La₂O₃ nanoparticle/polyacrylonitrile nanofibers for bacterial inactivation based on phosphate control

Abstract

La₂O₃ nanoparticle-doped PAN nanofiber mats were prepared by an electrospinning process. During the formation process, PAN nanofibers served as the matrix for immobilizing the La₂O₃ nanoparticles. Due to the great hydrophilicity of PAN nanofibers and the strong chemical affinity of phosphate/lanthanum bonds, the functional composite nanofiber mats exhibited high removal capacity for phosphate with a maximum adsorption capacity of 77.76 mg P per g (La), which guaranteed severe phosphate-deficient conditions. The effects of both the effective nutrient (phosphate) starvation and the close contact of E. coli with composites led to great antibacterial activity. The nutrient starvation (indirect antibacteria) was proved to be the primary mechanism. The intimate contact (direct antibacteria) was revealed as a supplementary role that only induced cell death without the existence of phosphate. The results demonstrated the feasibility of this antibacterial strategy through phosphate-starvation. In addition, the composite nanofiber mats implied a practical application by virtue of a simple fabrication and easy separation from the heterogeneous systems after using.

---

Introduction

As an essential element for living beings, phosphate has important implications for aquatic organisms in water.^1–3 It is widely accepted that limiting phosphate is a prerequisite for water quality improvement, especially in eutrophication treatment.^4–8 In addition, besides restraining blooms of excessive algae regrowth, phosphate shortage can also prohibit microbe growth effectively due to nutrient starvation.^2–9 However, this strategy of antibacteria is only effective under a severely low concentration of phosphate, which is unsatisfactory for traditional techniques for phosphate removal such as chemical precipitation, biological processes and adsorption.^10,11 The insufficient phosphate capture properties¹² and the leakage of phosphate residues¹ are the main bottlenecks of these methods, hindering the implementation of the antibacterial strategy. Because of its very strong binding affinity to phosphate (solubility product of lanthanum phosphate pK = 26.16),^13,14 lanthanum is utilized to control eutrophication in the lakes (Phoslock)^14–17 and to reduce excess phosphate in the human body (Fosrenol).^18,19 The results show remarkable effects in hampering the release of phosphate.¹ Furthermore, novel nano-lanthanum (La) species have been demonstrated with great adsorption capacity towards phosphate based on their appealing nanostructure. These phosphate adsorption improvements imply lanthanum-based materials are a bright prospect for phosphate scavenging.

Some prior work has successfully demonstrated the feasibility of the antibacteria strategy through phosphate-starvation based on La₂O₃ nanoparticles towards various microbe species (Gram-negative bacteria, Gram-positive bacteria and other organisms).² However, to the best of our knowledge, there are few studies on the antibacterial process and mechanism. As a metal oxide, the possibility for the existence of antibacterial properties by direct spatial contact between La₂O₃ nanoparticles and cells should not be neglected and it is meaningful to explore whether or not the interfacial interaction correlates with the antibacterial properties.

Although the nano-sized particles have prominent adsorption capacities due to their tiny size, they also face the separation problem in water treatment. Various lanthanum-based products, such as lanthanum-doped ordered mesoporous hollow silica spheres,²⁰ La(OH)₃-modified exfoliated vermiculites,²¹ hydroxyl–iron–lanthanum doped activated carbon fiber,²² and lanthanum hydroxide materials,²³ all reveal excellent capacity for phosphate removal, while the separation of nanoparticles from heterogeneous systems is still an inevitable problem, hindering their wide application.

Herein, we chose electrospinning, a facile and versatile route, to obtain La₂O₃ nanoparticle/polyacrylonitrile (PAN) nanofibers (LPNFs) containing both a functional nanostructured component and polymer matrix wrap, where PAN nanofibers were employed as the support of La₂O₃ nanoparticles due to the characteristics of prominent biocompatibility, spinnability and hydrophilicity.^24,25 The composite nanofiber mats presented great properties of phosphate removal. To assess the capability of antibacteria, the inhibition performance of bacterial growth through nutrient starvation and direct contact were analyzed. The antibacterial processes and mechanisms of LPNFs were discussed in detail from two perspectives, one was the process of antibacteria in the solution and the other one was the interaction between cells and nanoparticles on the nanofiber mats. Additionally, the influence of the initial bacteria concentration and the constituents of the nutrients were monitored to explore the influence of solution parameters on the antibacterial efficacy.

Materials and methods

Materials

N,N-Dimethylformamide (DMF) and nano lanthanum oxide (La₂O₃) were purchased from Aladdin Chemical Reagent Co., Ltd. (Shanghai, China). Polyacrylonitrile (PAN) (M_w = ∼80 000) was provided by Jilin Carbon Group, China. All of the used chemical reagents were analytical grade. E. coli (ATCC 25922) was incubated via a previous procedure reported in the literature.²⁶

Fabrication and characterization of the composite nanofibers

PAN (7 wt%) and La₂O₃ were dissolved in DMF, under vigorous stirring at 80 °C for 2 h. The mass ratio of PAN to La₂O₃ was 5 : 1. After being cooled down, the above mixtures were prepared for electrospinning. An electric potential of 15 kV was applied between the orifice and the ground at a distance of 20 cm, with a feed rate of 1.0 mL h⁻¹ by a syringe pump (Fisher Scientific, USA) under ambient conditions. For comparison, pure PAN nanofibers were also fabricated by electrospinning without addition of La₂O₃.

The surface chemical structures of the samples before and after adsorption were analyzed by Fourier-Transform Infrared (FT-IR) spectroscopy on a PerkinElmer Spectrum One B spectrometer with KBr used as the reference. X-Ray diffraction (XRD) was used to investigate the crystal structures of the samples using a Bruker D8 Advance diffractometer using Cu Kα radiation as the X-ray source. The surface morphologies of the products were observed by field emission scanning electron microscopy (FE-SEM, Helios Nanolab600i). High resolution transmission electron microscopy (HRTEM, FEI Tecnai G2 F30) was operated at a 300 kV accelerating voltage. The lanthanum contents in PAN/La₂O₃ nanofibers were observed using an inductively coupled plasma optical emission spectrometry (ICP-OES, Optima 5300 DV, PerkinElmer). The samples for ICP-OES were prepared by digesting 0.1 g of product with 6 mL of HNO₃, 2 mL of HClO₄ and 2 mL of HCl, then were followed by dilution with 2% HNO₃.

Phosphate adsorption experiments

A batch of tests were conducted to investigate the phosphate adsorption capability of the pure PAN nanofibers and our prepared samples. Potassium dihydrogen phosphate (KH₂PO₄) was chosen to be dissolved in DI water as the phosphate solution. All the experiments were carried out in 150 mL conical flasks with phosphate solution (50 mL) and adsorbents (3.0 g L⁻¹) added.

In equilibrium experiments, the initial concentration of the observed phosphate solutions ranged from 20 to 80 mg P L⁻¹. The conical flasks were shaken at 25 °C for 24 h. Langmuir and Freundlich equations were applied to describe the adsorption isotherm data by nonlinear regression forms.²¹

q_e = q_mK_LC_e/(1 + K_LC_e) (1)

q_e = K_FC_e^1/n (2)

where C_e (mg L⁻¹) is the concentration of phosphate in solution at equilibrium, q_e (mg g⁻¹) is the corresponding adsorption capacity, q_m (mg g⁻¹) and K_L (L mg⁻¹) are Langmuir constants related to adsorption capacity and energy or net enthalpy of adsorption, respectively, and K_F (mg g⁻¹) and n are the Freundlich constants.

In adsorption kinetic experiments, the initial phosphate concentrations were 20, 50, and 80 mg P per L. A 2 mL solution was taken out of the flask at each given time interval to detect the phosphate concentrations. Then pseudo-first-order and pseudo-second-order models, described by the following equations, were used to fit the experimental data.²¹

Pseudo-first-order equation:

ln(q_e − q_t) = ln q_e − k₁t (3)

Pseudo-second-order equation:

t/q_t = 1/(k₂q_e²) + t/q_e (4)

where q_t and q_e are the amount of phosphate adsorbed over a given period of time t (mg P per g) and at equilibrium (mg P per g), respectively; t is the adsorption time (min); k₁ (min⁻¹) and k₂ (g mg⁻¹ min⁻¹) are the adsorption rate constants of the pseudo-first-order adsorption and the pseudo-second-order adsorption, respectively.

In order to study the pH effect on phosphate removal the absorbents were added into a 50 mg P per L phosphate solution with an initial pH range from 2.0 to 11.0, which was adjusted by NaOH and/or HCl solutions. To investigate the effect of coexisting anions on adsorption, the adsorption capacities were evaluated by dissolving NaF, NaCl, Na₂CO₃ and Na₂SO₄ into a 50.0 mL phosphate solution. The concentrations of mixed solutions were 50.0 mg P per L and 0.01 M of F⁻, Cl⁻, CO₃²⁻ or SO₄²⁻.

An ammonium molybdate spectrophotometric method was used to detect the concentration of the phosphate solutions. The adsorption properties of the samples were characterized by UV-vis spectrophotometry (Shanghai Metash Instruments Co., Ltd., China). The pH of the solutions was analyzed using a Sartorius PB-10 (Sartorius, Germany). The concentrations of leakage were analyzed by ICP-OES.

The analysis of antibacterial mechanisms toward E. coli

The antibacterial mechanisms were investigated by observing the bacterial growth in the feedwater (25 mL) with LPNFs in the different periods of phosphate adsorption. The finial concentrations of E. coli were 10⁵ CFU mL⁻¹. The observed samples contained PAN nanofiber mats, pristine La₂O₃/PAN nanofiber (P-LPNF) mats, unsaturated La₂O₃/PAN nanofiber (U-LPNF) mats and saturated La₂O₃/PAN nanofiber (S-LPNF) mats. The S-LPNF mats were prepared by immersing LPNFs (1 g L⁻¹) into the phosphate solution (80 mg P per L) for 48 h, and then dried at 40 °C for 24 h. The PAN, P-LPNF, and S-LPNF samples with an area of 1 × 1 cm² were immersed in a solution with added glucose (0.1 wt%), NaCl (0.05 wt%), and HEPES buffer (10 mM, pH = 7.4). In order to observe the contamination of LPNF mats during the adsorption an unsaturated sample, 3 g L⁻¹ of LPNFs, was dispersed in the solution mixed with glucose (0.1 wt%), NaCl (0.05 wt%), potassium dihydrogen phosphate (10 mg P per L) and HEPES buffer (10 mM, pH = 7.4). The bacteria cultured in that solution without adding LPNFs was used as a control. After being incubated at 37 °C for 24 h, all the samples were analyzed by transmission electron microscopy (TEM) to study the morphology of the bacteria on the mats. TEM specimens were prepared by the following procedure.^27,28 The mats were rinsed with 0.01 M PBS several times and fixed with 2.5% glutaraldehyde in PB (0.2 M) at 4 °C for 12 h. Then the fixed samples achieved dehydration through a series of graded alcohol solutions (30, 50, 70, 80, 90, 95, and 100%, 10 minutes for each time). After being subjected to freeze-drying under vacuum at −50 °C for 12 h, the mats were observed using TEM. The number of surviving bacteria in each solution was evaluated by the colony forming count (CFU) method. The released concentrations of lanthanum and phosphorus were detected by using ICP-OES.

The influence of initial concentration of bacteria and the constituents of the nutrients on the antibacterial efficacy were detected. A dilution row (10⁴, 10³ and 10² CFU mL⁻¹) was obtained by adding different masses of bacteria in the media. The media also contained glucose (0.1 wt%), NaCl (0.05 wt%), potassium dihydrogen phosphate (10 mg L⁻¹), and HEPES buffer (10 mM, pH = 7.4). After adding 3 g L⁻¹ LPNFs, all the mixtures were rotated at 37 °C for 3 days. 50 μL of each sample was spread on the agar (LB) plates each day to obtain the cell counts after an incubation of 24 h.

Results and discussion

Characterization of LPNFs

In order to investigate the phosphate removal properties by LPNFs, the structural characteristics should be first analyzed. Fig. 1a and the inset show the typical electrospun SEM images of the nanofibers at different magnifications. As can be seen, the average diameter of the LPNFs was about 221 nm, with the diameters varying from 160 nm to 290 nm. Further observation was conducted by TEM to exhibit the structure of the composite nanofiber. The nanoparticles were found as loose dark block masses dispersed in the PAN nanofiber matrix in Fig. 1b, forming humps and even bigger beads as shown in Fig. 1c. Moreover, the Energy Dispersive X-ray (EDX) spectrum of the bead (Fig. 1d and Table S1†) indicated a high content of lanthanum, which is in accordance with the analysis of TEM images. The lanthanum content, detected by ICP-OES, was 14%. From the results it could be concluded that PAN nanofibers worked as the matrix to fix the functional part (La₂O₃ nanoparticles) through electrospinning.

Fig. 1 (a) SEM images of LPNFs at different magnifications (inset: diameter distribution of nanofibers). (b) TEM image of LPNFs (the selected part was a bead of La₂O₃). (c) SEM image of a La₂O₃ tubercle. (d) Energy Dispersive X-ray (EDX) spectrum of the selected part in (c).

The commercial lanthanum oxide nanoparticles (La₂O₃) and the LPNFs were analyzed by XRD to investigate the crystalline phases of the samples. As shown in Fig. S1,† the corresponding 2θ angles of the La₂O₃ and hexagonal (II-) La₂O₂CO₃ were located at 26.1, 39.5, 46.1° (JCPDS card no. 74-2430) and 22.3, 25.9, 27.7, 30.4, 33.8, 44.5, 47.5° (JCPDS card no. 25-0424), respectively. The results indicated the nanoparticles were composed of lanthanum compounds. These phenomena were due to the species transformation between La₂O₃ and La₂O₂CO₃ via hydrolysis or calcination when reacting with water and carbon dioxide, which happens in other commercial La₂O₃ as well.² The former study also reported that all the components of the lanthanum species possessed the ability to remove phosphate by forming LaPO₄.^2,29 Therefore, the commercial La₂O₃ nanoparticles were still referred to as La₂O₃ for better application here. The characteristic peaks of La₂O₃ such as (100), (102), (110) and the peaks of La₂O₂CO₃ such as (004), (101), (102), (103), (006), (110), (107) were observed for both PAN/La₂O₃ nanofibers and La₂O₃ nanoparticles, confirming the crystallinity was mainly based on the addition of lanthanum nanoparticles in the polymer.

Phosphate removal properties of the LPNFs

The phosphate adsorption properties of LPNFs were evaluated and are shown in Fig. 2. Compared with the phosphate removal of the pure PAN nanofibers, which was only 0.277 mg P per g under the same operating conditions with an initial concentration of 50 mg P per g, the adsorption capacities when using LPNFs demonstrated a great improvement through doping La₂O₃ nanoparticles in the samples. The results implied the La₂O₃ nanoparticles, as the main adsorption functional component, have an excellent ability to capture phosphate from solution. Fig. 2a shows the Langmuir and Freundlich isotherms with various initial concentrations from 20 mg P per L to 80 mg P per L at 25 °C. According to the correlation coefficient listed in Table S2,† both the two models fitted well to describe the phosphate adsorption isotherms. The R² of the Langmuir isotherm was slight higher than that of the Freundlich isotherm, indicating a monolayer adsorption on the homogeneous surface of LPNFs. Phosphate adsorption on other lanthanum doped adsorbents, such as lanthanum doped hydrochar,³⁰ hydroxyl–iron–lanthanum doped activated carbon fiber,²² and lanthanum-doped ordered mesoporous hollow silica²⁰ also fitted better to the Langmuir equation. Since the pure PAN did not contribute to the total adsorption capacities of phosphate and the content of La in the composite nanofibers accounted for 14 wt%, the final maximum adsorption capacity (q_m) of LPNFs was 77.76 mg P per g (La), demonstrating La₂O₃ nanoparticles possessed great phosphate removal properties. Moreover, different from other nanoparticles in the literature, the problem of retrieving them from solution is solved because the structure of LPNFs made the adsorbents easy to remove after the adsorption (Fig. 2b, inset).

Fig. 2 (a) The adsorption isotherms of the phosphate on the LPNFs (initial concentration = 20, 30, 40, 50, 60, 70, 80 mg P per L). (b) Adsorption kinetics of the phosphate on LPNFs (initial concentration = 20, 50, 80 mg P per L). The above experiments were carried out at T = 25 °C without pH adjustment.

As shown in Fig. 2b, the adsorption kinetics study using 3 g L⁻¹ of LPNFs as the adsorbent with three different initial concentrations (20 mg P per L, 50 mg P per L and 80 mg P per L) was carried out over 24 h to detect the whole time-dependent adsorption process. It could be seen that the adsorption amount increased by extending the contact time. The samples reached equilibrium after 6 h in a solution of 20 mg P per L, 12 h in a solution of 50 mg P per L and 24 h in a solution of 80 mg P per L. The initial phosphate concentration strongly effected the adsorption capacity, which increased from 43 mg P per g (La) in 20 mg P per L solution to 67 mg P per g (La) in 80 mg P per L solution. Additionally, the adsorption capacities were not affected by the increase of contact time after reaching equilibrium. The kinetic data analyzed by both the pseudo-first-order and pseudo-second-order models are shown in Table S3.† Comparing the correlation coefficients, the results fitted better with the pseudo-second-order model (R² > 0.97) compared to the pseudo-first-order model (R² > 0.93), suggesting a chemisorption process of adsorption. Moreover, the rate constants of the pseudo-second-order model (k₂) with initial concentrations of 20 mg P per L, 50 mg P per L and 80 mg P per L were 0.0006, 0.0003 and 0.0003 g mg⁻¹ min⁻¹ respectively, which illustrated that LPNFs were more suitable to be used in a solution of lower concentration for a faster removal rate.²¹ Because the solution with a lower concentration (less than 1 mg P per L) was more close to the actual water body, the results implied a high speed phosphate removal potential of LPNFs in practical application.

The morphologies and chemical compositions of LPNFs after adsorption were studied by SEM and EDX, shown in Fig. 3. Compared with the pristine composite nanofibers (Fig. 1a), the adsorbents still retained a fiber-like structure after adsorption, indicating the stability of the electrospun nanofibers. Apart from the unchanged structure, there were many blocks or particles formed and stuck on the pristine fibers (Fig. 3a and b). A further elemental analysis of the selected part showed obvious peaks of O, La, and P in Fig. 3c, which demonstrated the formed blocks were composed of these elements.

Fig. 3 (a and b) SEM image of LPNFs after adsorption. (c) Energy dispersive X-ray (EDX) spectrum of the selected part in (b).

In order to show the composition transformation of the LPNFs, Fourier-transform infrared (FTIR) spectra before and after adsorption were studied, shown in Fig. 4a. The obvious increase of the peak at 1040 cm⁻¹ corresponded to the phosphate P–O stretching and the peaks appearing at 616 and 540 cm⁻¹ were related to the presence of O=P–O bending and O–P–O bending modes, respectively.³¹ However, even though the interactions between phosphate and lanthanum clearly existed in the FTIR spectrum, there was no crystalline lanthanum orthophosphate observed in the XRD pattern (Fig. 4b), which suggested the binds were in almost amorphous forms.³⁰

Fig. 4 The (a) FTIR and (b) XRD spectra of the LPNFs before and after adsorption of phosphate.

The effect of pH on the adsorption of phosphate from 2–11 was investigated in Fig. S2a.† The experimental data showed the adsorption capacities decreased steadily with the increase of pH. The prominent point of the adsorption appeared at pH ∼ 3 (≥110 mg P per g (La)). The values of phosphate removal kept above 55 mg P per g (La) throughout the pH range from 2.0 to 7.0. The results illustrated a wide application scope of pH values. However, as the values increased from 8 to 11, the amount of adsorption sharply decreased, which indicated alkaline conditions were not beneficial to phosphate removal. Based on the analysis of a pH drift method,¹¹ the isoelectric point (pH_pzc) of LPNFs was 8. The surface of the nanofibers was positively charged when the pH was lower than 8, while when the pH > 8, the surface was negatively charged.²⁰ At pH ∼ 2–8, the electrostatic attraction facilitated the interaction between the positively charged surface and H₂PO₄⁻ or HPO₄²⁻. Simultaneously, the phosphate proton dissociation promoted the high phosphate adsorption capacities by ligand-exchange on the surface of the hydroxylated lanthanum oxides.^{20,23,30,32,33} When pH ∼ 8–11, the surface of adsorbents was negatively charged. Due to the high concentration of OH⁻ in the solution, the influence of ligand-exchange became weak and was replaced by the effect of electrostatic repulsion, resulting in the sharp reduction of phosphate removal.²²

Fig. S2b† showed the effect of competitive coexisting anions, such as F⁻, Cl⁻, SO₄²⁻ and CO₃²⁻, on the capacity of phosphate adsorption. The concentration of coexisting anions in each solution was 0.01 M. Comparing the adsorption amounts of all the samples, the results revealed no significant difference of phosphate removal, indicating a high selective phosphate adsorption capacity of the adsorbent.

Antibacterial activities of LPNFs

Because of the excellent affinity of phosphate and lanthanum, the adsorbents revealed no desorption of phosphate from the nanofibers. Thus a stable phosphorous restriction could be formed to realize indirect antibacteria through controlling essential elements in the solution. Escherichia coli (E. coli) was chosen as the model bacteria to investigate the antibacterial effect of LPNFs. LPNFs in the different periods of phosphate adsorption, containing pristine La₂O₃/PAN nanofibers (P-LPNFs), unsaturated La₂O₃/PAN nanofibers (U-LPNFs), and saturated La₂O₃/PAN nanofibers (S-LPNFs), were dispersed in the feedwater (10⁵ CFU mL⁻¹) to detect the inactivation mechanisms of LPNFs. After incubation of 24 h, the concentration of phosphorus, detected by ICP-OES, showed all the samples were lass than 20 μg P per L (Table 1). These results denoted that the occurrence of phosphorous starvation was formed in each sample.¹⁰ Except for the three samples, the PAN nanofibers used as biocompatible materials were also selected to detect in the same way. The cell numbers of PAN nanofibers exhibited a reduction of 10 570 CFU mL⁻¹ (Fig. 5). Compared with the control sample, attaining a slight growth from 22 900 CFU mL⁻¹ to 26 800 CFU mL⁻¹ in the solution with sufficient nutrients, the PAN nanofibers demonstrated a remarkable bacterial inactivation in phosphate-free solution. The comparison of PAN nanofibers and the three LPNFs indicated that except for P-LPNFs, which deactivated 15 500 CFU mL⁻¹, the unsaturated and saturated samples showed nearly the same antibacterial effects as the PAN nanofibers. These results are in agreement with the evaluation of released lanthanum (Table 1). The concentration of La (46 μg L⁻¹) in the solution of P-LPNFs suggested a direct antibacterial capability based on the analysis from previous research.¹⁰ Contrasting the amount of cell death between PAN nanofibers and P-LPNFs, more cells were killed with the indirect antibacterial action, implying nutrient starvation was the primary mechanism of antibacteria. In addition, to explore the direct contact between cells and the anchored La₂O₃ nanoparticles in PAN, SEM was carried out and the images of the bacterial survival on the membranes are shown in Fig. 6. Compared with the PAN nanofibers and U-LPNFs, shown in Fig. 6a and c, with no obvious bacterial survival, P-LPNFs mats (Fig. 6b) displayed a distinctively severe antibacterial efficacy. Conversely, there were scattered bacteria observed on the S-LPNFs mats (Fig. 6d and the magnified image inset) without damage of cell walls, suggesting the biological contamination of cells. From the various appearances of bacteria on different surfaces of membrane, the process of antibacteria probably occurred on three portions of the mats (Fig. 7). Step (1) a favorable adsorption occurred between negatively charged bacteria (E. coli)^34,35 and the positively charged surface of nanofibers (Fig. 2a). The morphological integrity deficiency of cells on the nanofiber mat surface (as illustrated by the arrows for some representative cells in Fig. 6b) was probably due to the toxicity of La₂O₃ nanoparticles, which caused the disruption of cellular membranes through direct contact between La₂O₃ nanoparticles and cells.³⁶ Step (2) a competitive adsorption generated between phosphate and bacteria on the La₂O₃ doped membrane. The high solubility constant of lanthanum phosphate,² both preventing lanthanum from releasing into the solution and hindering phosphate being assimilated by bacteria, caused neither death nor survival on the surface of U-LPNFs. Step (3) the oversaturated phosphate induced the growth of cells on the saturated samples. The ICP-OES detection showed the concentration of 17 μg P per L in solution (Table 1). The result also demonstrated there was available phosphorus for the growth of bacteria. Hence, from the exploration of the three processes, it could be concluded that the LPNFs would not be favorable for bacteria to survive in the presence of unsaturated LaPO₄ even though the accumulated phosphate was the essential element for their growth.

Table 1 Comparison of the residue of phosphorus and lanthanum in the solution

	P (μg L^−1)	La (μg L^−1)
a Below the MDLs (1 μg La per L).
PAN nanofibers	4	ND^a
P-LPNFs	3	46
U-LPNFs	15	1
S-LPNFs	17	ND
Control	9800	ND

---

Fig. 5 Cell viabilities in solutions of a PAN nanofiber membrane, pristine LPNF membrane, phosphate unsaturated LPNF membrane and saturated LPNF membrane after incubation with mixed feedwater for 24 h.

Fig. 6 SEM images of (a) PAN nanofiber membrane, (b) pristine LPNF membrane and (c) phosphate unsaturated LPNF membrane and (d) phosphate saturated LPNF membrane after 24 h incubation in a solution of E. coli (10⁵ CFU mL⁻¹).

Fig. 7 Illustration of the antibacterial activity processes of LPNFs.

The influence of initial concentration of bacteria and the constituents of the nutrients on the antibacterial efficacy are studied in Fig. 8. The first analysis was on the constituents of the nutrients in the solution. The death rates of bacteria in solution C and D were not the same even though their initial cell numbers were close. The bacteria were killed within 2 days in solution D with phosphorus nutrient only, while another containing glucose (0.1 wt%), NaCl (0.05 wt%), and HEPES buffer (10 mM, pH = 7.4) was more than 3 days. Therefore, the nutrients besides phosphate attributed to the retardation of cell death. The second aspect was the initial concentration of bacteria. The amount of bacterial colonies in solutions B, C and D reduced sharply from 260, 6980, and 7020 CFU mL⁻¹ to 12, 250, and 314 CFU mL⁻¹ for the first day while slowing down in the last two days. The remarkable decreases after the first day suggested high competition for the nutrients between bacteria when the counts of cells were larger than the nutrient solution could sustain. Additionally, although the initial cell amount in solution B (260 CFU mL⁻¹) was nearly the same as in solution D after the incubation on the first day (250 CFU mL⁻¹), the death rate of solution C was distinctly slower, which might be due to the larger amount of death in the solution, leaking nutrients for re-assimilation of living cells.³⁷

Fig. 8 Cell viability after incubation with different solutions. 25 mL of bacterial suspensions (10², 10³ and 10⁴ CFU mL⁻¹) with added glucose (0.1 wt%), NaCl (0.05 wt%), potassium dihydrogen phosphate (10 mg L⁻¹) and HEPES buffer (10 mM, pH = 7.4) corresponded to solution A, B, and C. Solution D was 25 mL of bacterial suspension (10⁴ CFU mL⁻¹) with added potassium dihydrogen phosphate (10 mg L⁻¹). All the solutions were exposed to LPNFs (3 g L⁻¹) during the experiments.

Conclusion

Here, La₂O₃ nanoparticle/polyacrylonitrile (PAN) nanofibers were synthesized by an electrospinning process. The La₂O₃ nanoparticles were wrapped in the polymer matrix as the functional component to achieve phosphate removal based on the tight affinity of lanthanum and phosphate, showing a great adsorption capacity of 77.76 mg P per g (La). The antibacterial mechanism of LPNFs combined both direct and indirect mechanisms in the solution, and the nutrient starvation (indirect mechanism) was proved to be the primary mechanism. Additionally, considering their simple fabrication and easy separation from heterogeneous systems, the LPNFs have remarkable prospects for solving bacterial contamination through phosphate control.

Acknowledgements

The authors gratefully acknowledge National Natural Science Foundation of China (Grant no. 51678181, 51573034), State Key Laboratory of Urban Water Resource and Environment in HIT of China (2016DX02), Fundamental Research Funds for the Central Universities of China (HIT.BRETIII.201417), Postdoctoral Science Foundation of Heilongjiang Prov. (LBH-TZ0606), National Water Pollution Control and Treatment Science and Technology Major Project of China (2012ZX07403004, 2012ZX07408001).

References

1. M. Lürling and F. V. Oosterhout, Controlling eutrophication by combined bloom precipitation and sediment phosphorus inactivation, Water Res., 2013, 47, 6527–6537.
2. L. C. Gerber, N. Moser, N. A. Luechinger, W. J. Stark and R. N. Grass, Phosphate starvation as an antimicrobial strategy: the controllable toxicity of lanthanum oxide nanoparticles, Chem. Commun., 2012, 48, 3869–3871.
3. M. A. H. Johir, T. T. Nguyen, K. Mahatheva, M. Pradhan, H. H. Ngo, W. Guo and S. Vigneswaran, Removal of phosphorus by a high rate membrane adsorption hybrid system, Bioresour. Technol., 2016, 201, 365–369.
4. M. Lürling and Y. Tolman, Effects of lanthanum and lanthanum-modified clay on growth, survival and reproduction of daphnia magna, Water Res., 2010, 44, 309–319.
5. J. S. Vrouwenvelder, F. Beyer, K. Dahmani, N. Hasan, G. Galjaard, J. C. Kruithof and M. C. M. Van Loosdrecht, Phosphate limitation to control biofouling, Water Res., 2010, 44, 3454–3466.
6. F. Xie, F. Wu, G. Liu, Y. Mu, C. Feng, H. Wang and J. P. Giesy, Removal of phosphate from eutrophic lakes through adsorption by in situ formation of magnesium hydroxide from diatomite, Environ. Sci. Technol., 2014, 48, 582–590.
7. H. K. Kwon, S. J. Oh and H. Yang, Growth and uptake kinetics of nitrate and phosphate by benthic microalgae for phytoremediation of eutrophic coastal sediments, Bioresour. Technol., 2013, 129, 387–395.
8. A. Ruiz-Martínez, J. Serralta, I. Romero, A. Seco and J. Ferrer, Effect of intracellular P content on phosphate removal in Scenedesmus sp. experimental study and kinetic expression, Bioresour. Technol., 2015, 175, 325–332.
9. A. C. C. Rotzetter, C. R. Kellenberger, C. M. Schumacher, C. Mora, R. N. Grass, M. Loepfe, N. A. Luechinger and W. J. Stark, Combining phosphate and bacteria removal on chemically active filter membranes allows prolonged storage of drinking water, Adv. Mater., 2013, 25, 6057–6063.
10. J. He, W. Wang, F. Sun, W. Shi, D. Qi, K. Wang, R. Shi, F. Cui, C. Wang and X. Chen, Highly efficient phosphate scavenger based on well-dispersed La(OH)₃ nanorods in polyacrylonitrile nanofibers for nutrient-starvation antibacteria, ACS Nano, 2015, 9, 9292–9302.
11. L. Zhang, Q. Zhou, J. Liu, N. Chang, L. Wan and J. Chen, Phosphate adsorption on lanthanum hydroxide-doped activated carbon fiber, Chem. Eng. J., 2012, 185–186, 160–167.
12. Z. Wang, D. Shen, F. Shen and T. Li, Phosphate adsorption on lanthanum loaded biochar, Chemosphere, 2016, 150, 1–7.
13. M. Lürling, G. Waajen and F. van Oosterhout, Humic substances interfere with phosphate removal by lanthanum modified clay in controlling eutrophication, Water Res., 2014, 54, 78–88.
14. V. Kuroki, G. E. Bosco, P. S. Fadini, A. A. Mozeto, A. R. Cestari and W. A. Carvalho, Use of a La(III)-modified bentonite for effective phosphate removal from aqueous media, J. Hazard. Mater., 2014, 274, 124–131.
15. D. Copetti, K. Finsterle, L. Marziali, F. Stefani, G. Tartari, G. Douglas, K. Reitzel, B. M. Spears, I. J. Winfield, G. Crosa, P. D’Haese, S. Yasseri and M. Lürling, Eutrophication management in surface waters using lanthanum modified bentonite: a review, Water Res., 2016, 97, 162–174.
16. Z. Wang, Y. Fan, Y. Li, F. Qu, D. Wu and H. Kong, Synthesis of zeolite/hydrous lanthanum oxide composite from coal fly ash for efficient phosphate removal from lake water, Microporous Mesoporous Mater., 2016, 222, 226–234.
17. L. Dithmer, U. G. Nielsen, M. Lürling, B. M. Spears, S. Yasseri, D. Lundberg, A. Moore, N. D. Jensen and K. Reitzel, Responses in sediment phosphorus and lanthanum concentrations and composition across 10 lakes following applications of lanthanum modified bentonite, Water Res., 2016, 97, 101–110.
18. F. Chen, Z. Gu, J. Yi, X. Xu, P. Kulkarni and Q. Tang, The evaluation of lanthanum trapped prussian blue as a phosphate binding agent with reduced bone uptake, New J. Chem., 2016, 40, 2644–2648.
19. M. R. Chan, Pilot study of the effect of lanthanum carbonate (Fosrenol®) in patients with calciphylaxis: a wisconsin network for health research (WiNHR) study, J. Nephrol. Therapeut., 2014, 4, 1–14.
20. W. Huang, Y. Zhu, J. Tang, X. Yu, X. Wang, D. Li and Y. Zhang, Lanthanum-doped ordered mesoporous hollow silica spheres as novel adsorbents for efficient phosphate removal, J. Mater. Chem. A, 2014, 2, 8839–8849.
21. W. Huang, D. Li, Z. Liu, Q. Tao, Y. Zhu, J. Yang and Y. Zhang, Kinetics, isotherm, thermodynamic, and adsorption mechanism studies of La(OH)₃-modified exfoliated vermiculites as highly efficient phosphate adsorbents, Chem. Eng. J., 2014, 236, 191–201.
22. J. Liu, Q. Zhou, J. Chen, L. Zhang and N. Chang, Phosphate adsorption on hydroxyl–iron–lanthanum doped activated carbon fiber, Chem. Eng. J., 2013, 215–216, 859–867.
23. J. Xie, Z. Wang, S. Lu, D. Wu, Z. Zhang and H. Kong, Removal and recovery of phosphate from water by lanthanum hydroxide materials, Chem. Eng. J., 2014, 254, 163–170.
24. W. Wang, X. Lu, Z. Li, X. Li, X. Xu, J. Lei, C. Wang, R. H. Baughman and S. Fang, Weak-acceptor-polyacrylonitrile/donor-polyaniline core–shell nanofibers: a novel 1D polymeric heterojunction with high photoconductive properties, Org. Electron., 2012, 13, 2319–2325.
25. W. Wang, Z. Li, X. Xu, B. Dong, H. Zhang, Z. Wang, C. Wang, R. H. Baughman and S. Fang, Au-Doped Polyacrylonitri–Polyaniline Core–Shell Electrospun Nanofibers Having High Field-Effect Mobilities, Small, 2011, 5, 597–600.
26. J. Yang, P. Yuan, H. Chen, J. Zou, Z. Yuan and C. Yu, Rationally designed functional macroporous materials as new adsorbents for efficient phosphorus removal, J. Mater. Chem., 2012, 22, 9983–9990.
27. S. Chen, Y. Guo, S. Chen, H. Yu, Z. Ge, X. Zhang, P. Zhang and J. Tang, Facile preparation and synergistic antibacterial effect of three-component Cu/TiO₂/CS nanoparticles, J. Mater. Chem., 2012, 22, 9092.
28. P. Gunawan, C. Guan, X. Song, Q. Zhang, S. S. J. Leong, C. Tang, Y. Chen, M. B. Chan-Park, M. W. Chang, K. Wang and R. Xu, Hollow fiber membrane decorated with Ag/MWNTs: toward effective water disinfection and biofouling control, ACS Nano, 2011, 5, 10033–10040.
29. P. Chen and T. Mnh, Synthesis and characterization of lanthanum phosphate sol for fibre coating, J. Mater. Sci., 1997, 32, 3863–3867.
30. L. Dai, B. Wu, F. Tan, M. He, W. Wang, H. Qin, X. Tang, Q. Zhu, K. Pan and Q. Hu, Engineered hydrochar composites for phosphorus removal/recovery: lanthanum doped hydrochar prepared by hydrothermal carbonization of lanthanum pretreated rice straw, Bioresour. Technol., 2014, 161, 327–332.
31. K. Rajesh, P. Shajesh, O. Seidel, P. Mukundan and K. G. K. Warrier, A facile sol–gel strategy for the synthesis of rod-shaped nanocrystalline high-surface-area lanthanum phosphate powders and nanocoatings, Adv. Funct. Mater., 2007, 17, 1682–1690.
32. L. Zhang, L. Wan, N. Chang, J. Liu, C. Duan, Q. Zhou, X. Li and X. Wang, Removal of phosphate from water by activated carbon fiber loaded with lanthanum oxide, J. Hazard. Mater., 2011, 190, 848–855.
33. J. Xie, Y. Lin, C. Li, D. Wu and H. Kong, Removal and recovery of phosphate from water by activated aluminum oxide and lanthanum oxide, Powder Technol., 2015, 269, 351–357.
34. I. Sondi and B. Salopek-Sondi, Silver nanoparticles as antimicrobial agent: a case study on E. coli as a model for Gram-negative bacteria, J. Colloid Interface Sci., 2004, 275, 177–182.
35. C. Xing, J. Guan, Z. Chen, Y. Zhu, B. Zhang, Y. Li and J. Li, Novel multifunctional nanofibers based on thermoplastic polyurethane and ionic liquid: towards antibacterial, anti-electrostatic and hydrophilic nonwovens by electrospinning, Nanotechnology, 2015, 26, 105704.
36. C. Kaweeteerawat, A. Ivask, R. Liu, H. Zhang, C. H. Chang, C. Low-Kam, H. Fischer, Z. Ji, S. Pokhrel, Y. Cohen, D. Telesca, J. Zink, L. Mädler, P. A. Holden, A. Nel and H. Godwin, Toxicity of metal oxide nanoparticles in Escherichia coli correlates with conduction band and hydration energies, Environ. Sci. Technol., 2015, 49, 1105–1112.
37. S. J. Blazewicz, R. L. Barnard, R. A. Daly and M. K. Firestone, Evaluating rRNA as an indicator of microbial activity in environmental communities: limitations and uses, ISME J., 2013, 7, 2061–2068.

---

Footnote

† Electronic supplementary information (ESI) available: See DOI: 10.1039/c6ra22374e

---