Development of a Cost-Effective Polyamide 6-Based Composite Ceramic Membrane for the Treatment of Cadmium-Contaminated Wastewater

Natural drinking water sources have been increasingly contaminated by industrial activities that release chemical effluents containing persistent pollutants. Among these, heavy metals are particularly problematic due to their non-degradable nature and their ability to bioaccumulate in living organisms, causing long-term ecological and health risks [1, 2]. As noted by Palisoc et al., heavy metals disseminate through aquatic ecosystems and food chains by accumulating in sediments and biota [3]. Cadmium (Cd), a toxic heavy metal of significant concern, is introduced into the environment primarily via anthropogenic activities such as sewage discharge, agricultural runoff, and various industrial processes. According to Sanaei et al., key sources of cadmium pollution include the combustion of biomass and coal, mining operations, use of liquid fuels, electroplating, nonferrous smelting, and municipal waste incineration [4]. Tobacco smoking represents the predominant route of cadmium exposure in humans [5]. Additional sources comprise industrial waste from leather tanning, garment manufacturing, painting, rechargeable battery production, pesticide application, as well as natural contributions from soil erosion and volcanic activity. Cadmium poses serious environmental and human health hazards, even at low concentrations over prolonged exposure [6]. It is linked to renal failure, cardiovascular diseases, and other systemic toxicities [7, 8]. Unlike many organic contaminants, cadmium is persistent and bioaccumulative, accumulating in tissues without environmental degradation [9, 10]. Chronic ingestion of Cd(II)-contaminated water has been associated with severe conditions such as Itai-itai disease, prostate cancer, osteoporosis, proximal tubular impairment, emphysema, decreased bone mineral density, testicular atrophy, and increased mortality [11]. Consequently, effective removal of cadmium from industrial wastewater before environmental discharge is essential to safeguard ecosystems and public health.

Therefore, designing a treatment unit that is both inexpensive and effective is crucial for removing cadmium from aqueous media. Numerous techniques of the treatment used to remove cadmium from drinking water, such as membrane filtration, electrochemical, adsorption, ion exchange, precipitation, reverse osmosis, electrolysis, and evaporation techniques, have been explored; however, membrane-based techniques have proven to be the most effective in this regard [12,13,14]. The membrane filtration method has beneficial features such as less consumption of energy, improved filtration efficiency, and a lack of phase change. While membranes have been made from inorganic and organic ingredients, ceramic filters have received greater attention than filters made from polymers because of their improved chemical endurance and superior fouling resistance. Even though polymeric membranes are commonly employed for water filtration because they are inexpensive and easy to generate pores, they have several drawbacks, including low mechanical strength and poor fouling resistance [15, 16]. Due to their many advantages, composite ceramic membranes have recently attracted considerable interest. Ceramic substrates have excellent mechanical, thermal, and chemical stability, making them excellent support. Similarly, because of its varied surface activity, the active polymeric layer can work as an excellent separating layer [17]. Various materials, including clays, titania, fly ash, silica, and zirconia, have created porous ceramic membranes. One of the most prevalent substances used to produce ceramic membrane filters and porous membrane supports for asymmetric membranes is alumina. Because of its superior metal ions adsorption capabilities, alumina was also employed to eliminate different metals from the environment [18].

In the present research, composite ceramic membranes for water treatment applications were developed from commercially available clays, feldspar, and sand to fabricate support ceramic membranes. Polyamide 6 was used to modify the ceramic membrane by formation of an active top layer via the dip coating technique. The effect of adding a thin film of polymer on the ceramic membrane surface was investigated. The performance of the fabricated composite membrane on cadmium ions (Cd(II)) removal from wastewater was evaluated. Also, the membrane characteristics, such as mechanical strength, water permeability, and porosity were investigated.

The goal of this research is to create a composite ceramic membrane that is both efficient and cost-effective for removing cadmium ions from industrial effluent. This study aims to address the challenges of cadmium contamination by developing, characterizing, and evaluating the performance of a PA6-coated ceramic membrane that was designed for efficient removal of cadmium ions from industrial effluents. The study's findings have the potential to expand water purification technology and provide a long-term strategy for cadmium ions removal that can be applied in large-scale wastewater treatment plants.

2.1 Materials

Clay samples were collected from Aswan, Egypt. Kaolin was obtained from South Sinai, Egypt. Quartz sand and feldspar (KAlSi₃O₈) were collected from the Eastern Desert of Egypt. Starch was purchased from the local Egyptian market. Deionized (DI) water was used in all experiments. Polyvinyl Alcohol (PVA) was supplied by Oxford Company, and polyamide 6 (PA6) was provided by Dop Organik Kimya. Formic Acid (FA) with a purity of 85% was obtained from El Nasr Pharmaceutical Chemicals Company. Cadmium nitrate tetrahydrate (Cd(NO₃)₂·4H₂O), with a purity greater than 98%, was purchased from Sigma-Aldrich.

2.2 Methods

2.2.1 Preparation of the Suggested Raw Mix

A ceramic raw mixture was prepared using a combination of Aswan clays and kaolin (65 wt%), feldspar (KAlSi3O8) (25 wt%), and quartz (10 wt%). The components were weighed in the specified proportions, and a total of 1000 g was used for each batch. Specifically, 650 g of a homogeneous blend of Aswan clays (150 g) and kaolin (500 g), 250 g of feldspar, and 100 g of quartz were combined. The raw materials were subjected to wet grinding using a laboratory ball mill for 30 min to ensure uniform particle size distribution and homogeneity of the mix. The resulting powder was dried and stored in a sealed container for subsequent use.

2.2.2 Solution Preparation

2.2.2.1 3 wt% PVA Solution

A 3 wt% poly(vinyl alcohol) (PVA) solution was prepared by dissolving 3 g of PVA powder in 97 g of boiling distilled water. The powder was added slowly with continuous stirring on a hot plate using a magnetic stirrer until complete dissolution. The solution, which turned clear and homogeneous, was then cooled to room temperature for an hour and stored for later use.

2.2.2.2 20 wt% PA6 Solution in Formic Acid

A 20 wt% polyamide 6 (PA6) solution was prepared by adding 20 g of PA6 powder in small portions to 80 g of formic acid (FA). The FA was cooled in an ice bath, and stirring was maintained throughout the addition. Once a clear solution was formed, it was kept in the ice bath and stored in a sealed container until needed.

2.2.2.3 Preparation of Synthetic Wastewater

To evaluate the membrane performance under realistic conditions, synthetic wastewater was prepared to simulate cadmium-contaminated industrial effluent. A 100 ppm Cd(II) stock solution was prepared by dissolving 274.5 mg of cadmium nitrate tetrahydrate (Cd(NO₃)₂·4H₂O) (analytical grade, ≥ 98%) in 1 L of deionized water. From this stock solution, working solutions with target concentrations of 25, 50, and 100 ppm Cd(II) were prepared by serial dilution using deionized water. These concentrations were selected to simulate various pollution levels commonly encountered in industrial wastewater from electroplating, battery manufacturing, and metal-finishing industries.

To mimic the composition of typical industrial wastewater, background electrolyte ions were added to the working solutions. The final synthetic wastewater composition included:

Cd(NO₃)₂·4H₂O: to provide Cd(II) ions at target concentrations (25, 50, or 100 ppm), NaCl (0.01 M): to simulate ionic strength and conductivity typical of industrial waste streams, CaCl₂.2H₂O (10 mg/L Ca(II)): to represent common hardness ions, MgSO₄.7H₂O (10 mg/L Mg(II)): to simulate divalent cation presence. The pH of all feed solutions was adjusted to 6.5 ± 0.1 using either 0.1 M HCl or 0.1 M NaOH, reflecting the typical pH range of untreated industrial wastewater and to ensure consistent testing conditions across all experiments. All solutions were freshly prepared prior to testing and stored in high-density polyethylene (HDPE) containers to prevent contamination or adsorption losses.

2.2.3 Fabrication of Support Ceramic Membrane

Cylindrical disks 50 ± 2 mm in width and approximately 5 mm in thickness were fabricated by compressing 20 g of a mixture of fine raw powder, 1 g of corn starch, and 4 g of the previously prepared PVA solution into a stainless-steel mold with a hydraulic press used in laboratories with a unidirectional load of thirty mega Pascal. Using a drying oven, the membrane specimens were dried in two stages. After 6 h of initial incubation at 60 °C, an additional 6 h at 110 °C were needed. A laboratory muffle was employed to sinter the green filter discs (M1 and M2) at 1000 °C for 30 min of soaking. At an invariable heat rate value of 5 °C min⁻¹ was maintained. A digital Vernier caliper was applied to estimate the membrane dimensions after production. Similar cubic specimens of approximately 50 ×ばつ 50 ×ばつ 50 mm³ were molded, dried, and fired in the same manner as the membrane specimen preparation to calculate the compressive strength of the created support ceramic filters.

2.2.4 Modification of the Ceramic Membrane Surface

The dip coating approach was used to modify membrane M2. The procedure used to coat the ceramic membrane with the polyamide 6 (PA6) layer included the following steps:

1. 1.

   Surface Preparation: The ceramic membranes were initially cleaned using ethanol and deionized water, followed by drying at 60 °C for 2 h to remove any surface contaminants and ensure good adhesion of the coating.

2. 2.

   Coating Process: The PA6 coating was applied using a dip-coating method. The ceramic membranes were immersed in a 20 wt% PA6 solution prepared in formic acid at room temperature for 24 h.

3. 3.

   Curing: After coating, the membrane was removed slowly to ensure uniform film formation, then upon removal from the solution, the membrane was immediately submerged in a cold-water bath for 1 h to initiate solidification of the PA6 layer. Following this step, the membrane was thoroughly rinsed multiple times with distilled water to remove any loosely adhered or unreacted material. The resulting composite ceramic membrane, coated with PA6, was labeled as M2.

Accordingly, two distinct groups of synthesized membranes were prepared for this study as follows:

M1: Unmodified ceramic membrane fabricated from raw powder, PVA, and corn starch, and sintered at 1000 °C.

M2: Composite ceramic membrane produced by modifying M1 via dip-coating in a 20 wt% PA6 solution, followed by cold water coagulation.

2.3 Characterization

2.3.1 Characterization of Raw Materials

A PANalytical AXIOS wavelength dispersive (WD-XRF) spectrometer was employed to evaluate the element composition of the raw powder using X-ray fluorescence (XRF) spectroscopy, raw materials were crushed and then ground in a Herzog-type mill to reach fine powder. The ground powder should pass a 0.063 mm sieve. The sample was prepared as pressed discs by thoroughly mixing 7 g of the fine powder of the sample with 1.6 g of binding wax in a small ball mill, at a speed of 380 rpm, for one minute. The sample was then put in a standard aluminum cup after which they were pressed in a uniaxial pressing machine under a total force of 130 kN. The yielded disk specimen was used in qualitative and quantitative analysis of the elements. A BRUKER D8 advanced computerized X-ray diffractometer with mono-chromatized Cu Kα radiation, operated at 40 kV and 40 mA was used to perform an X-ray diffraction (XRD) investigation to know the mineralogical constituents of the raw material that was finely ground (200 mesh), mounted randomly on an aluminum holder, and analyzed. Thermogravimetric analysis (TGA) and differential Thermogravimetric (DTG) were applied to the raw powder to investigate its thermal behavior. The Setaram THEMYS ONE + reactor was used in a nitrogen atmosphere and heated starting from room temperature and heated for 1000 °C at a heating level of 10 °C min⁻¹. The manufacturer's software was used to process the data. The standard screening process outlined in ASTM D422/2014 was utilized to assess the particle size distribution and ascertain the proportion of different grain sizes contained in a particular raw powder sample [19]. Following the reporting of the cumulative analysis, a semi-logarithmic graph is displayed. The median particle size (D₅₀) and the volume-surface mean diameter (D_s) are also estimated. The powder density was determined three times using a standard water pyrometer and the standard procedure ASTM D854/2024 [20].

2.3.2 Characterization of the Composite Membrane

The physical characteristics of the ceramic support (cold and boiling water absorption, bulk density, and apparent porosity) were determined for two times by ASTM C373/2024 [21]. The mechanical compressive strength of the cubic samples was evaluated two times according to ASTM C109/2023 [22]. The prepared ceramic filters were subjected to ATR-FTIR to characterize the membrane upper layer using a Bruker VERTEX 80 (Germany) joined with Platinum Diamond ATR, which comprises a diamond disk as an inner reflector in the run of 4000–400 cm⁻¹ with a refractive index of 2.4, and a resolution of 4 cm⁻¹. SEM was employed to examine the superficial and cross-sectional morphologies of the produced filters using a field emission gun (QUANTA FEG250) apparatus linked to an energy-dispersive X-ray analysis unit. Constructed filters' pore size distribution (PSD) was investigated using a pore sizer (Micromeritics 9320, USA). A zeta meter was applied to conclude the surface charge of the fabricated filters.

2.4 Membrane Performance

2.4.1 Filtration Test

The clean water permeability of the produced membranes was measured three times at room temperature using deionized water. The pure water flux (PWF) of the filters was estimated by Eq. (1) [23].

where V is the volume of filtrate (L), Δt is the operation period (h), and A is the effective membrane area (m²).

Filtration investigations were conducted using a dead-end filtration apparatus. Every experiment (6 runs) was conducted at an ambient temperature (25 °C) and pH of 6.5. The feed solution for the filtration experiment was synthetic wastewater containing cadmium with specific concentration and real wastewater samples that were obtained from electroplating industries. These effluents typically contain cadmium concentrations ranging from 0.2 to 8 mg/L along with other heavy metals and organic contaminants. An electroplating industrial effluent was gathered from 10th of Ramadan City, Egypt. Samples are acidified immediately with nitric acid (HNO₃) to pH < 2 to preserve metal ions. Stored at 4 °C and analyzed within 24 h to prevent changes in composition. Before membrane filtration the pH adjusted to be 6.5. A 3 L tank was filled with the feed solution (artificial wastewater/industrial effluent) and pumped through the test apparatus. The prepared filter discs were placed in the testing unit's housing module. The feed samples were pumped through the filtration system unit to evaluate the membrane separation efficiency. After that, the permeate was collected, the flux was calculated, and the metal ions examination was performed. Cadmium concentrations in both raw and treated wastewater samples were determined using Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES, PerkinElmer Optima 8000). Prior to analysis, samples were filtered through 0.45 μm membrane filters, acidified to pH < 2 using nitric acid, and diluted as necessary with deionized water. The instrument was calibrated using certified cadmium standards in the range of 0.01–5.0 mg/L. The analysis was conducted at the cadmium emission wavelength of 228.802 nm, with the following operating conditions: RF power of 1500 W, plasma gas flow rate of 15 L/min, nebulizer gas flow rate of 0.8 L/min, and axial viewing mode. The detection limit for cadmium was approximately 0.005 mg/L. All samples were analyzed in triplicate, and quality control standards were included after every ten samples. COD and TSS were measured using Standard Methods for the examination of water and wastewater.

Using Eq. (1), the permeate flux was determined. The percentage elimination of cadmium metal ions was estimated according to Eq. (2).

where C_f is the concentration of cadmium metal ions in the starting solution (mgL⁻¹), C_p is the strength of cadmium ions in the infiltrate (mgL⁻¹), and R is the calculated rejection (%) [24].

2.4.2 Fouling Test

Fouling experiments were carried out to evaluate the resistance of the membranes to flux decline under cadmium-containing synthetic wastewater and to assess their antifouling and self-cleaning capabilities.

Prior to fouling, the initial pure water flux of each membrane (water flux of the original membranes, (J_w1)) was measured by filtering deionized water at a constant transmembrane pressure (TMP) of 1 bar for 60 min. The volume of permeate was recorded at 15-min intervals to ensure steady-state flux behavior. Next, synthetic wastewater containing Cd(II) at 100 ppm (as described in Sect. 2.2.2.) was used as the foulant solution. The membrane was operated under the same pressure conditions for 60 min, and the permeate flux (J_p) was recorded every 15 min. This time frame was chosen to represent short-term fouling behavior, commonly used in membrane performance studies to observe initial fouling trends and establish reproducibility. Following the fouling phase, the membranes were rinsed with deionized water for 60 min under the same pressure to simulate hydraulic cleaning. After a stabilization period of another 60 min, the recovered water flux (J_w2) was measured. To quantify the fouling behavior and recovery efficiency, two key performance metrics were calculated, the flux decline ratio (FDR) using Eq. (3), and flux recovery ratio (FRR), as indicated in Eq. (4) [25]:

This parameter indicated the percentage of initial water flux recovered after rinsing. Both FDR and FRR were used to assess the antifouling performance of the modified (M2) membranes.

3.1 Examination of the Raw Materials

3.1.1 Elemental Analysis by XRF

The components that are employed to create ceramic membranes have a significant impact on their properties. The chemical components of the raw powder are as follows, silica 62.70%, alumina 26.20%, ferric oxide 1.4%, calcium oxide 1.11%, magnesia 0.22%, potassium oxide 2.20%, sodium oxide 0.37%, sulfate content 0.09%, and loss on ignition 5.71%. The major constituent of all the raw materials was silica, the second one in the composition was alumina. Alkali and alkaline-earth metal oxides are also in trace amounts, including MgO, CaO, K₂O, Na₂O, and transition metal oxides (Fe₂O₃).

3.1.2 Mineralogical Analysis (XRD)

The XRD peaks indicated that the primary constituents of the raw mix utilized are quartz, SiO₂, kaolinite, Al₂Si₂O₅(OH)₄, and mullite, (Al₂O₃)₃.2(SiO₂), with minor contributions from illite, KAl₂Si₃AlO₁₀(OH)₂, anatase, TiO₂, and hematite, Fe₂O₃. As shown in Fig. 1, the peaks in the XRD configurations indicated that the primary constituents of the inorganic admixture were formerly quartz, kaolinite, and mullite. The XRD pattern of the utilized corn starch, as examined by Chen et al. [26], the findings indicated that crystalline and amorphous components make up the structure of corn starch. Twenty-one percent of the corn starch was crystalline. The interior molecular chains of starch interacted to generate crystal peaks. Hydrogen bonds between the OH groups in the interior molecular sequence of starch particles allowed the molecules to form chain crystals. A hydrogen link among the starch molecule chain and water molecules forms the crystal observed at the crystal peak at 22.8° [26].

3.1.3 Thermogravimetric Analysis

TGA and DTG were employed to examine the thermal behavior of the raw mixed materials from the ambient temperature to 1000 °C. The findings are shown in Fig. 2. The proportion of weight loss and thermal stability of the ceramic membrane raw mix were investigated via thermal analysis. Thermal studies revealed that the produced membranes are highly stable at elevated temperatures. Two significant endothermic peaks were observed on the DTG curve. The first endothermic peak is associated with removing both absorbed and free water. At temperatures between 25 and 200 °C, this dehydration is accompanied by a 0.113% weight loss [27, 28]. This peak corresponds to the kaolin dehydration process. The octahedral layer reconfigures, initially at the surface's OH, leading to kaolin dehydration [29]. The α–β allotropic transition of the quartz was responsible for the second endothermic peak in the DTG curve, which was detected at approximately 580 °C [30]. The most significant weight loss of 5.216 weight percent that was noted between 400 and 580 °C is associated with the conversion of α-quartz to β-quartz and the loss of structural OH groups at 513 °C because of kaolinite transforming into metakaolinite, as per the following reaction [31].

Al₂Si₂O₅(OH)₄ (450 − 650 °C) = Al₂O₃.2SiO₂ + 2H₂O.

The overall weight loss through the thermal cycle was 5.329%, which agreed with the 5.71% loss on ignition.

According to the study of Gilman et al. to the thermal responses of PVA, the study approved that at approximately 480 °C, PVA decomposes entirely [32, 33]. According to Chen et al., corn starch begins to break down at approximately 297 °C [26]. This is because the α-1,4 glycosidic linkages are broken at the beginning of intermolecular dehydration. At 600 °C, the starch completely decomposed [26].

3.1.4 Particle Size Distribution

It is commonly known in ceramic membrane technology that the raw material particle size significantly impacts the membrane's pore size. Figure 3 illustrated the dispersion of the raw material's particle size. Particle sizes less than 0.1 mm (100 μm) made up 60% of the raw mixed powder, as observed in Fig. 3.

The diameter of a sphere having a comparable volume/surface area proportion as a particle of concern is known as the volume-surface mean diameter or Ds, is used to calculate the fineness of the designed mixtures. Using the following equation [34], the raw mix was calculated to be 0.09 mm (90 μm), and the corn starch was calculated to be 0.086 mm (86 μm).

D_pi is the mean aperture of two successive screens, and x_i is the mass portion in a particular increment.

For the raw mix, the median particle size (D₅₀) was 0.075 mm (75 μm), while for the corn starch, it was 0.043 mm (43 μm).

3.1.5 Powder Density

A standard water pycnometer was employed to estimate the actual density of the raw constituents. The mean values for the raw mix and corn starch were 2.22 ± 0.06 and 1.51 ± 0.03 g cm⁻³, respectively, after the procedures were repeated three times.

3.2 Composite ceramic Filter Characteristics

3.2.1 Mechanical and Physical Characteristics

The measured values of the physical characteristics and strength of the constructed support ceramic filter are displayed in Table 1. The fabricated ceramic support structure exhibited good mechanical properties correlated with its apparent porosity and bulk density. Thus, the material has good mechanical resistance. It should be highlighted that these findings are entirely consistent with the XRD and SEM investigations [18].

3.2.2 Fourier Transform Infrared (FTIR) Spectroscopy Analysis

The surface functional groups in the support and modified ceramic membranes were characterized using Fourier transform infrared (FTIR) spectroscopy. The vibrations of the polymer's molecular bands showed that it was present on the ceramic support's surface and that its chemical structure had not changed, as per the characterization provided by ATR-FTIR Fig. 4.

An ATR system was used to perform FTIR measurements on the ceramic support, as illustrated in Fig. 4; however, it did not exhibit absorption bands in the polymer region, revealing that the PA6 spectrum was observed on the composite membrane (M2) surface (Fig. 4).

The FTIR results of the support ceramic (M1) are presented in Fig. 4a; the band at 2986 cm⁻¹ was attributed to OH stretching. The peak detected at 1056 cm⁻¹ was attributed to quartz's Si–O stretching [36]. The approximately 778–796 cm⁻¹ peaks were ascribed to OH distortion associated with K⁺ and Al³⁺. The peak detected at 449 cm⁻¹ was attributed to Si–O–Si asymmetrical bending [31]. The bands at approximately 1056 cm⁻¹, 796 cm⁻¹, 778 cm⁻¹, 694 cm⁻¹, and 449 cm⁻¹ showed that all the samples had a high quartz content. The peak at 566 cm⁻¹ was ascribed to Si–O bending and Si–O–Al stretching, and the peak observed at 694 cm⁻¹ was assigned to Si–O stretching and Si–O–Al stretching [37].

Figure 4b displayed the modified composite membrane's FTIR spectra. The bands detected at 2938, 3297, and 3735 cm⁻¹ are characteristic of C–H axial deformation, stretching of the hydrogen bond N–H of PA, and the O–H band of formic acid, respectively. The bands at 1357, 1462, 1542, and 1639 cm⁻¹ correspond to C–O stretching vibrations and carbon skeletons (amides), asymmetric deformation of C–H, N–H in-plane deformation vibrations and C–N stretching vibrations of amides, and the stretching of C=O, respectively [13, 38, 39]. The functional groups of polyamide 6 (PA6) played a crucial role in the separation process by enhancing the membrane's ability to selectively interact with and filter out cadmium ions from contaminated wastewater.

3.2.3 SEM Analysis

Figure 5 displayed the microstructure analyses (SEM) of the support and composite ceramic filters. The membrane pore structure was random, as shown by SEM images of each membrane taken at different magnifications. The active organic filtration layer is effectively deposited onto the supports, as shown in the surface images of the composite filter. The penetration of the polymeric suspension into the ceramic support allowed for the verification of adequate coverage and pore filling. The PA top layer on M2 appeared denser (approximately 200 μm). In general, polymer deposition fills the voids in ceramic supports. The results of the water flux also verify the polymer's deposition and layer development on the ceramic substrate [40].

It is evident that there are no cracks and that the support and microlayers have excellent adherence. This finding justifies the favorable separation layer deposition conditions during the dip casting procedure [30]. The SEM images showed that the sintering temperature significantly impacted the ceramic membrane microstructure. The sample was sintered at 1000 °C, sufficient to achieve adequate particle cohesiveness. This morphology ensures a ceramic solid body. Good cohesion between particles can be achieved at a sintering temperature of 1000 °C. Furthermore, the partial verification of amorphous silica allows material densification. Although it hurts the porosity, this microstructure increases the membrane's mechanical strength [41]. As a result, the membrane has a denser form and a significant amount of pore closure. This finding explained why, compared to M1, which remained unmodified, M2, coated with a thin layer of PA, had a greater heavy metal separation efficiency [35].

3.2.4 The Distribution of Pore Size

The membrane pore size distribution is inversely correlated to the sintering temperature and decreases as the temperature increases. This could be because the melting of minerals at higher temperatures decreases pore size and membrane porosity overall [28, 31]. Thus, the pore radii on the cumulative curve had ranges of approximately 80 μm and 0.003 μm. Along the PSD curve, this broad range permits the identification of various membrane pore categories. Assuming that every pore is a cylinder, the average pore radius (r) is equal to 2 V/A when the volume (V = πr²L) is divided by the pore area (A = 2πrL). The ratio of the total volume of a sample to the total quantity of mercury that has been intruded at the most significant pressure is used to calculate the porosity.

Table 2 showed a summary of the intrusion data for the constructed membranes. Figure 6 displayed the cumulative pore volumes for M1 and M2.

3.3 Performance of Membranes

3.3.1 Water Permeability and Filtration Experiments

The water permeability Lp for the support and enhanced ceramic membranes. This illustration demonstrated how adding a thin layer of polyamide reduces the permeability from 591.4 to 558.5 L h⁻¹ m⁻² bar⁻¹ at the same sintering temperature. Because porosity and permeability are closely correlated, the decrease in porosity with the addition of polymers may account for this decrease in permeability [42]. Good adhesion between the separation layer and the ceramic support is essential for the structural stability of the composite membranes, according to Wei et al. [43]. The permeability results that were previously presented indicated that the composite membrane exhibited stable behavior throughout the testing.

As anticipated, the ceramic support showed a greater water flux. The pure water flux decreased due to the top layer of PA6 in the composite membrane, demonstrating that polymer deposition on the ceramic support top layer enhanced the resistance to water flux [44].

Both improved and original ceramic membranes were used in the rejection study of Cd(II) ions. The average permeate flux and average cadmium rejection efficiency are displayed in Table 3. A rejection study was conducted to evaluate the performance of the modified ceramic membrane (M2) in treating industrial effluent, particularly for the removal of heavy metals and other associated contaminants. The M2 membrane demonstrated a high removal efficiency for cadmium, achieving a rejection rate of up to 99.89%. The physicochemical characteristics of the industrial effluent before and after membrane treatment are presented in Table 4. Cadmium concentration was reduced to below 0.01 mg/L, meeting both Egyptian environmental discharge limits and WHO guidelines for safe water. Total Suspended Solids decreased by over 90%, enhancing water clarity. Chemical Oxygen Demand was reduced by approximately 75%, indicating removal of organic contaminants. Other heavy metals showed substantial reductions consistent with the membrane’s selective separation properties. These results demonstrate the membrane’s high efficacy and suitability for treating cadmium-contaminated industrial wastewater in real-world conditions. The enhanced removal performance is attributed to the strong interactions between the membrane surface and the effluent constituents, which facilitate effective retention of heavy metals [45]. The isoelectric point of the composite membrane, M2, (or a zero-point charge value (pH_zpc)) is 5.8. The PA composite membrane is positively charged at pH < 5.8 and negatively charged at pH > 5.8, according to this data. Consequently, the PA composite membrane has a high affinity for interacting with the Cd(II) ions at pH values greater than 5.8. At 6.5 pH, the feed solution pH, steric hindrance can occur when Cd(II) metal ions chelate with the surface amino groups of the PA filter, preventing metal ions from passing through even at high pressure [46]. It should be mentioned that because of the membrane's porous shape, the composite can adsorb some metal ions at pH values lower than pH_zpc. This effect was produced by the PA composite membrane's NH₂ groups protonated at low pH levels, which causes a repulsion with the Cd(II) ions.

The rate at which metal ions are removed increases when positively charged heavy metals adsorbed on negatively charged membrane surfaces. Metal cations tend to chelate with the -NH₂ groups on the membranes due to deprotonated amine groups, reducing the positive charge on the membrane surface. Thus, the Donnan effect may also explain why heavy metal cations are rejected at higher rates [17]. In the case of the unchanged membrane, silanol groups in fused silica are easily ionized under many conditions to create the negatively charged inner wall of pores, attracting the metal cations found within liquid solutions [47]. The membrane rejection rate increased with increasing starting concentration. This is because a significant portion of the adsorption sites in a system with low concentration remain unoccupied by cadmium ions. The vacant adsorption sites gradually fill up as the starting concentration rises and eventually reaches saturation [48]. Consequently, several possible removal mechanisms, including steric hindrance, charge repulsion, and chelation with surface functional groups, increased the overall rejection efficiency of the composite membrane. Nandi et al. demonstrated that rejection increased as membrane flux decreased [49]. The deposition of a thin film of PA6 over the upper surface of the ceramic support formed the ceramic composite membranes. The polyamide 6 layer on the ceramic substrate enabled the improved membrane to have smaller pores, as demonstrated by SEM examination. As illustrated in Figs. 7 and 8, the selective layer thickness produced a barrier to permeate flux passage, which explained why the membrane flux decreased, and the rejection of heavy metals increased [50].

3.3.2 Composite Membrane Fouling

Membrane fouling in membrane bioreactor (MBR) systems is a complex phenomenon influenced by various interacting mechanisms. It is commonly categorized into two types based on the ease of foulant removal: reversible and irreversible fouling. Reversible fouling refers to the accumulation of materials loosely bound to the membrane surface or pores, which can be effectively removed through physical cleaning methods such as backwashing, air scouring, or relaxation. This process typically restores membrane flux to its original state. In contrast, irreversible fouling involves strongly adhered or deeply embedded foulants that cannot be eliminated by simple physical cleaning and often require chemical treatment or membrane replacement. The classification is typically determined following a standard cleaning procedure [51,52,53,54]. During filtration tests, membrane fouling was observed due to the accumulation of Cd(II) ions on the membrane surface. This fouling manifested as a decline in permeate flux, but it was found to be reversible. The fouled membranes were restored to their original flux levels after rinsing with distilled water, indicating that the interaction between Cd(II) and the membrane was primarily through weak electrostatic forces and coordination bonds with surface functional groups such as –NH₂ in the PA6 layer. Unlike irreversible organic or biofouling, which often requires chemical cleaning, the metal ions fouling observed here did not involve strong precipitation or permanent binding and thus could be effectively mitigated by simple water washing. The FRR is 60.6% and the FDR is 40.9% for M2. As shown in Fig. 9, cleaning the membrane with pure water was sufficient to restore the membrane flux up to 61% of fouling and thus due to cadmium metal ions that accumulated on its pores and surface [55].

3.3.3 Economic Evaluation of the Prepared Membranes

Evaluating the economic aspects of membrane fabrication is essential to assess its practical and industrial viability. Commercial ceramic membranes are generally priced between USD 500 and 3000 per square meter [56]. In the present study, the costs associated with raw materials and energy consumption for manufacturing both unmodified and modified ceramic membranes were analyzed. The utilization of low-cost materials and sintering processes at reduced temperatures contributes significantly to lowering overall production costs. Nonetheless, it is imperative that any economically viable membrane also exhibits sufficient stability, durability, and performance under operational conditions. Based on the estimated costs of raw materials (Table 5) and energy consumption (Table 6), the production costs for membranes M1 and M2 were approximately USD 2.09 and 108.55 per square meter, respectively. These estimations indicate the feasibility of producing ceramic membranes at substantially lower costs than commercial counterparts, thereby enhancing their accessibility for a broad range of wastewater treatment applications. Lowering the production cost is a critical step toward promoting the large-scale implementation of ceramic membranes, enabling diverse industries to benefit from their inherent advantages such as high chemical resistance, mechanical strength, and operational stability in harsh environments.

3.3.4 Distinction with Previously Published Documents

Table 7 outlined the performance of various membranes in terms of heavy metal rejection, water permeance, and fouling resistance to eliminate cadmium metal ions from various water samples.

In comparison to previously reported membranes, the fabricated composite ceramic membrane demonstrated high separation efficiency for the removal of Cd(II) ions from artificial wastewater, as well as a high flux recovery ratio and pure water flow.

In this study, effective and low-cost composite ceramic membranes were successfully developed using locally available raw materials, including clay, kaolin, quartz, and cornstarch. The ceramic support membranes were modified via dip-coating with a 20 wt% polyamide 6 (PA6) solution, forming a dense and selective layer on the surface. Characterization using SEM, ATR-FTIR, and pore size distribution (PSD) analysis confirmed the successful formation of a defect-free PA6 layer, with a morphological transition from a porous to a denser structure observed in cross-sectional images. The average pore diameter was reduced from 84 to 59 nm under optimized fabrication conditions, indicating improved separation potential. In cadmium ions removal tests, the modified membrane (M2) exhibited a high removal efficiency of up to 98.85% under optimal operating conditions, while also demonstrating favorable flux recovery and decline ratios of 61% and 41%, respectively. These results confirmed that the PA6-coated composite membrane is both efficient and eco-friendly, with low fouling tendencies and good reusability. Moreover, the unmodified ceramic support membrane showed potential for use in integrated treatment systems. The overall performance suggested that the modified membrane is a promising solution for wastewater treatment applications, particularly for heavy metal removal, and supports broader efforts toward environmental protection and sustainable development.

To further validate the applicability of the method in complex real-world conditions, future studies will focus on investigating the effects of interfering species commonly found in wastewater matrices, including major inorganic ions and organic compounds. This will involve collecting and characterizing real wastewater samples to identify major inorganic ions (e.g., chloride, sulfate, nitrate) and organic compounds. Controlled spiking experiments with these species will be conducted to assess their impact on the analytical method’s performance. Additionally, potential mitigation strategies will be explored to enhance the robustness and accuracy of the method under practical conditions.

6.1 Sample Collection

Collect representative samples of real wastewater from different sources (e.g., municipal, industrial).

6.2 Characterization of Wastewater

Perform preliminary characterization to identify and quantify major inorganic ions (e.g., Cl⁻, SO₄²⁻, NO³⁻, Na⁺, Ca²⁺) and organic compounds present in the samples using standard analytical techniques such as ion chromatography and TOC analysis.

6.3 Spiking Experiments

Prepare synthetic wastewater samples by spiking known concentrations of typical interfering ions and organic molecules individually and in combination.

6.4 Method Application

Apply the developed treatment technique to both real wastewater samples and spiked synthetic samples to evaluate performance changes.

6.5 Data Analysis

Compare the analytical results (e.g., removal efficiency, detection limits) between controlled samples and real/spiked samples to assess the influence of interfering species.

6.6 Optimization and Mitigation

Investigate potential strategies to mitigate interference effects, such as sample pretreatment, calibration adjustments, or method modification.

Although the current study focuses on the removal of Cd(II) ions and the associated inorganic fouling behavior. Future studies will focus on evaluating the fouling behavior of the PA6/ceramic composite membrane using model organic foulants such as Bovine Serum Albumin and humic acid, to simulate more realistic wastewater conditions and further assess long-term performance.

The MWCO is the molecular weight at which 90% or more of the solute is retained by the membrane. It represents the membrane's molecular sieving performance during actual filtration. Therefore, a molecular weight cut-off (MWCO) analysis using neutral solutes such as polyethylene glycol or dextran is planned in future studies. This will allow for precise determination of the membrane’s solute rejection threshold and support its application in separating organic pollutants or biomolecules alongside heavy metals. These extended investigations will support the broader application of the PA6/ceramic composite membrane and enable its optimization for more comprehensive wastewater treatment scenarios, contributing to sustainable water management and resource recovery.