Biodegradation is known as a green technique to control the exposure of ECs [114]. However, the nature of ECs plays a significant role in ascertaining its complete biodegradability. This can be explained in terms of the rate constants of the biodegradation of the corresponding materials. Caffeine, acetaminophen, estradiol, and ibuprofen are some of the examples of ECs, which possess high biodegradation rate constants and hence can be degraded easily into the corresponding elemental precursors losing their bioactivity [115]. However, tetracycline, carbamazepine, and iopamidol possess very low biodegradation constants, thereby leading to incomplete or slow degradation of such compounds [116].

The factors influencing the degradation are redox potential, structural features, microbial diversity, temperature, pH, toxicity of the ECs and primary substrates [117]. Generally, biodegradation/bioremediation of these hazardous materials proceeds through highly specific enzymes [118]. The catalytic amount of such bioactive enzymes is very specific and efficient for transformation of these hazardous materials into their non-active precursors and this transformation can be viable for commercial adaptation, which is known as one of the ‘green’ bioremediations. These enzymes are mostly from oxidoreductase families. Some of the important enzymes are as follows [119]:

• Lignin peroxidase (1,2-bis(3,4-dimethoxyphenyl) propane-1,3-diol;
• Manganese peroxidase (Mn (II): hydrogen-peroxide oxidoreductase;
• Laccases;
• Tyrosinases: o-diphenol;
• Horseradish peroxidase.

The porous structure of the sludge or the biomass present in biotic or abiotic sludge resulted either in entrapment of the hazardous materials through physical adsorption or chemisorption followed by biodegradation. This approach to remove ECs is attractive.

Kamaz et al. [155] have reported the adsorption of Congo red, Remazol Brilliant Blue R and Eriochrome Black Ton activated municipal sludge. A Freundlich isotherm and pseudo-second-order kinetics were reported to be predominating during the sorption of the dyes. A thermodynamic analysis of the sorption process revealed that the processes were spontaneous. Enhancement in entropy leading to spontaneity of sorption was reported for Remazol Brilliant Blue R and Eriochrome Black T. The activated and deactivated sludge under aerobic and anaerobic conditions exhibited different sorption capacities indicating the involvement of different microorganism [155].

Streit et al. [156] have reported the adsorption of ibuprofen, ketoprofen, and paracetamol on effluent treatment plant sludge in the beverage industry. The porous structure with high surface roughness, surface area (642 m² g⁻¹), and total pore volume (0.485 cm³ g⁻¹) was responsible for achieving 145, 105, and 57 mg g⁻¹ sorption capacity for these pharmaceutical ECs. Coimbra et al. [157] have reported the adsorption of pharmaceuticals (diclofenac, salicylic acid, ibuprofen and acetaminophen) from municipal wastewater streams using a pulp mill sludge. Although 200 min was required to attain complete equilibrium sorption for all the ECs, their pseudo second-order rate constants followed the trend: salicylic acid > diclofenac > ibuprofen > acetaminophen.

The Sip isotherm was reported to be suitable for explaining the sorption processes with the trend in sorption capacity: diclofenac > ibuprofen ~ acetaminophen > salicylic acid. Removal of 17α-ethinylestradiol, 4-nonylphenol, and carbamazepine in wastewater using an aerobic granular sludge was found to initiate through adsorption followed by degradation with sorption capacity of 16.09 μg/g and 20.05 μg/g, for 17α-ethinylestradiol, and 4-nonylphenol, respectively [158]. Both Langmuir and Freundlich isotherm model has been used to describe the sorption processes. Chiavola et al. [159] reported the adsorption followed by biodegradation of EDCs: BPA, 17α-ethinylestradiol (EE2), estrone(E1) and 17β-estradiol (E2) using activated and inactivated sludge. Pseudo second-order kinetics were reported to be predominate. Temporary inhibition of the biological process was observed at an initial higher concentration of EDCs resulting in a reduction in the mineralization process till the non-inhibiting value of their concentration was reached. They reported the removal of EDCs from wastewater along with simultaneous nitrification.

Recently, activated sludge has been used for 98–99% removal of ibuprofen and paracetamol by adsorption. The kinetics of sorption were found to follow pseudo-first and pseudo-second order models at all concentrations of the pharmaceuticals. Mesoporous biochar obtained from textile mill sludge has been exploited for the removal of ofloxacin pharmaceutical ECs with a sorption capacity of 9.74 mg g⁻¹ following a π-π electron donor-acceptor and H bonding mechanism [160]. The sorption processes were spontaneous and exothermic in nature, best described by a pseudo second-order kinetics model and Redlich–Peterson and Freundlich isotherm models through a multilayer sorption process.

Although adsorption of ECs by the sludge is very common and a widely used cost-effective method, several other sorbents have also been evaluated. Some of them are porous materials obtained naturally, some are modified with specific surface functionalities to capture the ECs. Hence, not only efficient separation but also selective separation can be achieved. More than 90% separation of hormones (17α- dihydrouridine, 17α-Estradiol, 17α-ethinylestradiol, progesterone, estriol, and estrone) can be achieved using cyclodextrin coated silica [161]. However, the same sorbent showed only ~75% efficiency for norgestrel hormone.

The antidepressant, fluoxetine, was found to be adsorbed on zeolite, olive stone, sunflower, and walnut shell with sorption capacity in the range of 10 to 44 mg g⁻¹ [162]. The same sorbents were also reported to separate nicotinic acid and pharmaceutical compounds, with higher sorption capacity in the range of 57 to 92 mg g⁻¹. Avocado seed activated carbon was also reported to be very sorbent materials for sodium diclofenac analgesics with a capacity of 395 mg g⁻¹ [163].

Zn based metal-organic frameworks have been reported to adsorb amodiaquine, whereas carbon nanotubes, graphene and its derivatives were also utilized for adsorption of diclofenac, carbamazepine, and ciprofloxacin [164].

The retention of ECs by the membrane could be by adsorption or size exclusion. Retention by the membrane depends on the membrane pore size and pore size distribution, surface hydrophilicity, morphology, and roughness which in turn are affected by the membrane polymer [165]. Since the ECs possess a wide range of physio-chemical properties, the level of retention by the membrane can be highly variable depending on the specific EC. As indicated in Figure 7, if size exclusion is the main mechanism of retention, then the size of the EC relative to the membrane pore size is very important. If surface adsorption is the main mechanism of retention, the membrane surface interaction with the EC will be more important. Some general observations are as follows.

• Physical sieving can be used for the retention of non-ionic hydrophilic ECs (e.g., paracetamol, caeine, methylparaben);
• Surface interaction and initial adsorption is the major phenomena during retention of hydrophobic non-ionic ECs (e.g., carbamazepine, estrone). It was also reported that, there is a reduction in ECs rejection after the absorption saturation;
• For ECs with charged surface (either positive: propranolol, metoprolol or negative: ibuprofen, naproxen, diclofenac), the retention depends on electrostatic interaction between ECs and membrane materials in combination with sieving.

NF membranes were used for the retention of steroids [166]. Size exclusion and surface adsorption were reported to be the main mechanism of retention. Reverse osmosis (RO) has also been used for the removal of ECs [167]. Depending upon the nature of ECs, 40–100% rejection of ECs can be achieved by NF and RO processes [168,169,170].