Enhancing the Cu (II) Removal in Water Using Functional Hybrid Membranes

: This work developed hybrid EIPS/NIPS membranes of poly (butylene adipate-co-terephthalate) (PBAT) with nanocurcumin (NC) and/or Cloisite 20A (C20A). The dispersed phases were characterized by dynamic light scattering (DLS), ζ potential, X-ray diffractometry (XRD), scanning electron microscopy (SEM), and fourier-transform infrared spectroscopy (FTIR), while XRD, SEM, FTIR, mechanical properties, contact angle, and copper sorption evaluated the composite membranes. DLS analysis indicated that the dispersed phases present a nanometric size distribution; ζ potential measurements showed low electrostatic stability, explaining the agglomeration effects observed. Pure PBAT membranes presented macro-pores throughout their structure, which showed a slight size reduction with the inclusion of NC and C20A. The membrane’s mechanical performance was affected by the presence of the pores that functioned as stress-concentrating defects, and the inclusion of the dispersed phases increased the elastic modulus and tensile strength, especially for PBAT/0.5% C20A with values 18.7 and 8.9% higher than those of pure PBAT, respectively. Hybrid EIPS/NIPS membranes showed a hydrophilic nature with all the contact angle measurements lower than 90°. The sorption tests using a high-concentration copper solution (1,000 mg/L) showed a removal of around 25%. These results highlighted the development of new eco-friendly membranes using NC and C20A as dispersed phases with the potential to remove toxic elements from water.


Introduction
2] Heavy metals refer to elements with atomic densities above 4.5 g cm -3 and are derived from anthropogenic sources, such as petroleum, metal smelting, and electroplating. 3A more up-to-date nomenclature can refer to heavy metals and other harmful elements as potentially toxic elements (PTEs) since, above a specific concentration, they can be harmful to the ecosystem and living organisms. 4here are several PTEs related to anthropogenic activity, such as antimony (Sb), arsenic (As), beryllium (Be), cadmium (Cd), fluoride (F), gold (Au), lead (Pb), mercury (Hg), molybdenum (Mo), nickel (Ni), rubidium (Rb), selenium (Se), silver (Ag), uranium (U), vanadium (V), zinc (Zn), and copper (Cu). 5s a transition metal, copper can be found in nature in different oxidation states, such as a monovalent and bivalent ion, Cu (I) and Cu (II), respectively. 6While Cu (I) is absorbed into the intestine through food intake, Cu (II) can originate from water sources and may not be absorbed, increasing the concentration of copper in the human organism. 7ue to its industrial applications, Cu (II) can also be found in high wastewater concentrations.9] Above these limits, the literature reports toxic effects on human health, such as carriers of the gene for Wilson disease, gastrointestinal effects, 10 metabolic disorders, 11 and even death. 10ver the last decades, different techniques have been used to remove PTEs, such as chemical precipitation, 12 ion exchange, 13 ion flotation, 14 coagulation/flocculation, osmosis, and electrochemical removal.However, these methods have disadvantages, such as erratic removal of metal ions, creation of toxic sludge, or high energy consumption, 15  Sorpion efficiency PBAT/0.5 C20A → 27% PBAT/0.5 NC → 28% and, thus stimulate the search for cheaper and high-efficiency modern technologies.Among the PTE removal trends, adsorption is highlighted as a promising and efficient technique. 168][19] Several materials can be used to prepare the membranes, such as poly (vinylidene fluoride)/polyaniline, polyethylene, polyacrylonitrile, polyvinyl alcohol, and poly (N, N-dimethylacrylamide-co-2-hydroxyethyl methacrylate), 20 and also 2D materials such as boron nitride, 21 graphene 22 and MXene. 23However, due to environmental concerns, biodegradable polymers, e.g., cellulose, 24 polylactic acid, 25 and polybutylene adipate-co-terephthalate, 26 have been widely used for membranes.
][32] Montmorillonite (MMT) is a clay and one of the most studied inorganic materials for this application due to its high cation exchange capacity, natural electronegativity, controllable interlamellar spacing, and green and non-toxic properties. 33The interlamellar space can be functionalized to enhance MMT potential application further and offer newly available interactions.The insertion of hydrogenated ammonium organic modifiers deserves a highlight since MMT modified with those modifiers is known as Cloisite. 34Mohajeri and co-workers studied nano Cloisite Na + to remove Pb (II) and Cu (II), obtaining 40% and 16% adsorption efficiency, respectively. 31Another type of Cloisite, the Cloisite® 20A, presents a potential application for heavy metal removal in wastewater, which has not been extensively explored in literature for Cu (II) removal.Although it is not commonly used to remove PTEs, justifications for using Cloisite® 20A can be found in the literature. 35esides inorganic-based materials, organic-based materials such as cellulose, 36 natural fibers residues, 37 chitosan, 38 and curcumin 39 also can remove PTEs from aqueous media.] Still, the complexes formed between curcumin-metal ions lack further investigation of their complexes' properties.Both clays and curcumins can be dispersed phases in polymeric matrix membranes to remove PTEs from contaminated water.Moradi and co-workers used boehmite clay particles modified with curcumin in poly (ether-sulfone) membranes obtained by phase inversion for retention of multiple bivalent ions such as Cu (II), where the addition of 0.5% loading of clay and curcumin combined decreased the content of the contaminants in water. 44ccording to Hashim and co-workers, hybrid materials are nanocomposites with at least one element at a nanoscale. 45Therefore, this article is a follow-up to a previous study 35 and investigated the adsorption-hybrid membranes prepared using PBAT as a polymeric matrix, nanocurcumin (0.5% m/m), and nanoclay Cloisite® 20A (C20A (0.5% m/ m) as sorbents, aiming their potential application in reducing PTEs concentration in wastewater.Besides, the efficacy of the hybrid EIPS/NIPS membranes was investigated in reducing the Cu (II) concentration in a solution simulating an industrial effluent.

Curcumin preparation:
Firstly, 40 g of turmeric powder was added to 200 mL of ethanol and kept under magnetic stirring for 1 h to obtain curcumin. 44Then, the mixture was vacuum-filtered and dried at 30 °C for 24 h.Nano-sized curcumin (NC) structures were obtained by mechanical ball milling using a rotational mill (Model Q298, brand QUIMIS, 110 V, 60 Hz, 25 W, 150 rpm rotation on the drive roller).The grinding method was adapted from Jha and co-workers, 46 using a metal pot (300 mL) attached to a 1 L porcelain jar model 21 A (Chiarotti).200 g of alumina beads (10 mm diameter) and 10 g of curcumin (ratio 20:1 m/m) were added to the metal pot.Before grinding, the curcumin was dried at 30 °C for 24 h.The milling process took place for 24 h at room temperature.
Polymer solution preparation: A polymeric solution of PBAT 15 wt% in chloroform (85 wt%) was prepared under constant magnetic stirring for 6 h.A saturated aqueous sodium chloride (NaCl) solution (porogen with a volume corresponding to 6 wt% relative to the PBAT 47) was added.The mixtures were magnetically stirred for 30 min and then subjected to high-intensity ultrasound for 1 min (Sonics Vibra-cell VCX 500-VCX 750 sonicators) under the operating conditions: energy of 92,000 J and amplitude of 50%.The solutions with phase insertion of C20A and/or NC had an extra step in the mixing process before ultrasonication by adding the pre-set proportions of the dispersed phase(s), e.g., 0.5 wt% of C20A and/or 0.5 wt% NC.After adding the dispersed phase(s), the mixture was stirred for 1 h.

Membrane preparation by EIPS/NIPS:
The polymer solution previously described (with porogen and with or without dispersed phase(s)) was applied to glass plates (13 × 18 cm) and spread out using a 200 μm Erichsen wire (200 μm).The procedure was done with 1 min drying time for partial solvent evaporation (EIPS).After evaporation, the plates were immersed in polyethylene trays in a non-solvent (NIPS) (distilled water) coagulation bath to induce phase separation.Then, the membranes were separated from the glass plates and remained in a distilled water bath for 24 h for salt leaching, with a subsequent dry process in an oven at 60 °C for 6 h. 47The EIPS/NIPS membranes were obtained, as shown in Figure 1, and the compositions were based on a previous study 35 and the literature. 44The composition of the polymer solution and dispersed phases (NC and C20A) are described in Table 1.PBAT

Dynamic light scattering (DLS) and ζ potential
The Dynamic light scattering (DLS) recorded the average particle diameter of the adsorbents (C20A and NC).It was done with a stable 90° scattering angle using Zetasizer Nano-ZS (Malvern Panalytical Ltd., Malvern, U.K.).The adsorbents were prepared with 1 mL aliquots and diluted in 100 mL distilled water.The tests were done in triplicate.The ζ potential used the same equipment and sample preparation.

X-ray diffractometry (XRD)
The adsorbents (C20A and NC) and EIPS/NIPS membranes were analyzed by X-ray diffraction measurements using a D8 Focus diffractometer (Bruker AXS Advanced X-ray Solutions GmbH, Karlsruhe, Germany), operating at 40 kV and 40 mA.It used monochromatic CuKα radiation (λ = 1.54056Å) at a step width of 0.01° and counting time of 100 s at each 0.5°, from 2θ = 10° to 60°.

Scanning electron microscopy (SEM)
The adsorbents (C20A and NC) and EIPS/NIPS membrane samples (surface and cross-section) were coated with a 25 nm thick gold layer using Sputtering Leica EM ACE 200 (Leica Microsystems, Wetzlar, Germany).The cross-section membrane samples were obtained by fracturing the membrane after nitrogen immersion.Micrographs were performed using a microscope FEI Quanta 250 (Thermo Fisher Scientific, Hillsboro, Oregon, USA), using an accelerating voltage of 10 kV and a spot size of 4 nm.

Fourier-transform infrared spectroscopy
The adsorbents (C20A and NC) and EIPS/NIPS membranes were analyzed by Fourier-transform infrared spectroscopy (FTIR) using a Frontier 94.942 (PerkinElmer Inc, Massachusetts, USA) with attenuated total reflectance (ATR) mode in a range of 500-4,000 cm -1 , 32 scans, and spectral resolution of 4 cm -1 .

Mechanical testing
Tensile tests analyzed the EIPS/NIPS membranes using an Instron 3367 (Instron®, USA Norwood, USA).All experiments were conducted according to the ASTM D638-14 standard.Five specimens were evaluated with a load cell of 50 N and 50 mm min -1 test speed.

Contact angle (CA)
The EIPS/NIPS membranes were analyzed by contact angle using a dynamic contact angle tensiometer (SEO) model Phoenix 300 (Kromtech Alliance Corp., London, UK), with three phases (air/water/sample).The static contact angles (θ est ) were determined according to the Young-Laplace equation by observing a water droplet of 4.00 μL deposited on the membrane surface.The contact angle values were collected in triplicates for 5 minutes, in 3-second intervals of each measurement, and a mean value of the collected values was considered.

Copper adsorption
A Cu (II) solution containing 1,000 ppm of Cu (II) was prepared to simulate the effluent from a plating industry for the adsorption experiments. 48Cu(NO 3 ) 2 •3H 2 O was dissolved in a solution of 0.01 M HNO 3 , and a sodium hydroxide solution of 0.01 M NaOH was used for pH 5 correction.The pH was set to 5 due to precipitation and low stability of aqueous solutions with pH > 5. 49 Once the solution was prepared, 0.1 g of each EIPS/NIPS membrane was placed in a plastic tube and mixed with 50 mL of Cu (II) solution, on the other hand, an adsorbent dosage of 2 g/L. 50Adsorption measurements in triplicate were performed from the beginning (t = 0) and at times (t) of 10 min, 45 min, and 1,440 min.
2] The adsorption efficiency was calculated using Equation 1, where C o (mg/L) is the initial concentration of the metal ion in the wastewater before adsorption, and C f (mg/L) is the final concentration of the metal ion in the wastewater after adsorption.

Dynamic light scattering (DLS) and ζ potential
Dynamic light scattering (DLS) was used to evaluate the distribution profile and hydrodynamic radius (RH) of C20A and NC, considering these nanomaterials to be spheres.The C20A showed a trimodal distribution profile (Figure 2), with average peak values around 10.1, 342, and 5,560 nm, indicating a polydisperse and heterogeneous sample.The first two peak values are characteristic of nanoclays, as reported in the literature, while the higher radius values (5,560 nm) indicate lamellae agglomerations due to van der Waals bonds.According to Zdiri and co-workers, although individual nanoclays have nanometric dimensions, MMT aggregates have micrometric dimensions, between 0.1-10 µm, consisting of lamellae in random directions. 53The presence of agglomerates justifies using low C20A contents in the membranes, aiming to achieve better dispersion and adequate performance for application in water treatment.The nanocurcumin DLS results showed a broad peak, and the wide distribution is associated with the mechanical grinding process that results in irregularly shaped and amorphous particles. 54It was observed that two peaks were located at 255 and 4,150 nm.The ball milling process is a top-down mechanical method of green synthesis that allows the conversion of microparticles into nanometric dimensions, 55 as observed in Figure 3.

X-ray diffractometry
The XRD technique was used to determine the materials' crystalline pattern, and Figure 4 shows the obtained diffractograms.
C20A showed two broad peaks between 2θ = 0-10°, at i) 3.6° and ii) 7.1°, where peak i) refers to the crystallographic plane (001), and peak ii) refers to the crystallographic plane (002). 57The peak i) showed an interplanar distance (d) of 24.2 Å and is assigned to the montmorillonite-modified layers, [58][59][60][61] corroborating the organomodified nature of Cloisite® 20A.The peak ii) showed d = 12.4 Å, associated with the layers that did not suffer the influence of the organic modifier. 62Other peaks resulting from the modification of MMT were also seen at iii) 19°, iv) 24° and v) 35°6 3 (Figure 4a).According to Patiño-Soto and co-workers, the peak at 3.6° (Figure 4b) is the most characteristic for organomodified clays, corresponding to the original basal reflection peak for Cloisite® 20A. 64Abdolrsaouli and co-workers reported the second peak of Cloisite® 20A at 8.2°, reflecting the smaller interplanar distance, i.e., more excellent packing of the MMT layers. 59The crystallinity index (CI) of the C20A was 85.6%, which agrees with literature reports. 65igure 4c and d-e demonstrate the curcumin XRD patterns before and after the milling process, respectively.Figure 4c shows a broad band with two peaks at 17.4 and 22.1°, highlighting the amorphous characteristics of curcumin associated with the turmeric organic nature. 66After the mechanical milling (Figure 4d-e), the amorphous character of the sample was accentuated due to the ball grinding friction. 67In the nanometric particles, there is an increase in surface defects and random bonds to facilitate adsorption, as illustrated in Figure 4f. 68Zhang and co-workers reported an amorphization of curcumin nanoparticles milled for 24 h at 157 rpm. 54According to Qiu and co-workers, the mechanical energy in the grinding process can disorganize crystalline areas and result in amorphization, increasing samples' adsorption capacity due to increased particle surface area. 69However, the agglomeration process can mask the decrease in particle size after long grinding times. 70The authors reported that the relationship between amorphization and particle agglomeration still requires further studies to understand the linkage of factors.Other authors reported increased particle sorption capacity after amorphization. 71 The lack of atomic periodicity of adsorbent materials also influences diffusion processes involving adsorption. 746] Joly and Latha identified a broad band centered at 2θ = 17°, 77 which was used to calculate the CI value of 30.6% in this work.Figure 5a shows a standard PBAT diffractogram with characteristic peaks between 2θ = 15-25° (Figure 5b, reflections from i to v). [78][79][80][81] Kashi and co-workers described that the peaks in the mentioned range correspond to the characteristic peaks of the PBAT due to the basal reflections in the Miller index planes 2θ = 16.3°(001), 2θ = 17.6° (010), 2θ = 20.3°(110), 2θ = 23.3°(100), and 2θ = 24.6°(111).Figure 5c shows the XRD patterns of the membranes, which showed high similarity with the pure polymer.
The diffractograms with lower peak intensity in the hybrid EIPS/NIPS membranes show more excellent exfoliation of C20A and dispersion of NC along the membrane weight, despite the tendency to agglomerate seen in DLS and ζ potential.

Scanning electron microscopy
The micrographs obtained by SEM of the surface and cross-section of EIPS/NIPS membranes are shown in Figure 6.The results demonstrated that the surface of the pure PBAT membrane presents pores with a more homogeneous cross-section (less surface roughness) than the hybrid membranes.For PBAT/0.5 C20A, it was noted that adding nanoclay decreases the pore diameter by changing the surface, leaving it more irregular and its cross-section, while for PBAT/0.5 NC, the surface becomes less rough.Such differences between the dispersions of the phases may result from different interactions with the matrix since C20A tends to exfoliate when in polymer solutions.The PBAT/0.5 C20A + 0.5 NC sample has more evidence of a dispersed phase due to its irregularity, agglomerations, and higher additive fractions.
Table 2 shows the average diameter and maximum pore size measurements of the EIPS/NIPS membranes and the average dry thickness measurements (cross-section after rupture).Compared to pure PBAT, all hybrid EIPS/NIPS membranes showed smaller average pore diameters ranging from 0.4 to 0.5 µm.Barbosa and co-workers obtained pores ranging from 0.25 to 0.6 µm in PBAT membranes with 6% NaCl as porogen. 84The pores of the membranes for both preparation methods are classified as macropores because they exceed the size of 50 nm (according to the standard of the International Union of Pure and Applied Chemistry (IUPAC)). 85lthough PBAT/0.5% C20A membrane has lower dry thickness than pure PBAT (15.4 and 22.9 µm, respectively), most EIPS/NIPS membranes have thicknesses above 21 µm.According to Liu and co-workers, Young's modulus and tensile strength limit depend on a membrane's thickness. 86Lima and co-workers observed that increasing the thickness of elastomeric membranes decreased both the wettability seen by the contact angle and the hardness of the membranes. 87

Fourier-transform infrared spectroscopy
The FTIR analysis was also used to investigate the developed EIPS/NIPS membranes, and the obtained spectra are presented in Figure 7.The prominent peaks associated with PBAT structure are observed at 3,065 cm -1 (C-H stretching), 2,958, 2,875 cm -1 (CH 3 and CH 2 stretching), 1,710 cm -1 (C = O bond of ester group), 1,410 cm -1 (bending mode of O-CH 2 ), 1,366 cm -1 (bending mode of CH 2 ), 1,267 cm -1 (C-O group) and 726 cm -1 (deformations of the benzene ring).][92] The bands corresponded to the chemical composition of phenolic groups for NC, 93 and these bands are essential for the sorption process of contaminant metal cations, as they can complex bivalent cations due to the direct binding of curcumin molecule with the ion or by hydrogen bond formation. 94] Similarly, the silicates, aluminate, and alkylammonium groups are relevant to promote the complexation and removal of contaminants.The peaks of C20A and NC can be seen in Figure 8.
Comparing the FTIR spectra of hybrid EIPS/NIPS membranes, it was observed that they were mainly related to PBAT typical peaks, which could be attributed to i) the overlapping of some peaks of PBAT that are observed in similar regions of those from NC and C20A; ii) the low concentration of NC and C20A used in the membrane's formulation.

Mechanical testing
The mechanical test was employed to investigate the mechanical properties of EIPS/NIPS membranes (Figure 9).The strain of a membrane is the elongation at break (%), which measures the flexibility, i.e., the ability of the film to deform before cracking and eventual rupture. 97In Figure 9a, it was observed that the pure PBAT membrane obtained the highest elongation at break (154%).In the literature, the elongation at break for PBAT is higher, with records of up to 508%. 98The hybrid EIPS/NIPS membranes are smaller, which is associated with the fact that it is a porous membrane that may have affected the deformation since the pores may have acted as stress-concentrating defects that facilitate rupture. 47The dispersed phase can reinforce the PBAT structure, increasing its rigidity and reducing the membrane's elongation.Both C20A and NC influenced this property, and among the hybrid EIPS/NIPS membranes, the one with the highest strain was PBAT/0.5 C20A (107.5%), and the lowest value was for the PBAT/0.5% NC membrane (72.5%).Thus, PBAT/0.5 C20A + 0.5 NC had an intermediate value compared to membranes with only one dispersed phase (96.1%).A polymer's stiffness is determined by the elastic modulus, obtained by the slope of the linear portion of the stressstrain distribution for a polymer under compression.It is a widely popular metric for choosing materials with superior mechanical attributes.Figure 9b shows that membranes with only one dispersed phase added had a higher elastic modulus.Correia and co-workers also observed this behavior in PBAT membranes with charcoal addition, showing that adding dispersed phases increased membrane integrity. 97In this work, compared to pure PBAT elastic modulus of 62.5 MPa, PBAT/0.5 C20A had an increase of 18.72% (74.2 MPa), and PBAT/0.5 NC had an increase of 9.44% (68.4 MPa).This fact corroborates the increase in IC seen in the XRD analysis, which increases the membrane's stiffness.
In Figure 9c, it was observed that the insertion of C20A in the PBAT/0.5 C20A (with 4.9 MPa) was the one that most favored the increase in tensile strength (8.9%) when compared to that of pure PBAT (4.5 MPa), followed by the PBAT/0.5 NC (4.6 MPa).Crystalline particle loading in polymeric matrix increases the stiffness of materials and thus can influence their tensile strength. 97The PBAT/0.5 C20A + 0.5 NC membrane, on the other hand, showed no increase and had the lowest tensile strength limit (4.0 MPa), which may be an indication that both dispersed phases (C20A and NC) in the matrix did not favor the matrix/reinforcement interaction.Ludwiczak and co-workers studied the mechanical behavior of MMT in a PBAT and PLA polymeric matrix.The authors concluded that the role of clay as an active interfacial modifier in a polymeric composite still needs further research to understand the effect of the dispersed phase on material stiffness and deformation.Thus, the insertion of C20A and NC both separately and simultaneously is still a research object that is not fully understood. 98

Contact angle
The contact angle (CA) is an essential measurement to investigate the surface behavior of the EIPS/NIPS membranes and their interaction with water.Figure 10 shows the dynamic CA over 5 min (Figure 10a), the representation of different CA classifications (Figure 10b), and the average CA (Figure 10c).
The dynamic measurement of the CA can investigate the interaction of the water droplet with the membrane surface as a function of time.Figure 10a showed that EIPS/NIPS membranes decreased in CA over the test time (5 min).It was also observed that CA values were lower than 90° throughout the assay.As illustrated in Figure 10b, hydrophilic materials have CA surfaces less than 90°, and hydrophobic materials have CA above 90°.Therefore, the EIPS/NIPS membranes obtained are hydrophilic. 99That behavior was expected due to the hydrophilic nature of the biopolymer PBAT.Yasin and co-workers observed that the water droplet on poly (vinyl alcohol) membranes with chitosan particles exhibited a decrease in CA concerning time, i.e., the biopolymer matrix surface with a dispersed phase exhibits a hydrophilic nature. 100Lin and co-workers also observed decreased CA with time for a polyhydroxybutyrate (PHB)/ PBAT membrane. 101Furthermore, the porous surface of the EIPS/NIPS membranes seen in the SEM micrographs may have reduced CA. 102evertheless, the dispersed phases' insertion influenced the membrane's CA.The average values of CA in Figure 10c demonstrated that pure PBAT has CA = 55.9°.Botta and co-workers found a CA of 55° for PBAT membranes, like this work.However, the authors observed that the insertion of biochar into the PBAT matrix increased the CA, increasing hydrophobicity, 103 unlike what occurred for the dispersed phases of this work, as seen in Figure 10c.The 0.5% C20A and/or NC loading was sufficient to decrease the CA.The EIPS/NIPS membranes with one dispersed phase had the lowest CA's, with PBAT/0.5 NC with CA = 46.5°and with prominence for PBAT/0.5 C20A with CA = 37.9°.Although both curcumin 104 and organomodified Cloisites 105 are hydrophobic, in the literature, it has already been discussed that both unmodified and modified MMTs, such as Cloisite, [106][107] and curcumin-based materials 108 can increase the hydrophilic character of polymeric films.Therefore, the insertion of nanoparticles may have increased the surface roughness by increasing the surface capillary action and, consequently, the interaction with water. 109This behavior of composite membranes may favor the application of the membranes for adsorption and filtration.In the case of PBAT/0.5 C20A + 0.5 NC, the CA = 54.4° was similar to the pure PBAT membrane, showing that the interaction of this hybrid EIPS/NIPS membrane surface with the aqueous medium is negatively affected by the simultaneous action of the dispersed phases.

Copper adsorption
The adsorption test performed on the EIPS/NIPS membrane demonstrates the ability of the membrane to be an adsorbent material having a highly concentrated solution of contaminant, simulating the industrial effluent from a plating industry and the leak from the Brumadinho disaster (~ 1,000 mg/L). 48,110Figure 11 shows the sorption efficiency (%) aiming to decrease the initial concentration of 1,000 mg/L of Cu (II).
Figure 11 demonstrated that the EIPS/NIPS membrane could decrease the contaminated solution's initial concentration.The times studied showed variation in efficiency up to 24 h.The pure PBAT membrane, e.g., showed an efficiency of 20.5% in 10 min and 21.2% in 24 h, i.e., an increase of only 3.4%.The minor variation in sorption efficiency may be due to the saturation of the monolayer of the EIPS/NIPS membranes surface in contact with a solution of high Cu (II) concentration, reaching maximum adsorption. 111he hybrid EIPS/NIPS membranes improved sorption efficiency, with membranes with only one dispersed phase being the most prominent in sorption efficiency at the end of the analysis (24 h), in the following order PBAT/0.5% NC (28.0%) > PBAT/0.5% C20A (27.0%) (Figure 11).Gaikwad and co-workers achieved a 27% removal efficiency of Cu (II) from mining wastewater at pH 5. 112 In the case of the PBAT/0.5 C20A + 0.5 NC, no increase in adsorption efficiency was observed, having diminished their adsorption efficiency from T1 to T3 (24.4% to 16.7%, respectively).Thus, between 10 min and 24 h, there was a 31.5% reduction in efficiency.4] Xu and co-workers observed a 17% drop in Cu (II) sorption efficiency due to the breaking of active sites of chitosan in contact with contaminated industrial effluent. 113arimi obtained a similar adsorption efficiency to this work using Ca(OH) 2 , with a 20% removal efficiency of Pb (II) at pH 6 from industrial wastewater. 115Kumar and co-workers mentioned a 10% and 20% removal efficiency of Cu (II) and Ni (II) in contaminated water using Trien. 116Therefore, the results are attractive since they illustrate the potential application of hybrid EIPS/NIPS membranes for removing PTEs, for which the inclusion of NC and C20A showed a potential to increase the membrane's potential.Additionally, using eco-friendly materials is appealing to reduce environmental impacts and promote sustainable treatment alternatives to increase water quality.

Conclusion
This work proposes the development of poly (butylene adipate-co-terephthalate) (PBAT) membranes with the inclusion of nanocurcumin (NC) and Cloisite® 20A (C20A) for the removal of toxic contaminants from water.Both dispersed phases presented nanometric size but with agglomeration effects observed due to their low ζ potential.C20A showed a crystallinity index (85.6%),while NC presented amorphization due to mechanical milling.The hybrid EIPS/ NIPS membranes showed smaller pores, also impacting the membrane's mechanical performance, with the C20A promoting an increase in elastic modulus and tensile strength.In contrast, the PBAT/0.5 C20A + 0.5 NC showed reduced properties due to the amorphous nature of NC.EIPS/NIPS membranes present a hydrophilic nature, which is attractive for water contact applications.The sorption results showed that the membranes effectively reduced copper concentration even at low contact time, removing 25.7% for the PBAT/0.5 NC and 25.2 for PBAT/0.5 C20A.Thus, the PBAT/0.5 C20A + 0.5 NC membrane demonstrated a reduction in efficiency over time.The significant reduction in initial high concentration highlights the membrane's capability to eliminate toxic elements from water.Moreover, these findings emphasize using eco-friendly materials for water decontamination, providing a promising substitute for conventional treatment approaches.With the results obtained in this work, in future work in Cu (II) adsorption experiments using

Figure 1 .
Figure 1.Illustrative scheme of methodology preparation for the functional hybrid EIPS/NIPS membrane, (1a-c) preparation of PBAT solution and (1d) PBAT membrane with C20A and/or NC loading; (2a-e) EIPS/NIPS membranes structure obtaining and (2f) the final drying step are named EIPS/NIPS membranes, and the membranes with NC and/ or C20A are named hybrid EIPS/NIPS membranes

Figure 2 .
Figure 2. Distribution profile and hydrodynamic radius (RH) of C20A ζ potential was used to evaluate the electrostatic stability of the nanoparticles, and the samples showed ζ potential values of -11.5 and -3.2 mV for C20A and NC, respectively.According to the literature, particles with high ζ potential values in the module present higher electrical stability, while particles with lower ζ potential could present agglomeration tendency.56Since the dispersed phases have low ζ potential, they can present agglomeration effects, corroborated by DLS results, with C20A lamellae clusters and poor NC dispersion.

Figure 3 .
Figure 3. Distribution profile and hydrodynamic radius (RH) of NC

Figure 7 .
Figure 7. FTIR spectra EIPS/NIPS membranes, highlighting the prominent peaks associated with C20A and NC

Figure 8 .
Figure 8. FTIR spectra of C20A and NC

Figure 10 .
Figure 10.(a) Dynamic contact angle for the EIPS/NIPS membranes in the range of 0 to 5 min, (b) representation of a water droplet in a hydrophilic and a hydrophobic surface, and (c) mean values of the contact angle of the membranes

Table 1 .
Composition of EIPS/NIPS membranes developed, with the percentages of PBAT solution, C20A, and/or NC powders