The materials used in producing the filtering membranes include isopropanol (C₃H₈O, 99.5%, from Shaanxi Huaxing Experiment Technology Co., Ltd.), powdered activated carbon (AC), silver nitrate solid (AgNO₃), polyvinyl alcohol (EVOH, from Sigma-Aldrich. Batch number: 12822PE).

A commercial electrospinning device was used to fabricate fiber membranes. A scanning electron microscope (SEM), energy-dispersive X-ray spectrometer (EDX), and an inductively coupled plasma emission spectrometer were used to characterize the fibers’ physical and chemical features.

A solution mixed with 7 ml isopropanol and 3 ml deionized water was used as the solvent. EVOH particles of 1.25, 1.0, and 0.75 g were added to the solution to obtain three different concentrations, respectively; the solutions were heated at 75°C by a magnetic stirrer, while silver nitrate solids (0.20, 0.50, and 0.80 g) were added, followed by addition of activated carbon powders (0.15, 0.55, and 0.85 g). In total, 27 compositions are used as shown in Table 1.

The principle and layout of the electrospinning process are indicated in Figure 2. A voltage of 18 KV was provided between the syringe injector and the fiber collector. The standoff distance from the syringe needle to the collector was 15 cm, and the rotation speed of the receiving roller was kept at 500 r/min. The injection speed (the speed of plunger movement) was set at 0.0063–0.0126 ml/min for different fiber diameters. Meanwhile, the environmental temperature was kept between 40 and 45°C to avoid needle blockage by crystallization of the solution. An AgNP-AC composite nanofiber membrane was formed on the roller following electrospinning over a certain time and was finally removed for assessment and tests after irradiation by a UV lamp (20–40 W) installed in the device.

A scanning electron microscope (SEM) was used to observe the morphology and silver nanoparticles and activated carbon particles on the surface of AgNP-AC composite nanofibers. Energy-dispersive X-ray spectroscopy (EDX) was also applied to observe the constituent elements of the particles on the fiber surface.

In aerosols that make up haze, the most harmful substances are industrial pollution, including airborne heavy metal particles and organic matters. Activated carbon is proven to be effective for the adsorption of heavy metal ions due to its specific microporous/ultra-microporous structure (Zaini et al., 2021). Cu(II) is a highly toxic pollutant harmful to humans, which is often seen being discharged by the metallurgy, printing, fertilizer, and other industries (Yu et al., 2021). Therefore, it is of interest to assess the effectiveness of AgNP-AC composite nanofiber membranes for Cu(Ⅱ) adsorption.

Yan et al. (2018) found that a higher temperature is beneficial in promoting the adsorption capacity of the composite fibers with larger nanosilver particles easily activated. While a low pH value of the solution reduces the adsorption capacity of the fiber. This is due to the protonation reaction of functional groups such as the amino and carboxyl groups on the fiber surface with a strong acid condition. Also the competition between heavy metal ions and H⁺ is intensified, which affects the fixation of heavy metal ions onto the fiber membranes (Min et al., 2012; Martín et al., 2018). Based on these, the copper solution of 1000 mg/L concentration was diluted with deionized water to 110 mg/L, 90 mg/L, 60 mg/L, and 25 mg/L, respectively, then adjusted to pH 8 with 0.1 mol/L HCl or 25% NH₃·H₂O solution. The membranes of AgNP-AC composite were put into the diluted solutions containing copper ions in a water bath oscillator and kept at 45°C for 240 min. The concentrations of copper ions in the solution were then measured by an inductively coupled plasma emission spectrometer (ICP-OES).

The antibacterial performance of the membrane was evaluated by the bacteriostatic ring test using Staphylococcus aureus (Ryšánek et al., 2019; Wang et al., 2018). Resuscitated bacteria liquid was divided into three tubes, each containing 3 ml of culture medium. The tubes were then kept overnight in a water bath in a shaker at 37°C. The bacterial liquid was then put into a petri dish (20 ml/dish) and kept sealed at 4°C after cooling, ready for use. For the ring test, 0.5 ml of bacterial liquid was spread onto the petri dish evenly with an inoculation loop or a sterile pipette tip. Four circular cuts of fiber membranes with different AgNP or activated carbon concentrations were placed with distance into a dish, which was then kept in an incubator at 37°C for 4, 8, 12, 24, 36, and 48 h, respectively, diameters of bacteriostatic circles in the dish were then measured as the quantitative assessment of the inhibitory effect.

SEM images of fibers corresponding to different concentrations of EVOH (7.5, 10.0, 12.5%) are shown in Figures 2A–C. It can be seen in the 3D microstructure network of the random and disorienting arrangement of fibers. The diameter distribution of the AgNP-AC composite fiber was measured in the range of 750–1900 nm. Also, there are a large number of AgNPs on the surface and inside of the composite fibers.

Figure 3 1) SEM image of the composite fiber with an EVOH concentration of 7.5%, a silver concentration of 0.03 g/ml, and an activated carbon concentration of 0.02 g/ml. The surface of the fiber characterized by energy spectrum and the EDX diagram of the elements contained in the composite fiber are shown in Figures 3B,C. It can be found that the composite fibers contain four main elements: C, O, Ag, and Pt, while the high Al from the aluminum foil is used to collect the fibers. Among them, Pt was attached to the fiber surface due to the gold sputter coaters for observation. Figures 4A–D are element mapping diagrams of AgNP-AC composite fiber, which verifies the distribution of elements in the composite fiber. Among the four elements contained in the fiber, the carbon element has the largest proportion at 63.71% (wt%). A part of the carbon elements is distributed as a block structure, while most are evenly distributed on the surface of the composite fibers. The silver particles on the fiber surface indicate that the connection with activated carbon may be reliable (Duan et al., 2018).

Diameters of nanofibers were measured using ImageJ (Collins, 2007) software to analyze the influence of different material concentrations on the fiber diameter. Figures 5A,B show SEM images of the composite fibers with different concentrations of silver and activated carbon with the same EVOH concentration.

As shown in Figure 5C, the diameter of the composite fiber increases as the concentration of activated carbon increases. This is because of the increased adhesion on the surface of the droplets/jet, resulting from the addition of activated carbon to the spinning solution. The degree of stretching is thus gradually weakened, resulting in increases in fiber diameters with the same electric field force.

Figure 5D shows the influence of the silver concentration on the average diameter of fibers. As the silver concentration increases, fiber diameters increase. Due to the electrical conductivity of silver, the electric field is weakened with less spinning forces, resulting in larger fiber diameters.

It can also be seen from both Figures 5C,D that a higher concentration of EVOH leads to larger fiber diameters. EVOH increases the viscosity of the solution, reducing the stretching effect of the electric field force during the fiber formation, thus yielding in larger fiber diameters. This has also been verified in previous work (Zhang, et al., 2018).

From these figures, the size of silver particles was also estimated and is found to be in the range of about 0.1 μm (100 nm) to 1.5 μm, with the average being close to 1 μm. A more detailed measurement of the silver particle size and its effect on the antibacterial function will be conducted in future works.

The adsorption rates of Cu(Ⅱ) by tested AgNP-AC composite nanofiber membranes are shown in Figure 6. As can be seen, the best adsorption performance, with an adsorption rate of 94.07% for Cu(II), is achieved by membranes made from 10% EVOH, 0.05 g/ml silver nanoparticle, and 0.085 g/ml activated carbon concentration. The adsorption pattern of this composite fiber membrane can be obtained by the curve fitting of Langmuir and Freundlich isotherm models in Eqs 1, 2 (Nnadozie and Ajibade, 2020; Dandil et al., 2019) using the experimental data, as shown in Figure 7. C e q e = C e Q m + 1 k L Q m (1) ln q e = ln k F + 1 n ln C e (2) where Q m is the maximum adsorption capacity, k L the Langmuir equilibrium constant, k F the Freundlich equilibrium constant, n is the constant related to the adsorption capacity, and C e is the equilibrium concentration.

Figure 7 shows the two isotherm model fitting diagrams of Langmuir (R ² = 0.9573) and Freundlich (R ² = 0.8912), respectively, With a higher coefficient of determination (R ²), the Langmuir model appears to be the better one to describe the adsorption of Cu(Ⅱ) by AgNP-AC composite fiber membranes. As reported by Ghosal and Gupta (2017), following the Langmuir model means a more uniform distribution. In addition, the balance parameter R L , also called the separation factor as a dimensionless constant, can express the basic characteristics of Langmuir isotherms. The value of R L can be calculated by Eq. 3 (Stromer et al., 2018). R L = 1 / ( 1 + K L C 0 ) (3) where C 0 (mg/L) is the initial concentration of heavy metal ions in the solution. Thus, the value of the equilibrium parameter R L (irreversible: R L = 0, favorable: 0< R L <1, linear: R L = 1, unfavorable: R L >1) in the Langmuir model was calculated as 0.05 for this test, which indicates that the adsorption is developing in a favorable direction.

Figure 8 illustrates the influence of the contact time on adsorption. As the contact time increases, the adsorption of Cu(II) by the composite fiber membranes increases. The adsorption shows a higher rate in early time. After fitting the raw data, the adsorption equilibrium time can be regarded as approximately 240 min with the absorption rate reaching a plateau value.

The antibacterial experiment results of fiber membranes at different times are shown in Figure 9 for different AgNPs or activated carbon concentrations. Diameters of bacteriostatic circles in the dish were measured as an indicator of the inhibitory effect. It can be observed from Figure 9 that various bacteriostatic rings were formed around the circular patches of membrane containing silver nanoparticles, while there was no such ring around the composite nanofiber patch without silver nanoparticles.

Figures 10, 11 show the influence of different concentrations of silver and activated carbon on the measured diameters of antibacterial rings against the incubation time. It can be seen from Figure 10 that ring diameters increase with the silver concentration. And the rate of increase was fast in the early hours, and then slowed down with time, which is similar to the results of other studies (Ahmed et al., 2019; Mostafa et al., 2022).

In contrast, as the concentration of the activated carbon increases, the diameters of the bacteriostatic ring gradually decrease as shown in Figure 11. As the content of activated carbon increases, more adsorption sites exist which cause part of the silver particles are embedded in the fiber. Therefore, the number of silver particles exposed on the surface of the fiber is reduced which ultimately affects the antibacterial capacity.