Investigation into the impact of anionic substitution on modulating the optical and catalytic properties of bismuth ferrite nanoparticles

Abstract

Bismuth ferrite (BFO) nanoparticles have emerged as a non-toxic catalyst with remarkable potential for the photodegradation of various environmental pollutants. A notable departure from conventional approaches, where cations are added as dopant, this study achieved enhanced catalytic performance through anion substitution. Specifically, replacing oxygen atoms with nitrogen introduces spin-polarized defect states within the BFO’s energy gap, resulting in a notable reduction in the energy band gap. Nitrogen doping of bismuth ferrite yields a novel material with exceptional capabilities for the photodegradation of methylene blue dye and the reduction of 4-nitrophenol. Comprehensive characterization, including X-ray diffraction, Fourier-transform infrared spectroscopy, energy-dispersive X-ray spectroscopy, and X-ray photoelectron spectroscopy, has unequivocally confirmed the successful incorporation of nitrogen into the BFO nanoparticle lattice. Interestingly, field emission scanning electron microscopy analysis revealed no significant alteration in nanoparticle size after nitrogen doping. Meanwhile, UV-diffuse reflectance spectroscopy unveiled a distinct decrease in the energy gap upon nitrogen incorporation. The observed improvements in catalytic activities can be attributed to nitrogen ions, introduced as substitutes, effectively occupying the oxygen defects within the sample, thereby diminishing recombination centers for photogenerated charge carriers and decreasing recombination rates. Additionally, adsorption kinetics studies underscore the efficacy of the catalyst surface in adsorbing methylene blue and/or 4-nitrophenol, conforming to the Ho pseudo-second-order model. This study not only highlights the exciting potential of nitrogen-doped bismuth ferrite nanoparticles in environmental remediation but also sheds light on the intricate interplay between anion substitution, band structure modification, and catalytic performance enhancement.

Graphical Abstract

Introduction

Multiferroic materials have garnered significant attention, not just for their simultaneous display of ferromagnetism and ferroelectricity but also due to the remarkable magnetoelectric coupling they offer. In this intriguing realm of materials, where compounds like BiFeO₃ (BFO), BiMnO₃, and YMnO₃ have been thoroughly explored, BFO emerges as a standout candidate with exceptional promise for diverse practical applications due to its unique blend of room-temperature ferroelectric and antiferromagnetic orders, a narrow energy band gap, robust chemical stability, and cost-effectiveness (Bismibanu et al. 2018; EL-Bassuony AAH, Hafez RS, Matter NMS, Abdelsalam HK 2024). It has potential applications ranging from spintronics, memory devices, and data storage to ferroelectric random-access memory devices, sensors, digital recording, microwave and satellite communications, photovoltaics, and even the eco-friendly photodegradation of organic pollutants (Wang et al. 2020; Matinise et al. 2023). Although the magnetoelectric coupling in BFO may be relatively weak, it showcases substantially intensified multiferroic properties (Bismibanu et al. 2018; Li et al. 2019; Mubarak et al. 2014).

BFO is a ferroelectric and an antiferromagnetic material with a Curie temperature T_c of 1103 K and a Neel temperature T_N of 643 K (Mubarak et al. 2014). The structural arrangement of BFO is rhombohedrally distorted perovskite, adhering to the R3c space group, with lattice parameters of a = 5.58 Å and c = 13.9 Å (Godara et al. 2014). The ferroelectric behavior in BFO is primarily attributed to the presence of Bi³⁺ ions, while its G-type antiferromagnetic characteristics are associated with Fe³⁺ ions. This antiferromagnetic spin structure exhibits a modulation period of 620 Å, resulting in a spiral modulated spin structure. At the nanoscale, BFO exhibits intriguing magnetic, electrical, and optical properties due to size effects that distinguish it from its bulk counterparts (Mubarak et al. 2014).

Numerous investigations have highlighted the potential of BFO as a formidable material for the photodegradation of organic dyes (Wang et al. 2020; Pattnaik et al. 2018; Siddique et al. 2018; Volnistem et al. 2018). The narrow energy gap of BFO, typically ranging from 2.1 to 2.7 eV, enables its activation in the presence of visible light and leads to the generation of electron–hole pairs. Additionally, the inherent ferroelectric behavior of BFO, characterized by spontaneous polarization, induces band bending, directing electron–hole pairs in opposite directions (Pattnaik et al. 2018; Siddique et al. 2018; Volnistem et al. 2018). These photo-generated holes play a pivotal role in oxidizing various organic pollutants, effectively transforming BFO into a photocatalyst under visible light conditions (Bagvand et al. 2018; Sarraf et al. 2024). Nevertheless, the rapid recombination of these photogenerated electron–hole pairs limits the catalytic efficacy of BFO in practical applications (Di et al. 2017).

Various strategies have been explored thus far to enhance the effective separation of electron–hole pairs and augment the catalytic performance of BFO, including modifications to its structural parameters, such as particle size and surface morphology (Gao et al. 2007; Wang et al. 2011; Bai et al. 2016), elemental doping with both metals and non-metals (Maleki 2018; Hu et al. 2017; Reddy et al. 2018), the creation of hetero and homo junctions with materials possessing low band gaps (Zhang et al. 2017; Humayun et al. 2016; Guo et al. 2011; Datta et al. 2017), and the introduction of noble metal dopants (Wang et al. 2020; Niu et al. 2015). Doping BFO with rare earth and transition metals significantly enhances its catalytic performance (Arti et al. 2023). However, the leaching of metals and potential toxicity concerns impose limitations on the utilization of metal-doped BFO in applications related to water purification or drinking water treatment (Jia et al. 2018; Pelaez et al. 2009).

Accordingly, in our current study, we have introduced nitrogen ions to explore its impact on the catalytic behavior of BFO nanoparticles. The adoption of nitrogen doping has been deemed productive and advantageous due to its atomic size, which closely approximates that of oxygen within the ABO₃ structure (Jia et al. 2018; Pelaez et al. 2009; Ansari et al. 2016). The incorporation of nitrogen brings about alterations in the crystal structure and/or suppresses the recombination rate of photogenerated electron–hole pairs, resulting in a notable enhancement in catalytic efficiency compared to pure BFO nanoparticles (Pelaez et al. 2009). The mode of nitrogen doping can be either substitutional, interstitial, or both. Substitutional doping replaces oxygen ions, and the surface of the nanoparticles is modified by nitrogen bonding via interactive forces such as dipole–dipole interaction, electrostatic interaction, van der Waals interaction, or London forces. The introduction of nitrogen into the interstitial position, on the other hand, alters the lattice parameters or crystal structure of the nanoparticles. Notably, substitutional doping reduces the energy band gap to a relatively lesser extent than interstitial doping (Jia et al. 2018; Pelaez et al. 2009; Ansari et al. 2016; Dunnill and Parkin 2011). While doping at cation sites has been extensively reported, there is notably limited research on doping at oxygen sites. Recently, Jia et al. synthesized nitrogen-doped bismuth ferrite (N-BFO) using melamine as the nitrogen precursor and found enhanced degradation of bisphenol A under visible light (Jia et al. 2018). However, no studies have yet addressed the photodegradation of organic dyes in the presence of N-BFO. Additionally, there has been no concerted research in understanding the role of nitrogen as an anionic substituent and its capacity for enhancing the catalytic activity in BFO. The main purpose of this work was to employ an economical and energy-efficient auto-combustion method to synthesize stoichiometrically pure BFO nanoparticles at low temperatures. Subsequently, we aimed to introduce nitrogen into these BFO nanoparticles (xN-BFO) and investigate their impact on catalytic activity. The role of nitrogen in suppressing oxygen vacancies of BFO is systematically investigated by X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), and UV–Vis diffuse reflectance spectroscopy (DRS) techniques. Specifically, we evaluated their effectiveness in the photodegradation of methylene blue (MB). Additionally, we studied their potential in the reduction of 4-nitrophenol (4-NP), a toxic and environmentally significant organic pollutant, into aminophenol (4-AP). Our findings revealed that nitrogen-doped BFO exhibited enhanced capabilities for both reduction and degradation compared to pure BFO.

Experimental

Materials and methods

Bismuth nitrate, Bi (NO₃)₃.6H₂O (Avra, Hyderbad), iron nitrate, Fe (NO₃)₃.9H₂O (Loba Chemie, Mumbai), ascorbic acid (Fisher Scientific, Mumbai), nitric acid, ammonium chloride (Nice Chemicals-cochin), methylene blue, 4-nitrophenol, and methanol (Molychem, Mumbai) were the chemicals used for the synthesis of pure and nitrogen-doped BFO (xN-BFO, x = 0.01,0.25,1) and deionized water was used throughout the experiment.

Synthesis of bismuth ferrite nanoparticles

The auto-combustion method was employed to synthesize nano-sized BFO particles at low temperature, as outlined in Fig. 1. Bismuth nitrate and iron nitrate were used as Bi and Fe precursors, with ascorbic acid as the chelating agent. Equimolar solutions of the nitrates were prepared, stirred, and combined with ascorbic acid in a 2:1 molar ratio. The mixture was heated until auto-ignition occurred, producing a brownish powder, which was then calcined at 550 °C for 1 h to yield nano-sized BFO.

Fig. 1

Synthesis of BFO using auto combustion method

Nitrogen doping

In this process, ammonium chloride, dissolved in 30 mL of methanol, served as the nitrogen source, with methanol as the solvent. The solution was sonicated for 15 min before adding 1 g of BFO nanoparticles, followed by another 15 min of sonication. The mixture was then heated at 60 °C for 24 h, and the resulting product was ground into a fine powder. Different levels of nitrogen-doped BFO nanoparticles were produced by varying the amount of ammonium chloride, denoted as xN-BFO (x = 0, 0.01 g, 0.25 g, 1 g). The synthesis procedure is summarized in Fig. 2.

Fig. 2

Synthesis of xN-BFO nanoparticles

Structural characterizations

The pristine and nitrogen-doped BFO nanoparticles underwent comprehensive characterization through a combination of microscopic and spectroscopic techniques. The crystal structure and lattice parameters were determined using X-ray diffraction (XRD) analysis conducted on a Rigaku Miniflex 300/600 powder X-ray diffractometer. Cu Kα radiation with a wavelength of 1.5407 Å, operating at 40 kV and 15 mA, was utilized. The scan covered a 2θ range of 3–80° at a rate of 2°/min. FTIR spectroscopy was employed (Shimadzu) to confirm the presence of different chemical bonds within the synthesized nanoparticles. The structural features and morphologies of the nanoparticles were examined using field emission scanning electron microscopy (FESEM) (Carl ZEISS 03–81) at a magnification of 90.00 KX, a resolution of 200 nm, and an EHT of 5 kV. To determine the stoichiometric ratio, energy-dispersive X-ray spectrometry (EDS) (Oxford Instruments) was employed. Additionally, UV–Vis (DRS) were obtained at room temperature using a UV–Vis spectrophotometer (Holmarc UV–Vis-NIR spectrophotometer, model; HO-SPA-1990P) with a wavelength range of 350–800 nm, using BaSO₄ as a reference. XPS analysis, conducted with a PHI 5000 Versa Probe II (ULVAC-PHI Inc., USA) equipped with a micro-focused (100 μm, 15 kV) monochromatic Al-K X-Ray source (hυ = 1486.6 eV), was utilized to confirm the elemental composition. Both survey and high-resolution spectra were recorded, with survey scans utilizing a 50W X-ray source and a pass energy of 187.85 eV. High-resolution spectra for major elements were acquired with a pass energy of 46.95 eV. Data analysis was performed using PHI’s Multipak software, and the binding energy was referenced to the C1s peak at 284.8 eV. Furthermore, UV–Vis spectrophotometry (Shimadzu) was employed to collect absorption spectra for the degradation of MB dye and the reduction of 4-NP.

Photocatalytic degradation of methylene blue dye

To assess the degradation behavior, either 25 mg of BFO or xN-BFO was mixed with a 100 mL solution containing 2 ppm of MB dye. Subsequently, the mixture was subjected to 45 min of magnetic stirring in the dark to establish adsorption–desorption equilibrium. Following this, the solution was exposed to sunlight, specifically at 11 a.m. on January 18, 2022, at Mangalore University’s campus in Karnataka, India, located at coordinates 12° 48′ 56.72″ N and 74° 55′ 26.67″ E. Absorbance measurements were taken every 15 min to monitor the degradation process. The degradation or reduction percentage of both pure BFO and xN-BFO was evaluated using the following equation:

$$\text{i}.\text{e}.,\text{ Degradation}/\text{Reduction }(\%) = \frac{{A}_{0}-{A}_{t}}{{A}_{0}} \times 100$$

(1)

where A₀ represents the initial absorbance at t = 0, and A_t denotes the absorbance at any time interval t during the degradation process.

Catalytic reduction of nitrophenol

The catalytic activity of both BFO and xN-BFO in the reduction of 4-NP was investigated in the presence of sodium borohydride (NaBH₄) under room temperature. In a standard reaction, 10 mg of either BFO or xN-BFO was introduced into a 25 mL aqueous solution containing a mixture of 1 mL of 4-NP (2 mM) and 1 mL of NaBH₄ (0.25 M). The solution was thoroughly stirred, and at regular 2-min intervals, samples were extracted to monitor the absorption behavior using a UV–Vis spectrophotometer. The reduction efficiency of the samples was quantified using Eq. 1.

Adsorption kinetics

Adsorption plays a crucial role in the catalytic activity of nanoparticles, as this activity is contingent on the adsorption of organic pollutants onto the catalyst surface (Kodoth and Badalamoole 2017). To gain a deeper insight into the adsorption mechanisms, kinetic studies were conducted. Specifically, Lagergren’s pseudo-first-order model (Langmuir 1918) and Ho’s pseudo-second-order model (Ho and McKay 1999) were employed to analyze the adsorption kinetics during the photodegradation of MB and reduction of NP in the presence of both BFO and xN-BFO.

The Lagergren pseudo-first-order model operates under the premise that the rate at which adsorption changes with time is directly proportional to the disparity between the saturation concentration and the amount of adsorption taking places over time.

i.e.,

$$\frac{{dQ}_{t}}{dt}={k}_{1}({Q}_{e}-{Q}_{t})$$

(2)

Integrating the above equation by considering Q_t = 0 when t = 0.

$$\text{log }({Q}_{e}- {Q}_{t}) =\text{ log }{Q}_{t} - \frac{{k}_{1}}{2.303} t$$

(3)

In this equation, Q_e and Q_t stand for the amount of organic pollutant adsorbed per unit of adsorbent at equilibrium and at a given time “t,” respectively (in mg/g). The parameter k₁ represents the pseudo-first-order rate constant (in min⁻¹), while t denotes the time duration (in minutes). By plotting log (Q_e – Q_t) against t, the intercept and slope of the resulting graph provide the values for Q_e and k₁, respectively.

The Ho pseudo-second order can be constituted by the equation:

$$\frac{{dQ}_{t}}{dt }= {k}_{1 }\left({Q}_{e}- {Q}_{t}\right)2$$

(4)

On integrating the above equation at, Q_t = 0 when t = 0, we get:

$$\frac{t}{{Q}_{t}}= \frac{1}{{k}_{2}{Q}_{e}^{2 }}+ \frac{t}{{Q}_{e}}$$

(5)

In this equation, k₂ represents the pseudo-second-order rate constant (in g/mg/min). When plotting t ∕Q_t vs t, the slope and intercept of the resulting graph provide the values for Q_e and k₂, respectively.

Result and discussion

XRD analysis

Figure 3 displays an X-ray diffraction pattern for both pure BFO and xN-BFO, with varying nitrogen doping levels (x = 0.01, 0.25, 1). The presence of high-intensity sharp peaks indicates the excellent crystalline nature of the synthesized nanoparticles. In the context of the distorted rhombohedral structure of BFO with the R3C space group, all prominent peaks can be attributed to different (h k l) planes [ICSD180377]. Notably, the sharp peak at approximately 32° signifies that the BFO nanoparticles are oriented along the (110) direction, and the peak splitting in this region, specifically along the (104) plane, confirms the distorted rhombohedral structure of both BFO and xN-BFO (x = 0.01, 0.25, 1). Interestingly, in contrast to the typical behavior observed in doping, which often results in the merging of the (110) and (104) peaks into a single peak (Wrzesińska et al. 2019), nitrogen doping does not result in the merging of these two peaks, as shown in Fig. 3. This suggests that the addition of nitrogen has no discernible impact on the rhombohedral structure of BFO. Due to nitrogen ions being larger than oxygen ions, the addition of nitrogen results in an increase in lattice parameter consistent with nitrogen concentration. This minor structural alteration is probably due to the relatively low concentration of nitrogen. Table 1 presents the lattice parameters for both BFO and xN-BFO. The observed shift toward higher 2θ values for the (012) diffraction peak with increasing nitrogen concentration is likely a result of nitrogen ion substitution at the oxygen sites within the BFO lattice. Additionally, besides the characteristic peaks, several impurity phase peaks of Bi₂O₃ are also present, and their intensity increases with higher nitrogen concentrations. This phenomenon is consistent with previous studies on N-doped BFO (Jia et al. 2018). To determine the average crystallite size shown in Table 1, the Scherrer equation was employed for both pure BFO and xN-BFO.

Fig. 3

XRD pattern of pure BFO and xN-BFO

Table 1 Lattice parameters and particle size of the pure and xN-BFO nanoparticles

FTIR analysis

The synthesized nanoparticles were subjected to characterization through FTIR spectroscopy at room temperature, varying the wavenumber range from 399 to 3999 cm⁻¹, to explore their chemical composition and bonding, as illustrated in Fig. 4. The two significant peaks observed at 416 cm⁻¹ and 513 cm⁻¹ correspond to the stretching vibrations of the Bi-O and Fe–O bonds, as reported in previous studies (Song et al. 2014; Soltani and Entezari 2013). The peak observed within the range of 1046–1107 cm⁻¹ serves as confirmation of the existence of C-O and C-N bonds (Smith 2019, 2020). As the nitrogen concentration increases, there is a noticeable rise in the intensity of the peak in this region, and the peak also shifts toward the higher wavenumber side. This phenomenon may arise due to the elevation in nitrogen ion concentration, leading to the substitution of a greater number of oxygen ions by nitrogen ions, consequently augmenting the quantity of C-N bonds. Additionally, the shift to a higher wavenumber may be due to the replacement of oxygen ions with lighter nitrogen ions. The sharp peaks observed at 1232 cm⁻¹ and 1390 cm⁻¹ in the 1N-BFO samples, and at 1397 cm⁻¹ in the 0.25 N-BFO samples can be attributed to the C-N stretching vibrations. The peak observed at 3000 cm⁻¹ is associated with the bonding of O–H and N–H in the sample (Smith 2019, 2020). Due to the absence or extremely low levels of nitrogen in the composition, this peak is nearly non-existent in pure BFO and exhibits nearly negligible intensity in 0.01N-BFO. Moreover, with the escalation of nitrogen concentration, the broadening of the peak becomes increasingly conspicuous. This phenomenon may arise from alterations in the chemical environment, potentially influenced by interactions with specific elements like hydrogen or carbon binding with nitrogen (Coates 2006). These observations substantiate the substitutional doping of nitrogen within the BFO lattice at oxygen sites.

Fig. 4

FTIR spectra of pure and xN-BFO

FESEM and EDS analysis

The surface morphology and average particle size of the samples were examined through FESEM. Figure 5 depicts FESEM images of BFO as well as xN-BFO. All the samples exhibit a consistent and uniform nature. The average nanoparticle sizes were determined using ImageJ software, and the corresponding histogram plots are provided in Table 2. The presence of nitrogen does not exert a notable impact on the nanoparticle size. Due to the presence of multiple crystallites within each particle, the particle size determined through FESEM appears larger than the size calculated from XRD analysis (Ibrahim et al. 2017). Nonetheless, the FESEM images also reveal the existence of substantial structures, which could potentially signify the presence of agglomerated particles.

Fig. 5

The FESEM images of pure BFO and xN-BFO

Table 2 Particle size calculated from FESEM for pure BFO and xN-BFO

The EDS analysis, as depicted in Fig. 6, confirms the stoichiometric ratio and the presence of elements such as Bi, Fe, O, and N in the synthesized samples. It is noteworthy that the atomic percentages of both Bi and Fe remain nearly identical across all the samples, while the concentration of O ions is approximately three times greater than that of the metal ions. A slight variation in values could potentially arise from the volatilization of ions during the heat treatment process. Nitrogen doping leads to an increase in the atomic percentage of N ions. A little amount of nitrogen concentration is also observed in pure BFO, possibly originating from the nitric acid utilized during the synthesis of BFO nanoparticles.

Fig. 6

a–d The EDS data of pure BFO and xN-BFO (x = 0.01,0.25,1) respectively

UV–Vis analysis

The UV–Vis DRS were employed to investigate the optical characteristics of both pristine and nitrogen-doped samples. Figure 7 showcases the UV–Vis spectra of pure BFO and xN-BFO, illustrating that both BFO and nitrogen-doped BFO exhibit light absorption across both the UV and visible spectral regions. Furthermore, it is worth noting that nitrogen-doped BFO nanoparticles exhibited lower light reflectivity compared to pure BFO, indicating a higher potential for light absorption in this spectral range. Moreover, as the nitrogen level rises, so does the strength of the absorption.

Fig. 7

UV-DRS spectra of pure and xN-BFO

In general, the optical energy band gap serves as a pivotal factor in elucidating the observed optical characteristics of the material. By analyzing the Tauc plot, i.e., (F(R) hυ)² plotted against energy (hυ) as shown in Fig. 8, it becomes evident that the energy gap (E_g) values exhibit a decline with increasing nitrogen concentration, as indicated in Table 3. This phenomenon suggests the potential for enhanced absorption efficiency. The reduction in energy gap may result from multiple factors such as structural defects on the surface, distortion of FeO₆ octahedral (spin polarization), formation of impurity bands just above the valence band of BFO, or lattice modifications, such as oxygen ion substitution with nitrogen ions or interstitial nitrogen doping or rearrangement of molecular orbitals upon doping. Typically, charge transfer in BFO involves the direct band gap transition between O-2p (valence band) and Fe-3d (conduction band) states (Bai et al. 2016). As shown in Fig. 9, nitrogen doping shifts the valence band upward, thereby decreasing the energy band gap (Jia et al. 2018; Lee and Wu 2017).

Fig. 8

Tauc plot for pure BFO and xN-BFO

Table 3 List of energy gap of BFO and xN-BFO samples obtained from Tauc plot

Fig. 9

Energy band diagram for BFO and xN-BFO

X-ray photoelectron spectroscopy

The elemental composition and chemical states of various elements in both BFO and 1N-BFO nanoparticles were rigorously analyzed through XPS examination (Fig. 10a and b). The XPS spectra revealed the presence of distinct peaks corresponding to Bi, Fe, O, N, and C, with the C 1 s peak at 284.8 eV serving as a reference point for calibration. The N 1 s spectra in 1N-BFO are identified in Fig. 8. Specifically, two pronounced peaks discerned at 156.31 and 166.50 eV in pure BFO and 156.99 and 165.39 eV in 1N-BFO can be attributed to the binding energies of Bi-4f_7/2and Bi-4f_5/2, respectively (Fig. 10c). This observation provides confirmation of the + 3- oxidation state of Bi ions in both BFO and 1N-BFO nanoparticles. Likewise, the presence of peaks at 710 and 725 eV for BFO and 714 and 724 eV for 1N-BFO validate the presence of Fe³⁺ and Fe²⁺ states for iron ions in both samples, as depicted in Fig. 10c. Notably, the presence of Fe²⁺ species might be attributed to the reduction of Fe³⁺ species, potentially facilitated by a chelating agent (ascorbic acid) in a high-temperature environment (Li et al. 2019). Furthermore, another set of peaks observed at 532.08 eV for BFO and 530.03 eV for 1N-BFO is indicative of oxygen species, encompassing surface-adsorbed oxygen (O_ads) and lattice oxygen ions (O_latt) integrated within the lattice structure, as delineated in Fig. 10c (Li et al. 2019; Jia et al. 2018; Macías-Sánchez et al. 2015; Zhang et al. 2014; Bharathkumar et al. 2016; Long et al. 2016). Importantly, nitrogen incorporation led to noticeable chemical shifts toward lower binding energy regions in the Bi-4f, Fe-2p, and O-1 s states. These chemical shifts provide compelling evidence for the partial breaking of Bi-O and Fe–O bonds, concomitant with the formation of Bi-N and Fe–N bonds, primarily driven by the slightly lower electronegativity of nitrogen (3.04) in contrast to oxygen (3.44) (Baskaran et al. 2018). High-resolution XPS spectra of the N 1 s region for the 1N-BFO nanoparticles (Fig. 11) unveiled a characteristic broad peak spanning from 396 to 414 eV, with its maximum at 399 eV, assigned to the N 1 s state. This finding serves as compelling evidence for the existence of nitrogen substitution at oxygen site and multiple oxidation states of nitrogen (Kim et al. 2005; Liu et al. 2010; Chen et al. 2009; Paliwal et al. 2017; Irfan et al. 2017).

Fig. 10

a XPS survey scan of pure BFO. b XPS survey scan of 1N-BFO. c The high-resolution XPS spectra of Bi 4f, Fe 2p, and O 1 s states of BFO and 1N-BFO

Fig. 11

High-resolution spectra of N1s state in 1N-BFO

Catalytic activity studies

Photodegradation of methylene blue dye

MB, with the molecular formula C₁₆H₁₈N₃SCl, is a well-known organic chloride compound widely recognized for its versatile utility as a cationic dye, acid–base indicator, and cationic stain. Additionally, MB possesses notable pharmacological attributes, including antimalarial, antidepressant, antioxidant, and cardioprotective properties, which have led to its incorporation in medicinal applications. However, the inadvertent discharge of MB into aquatic environments has emerged as a significant concern, giving rise to water pollution, and presenting potential threats to diverse forms of aquatic life (Paliwal et al. 2017). Contemporary environmental remediation strategies increasingly emphasize advanced oxidation processes (AOPs), with photocatalysis emerging as a prominent technique for addressing the challenge posed by organic pollutants such as MB in contaminated water.

Maleki (2015) and Reddy et al. (2018) observed improved photocatalytic performance with lanthanum doping in the BFO lattice but noted a decline at high concentrations due to increased defect concentration, which acts as recombination centers for charge carriers. Similarly, Hu et al. (Hu et al. 2017) reported that samarium (Sm) doping in BFO introduces new trapping levels that serve as electron acceptors, reducing the recombination rate of photogenerated charge carriers. However, due to their potential toxicity and metal leaching issues, elemental dopants are not suitable for drinking water purification. Furthermore, none of the studies has explored the effects of simple anionic doping, such as nitrogen doping in BFO, on photodegradation of organic dye.

The possible photocatalysis mechanism mediated by BFO, depicted in Fig. 12, necessitate three stages: (1) photon absorption by photocatalyst; (2) production of photogenerated charge carries, its effective separation, relocation, or recombination; and (3) the redox reaction on the surface of the photocatalyst, i.e., when a photon equal to the energy band gap of BFO falls on its surface, the BFO gets activated and produces electron–hole pairs in the conduction band and valence band, respectively. Within the valence band, the generated hole initiates a reaction with water, yielding hydroxyl radicals (•OH), known for their potent oxidizing capabilities. Simultaneously, electrons residing in the conduction band engage in reactions with molecular oxygen, leading to the formation of anionic superoxide radicals (O₂^•–). These ions actively participate in oxidation reactions, effectively curbing the recombination of electron–hole pairs. Furthermore, the hydroxyl radicals (•OH) and superoxide ions (O₂^•–) generated in the process exhibit dual functionality, wherein they target both adsorbed dye molecules and those dye molecules in close proximity to the surface of the BFO nanoparticles. This concerted action leads to the degradation of the dye molecules (Siddique et al. 2018; Chen et al. 2009; Irfan et al. 2017). The comprehensive reaction is elucidated as follows:

Fig. 12

Photocatalytic effect

$$\begin{array}{c}\text{i. e., BFO}+\text{h}\upsilon \to \text{BFO} ({\text{e }}^{-} + {\text{h}}^{+})\\ {\text{H}}_{2}\text{O}+ {\text{h}}^{+} \to {\text{OH}}^{\bullet } + {\text{H}}^{+}\\ {\text{O}}_{2} + {\text{e}}^{-} \to {\text{O}}_{{2}{}}^{{\bullet }-} (\text{super oxide ion})\\ {\text{O}}_{{2}{}}^{{\bullet }-} + {\text{H}}^{+}\to \text{ OOH}\\ 2\text{ OOH }\to {\text{H}}_{2}{\text{O}}_{2}+ {\text{O}}_{2}\\ {\text{H}}_{2}{\text{O}}_{2} \to {2\text{ OH}}^{-} \\ \text{Dye}+\text{ OH}^{\bullet } \to {\text{CO}}_{2} + {\text{H}}_{2}\text{O}\\ \text{Dye}+ {\text{O}}_{{2}{}}^{{\bullet }-} - \to {\text{CO}}_{2} + {\text{H}}_{2}\text{O}\end{array}$$

The investigation into the photocatalytic performance of both pure BFO and 1N-BFO in the degradation of methylene blue dye was carried out under natural sunlight; the illustration is provided in the Supporting Information (SI). The primary absorption band of MB at λ = 660 nm was used to assess the impact of the nanocatalyst on the photodegradation process. During the degradation process, there was a notable rapid reduction in the intensity of the absorption peak over increasing irradiation time. For comparison, a blank experiment was conducted under identical conditions, omitting the presence of a photocatalyst. However, this control experiment did not exhibit any substantial degradation of the dye. Upon introducing pure BFO into the dye solution, the degradation efficiency reached 53.6% within a duration of 105 min. Similarly, when 1N-BFO was incorporated into the reaction environment, the photocatalytic activity displayed a notable enhancement, reaching up to 63% (refer to the SI for the graph) over the same time interval. The kinetics of the degradation reaction followed a pseudo-first-order Langmuir–Hinshelwood reaction model.

$$\text{i}.\text{e}.,\text{ ln }(\frac{{A}_{t}}{{A}_{0}}) = -\text{kt}$$

(6)

where k is the rate constant and t is the exposure time (Li et al. 2019; Loghambal et al. 2018). To explore the reaction kinetics, a plot of ln (A_t/A₀) vs time (Fig. 13) was constructed. The determined rate constants for pure BFO and 1N-BFO were found to be 0.00595/min and 0.00945/min, respectively. These findings clearly indicate a substantial enhancement in photocatalytic activity upon the nitrogen doping of pure BFO. This improvement can be attributed to the governing factors of nanoparticle photocatalytic activity, particularly light absorption efficiency and reduction in the rate of recombination of photogenerated electron–hole pairs (Di et al. 2017; Caroline et al. 2017). As a conclusion, the role of photocatalyst is effectively shown up here. Notably, the incorporation of nitrogen in 1N-BFO led to a significant reduction in the energy band gap by increasing the valence band edge, as evident from the Tauc plot in Fig. 8, which revealed enhanced absorption of visible light in 1N-BFO. Furthermore, the addition of nitrogen may decrease the oxygen vacancies in 1N-BFO, which serve as electron–hole pair recombination centers. Consequently, 1N-BFO exhibited markedly intensified visible light activity.

Fig. 13

Plot of ln (A_t/A₀) vs time for BFO and 1N-BFO

Adsorption kinetics of the photodegradation of MB

The adsorption of MB onto the surfaces of BFO and 1N-BFO was investigated, employing both the Lagergren pseudo-first-order model and the Ho pseudo-second-order model. The respective values of equilibrium adsorption capacity Q_e and rate constants k₁ for the Lagergren pseudo-first-order model were determined through the plot of log (Q_e – Q_t) vs time, as presented in Fig. 14. Similarly, the Q_e and rate constants k₂ for the Ho pseudo-second-order model were extracted from the linear plot of $\frac{t}{{Q}_{t}}$ versus time, as depicted in Fig. 15. A comprehensive tabulation of the experimental and calculated values of Q_e, k₁, and k₂ is provided in Table 4.

Fig. 14

log (Q_e – Q_t) vs time for BFO and 1N-BFO

Fig. 15

$\frac{t}{{Q}_{t}}$ vs time plot for BFO and 1N-BFO

Table 4 Adsorption kinetics values of BFO and 1N-BFO

It is obvious from the kinetics data (Table 4) that the maximum value of regression coefficient (R²) and the calculated Q_e value are nearly identical to the experimental value attained in a pseudo-second-order model. Hence, the adsorption of MB to the surface of BFO and 1N-BFO obeys pseudo-second-order kinetics.

Reduction of 4-nitrophenol

4-NP is another prevalent organic pollutant commonly found in wastewater sources. Its extensive application spans various agricultural and industrial activities, encompassing the production of insecticides, dyes, and pharmaceuticals (Shi et al. 2019). Due to their pronounced potential for toxicity, even at minuscule concentrations, nitrophenols pose significant environmental hazards. These compounds have the capacity to inflict harm upon multiple physiological systems, including the central nervous, circulatory, visual, and respiratory systems, among others. Consequently, the removal of these pollutants from wastewater constitutes a pressing environmental concern. Remarkably, 4-AP, a compound derived from the reduction of 4-NP, finds application in diverse domains, such as photographic development, corrosion inhibition, antipyretic and analgesic medications, and hair dyeing agents, among others. Consequently, the conversion of 4-NP to 4-AP emerges as an efficient approach for purifying nitrophenol-contaminated wastewater. Even though NaBH₄ stands out as a potent and efficient reducing agent for the reduction of nitroaromatic compounds, it exhibits limited capacity for reducing nitro groups in the absence of a catalyst (Goyal et al. 2015). Catalysis is the most cost-effective and efficacious method for converting deleterious 4-NP to beneficial 4-AP. The catalytic activity predominantly involves a series of critical reactions, including (1) adsorption of 4-NP onto the nanocatalyst surface; (2) diffusion of 4-NP onto the active site of nanocatalyst; (3) conversion of 4-NP to 4-AP; and (4) desorption of adsorbed compound from the surface of the nano catalyst (Kong et al. 2017).

Noble metal catalysts like silver, gold, and palladium show high efficiency in 4-NP reduction but are costly due to their scarcity, limiting industrial application (Kästner and Thünemann 2016; Thawarkar et al. 2018; Strachan et al. 2020; Rodriguez et al. 2022; Ehsani et al. 2023). Copper catalysts require complex stabilization procedures (Ahlam et al. 2020; Garba et al. 2021). Nickel offers higher catalytic activity, low cost, and easy availability but suffers from aggregation, reducing its surface area and active sites (Vivek et al. 2016). Therefore, developing inexpensive, easily synthesized, stable, and reusable catalysts is increasingly important.

BFO shows outstanding catalytic efficiency in reducing nitrophenol compounds, with NaBH₄ playing a vital role as the primary hydrogen source for the catalytic conversion of 4-NP to 4-AP.

The comprehensive process involved in the reduction of 4-NP can be conceptualized as a combined mechanism encompassing the adsorption of hydrogen atoms and the subsequent elimination of water molecules (Goyal et al. 2015). A succinct overview of this intricate 4-NP reduction process is provided below and is visually depicted in Fig. 16.

Fig. 16

Reduction of 4-NP to 4-AP

The initial phase involves the migration of a free hydrogen atom toward the oxygen atoms of the nitro group within the 4-NP molecule, leading to its adsorption on these oxygen atoms. Subsequently, another hydrogen atom approaches, resulting in the formation of a hydrogen cluster. Consequently, the assembly of this cluster leads to the generation and subsequent elimination of a water molecule, facilitated by the residual cluster, thereby producing a 4-nitrosophenol molecule. In the subsequent step, an additional two hydrogen atoms approach the remaining oxygen and nitrogen atoms within the nitro group, giving rise to the formation of a second water molecule, which is then released. Finally, a pair of hydrogen atoms approaches the initially formed cluster, leading to the creation of an amino group (Shi et al. 2019).

To investigate the catalytic effect of BFO and xN-BFO in the reduction of 4-NP to 4-AP, the study initially employed NaBH₄ as the reducing agent. In this experimental procedure, a solution comprising 1 mL of 4-NP and 1 mL of NaBH₄ within a 25 mL aqueous medium was prepared and subjected to continuous agitation. The addition of NaBH₄ led to the formation of phenolate ions, resulting in a notable shift in the absorption peak from 403 to 314 nm, accompanied by a change in the solution color from pale yellow to dark yellow. However, despite prolonged stirring for a duration of 30 min, no discernible reduction of 4-NP was observed (provided as SI).

Subsequently, to assess the involvement of NaBH₄ in the reduction process, the experiment was carried out in the absence of NaBH₄. However, BFO and 1N-BFO could only reduce 8% and 25% of 4-NP respectively even after 30 min of stirring (provided as SI). Hence, it is apparent that NaBH₄ is important for the reduction process and serves as a major source of hydrogen during the reaction.

In the final phase of experimentation, the influence of NaBH₄ in conjunction with various catalysts was investigated under different conditions. When 10 mg of pure BFO was introduced into an aqueous solution containing 4-NP and NaBH₄, a conspicuous reduction in the intensity of the absorption peak was observed. Moreover, the solution transitioned from its initial coloration to a colorless state within a span of 22 min. This transformation indicated that pure BFO effectively and completely converted 4-NP into 4-AP. The introduction of the same quantity of 0.01N-BFO, 0.25N-BFO, or 1N-BFO as catalysts led to a substantial reduction in the reaction time, with reduction times of 12 min, 10 min, and 4 min, respectively (given as SI).

The catalytic reduction of 4-NP by BFO and xN-BFO was subjected to a comprehensive evaluation using a pseudo-first-order Langmuir–Hinshelwood reaction model, which adheres to the equation ln (A_t/A₀) = -kt (Fig. 17), wherein “k” represents the rate constant and “t” signifies the time at any instant (Li et al. 2019). Deviations from the linear trend in the plot indicate that the reduction of 4-NP into 4-AP is not a constant process over time. This observation corroborates that the reduction of 4-NP to 4-AP is not a direct conversion but instead involves varying rate kinetics during the intermediate state, notably the nitroso phenol, in the course of the reduction reaction (Kong et al. 2017). The rate constants for pure BFO and xN-BFO are tabulated in Table 5, revealing a discernible increase in the rate of reaction with an increase in nitrogen concentration.

Fig. 17

First-order Langmuir–Hinshelwood plot for BFO and xN-BFO

Table 5 First-order Langmuir–Hinshelwood rate constant for BFO and xN-BFO (x = 0.01,0.25,1)

The preceding data set unambiguously establishes BFO as an excellent catalyst, with the catalytic activity exhibiting a notable enhancement with concentrations of nitrogen. This augmentation in catalytic performance can be attributed to several key factors, including the size, surface area, chemical composition, and redox potential of the metal ions present on the surface of BFO nanoparticles, all of which exert considerable influence on their catalytic capabilities (Kong et al. 2017). The increase in catalytic activity with the increase in the introduction of nitrogen into BFO may be due to the high positive charge density and high spin density of nitrogen (Kong et al. 2014).

Adsorption kinetics study for the reduction of 4-NP

The adsorption kinetics of 4-NP onto the surface of BFO and xN-BFO were rigorously investigated using both Lagergren’s pseudo-first-order model and Ho’s pseudo-second-order model. This analysis involved plotting log (Q_e – Q_t) vs time (Fig. 18) and t/Q_t vs time (Fig. 19), respectively. The obtained results, encompassing the calculated and graphical values of Q_e, k₁, and k₂, have been comprehensively tabulated in Table 6. Notably, the pseudo-second-order model demonstrated the highest value of regression coefficient (R²) and exhibited remarkably close agreement between the experimental and graphical values of Q_e. Consequently, it is certainly established that the adsorption of 4-NP onto the surfaces of BFO and xN-BFO adheres to pseudo-second-order kinetics.

Fig. 18

Plot of Log (Q_e-Q_t) vs time for pure BFO and xN-BFO

Fig. 19

t/Q_t vs time curve for BFO and xN-BFO

Table 6 Kinetic parameters of adsorption of 4-NP to BFO and xN-BFO (x = 0.01,0.25,1)

To address the issue of the regression coefficient being absent for 1N-BFO, the experiment was repeated with an increased concentration of 4-NP (5 mM). The reduction of 4-NP to 4-AP was achieved within 8 min (The corresponding plot is provided as SI) and the rate constant was determined to be 0.2564/min.

The data presented in Fig. 20 illustrates the reaction studies involving the reduction of 5 mM 4-NP using 1N-BFO. The corresponding Table 6 provides the values of R² and the rate constants for both first-order and second-order reactions. Notably, the data clearly indicates that the reduction reaction follows pseudo-second-order kinetics.

Fig. 20

a and b Plot of Log (Q_e-Q_t) vs time and t/Q_t vs time curve for 1 N-BFO, 5 mM 4-NP

The recyclability of a catalyst is crucial for its practical application. In this study, the cycling procedure for the reduction of 4-NP was repeated up to 5 cycles for both BFO and 1N-BFO. After each cycle, the catalyst was removed from the 4-NP solution, centrifuged, cleaned, and dried. Each cycle was conducted under identical conditions. The reduction efficiency remained nearly constant, with 97.5 to 97.3% for BFO and 99.3 to 99.1% for 1N-BFO from the 1st to the 5th cycle, respectively (the illustration is presented as SI). This indicates that there was no significant loss in catalytic activity during the reduction process. Therefore, BFO and 1N-BFO demonstrate excellent stability and can be considered viable catalysts for the reduction of organic pollutants, suitable for practical applications.

Conclusion

In summary, pure BFO nanoparticles were synthesized via chemical auto-combustion method, while varying concentrations of NH₄Cl solutions served as precursors for nitrogen doping into the BFO lattice. Despite similar nanoparticle sizes for both pure and nitrogen-doped BFO observed in FESEM and XRD data, EDS analysis confirmed the presence of nitrogen in xN-BFO, with the nitrogen content increasing proportionally to the precursor concentration. This nitrogen incorporation, which resulted in the displacement of oxygen ions, was further corroborated by XRD, FTIR, UV-DRS, and XPS data. Notably, while nitrogen doping had minimal effects on nanoparticle size and lattice parameters, it significantly reduced the energy gap of BFO nanoparticles. This reduction could be attributed to factors like the introduction of structural defects at the surface and the substitutional doping of nitrogen ions into the BFO lattice.

Furthermore, both pure BFO and 1N-BFO exhibited effective degradation of MB dye under visible solar radiation, with 1N-BFO showing a 1.17-fold enhancement in photodegradation efficiency compared to pure BFO. The catalytic reduction of nitrophenol demonstrated that catalytic activity increased with nitrogen concentration, resulting in faster reduction reactions. These processes followed pseudo-first-order kinetics. The observed decrease in the energy gap and the increase in local charge density and spin density introduced by nitrogen doping, also reduction in recombination centers like oxygen vacancies by addition of nitrogen ions, are likely contributed to this enhanced catalytic behavior.

Moreover, adsorption kinetics analysis revealed that the adsorption of organic pollutants (MB and 4-NP) onto the surfaces of BFO and xN-BFO followed a pseudo-second-order kinetics model, highlighting the effectiveness of these materials in pollutant removal processes.

The recyclability study confirmed the stability of both BFO and nitrogen-doped BFO, with no significant performance loss after five cycles. Therefore, these nanoparticles hold great potential for practical applications in environmental remediation.