metal oxide sorbents for water and wastewater treatment. A thorough comparison of reported nano-metal oxide sorbents was carried out in order to rate them and forecast knowledge gaps in this field.

Adsorption is one of the most successful procedures compared to any other technologies (coagulation, advanced oxidation, membrane separation) that have been studied and proven to produce excellent results (Rangabhashiyam et al. 2013, 2014; Bhatnagar et al. 2014, 2015; Tran et al. 2015). The use of this process is now in progress for purification and separation purposes. Adsorbents with small pore sizes, moderate porosity, high micropore volume and a broad pore network are required for effective adsorption. Generally, the nature of any kind of bonding builds on details of the involved species. In the case of adsorption it can be classified as physisorption (the attraction having physical nature between a solid surface and adsorbed molecule happens because of electroweak interactions including dipole-dipole forces, London forces, Van der Waals interactions) and chemisorption (the attraction due to chemical bonding such as covalent bonding (Sadegh et al. 2016) or electrostatic attraction (Gupta et al. 2016) between a solid surface and adsorbed molecule by sharing or transferring of electrons). In physisorption, adsorbed molecules form multilayers under ideal conditions, but in chemisorption, the adsorptive molecules make direct contact with the active surface, resulting in a single layer process. Chemisorption has a high enthalpy, happens at all temperatures, and is an irreversible process, whereas physisorption has a low enthalpy and occurs at low temperatures. Chemisorption is two times greater in magnitude than physisorption (Al-Ghouti & Da'ana 2020). The adsorption procedure is one kind of surface phenomenon where adsorbate is amassed on the adsorbent's active surface portion. The intermolecular force attraction of liquid-solid occurs when adsorbate solute comes into contact with a highly porous surface structured solid (Shen et al. 2009; Singh et al. 2011; Wang et al. 2012). Adsorption techniques for removing metallic contaminants from industrial effluents or aqueous solutions are highly valued (Jayaprakashvel et al. 2014) because of their cost effectiveness, high efficiency, adaptability, and effectiveness in removing water pollutants (Navakoteswara Rao et al. 2019).

The adsorption isotherm is a significant parameter of an adsorption system which gives information about the interaction between adsorbate species and adsorbent materials and also predicts the theoretical maximum adsorption capacity (Crini & Badot 2010). Adsorption process is important for determining the rate of solvent adsorbed and the amount of time required for adsorption. For adsorption, linear and non-linear pseudo-first- and second-order kinetics are commonly used at different intervals. The impact of pH, removal efficiency, and initial contact time are all important factors in adsorption. In the acidic range of pH, the percentage of removal effectiveness is considerably lower. The creation of negative charge on the adsorbent surface occurs when pH rises, whereas the surface of these nanoparticles was positively charged at low pH due to the presence of excess H⁺ ions. The electrostatic attraction between the adsorbent and the positively charged adsorbate allows for excellent removal efficiency, but the decolorisation process slows down at low pH due to the strong repulsive force between the positively charged dye and the adsorbent. Adsorption isotherms are also necessary for describing how pollutants interact with adsorbent surfaces and how adsorbents are employed to enhance pollutant removal from aqueous solutions. As more applications are developed, finding the ‘best-fit’ isotherm is necessary for adsorption studies. There are many isotherm models, such as the linear isotherm model, Langmuir, Freundlich, Temkin, Dubinin-Astakhov, Harkin-Jura, and Redlich-Peterson; among them, Freundlich and Langmuir are the two oldest models, which are commonly used and popular due to their simplicity and accuracy (Ho & McKay 1998; Wang et al. 2012; Abdel Ghafar et al. 2015; Agarwal et al. 2016) (Table 1). The Langmuir isotherm presupposes that the surface of adsorbent is homogeneous and is all equally likely to adsorb, resulting in the creation of an adsorbed molecule monolayer (Langmuir 1918), whereas the Freundlich isotherm is used to represent heterogeneous surface adsorption with different affinities (shown in Figure 1) (Nethaji et al. 2013).

The Langmuir isotherm (Al-Ghouti & Da'ana 2020) model (Equation (2)) was originally established to present the adsorption of gases on metal, where the adsorption onto solid surface is based on the kinetic principle, which is a continuous process of interaction between adsorbent and adsorbate molecules. The rates of adsorption and desorption should be equal (Langmuir 1916). There is some limitation of sorption sites, and the speed of sorption is proportional to the concentration of molecules that come into contact with the solid phase. The Langmuir isotherm is an empirical model, and some basic assumptions of this model are:

• Each active site of adsorbed layer interacts with only one molecule

• This is a homogeneous process, i.e., energetically equivalent, and all sites of adsorbent molecules have equal affinity towards adsorbate.

The adsorption nature may be linear (R_L = 1), unfavourable (R_L > 1) or favourable (0 < R_L < 1), or irreversible (R_L = 0), all these conditions are indicated by the separation factor R_L.

On the other hand, in 1906 Freundlich proposed the fast isotherm model for sorption process, which describes reversible, nonlinear sorption on heterogeneous surfaces and multilayer adsorption. There is no need for distribution of adsorption heat uniformly in this isotherm, and it is defined as a heterogeneous system like adsorption of organic compounds, and highly reactive on molecular sieves or activated carbon, which can be expressed by the Freundlich model.

To recognise the adsorption method and determine the potential steps related to adsorption rates, such as chemical reaction and mass transport, kinetics study is a very important part of the adsorption system. The study of adsorption kinetics is important, which determine the required time for reaching equilibrium in adsorption process and indicate the development process and adsorption system design on the reaction pathway. The rate at which the adsorption occurs is evaluated by the kinetics. There are various kinetic models: namely, pseudo-first-order (PFO), pseudo-second-order (PSO), Elovich and intra particle (IP) diffusion are some of the kinetics which are generally used to recognise the dynamics of the adsorption process. Among these kinetics, PFO and PSO kinetics are frequently used to disclose the adsorbate adsorption phenomenon (Table 2). The linear regression correlation coefficient (R²) indicates that the error level of the model and the values are compared to evaluate the best fit model.

PFO kinetic model: Eminent scientist Lagergren first proposed the PFO model in 1898, which states that the solute uptake changing rate is directly proportional to the difference between the current concentration and the concentration at saturation. This model is mostly applied in the initial state of the sorption process.

Here Q_t and Q_e are the adsorption amounts at time t and at equilibrium respectively. The rate constant is K₁ at equilibrium.

This equation is frequently used to fit kinetic data and Q_e and K₁ parameters are obtained by plotting in (Q_e−Q_t) vs. t graph.

PSO kinetic model: PSO rate equation is the popular model which is used to illustrate the adsorption kinetics, which is popularised by Ho & Mckay (1999). The best thing is that the equilibrium capacity can be evaluated by this model without being calculated from experiment.

In addition, the equation of the intra-particle diffusion model can be expressed as follows (Abdel Ghafar et al. 2015; Agarwal et al. 2016).

Q_t = K_pt^1/2+c, where the intercept is c and the intra-particle diffusion rate constant (mg/g h^1/2) is Kp, which can be evaluated by slope of Q_t vs. t^1/2 graph.

A linear relationship example of Langmuir and Freundlich isotherm and PSO kinetic equation is represented below in Figure 2 for the application of waste water treatment by green synthesis of Fe₃O₄ nanoparticles (Das et al. 2020).

Metal oxides are important in a variety of chemistry, physical, and material science domains (Maier 1990; Henrich & Cox 1994; Noguera 1996; Fernández-García et al. 2004; Rodríguez & Fernández Garcia 2007; Wells 2012; Arora 2018; Priyadarshini et al. 2018). Researchers are increasingly interested in metal oxides and their different physical, magnetic, optical, and chemical properties, as these are extremely sensitive to change in composition and structure. There are many metal oxide based nanoparticles that are attracting their attention, as their applications are taking place in several fields. Due to their high surface to volume ratio, chemical stability and ease of interaction with adsorbents, as well as any high activity generated by the size quantisation effect, metal oxides are more effective than any other sorbent (Henglein 1989; El-Sayed 2001). Only a few reviews in the literature have addressed the application of metal oxide nanoparticles for water treatment. Yang et al. investigated the impacts of four kinds of metal oxide nanoparticles in wastewater treatment: zinc oxide (ZnO), titanium oxide (TiO₂), silver nanoparticle (Ag nanoparticle), and zero-valent iron (Yang et al. 2013). Singh et al. discussed three types of metal oxide nanoparticles: ZnO, TiO₂ and iron oxide are essential adsorbents for wastewater, as well as to purify industrial effluents (Singh et al. 2019), while Junbai et al. highlighted various kinds of metal oxide nanoparticles such as TiO₂, Fe₃O₄/Fe₂O₃, MnO₂, Al₂O₃, MgO, Fe₃O₄/Fe₂O₃ and CeO₂ and their uses in wastewater treatment (Fei & Li 2010). The reason why these nanoparticles are chosen for wastewater treatment is because of their capability to be oxidised or dissolved in water easily. These metal oxide nanoparticles are chemically stable (with no side effects) and have a diversity of uses, including adsorption, photocatalytic activity, and antifungal and antibacterial activities. Metal oxides are extensively used to remove heavy metal ions and dye from the wastewater. As the applications of metal oxide nanoparticles in wastewater treatment have gradually increased, on this basis the number of publications on nanoparticles in wastewater treatment in the last ten years (shown in Figure 3) is provided. There are some nanosized metal oxides, such as TiO₂, Fe₃O₄, MnO₂, MgO, CdO, ZnO, CuO, Al₂O₃, CeO₂, ZrO₂ and so on, which are widely used as adsorbents. In this review, five types of metal oxides and their recent applications are discussed.

Among the numerous metal oxides that are commonly employed as adsorbents, TiO₂ nanoparticles (Figure 4) have received a lot of attention in recent decades. TiO₂ nanometal is less expensive than other nanomaterials, and it has superior thermal and chemical stability, light stability, and biological stability (Imamura et al. 2013; Rawal et al. 2013; Guesh et al. 2016) as well as minimal toxicity in humans. TiO₂ nanoparticles have strong adsorbent activity (Mironyuk et al. 2019), a developed surface area (Bankmann et al. 1992) and resistance to acidic and alkaline media. Another benefit is that their surface has a greater concentration of hydroxyl groups (-OH), which can absorb contaminants from water by interaction with the OH surface. TiO₂ nanoparticles or nanocomposites have been employed in a variety of applications, with the majority of studies focusing on photocatalytic activity; the adsorption study is comparatively rare. Because of their insolubility and point of zero charge (pHzpc), TiO₂ nanoparticles are a suitable adsorbent for studies over a spacious range of pH. Fazal and group investigated the removal of Malachite green (MG) colour using a hybrid macro algae-based biochar-TiO₂ composite (BCT) produced using a wet precipitation technique. The adsorption capacity of biochar-TiO₂ (74.30 mg/g) was better than that of biochar (30.40 mg/g) and pure TiO₂ (1.50 mg/g). The results showed that the BCT composite can perform in the intended application of the removal of organic industrial pollution (Fazal et al. 2020). Binaeian et al. used an in-situ approach to produce TiO₂ in a chitosan grafted matrix (TiO₂-PAM-CS) to absorb Sirius yellow K-CF dye from an aqueous solution. At 40 °C, the maximum adsorption capacity was 1,000 mg/g at pH 2, with 96.81% colour removal. The equilibrium data are well-suited to the Langmuir isotherm and second-order kinetic model (Binaeian et al. 2020). Multi-walled carbon nanotubes (MWCNTs) supported TiO₂ was prepared by Kariim et al., where MWCNTs/TiO₂ nanocomposite adsorbents remove cyanide and phenol from refinery effluent with a contact duration of 70 minutes, an adsorbent dose of 0.3 g, and a temperature of 40 °C. The investigation revealed that absorption capacity of MWCNTs/TiO₂ composite was superior to that of pure MWCNTs. The highest adsorption capacity for cyanide and phenol was 86.207 and 57.4741 mg/g, respectively, according to the experimental results fitted with Freundlich isotherm and PSO kinetic model with multilayer adsorption (Kariim et al. 2020). Elbarbary et al. used γ-irradiation to make polyvinyl alcohol (PVA) and acrylamide (AAm) blends with TiO₂/SiO₂ nanopowder to increase the removal of basic blue 3 dyes (BB3) and Cu (II) ions from aqueous solution. After 7 h and 6 h at pH 11 and pH 6, respectively, the maximal adsorption capacities were 140.9 and 190.3 mg/g, with a removal efficiency of 93.5 and 95.2%. The Langmuir isotherm and the PSO kinetic model suit the experimental results well (Elbarbary & Gad 2021). Akl M. Awwad et al. reported the adsorption of Pb(II) and Cd(II) ions by the nanocomposite TiO₂/kaolinite, which was prepared from raw Jordanian kaolin clay by batch adsorption treatment. The maximum adsorption capacity was 250 mg/g and 333.3 mg/g at pH 5.5–6.0 of Cd(II) and Pb(II) ions respectively, determined by the Langmuir isotherm model (Awwad et al. 2020). Jena et al. also prepared TiO₂-SiO₂-sulfur (Ti-Si-S) nanohybrid material using TiO₂ nanoparticle by sol-gel process and hydrothermal method at 120 °C as adsorbent, which was tested to adsorb methylene blue (MG) from aqueous solution. The adsorption followed the Langmuir isotherm with a maximum adsorption capacity 804.80 mg/g and PSO kinetic model (Jena et al. 2019). Besides, Abdulhameed and colleagues developed a chitosan based ethylene glycol diglycidyl ether/TiO₂ nanoparticles (CS-EGDE/TNP) composite for adsorbing reactive orange 16 (RO) dye from aqueous solution. Both the Freundlich and PSO kinetic models are well suited to the adsorption isotherm. The greatest adsorption capacity of CS-EGDE/TNP for RO dye was 1,407.4 mg/g with an adsorbent dosage of 0.02 g, a solution pH of 4 and a temperature of 40 °C. The use of CS-EGDE/TNP as a wastewater treatment adsorbent was backed up by this discovery (Abdulhameed et al. 2019). Mironyuk et al. employed a complex [Ti (OH₂)₆³⁺.3Cl⁻ and a modifying agent Na₂CO₃ as an adsorbent to remove Sr(II) ion from aqueous solution in a sol-gel method to produce mesoporous carbonated TiO₂ nanoparticles. PSO kinetics and the Langmuir isotherm model were used to match the adsorption data. The adsorption capabilities of the materials with varied wt percent carbonate groups, namely 2C/TiO₂, 4C/TiO₂, and 8C/TiO₂, were 170.4, 204.4, and 190.8 mg/g, respectively. The best adsorption capability towards Sr(II) cations is found when 4% (wt) of carbonate group is used (Mironyuk et al. 2019). By utilising the biomass from lemon peels, Herrera-Barros et al. demonstrated chemically titanium dioxide (TiO₂) nanoparticles via siloxane linkages to adsorb nickel ions from water. The adsorption capacity (15.36 mg/g) of Ni(II) ions by biomass modified with TiO₂ nanoparticles increased from 78.2% to 90.1%, demonstrating that adding TiO₂ nanoparticles to biomass matrix can improve the removal capacity of Ni(II) ions from aqueous solution (Herrera-Barros et al. 2020). Sharma et al. stated that mesoporous ZnO and TiO₂@ZnO nanoliths with useful surface area (120–332 m²/g) were manufactured using a nano casting process to adsorb Pb(II) and Cd(II) ions from aqueous solution. The Freundlich isotherm model and the PSO kinetic model suit the adsorption data better. Pb(II) ions had maximum adsorption capacities of 790 and 978 mg/g, and Cd(II) ions had maximum adsorption capacities of 643 and 786 mg/g of ZnO and TiO₂@ZnO monoliths, respectively, at pH 6, contact period of 150 minutes, and temperature of 30 °C (Sharma et al. 2019). Zain et al. created activated carbon (AC)/TiO₂ composite by depositing two different sizes of TiO₂ (small-sized TiO₂ (TS) and big-sized TiO₂ (TB)) onto AC. As a result of a synergistic effect, AC/TiO₂ (AC/TS and AC/TB) composites showed improved ruthenium N-3 dye removal by both adsorption and photo degradation. The AC/TB composite was more efficient in this investigation, with a maximum amount of N-3 dye adsorbed per unit mass of 523 mg/g after 180 minutes, accounting for almost 99% of N-3 dye. The composite has overall PSO reaction kinetics and prefers a Freundlich isotherm in which the sorption process is conducted in a heterogeneous environment (Zain et al. 2021). A hydrothermal followed by sol-gel technique was successfully used by Chang et al. to synthesise an efficient nanocomposite adsorbent comprising magnetic core-shell MnFe₂O₄@TiO₂ nanoparticles loaded on reduced graphene oxide (MnFe₂O₄@TiO₂-rGO) by sol-gel technique. The adsorption properties of MnFe₂O₄@TiO₂-rGO were comprehensively learned using ciprofloxacin (CIP) and Cu²⁺ as target pollutants, as well as the parameters that influence the adsorption properties, such as temperature, adsorption time, initial pollutant concentration, and pH. The maximal adsorption capabilities for CIP and Cu²⁺, according to the experimental data, are 122.87 mg/g and 225.99 mg/g, respectively. At all temperatures, their adsorption behaviours match the Langmuir isotherm and a PSO kinetic model. CIP and Cu²⁺ adsorption capabilities could still reach 76.56 and 118.45 mg/g, respectively, when the adsorbent was regenerated six times. MnFe₂O₄@TiO₂-rGO adsorbent has a lot of potential applications in the field of water purification because of its outstanding adsorption, magnetic separation, and recycling capability (Chang et al. 2021). Zhou et al. reported that the GO-TiO₂ composite material is created in this study by loading TiO₂ on the surface of GO and binding with C–O–Ti chemical linkages, resulting in a material with numerous oxygen-containing functional groups. The anatase-type TiO₂ produced is derived from hydrolysing butyl titanate in a moderate water bath (80 °C), which allows the oxygen-containing functional groups on GO- TiO₂ to be preserved to improve its efficacy for adsorbing Cr³⁺ in wastewater. The adsorptive process of Cr³⁺ on GO- TiO₂ is described by the quasi-second-order kinetic equation and the Langmuir equation (adsorption potential up to 149.24 mg/g). Thermodynamic investigations further suggest that the adsorption process toward Cr³⁺ is spontaneous and endothermic. The removal effectiveness is still over 85% after five cycles, indicating that the developed adsorbent has a lot of promise in extracting Cr³⁺ from wastewater (Zhou et al. 2021).

This review of TiO₂ nanoparticles reveals that TiO₂ nanoparticles not only have photocatalytic activity, but also have high adsorption efficiency in various nanocomposites. Furthermore, these nanoparticles can be reused for up to seven cycles. CS-EGDE/TNP nanocomposite has the highest adsorption capacity among a variety of adsorbents. A summary of all these studies is given in Table 3.

Iron is one of the most widely used materials on the planet. Iron oxide nanoparticles (Figure 5) are attracting a lot of attention because of their ease of use and availability in removing hazardous heavy metals. Maghemite (Fe₂O₃), hematite (Fe₂O₃), and magnetite (Fe₃O₄) are among the most studied iron oxides and are frequently used as nano adsorbents. Because of their low cost, ease of use, and environmental friendliness, magnetite (Fe₃O₄) based nanoparticles are widely employed as adsorbents. Hematite (α-Fe₂O₃) is the most durable iron oxide and it has proven to be an effective adsorbent towards heavy metals (Ahmed et al. 2013; Adegoke et al. 2014; Dickson et al. 2017). Maghemite (γ-Fe₂O₃) and magnetite (Fe₃O₄) can easily be separated and lifted from the system with the presence of an external magnetic field. To enhance the adsorption efficiency and avoid the resistance against interference of several other mixed types of ions in the wastewater, various types of ligands or polymers were added to improve the dispersion stability and act as a binder for metal ions, thus becoming a carrier of metal ions from treated water (Khaydarov et al. 2010; Warner et al. 2010; Ge et al. 2012). Far et al. used a simple solvothermal approach to make a magnetic zirconium-based metal-organic framework nanocomposite to utilise as an adsorbent for the dyes Direct red 31(DR 31) and Acid blue 92(AB 92) in aqueous solution. PPI (polypropylene imine) dendrimer was added to boost the activity of the surface of Fe₃O₄@UiO-66 nanocomposite by combining Fe₃O₄ nanoparticles with the nanoporous structure of Zr-based MOF (UiO-66). The adsorption capacities of DR 32 and AB 92 were 173.7 and 122.5 mg/g, respectively, as determined using the Freundlich isotherm and the PSO kinetic model (Far et al. 2020). Tran et al. synthesised the magnetic Fe₃O₄/zeoliteNaA(Fe₃O₄/ZA) composite by facile hydrothermal method to use as an effective adsorbent for the removal of methylene blue (MB) dye from aqueous solution having maximum adsorption capacity of 40.36 mg/g where that of zeolite NaA was 32.99 mg/g, which showed that by implementation of Fe₃O₄ the adsorption capacity became higher (Tran et al. 2021). Magnetic nanoparticles of Fe₃O₄-chitosan nanocomposite (CTS-MNPs) were synthesised by Asgari et al. to remove metronidazole (MTZ) from aqueous solution. The maximum adsorption capacity was 98 mg/g at pH 3 and the equilibrium data well fitted by the Freundlich adsorption isotherm and PSO kinetic model (Asgari et al. 2020). As humic acid has an affinity for magnetic nanoparticles, Xue et al. modified Fe₃O₄ nanoparticles with oxidised humic acid (Fe₃O₄/HA-O) for superior metal adsorption results. Pb(II), Cu(II), Cd(II), and Ni(II) ions were removed from aqueous solution using Fe₃O₄/HA-O as an adsorbent. Following the Langmuir adsorption isotherm model, the highest adsorption capacities were 111.10, 76.92, 71.43, and 33.33 mg/g, respectively (Xue et al. 2021). Alorabi and co-workers prepared Fe₃O₄-CuO-activated carbon composite as an adsorbent by hydrothermal method for the removal of bromophenol blue (BPB) dye from wastewater and the composition showed higher adsorption capacity (97% removal efficiency) of 88.60 mg/g for BPB dye within the optimum conditions following the PSO kinetic model (Alorabi et al. 2020). Joshi et al. used a co-precipitation approach to make Fe₃O₄ loaded activated carbon (Fe₃O₄@AC) as an adsorbent to remove methylene blue (MB) and brilliant green (BG) dye from aqueous solution. For the MB and BG dyes, the maximum adsorption capacity of Fe₃O₄@AC was determined to be 138 and 166.6 mg/g, respectively. PSO kinetics and the Langmuir isotherm model fit the experimental results well for both dyes (Joshi et al. 2019). Sheikhmohammadi and group synthesised Fe₃O₄ @ SiO₂ nanoparticles functionalised with amine (NH₂), i.e., FSN nanocomposite, for removal of Ethylparaben (EtP) chemical from aqueous solution. The surface of the FSN nanocomposite was modified by 3-amino propyl triethoxysilane (APTES) polymer to increase the adsorption efficiency. The maximum adsorption capacity of FSN toward EtP was 22.22 mg/g with a removal efficiency of 93% at pH 7. The kinetics and adsorption isotherm followed PSO and Freundlich isotherm model, respectively (Sheikhmohammadi et al. 2019). Chitosan-based magnetic composite CTS@SnO₂@ Fe₃O₄ was prepared by Yu et al. for the removal of reactive brilliant red (RBR) dye from aqueous solution. The maximum adsorption capacity was 981.23 mg/g at pH 2 and the equilibrium data best fitted with PSO kinetics and Langmuir isotherm model (Yu et al. 2020). Tahir et al. also prepared hydroxy propyl-β-cyclodextrin graphene loaded with Fe₃O₄ nanoparticle (HP-β-CD-GO/Fe₃O₄) by hydrothermal method for removal of Pb(II) and Cu(II) ions from aqueous solution. The maximum adsorption capacities were 50.39 and 17.91 mg/g, respectively, followed by the Freundlich isotherm model (Tahir et al. 2019). With the use of crude latex from Jatropha curcas (JC) and Cinnamomum tamala leaf extract, Das et al. created magnetic nanocomposites JC-Fe₃O₄ and CT-Fe₃O₄, respectively. Methylene blue (MB) dye, as well as Cu(II) and Co(II) ions, was removed from the aqueous solution using the JC-Fe₃O₄ and CT-Fe₃O₄ adsorbents. Maximum adsorption capacities for Cu(II) and Co(II) ions of CT-Fe₃O₄ were 463.2 and 513.7 mg/g, respectively, and JC-Fe₃O₄'s maximum capacities for Cu(II), Co(II), and MB dye were 543.3, 501.3, and 466.6 mg/g. These two adsorbents have also been shown to be potent antibacterial and antioxidant agents (Das et al. 2020). Rehman et al. used a simple, one-pot, template-free hydrothermal process to make magnetite (Fe₃O₄) nano hollow spheres (NHS), which were utilised as an adsorbent for treating industrial trinitrotoluene (TNT) wastewater to minimise chemical oxygen demand (COD) values. The NHS produced has an adsorption capacity (Qe) of 70 mg/g, indicating that attractive forces exist between the adsorbent (Fe₃O₄ NHS) and the adsorbate (TNT wastewater). At room temperature, the COD value of TNT wastewater was reduced to > 92% in 2 hours. By simply adjusting the magnetic properties of the applied Fe₃O₄ NHS with a little variation in the pH of the solution, the Fe₃O₄ NHS was retrieved for reuse. After each cycle, there was a little drop in Qe (5.0–15.1%). The capacity of a polypyrrole-iron oxide-seaweed (PPy-Fe₃O₄-SW) nanocomposite to remove Congo red (CR) from aqueous solution has been studied. pH 3, starting CR concentration of 40 mg/L, nanocomposite dosage of 20 mg, contact time of 40 minutes, and temperature of 40 °C are the ideal conditions for efficient CR elimination. Adsorption isotherm and kinetic investigations were carried out.The Langmuir isotherm has a high coefficient of determination (R² = 0.98) and a maximum dye uptake of 500 mg/g. The PSO model (R² = 0.994) was followed in kinetic investigations. Thermodynamic parameters were investigated, and it was discovered that the process is spontaneous, endothermic, and random in nature.

Iron oxide nanoparticles are also excellent adsorbents and we can see from the table that CTS@SnO₂@Fe₃O₄ has the highest adsorption capacity and can be reused for up to five cycles. Furthermore, at normal temperature, green synthesised Fe₃O₄ NPs showed excellent adsorption capabilities for heavy metals and dye. All these studies have been summarised in Table 4.

Like any other metal oxide nanoparticle, copper oxide nanoparticle has gained extensive popularity, as it can easily be used for the removal of heavy metal ions and toxic dyes from aqueous solutions. CuO nanoparticles (Figure 6) are largely involved in various applications due to their high efficiency and low cost. Besides CuO nanoparticles having a large surface area and an abundance of active sites, they has been a great research topic due to their remarkable properties for obtaining a p-type semiconductor with compressed bandgap (1.2–2.0 eV) and high communication features. Ighalo et al. reported that conventional chemical procedures such as microwave heating and precipitation are the most often used techniques for the synthesis of CuO NPs. CuO NPs have the greatest documented fluoride adsorption capacity of 3,152 mg/g, indicating that they can remove more than three times their weight in fluoride from the aqueous phase (Ighalo et al. 2021). Baylan et al. synthesised copper oxide nanoparticles as an effective adsorbent for the removal of acrylic acid from an aqueous solution by using a simple precipitation method. The maximum adsorption capacity was 202.67 mg/g at a contact time of 180 min and a temperature of 25 °C, and the equilibrium data well fitted the Langmuir isotherm model and first-order kinetics (Baylan et al. 2020). Vidovix et al. reported the green synthesis of copper oxide nanoparticles obtained from pomegranate (Punica granatum) leaf extract, which were shown to be an effective adsorbent for the removal of safranin-O (SO) dye. The maximum adsorption capacity was 189.54 mg/g at 298 K. The experimental data best fitted the Langmuir isotherm model and PSO kinetic model (Vidovix et al. 2021). Wei et al. prepared copper oxide and modified hickory wood chips biochar composite through a simple ball milling method, which was used as an efficient adsorbent for the removal of reactive red (RR) dye from aqueous solution. The removal efficiency of 10% CuO-BC nanocomposite was higher (percentage of removal 46%) than that of pristine BC (percentage of removal 20%). The maximum adsorption capacity of CuO-BC was 1,399 mg/g following Langmuir isotherm model, and kinetics followed the PSO kinetic model (Wei et al. 2020). Jain et al. synthesised copper oxide nanoparticles by a chemical precipitation method for the removal of Ni (II) ions from aqueous solution. The maximum adsorption capacity of CuO was 15.4 mg/g under optimum conditions (Jain et al. 2020). Baylan and group reported CuO nanoparticles as a novel adsorbent for the removal of tartaric acid from aqueous solution. CuO nanoparticles showed higher removal performance with an adsorption capacity value of 619.23±2.08 mg/g. The equilibrium data best fitted the Freundlich isotherm model and PSO kinetics (Baylan et al. 2021). Zaidi et al. synthesised copper oxide nanoparticles using a co-precipitation approach to ingest Pb(II) ions from aqueous solution. The maximal adsorption capacity was 97 mg/g, with the Freundlich isotherm model best fitting the equilibrium data and the PSO kinetic model better fitting the kinetic investigations (Zaidi et al. 2021). CuO nanoparticles were prepared by Mahmoud and co-workers through a green synthesis route with the extracts of mint leaves and orange leaves, named CuO NPs 1 and CuO NPs 2, which were used as adsorbents to take up Pb(II), Ni(II), Cd(II) ions from aqueous solution. The maximum uptake capacity followed the order of Pb(II) > Ni(II) > Cd(II) for both, and CuO NPs 1 showed better adsorption performance than that of CuO NPs 2, where the maximum capacities of CuO NPs 1 were 88.80, 54.90 and 15.60 mg/g, respectively. Freundlich and PSO kinetic models best fitted the equilibrium data (Mahmoud et al. 2021). However, Panda et al. synthesised copper oxide nanoparticles loaded with boehmite (CuO/Boehmite) as an adsorbent to take up As(III) and As(V) ions from aqueous solution through an ultrasonication aided hydrothermal method. The maximum adsorption capacities were 79.05 and 97.35 mg/g for As(III) and As(V), respectively (Panda et al. 2020). Srivastava et al. synthesised CuO nanoparticles through a green route using plant extract of Ficus retusa as adsorbent for removal of organic dyes Congo red (CR) and eriochrome black T (EBT) from the aqueous solution. The maximum adsorption capacity was found to be 119.70 and 235.70 mg/g for CR and EBT dye, respectively. The adsorption phenomenon followed the PFO kinetic model and Freundlich isotherm model (Srivastava & Choubey 2021). Syarif and group prepared CuO nanoparticles by a precipitation method to take up methylene blue (MB) dye from aqueous solution. The maximum uptake capacity was 61 mg/g at 400 °C temperature, pH 7, and contact time 2 h. The process followed second-order kinetics and Langmuir isotherm model (Syarif et al. 2020). Alhalili et al. synthesised copper oxide nanoparticles using Eucalyptus globulus leaf extract. Using methyl orange as a test, the adsorption efficiency of the nano-adsorbents was found to be 95 mg/g at room temperature. When methyl orange (MO) dye is administered in quantities of 0.04 g/50 mL, copper oxide nanoparticles (CuO NPs) absorb the colour most effectively at pH 4.5. The Langmuir adsorption isotherm model and experimental results showed high agreement. Despite this, the adsorption of MO to CuO nanoparticles was accompanied by a PSO kinetic model, indicating that the adsorption behaviour might be regulated by hydrogen bonding between the adsorbent and the adsorbate (Alhalili 2022).

CuO NPs have an important role as an adsorbent, as shown in Table 5. CuO with hickory wood chips biochar (CuO-BC) nanocomposite has the highest RR 120 dye adsorption capability. A summary of all these studies is given in Table 5.

Zinc oxide (ZnO) nanoparticles are known to be the best adsorbent due to their high surface area, stability (Heinonen et al. 2017), biocompatiblity (Sruthi et al. 2018), cost-effectiveness and extraordinary removal capacity; besides this, ZnO nanoparticles are unique in the removal of anionic organic compounds due to their high point of zero charge (pHzpc) of ZnO = 9.3–10.5 and also their propitious electrostatic attraction to various anionic particles (Fatehah et al. 2014) which can be applied in various fields. In this regard, Gusain et al. presented a review of metal oxides and composites for organic pollution removal, in which zinc oxide-based adsorbents are briefly mentioned (Gusain et al. 2019). Primo et al. synthesised ZnO nanoparticles by two simple eco-friendly routes, green synthesis and gelatinisation, using Aloe vera and Cassava starch, which were Zn-AL and Zn-ST, as adsorbents for the removal of Cu (II) ions from aqueous solution. The maximum adsorption capacity of Zn-AL and Zn-ST was 20.42 and 10.95 mg/g, respectively. Results followed Langmuir isotherm model. The removal capacity of Zn-AL was higher than Zn-ST (Figure 7, Primo et al. 2020). Gu et al. prepared ZnO nanoparticles by a traditional hydrothermal method to take up Cr (II) ions for dental purification treatment. The maximum adsorption capacity was 88.547 mg/g, which followed Langmuir isotherm, and kinetics followed PSO kinetics (Gu et al. 2020). Muinde et al. also reported the synthesis of chitosan–zinc oxide (CS-ZnO) nanocomposite to remove malachite green (MG) dye from aqueous solution. The maximum adsorption capacity of CS-ZnO was achieved at 11 mg/g with a high removal capacity of 98.5%. The equilibrium data best fitted the Langmuir isotherm and second-order kinetic model (Muinde et al. 2020). Furthermore, Debnath et al. prepared ZnO nanoparticles by a green synthesis route from Hibiscus rosa-sinensis leaf extract for the removal of Congo red (CR) dye from aqueous solution. The maximum adsorption capacity was 9.615 mg/g with a high percentage of removal efficiency of 95.55% following the Langmuir isotherm model (Debnath & Mondal 2020). Amde and group synthesised nano selenium functionalised zinc oxide nanorod (Se@ZnO-NR), which was used as an effective adsorbent for the removal of Hg(II) ion from water. The maximum adsorption capacity was 1,110 mg/g and the highest percentage of removal 99.2% (Amde et al. 2020). Rodriguez et al. synthesised ZnO nanoparticles functionalised by graphene oxide (GO), as GO has a high surface area. The GO-ZnO nanocomposite acted as an effective adsorbent for the removal of Al and Cu ions, from acid mine drainage water. The maximum removal capacities were 19.9 and 33.5 mg/g with a high removal efficiency of 95.6% and 92.9% for Al and Cu ions, respectively (Rodríguez et al. 2020). ZnO nanoparticles were synthesised by Samad et al. with silty clay (SC), i.e., ZnO-SC composite, to use as an adsorbent for removal of Pb(II) ions from aqueous solution. The adsorption capacity was increased by introducing ZnO nanoparticles to SC. The maximum adsorption capacity of ZnO-SC (14.54 mg/g) was higher than normal SC (12.43 mg/g). The process of adsorption of ZnO-SC followed Langmuir isotherm and PSO kinetic model (Samad et al. 2021). Besides, Nayak et al. synthesised ZnO nanoparticles from Ocimum sanctum (Tulsi leaf) leaf extract (ZnO-T) through a green approach as an adsorbent for the removal of Congo red (CR) dye. The removal capacity of ZnO-T was higher (97%) than commercially available ZnO (78%), the maximum adsorption capacity of ZnO-CT was 74.07 mg/g following the Langmuir isotherm condition, and kinetics followed the PSO kinetic model (Nayak et al. 2020). Rashtbari and co-workers synthesised ZnO nanoparticles doped on activated carbon (AC) from walnut peel extract by green synthesis route to use as an adsorbent for the removal of Erythrosine-B (Er-B) and Eosin-Y (Eo-Y) dye from aqueous solution. The maximum adsorption capacity was found to be 144.92 and 163.9 mg/g with a removal efficiency of 98.31% and 95.11% for Er-B dye and Eo-Y, respectively, which followed Langmuir isotherm and PSO kinetics (Rashtbari et al. 2022). Mittal et al. synthesised ZnO nanoparticles through in-situ process by using gum Arabic grafted polyacrylamide (GA-Cl-PAM) hydrogel as a self template. The GA-Cl-AAM-ZnO composite was used as an effective adsorbent to take up malachite green (MG) dye from aqueous solution. The maximum adsorption capacity was 766.52 mg/g with 99% high adsorption efficiency (Mittal et al. 2020).

From this review of ZnO NPs we can see that Se@ZnO-NR has the maximum adsorption capacity and GA-Cl-PAM/ZnO has the best reusability result, with up to ten consecutive cycles at room temperature and pH 7. A summary of all these studies is given in Table 6.

Manganese dioxide (MnO₂) nanoparticles (Figure 8) are one of the most preferable metal oxides and essential as an adsorbent for wastewater treatment due to their large specific surface area, high activity, good stability, environment-friendly nature, structural flexibility, and effectiveness as well as strong oxidising properties. MnO₂ nanoparticles also have the great feature that their structure can gain different morphologies, such as nanoparticles, nano-sheets, nano-flowers, nano-wires, nano-tubes, urchins/orchids, mesoporous and branched structures, etc. (Julien & Mauger 2017). Studies of MnO₂ adsorbents are a new scope for researchers, and this is gaining great attention day by day. Xia et al. synthesised MnO₂ modified magnetic graphitic carbon nitride (MMCN) composite via in-situ deposition for use as an adsorbent for removal of Pb(II) ion from aqueous solution. The maximum adsorption capacity was 187.6 mg/g at pH 6, higher than MCN and original CN, with a removal efficiency of about 99%. The adsorption kinetics and isotherm followed the PSO model (Xia & Liu 2021). Verma et al. used a hydrothermal technique to make graphene oxide-manganese dioxide (GO-MnO₂) nanocomposite as an adsorbent for removing cationic methylene blue (MB) dye and anionic methyl orange (MO) dye from aqueous solution. The maximal adsorption capacities for MO and MB dye were 149.253 and 178.253 mg/g, respectively, according to the Langmuir isotherm model, and the kinetics fitted well with the PSO kinetic model (Verma et al. 2021). Liu et al. also synthesised hierarchical Mg (OH)₂-MnO₂ nanoparticles (MMNC) by a water bath method for the removal of Congo red (CR) and methyl orange (MO) dye from wastewater. The maximum adsorption capacity was 17,100 and 7,300 mg/g for CR and MO dye, respectively, exhibiting higher removal efficiency (Liu et al. 2021). Huo and group prepared polyvinyl alcohol, graphene oxide, MnO₂ (PVA/GO/ MnO₂) composite by a co-inducing and in-situ oxidising process to implement as adsorbent for the removal of Co(II) and Sr(II) ions from aqueous solution. The functionalised MnO₂ improved the adsorption capacity of the nanocomposite. The maximum adsorption capacities were 60.3 and 26.8 mg/g for Co(II) and Sr(II) ions, respectively (Huo et al. 2021). On other hand, Dinh et al. synthesised MnO₂-chitosan (MnO₂/CS) nanocomposite adsorbent to take up Cr(VI) ion from aqueous solution. The maximum adsorption capacity was 61.56 mg/g at pH 2.0 for Cr(VI) ion (Dinh et al. 2020). Abdullah et al. prepared MnO₂ nanoparticles by a hydrothermal method using KMnO₄ as a precursor, which were annealed at different temperatures (250 °C, 450 °C, 750 °C) for 2 h, and used the nanoparticles as adsorbent for the removal of methylene blue (MB) dye from aqueous solution. The highest MB dye removal efficiency was 100% after annealing it at 750 °C temperature, and the maximum adsorption capacity was 22.22 mg/g (Abdullah et al. 2021). Graphene nanosheet-manganese dioxide (GN-MnO₂) composite was prepared by Yusuf et al. using a microwave-assisted method to adsorb Congo red (CR) and Acid red 25(AR 25) dye from aqueous solution. The maximum adsorption capacities were 270.06 and 324.26 mg/g for CR and AR 25 dye at 303 K, pH 3, respectively, and the experimental data best fitted the Langmuir isotherm model and PSO kinetic model (Yusuf et al. 2020). Edathil et al. prepared porous graphitic carbon (PGC) impregnated with δ-MnO₂ using the in-situ wet deposition method for the removal of dissolved sulphide from aqueous solution. The maximum adsorption capacity was 526.3 mg/g with adsorption efficiency of 90% within 20 min of contact time at 298 K, and experimental data followed Langmuir isotherm model and second-order kinetics (Edathil et al. 2020). Claros et al. synthesised MnO₂ nanowires by a hydrothermal method to use as an adsorbent for the removal of Pb(II) and Cu(II) ions from water. The maximum adsorption capacity of Pb(II) (3.730 mg/g with the percentage of removal 99.99%) was higher than Cu(II) (2.972 mg/g with the percentage of removal of 75.48%). Langmuir and Freundlich isotherms were fitted well for the adsorption of Cu(II) and Pb(II), respectively (Claros et al. 2021). Yusuf et al. also synthesised δ-MnO₂ on graphene nanosheet (GN-MnO₂) via the microwave-assisted technique for the removal of Cr(III) and Co(II) ions from aqueous solution. The maximum adsorption capacity of this nanocomposite was 491.98 and 403.4 mg/g at 303 K and pH 4 and 6 for Cr(III) and Co(II), respectively (Yusuf & Song 2020).

Table 7 demonstrates that Mg(OH)₂-MnO₂ is a better adsorbent than the other four metal oxide nanoparticles, with adsorption capacities 17,100 mg/g and 7,300 mg/g for CR and MO dye, respectively. Furthermore, it can also be observed that pure MnO₂ nanoparticles show lower adsorption capacity than other MnO₂ nanocomposites in this table. A summary of all these studies is given in Table 7.

Many studies have been conducted to improve the quality of drinking water, which is directly related to human and environmental health and safety. Pollutants can be effectively killed using nanomaterials with precise physical and chemical characteristics. The notion of nanomaterial production has been elevated in order to prioritise implementation options. Metal oxide nanoparticles are favoured for heavy metal and organic pollutant absorption because they have shown promising results in a variety of applications. Immobilisation carriers are a form of nanoparticles that can be employed as support carriers for biosensors and biosorbents, albeit they are rarely discussed. In nanomedicine and other biological applications, metal oxide nanoparticles will be vital. These nanoparticles can be made in a variety of ways and can be employed in a variety of nanomedical and biological applications. To cut costs, however, these nanoparticles must still be manufactured on a commercial scale. Sustainable, low-cost, environmentally friendly, and toxic-free natural resources for the manufacture of these nanoparticles are required. The production of monodispersed nanoparticles is critical for future study. The process for the formation of these nanoparticles, however, remains unknown at this time. The process by which nanoparticles’ size and form can be changed should be the focus of future research. Future applications will focus on efficient methods that use only modest amounts of metal oxide nanoparticles. Furthermore, more research is needed to create cost-effective synthesis methods, as well as large-scale testing, for the successful field deployment of metal oxide nanomaterials.

Contamination with heavy metals and dyes is the major reason for water pollution, which causes aquatic life and human life as well as the whole environment to suffer. Some metal oxides are highlighted in this review because they exhibit unique properties. Adsorption is a quick and easy way to get rid of organic and inorganic pollutants. Only a few metal oxides have attained the optimal conditions for usage as adsorbents so far.

We provided a comprehensive analysis of five metal oxide nanoparticles in this review: titanium oxide, iron oxide, copper oxide, zinc oxide, and manganese dioxide nanoparticles.

Manganese dioxide nanoparticles have the highest adsorption capacity among these five metal oxide nanoparticles. Despite the fact that the five metal oxide nanoparticles have a wide range of uses as nanoparticles and, in some cases, nanocomposite forms, they all have similar applications. The nanomaterials employed in various applications are summarised in these tables, and by following their activities and future research fields will receive assistance and new ideas. Finally, the influence and toxicity of metal oxide nanoparticles on living beings should be considered. It is difficult to choose which of the nanoparticles is superior, the essential thing is that they should be excellent for future growth.

The authors declare no competing financial interest.

All relevant data are included in the paper or its Supplementary Information.