RSM optimization and morphological investigation of bismuth ferrite for photodegradation of organic pollutant under visible light

Abstract

The current study has been conducted to assess the influence of the morphology of bismuth ferrite perovskite (BiFeO₃) on its photocatalytic performance. Nanorod and nanoparticle morphologies of BiFeO₃ were prepared using the hydrothermal process and analyzed with several techniques. The photocatalysts’ effectiveness was appraised by MB photo-degradation in the presence of visible light. Various techniques were applied to analyze the structure, morphology, together with photoelectrochemical and optical properties of the photocatalyst. Optimum photo-degradation parameters were recognized via RSM, resulting in a maximum MB degradation efficiency of 98.2% (0.2021 min⁻¹) in 20 min for nanorods. Whereas, the degradation yield for nanoparticles was achieved at 76% (0.075 min⁻¹).From the radical trapping experiments, superoxide reactive radicals played a main role in MB photo-degradation. Also, BiFeO₃ nanorods displayed superb stability during the five cycles of the photo-reaction procedure. This current survey suggests that the appropriate selection of nanomaterial morphology is crucial for achieving efficient photocatalysts.

Introduction

Rising water pollution, crowd expansion, and climatic alteration are considered significant risks to the availability of purified water for residential, commercial, and agricultural applications. Consequently, the reclamation of processed municipal or industrial effluent has attracted considerable interest from researchers due to increasing concerns regarding water scarcity¹. Given that organic contaminants in aquatic environments are predominantly extremely hazardous and poisonous, which may result in a variety of significant health concerns, the effective removal of these detrimental organic substances from the environment has emerged as a critical priority for the research community. Among the various industrial contaminants, pigmented substances are of particular concern in discussions about natural environments^2,3. Latterly, advanced oxidation processes (AOPs) have been recognized as viable alternatives to customary procedures⁴. In the last few years, the organic contaminants’ photodegradation through a radical oxidative mechanism has been extensively researched as a safe and sustainable method for their removal². Photocatalysis is a progressive oxidation method that involves the activation of a semiconductor under light exposure, resulting in the generation of active species that can degrade contaminant molecules^5,6. Regrettably, the majority of commonly utilized photocatalysts predominantly react to UV-visible light exposure, which restricts their effectiveness during light exposure. As a result, the search for extremely effective and durable photocatalysts driven by visible light has become a prominent area of interest within the domain of photocatalytic processes². Within the extensive range of materials, multiferroics represent a promising category for design and synthesis, owing to their applicability in multifunctional applications. They have garnered significant interest due to the intriguing physics that governs their behavior and their potential for innovative device applications⁷. Recently, BiFeO₃ photocatalysts have garnered interest due to their low energy band gap, exceptional chemical durability at ambient temperature, and superior charge carrier transport characteristics. BiFeO₃ demonstrates the potential for catalytic uses when exposed to visible illumination. Research indicates that BiFeO₃ possesses a band gap of almost 2.12 ~ 2.4 eV^7,8. It demonstrates a positive reaction to visible light and is also utilized for photodegradation of water contaminants, including various types of dye^9,10. Bismuth (Bi) containing photocatalytic composites were selected for their superior absorption of visible radiation. These photocatalysts remove organic contaminants by producing reactive oxygen species and exhibit elevated chemical durability and resistance to photo-corrosive degradation, ensuring their stability¹¹. A variety of synthesis techniques for BiFeO₃ perovskite are available, providing different strategies for adjusting its characteristics. These synthesis methods include solution combustion¹², co-precipitation¹³, sol-gel¹⁴, perching¹⁵, hydrothermal¹⁶, and solution evaporation¹⁷. Each method offers distinct benefits and obstacles, affecting parameters including particle dimensions, purity, degree of crystallinity, and shape¹⁸. Different synthesis methods generate products of diverse shapes and sizes, which influences the catalyst’s efficacy of efficacy of the catalyst is primarily determined by surface characteristics. Catalysts produced with exact regulation of the size and morphology of particles can effectively recognize specific catalytic sites¹⁹. The high temperature decomposition of bismuth salt compounds and the variations in the ionic chemical valence make it challenging to synthesize bismuth ferrite without impurities using typical solid phase reactions at elevated sintering temperatures²⁰. Among these techniques, the hydrothermal method is a widely recognized and straightforward approach for synthesizing highly crystalline, phase-pure bismuth ferrite nanostructures. Also, this synthesis technique offers the advantage of producing nanostructures exhibiting adjusted morphology, narrow size distribution, and excellent dispersion, which can be achieved by modifying experimental parameters such as pH, duration, temperature, and the incorporation of surfactants, among others²¹. The hydrothermal technique enables improved management of the material’s purity, microstructure, and morphology by adjusting the reaction parameters^22,23. Enhancing the morphology of the BiFeO₃ bulk structure is a crucial approach to augment the specific surface area and the active site count²⁴. The regulation of particle morphology and dimensions is crucial for optimizing photocatalytic efficiency, as it affects the reaction sites and the magnitude of energy of the band gap. In the last few years, various shapes and structures of bismuth ferrite have been synthesized, including wire²⁵, tubes²⁶, and Submicron spindle-shaped structures²⁷. In this context, significant advancements have been achieved in the regulated synthesis and self-assembly of bismuth ferrite particles exhibiting controlled geometric configurations. Understanding how morphology and dimensions influence the band gap energy and photocatalytic effectiveness is essential for grasping catalytic mechanisms and improving semiconductor catalysts²⁰. Previous research aimed at enhancing the multiferroic characteristics of BiFeO₃ has noted an improvement in electrical and magnetic characteristics within single-phase BiFeO₃nanostructures, which exhibit a pronounced dependence on size. Additionally, acid leaching is essential for achieving a pure phase, which results in the formation of coarser powders and reduced reproducibility^28,29. The solubilization of bismuth (Bi) and ferric(Fe) precursors as their corresponding hydroxides through the reaction with a base, for example, potassium hydroxide (KOH), along with the formation of not soluble oxide powders through precipitation from highly concentrated hydrothermal solution is considered the process responsible for the phase formation of bismuth ferrite samples³⁰. BiFeO₃ with various morphologies has been successfully produced through a hydrothermal method^30,31. Nevertheless, comprehensive studies examining the effect of nanostructure and morphology on the physicochemical characteristics of single-phase BiFeO₃ have been minimally explored. The photocatalytic characteristics of nanostructures are significantly affected by the specific shape of the nanostructure being examined^32,33. Size and shape are crucial for enhancing the photocatalytic characteristics of BiFeO₃. Thus, in this research, the successful synthesis of BiFeO₃ nanostructures was achieved. Moreover, their photocatalytic performance was examined for comparative purposes. Using analyses, the structural characteristics of the nanostructures were expounded, including DRS, XRD, TEM, PL, Mott-Schottky, and BET. The RSM-CCD was utilized to identify the optimal operating parameters, including pH, intensity time, LED power, catalyst dosage, and the MB concentration. This investigation also evaluated the influence of variables on the photodegradation effectiveness of MB dye using BiFeO₃ nanorods as the catalyst. The current research involved the synthesis of photocatalysts, called BFO, with morphological structures of both particles andnanorods. To comprehend how BFO morphology influences optical properties, a detailed comparative study was carried out. The novelty of this study is the impact of the morphology of the bismuth ferrite nanorod on the MB dye degradation.

Experimental

Materials

Bismuth nitrate [Bi(NO₃)₃·5H₂O, 98%] was procured from Alfa Aesar. Iron nitrate [Fe(NO₃)₃·9H₂O, 99%], potassium hydroxide (KOH, 85%), cetyltrimethylammonium bromide (CTAB, 99%), sodium sulfate (Na₂SO₄, ≥ 99%), sodium carbonate (Ca (NO₃)₂ >99%), nitric acid (HNO₃, 65%) and sodium chloride (NaCl, > 99%) were provided by Merck. All the chemical compounds were utilized as received without any additional purification. The solutions employed to identify species of active comprised 2 mM isopropyl alcohol (IPA, Scharlau), 1 mM p-benzoquinone (BQ, Merck), and 1 mM ethylenediaminetetraacetic acid (EDTA), which acted as scavengers for hydroxyl radicals (·OH), superoxide anions (·, and hole (), respectively.

Synthesis of BiFeO₃ (nanorods and nanoparticles)

BiFeO₃ nanorods were produced through a hydrothermal process as described by Remya et al.²¹, with certain modifications. In a usual method, the necessary molar concentrations of ferric nitrate nonahydrate [Fe(NO₃)₃·9H₂O] and bismuth nitrate pentahydrate [Bi(NO₃)₃·5H₂O] (in a 1.1 ratio) were dissolved individually. Bi (NO₃)₃·5H₂O was dissolved in diluted HNO₃, while Fe(NO₃)₃·9H₂O was dissolved in double-distilled water. The Fe(NO₃)₃ solution was gradually combined with the Bi(NO₃)₃ solution while maintaining steady stirring. The prepared solution was subjected to ultrasonication for 30 min. Ultimately, KOH was combined gradually, drop by drop, in the aforementioned blended solution to facilitate the co-precipitation of Fe₃⁺ and Bi₃⁺ ions. Subsequently, the entire solution was placed in a 75 ml Teflon-lined autoclave, where it remained at 200 °C for 12 h and then allowed to cool to the temperature of the ambient naturally. The prepared precipitates were rinsed multiple times using distilled water until reaching a pH of 7. Thereafter, it was dried at 73 °C for 12 h. After drying, the brown powder was heated to calcination in a furnace at 400 °C for three hours to produce phase-pure BiFeO₃ nanopowder. The conventional synthesis method implemented various modifications to generate nanostructures with distinct morphologies. CTAB and KOH were utilized to adjust the morphology of the synthesized catalyst further. Throughout this investigation, the BiFeO₃ nano-catalyst was identified by the abbreviation BFO.

Characteristics

Using a variety of techniques, the prepared BFO powders were analyzed. The phase structure of the prepared photocatalytic samples was characterized througha powder X-ray diffractometer with high-intensity Cu Kα radiation (XRD, PW 1730, PHILIPS). Utilizing transmission electron microscopy (TEM, Zeiss EM900), the morphology and nanostructure of the synthesized photocatalytic samples were analyzed. Also, using the Brunauer-Emmett-Teller (BET) instrument (BET, BELSORP MINI II, BEL), the surface area specific to the photocatalytic powders was determined.PL (photo-luminescence) spectra were analyzed using a luminescence spectrophotometer (Varian Cary Eclipse, Varian) with an excitation wavelength of 483 nm. Using a UV–vis spectrophotometer (Avaspec-ULS3648-RS, Avantes) was measured Ultraviolet-visible diffuse reflectance spectra (DRS). Mott-Schottky (MS, ZIVE-SP1) curves were achieved using an electrochemical workstation. LED lamps with varying wattages were utilized as the light source for irradiation.

Photocatalytic study

The methylene blue (MB) decolorization from aqueous solution was conducted utilizing the photocatalyst samples under the influence of exposure to visible light from home LED (λ > 420 nm) and natural sunlight sources. 20 mg of BFO sample was introduced in a volume of 100 mL of methylene blue dye solution (5 mg/L). Before activating the LED lamp, to achieve the adsorption balance of the dye and the photocatalytic material, the obtained solution was stirred for 30 min under dark conditions. Subsequently, the photocatalytic activity was scrutinized in the presence of exposure to LED light (200 W) and sunlight. The lamp was located 12 cm distant from the surface of the solution, and it was constant for all tests. Throughout the dye degradation process, 5 mL of the suspended solution was extracted at predetermined periods and placed to centrifugation to isolate the photocatalysts. The concentration of methylene blue was determined utilizing the UV-vis spectrophotometer at a wavelength of 664 nm. The degradation rate was determined as follows:

1

In Eq. (1), & represent the initial concentrations and concentrations of the dye solution at time t.

Photoelectrochemical testing

Mott − Schottky (MS) tests were conducted on an electrochemical work station. Ag/AgCl and Pt were used as the reference electrode and counter electrodes, respectively, in a three-electrode system. Na₂SO₄ (0.5 mol/L) was used as electrolyte solution. The amount of 15 µl of nafion and 10 mg of BFO catalyst was dispersed in an isopropanol/water mixture (1 mL, 0.5:0.5 mL) for the preparation of the working electrode, and then 0.03 mL of suspension was transferred onto a 1 2 cm² ITO glass.

RSM-CCD design procedure

The response surface methodology (RSM) was employed using Design- Expert software (version 7, https://www.statease.com/software/design-expert/) for designing experiments and optimizing the influence of process variables. Notably, the optimization of the process was accomplished through the strategic implementation of Central Composite Design (CCD), which is acknowledged as a constituent of RSM. CCD was selected from various RSM approaches as the most efficacious and widely recognized technique for developing a response surface model with five levels for five independent variables: photocatalyst dosage, MB concentration, time, LED power, and pH^34,35, as illustrated in Table 1. Furthermore, as outlined in Eq. (2), a second-order polynomial relationship was utilized to predict the correlation between the key factors and the efficiency of MB dye degradation.

Table 1.

CCD design for input variables.

Factor	Unit	Levels
		−2	−1	0	1	2
Catalyst amount	mg	5	10	15	20	25
pH	-	2	4	6	8	10
Concentration MB	mg/l	5	10	15	20	25
LED lamp power	W	50	100	150	200	250
Time	min	20	40	60	80	100

2

Here, ε and are the unanticipated errors and a constant amount, and are reaction variables. Furthermore, , , and illustrate the constants of reaction³⁶.

Results and discussions

The scrutiny of morphology formation

Singular phase bismuth ferrite samples exhibiting two distinct morphologies were produced through the hydrothermal method. Figure 1 presents a schematic representation of the empirical situation utilized to adjust the morphology of bismuth ferrite through the hydrothermal method. A spherical particle shape was achieved when potassium hydroxide (KOH) was employed as the structure-directing factor. The production of impurity-free perovskite oxides necessitates a crucial concentration of the alkaline cultivation environment. Throughout the synthesis, under elevated temperatures and pressures, the metal hydroxides were exposed to reactions with potassium hydroxide (KOH). The insoluble depositions composed of oxides of ceramic samples were obtained throughout this procedure. Effectuating the processes of dissolution and recrystallization that lead to the self-stabilization of the system and the production of impurity-free BFO nanoparticles, under regulated situations of temperature, pressure, and appropriate time. An increased molarity of potassium hydroxide (KOH) in the reaction ambient promotes the production of Bi₂₅FeO₄₀²¹.

Fig. 1.

Schematic representation of the synthesis of the morphology of BFO nanostructures.

The Reactions related to BFO nanostructure can be expressed as follows: Eqs. (3–7):

3

4

5

6

7

Based on the equations, and were converted into amorphous precipitate of and . Then, KOH as a mineralizer interacts with the hydroxides to dissolve them and generate groups of ions. After achieving super saturation, groups of ions through nucleation and growth, resulting in the formation of BFO crystals³⁷. During the nucleation trend, hydroxyl ions (OH) derived from potassium hydroxide (KOH) are adsorbed onto the surfaces of the nanoparticles, reducing their surface energy and resulting in the formation of spherical nanoparticles with minimal surface energy^31,38,39. During the crystal growth procedure, the use of 3 molar potassium hydroxide (KOH) results in the formation of particles of tiny size, typically measuring nanometers in size, which are generally spheroid-shaped in appearance²⁷. Achieving the phase of impurity-free bismuth ferrite (BFO) is accomplished by utilizing the precise concentration of potassium hydroxide (KOH), as it affects the solubility of the mediocre Bi₆O₆⁶⁺ and Fe(OH)₄, as well as the ratio of concentrations of bismuth (Bi) to iron (Fe). Until cetyltrimethylammonium bromide (CTAB) is employed as a cationic surfactant under identical conditions, BFO nanorods are produced. Surfactants create advantageous sites and also affect the processes of nucleation, formation, and agglomeration. It is well recognized that surfactant ions significantly impact the formation of various morphologies of nanostructures of oxide, which are compounds of polar direct interactions with the same substance particles. CTAB is a type of cationic surfactant characterized by a hydrophilic head and a hydrophobic tail. Cetyltrimethylammonium bromide operates through the action and reaction groups of the head CTA⁺, its positive charge with negatively charged surfaces of the BFO crystals, promoting a self-organization procedure of particles via molecular interactions. This interaction can facilitate the development of a template layer on specific crystal surfaces of BiFeO₃, steering the growth of nanorods in particular crystallographic directions. Also, a decrease in the attraction of the surface of the solution leads to diminished energy needed for the growth of a novel phase. The adsorption of CTAB on the surfaces of nanoscale crystals aligned with the c-axis constrains growth kinetics along the lateral facets, which may prevent solute particle diffusion and lead to one-dimensional nucleus growth, resulting in rod-like structures²¹.

Surface structural studies

X-ray diffraction (XRD) is a laboratory method that gathers data non-destructively and precisely with various aspects, including crystal structure, chemical composition, crystal orientation, lattice strain, crystallite size, layer thickness, and preferred orientation. Investigators are utilizing this analysis to identify the impurities within the samples^40,41. The phase of BFO nanostructures (JCPDS Card No. 86-1518) was exhibited by XRD analysis in Fig. 2. The major peaks at 2θ = 22.4, 31.7, 31.9, 39.6, 46.2, 51.4, 57.5, 57.6, and 67.4° corresponds to the reflections from (012), (104), (110), (202), (024), (116), (018), (214), and (220) planes respectively^21,42. Distinctly, taper diffraction peaks demonstrate the growth of crystalline structures. In contrast, the broadening and fluctuations in the intensity of the diffraction patterns imply variations in the size of the synthesized nanostructures, which can be attributed to the specific experimental conditions during the synthesis process. The small peak at 28° in Fig. 2(b), marked with (*), indicates a small amount of the secondary phase Bi₂Fe₄O₉, which may result from the formation kinetics and the metastable nature of BFO²¹.

Fig. 2.

Pattern of XRD for BFO (a) nanorods and (b) nanoparticles.

To achieve morphological information, form, size, and structural specification, transmission electron microscopy (TEM) is employed. The particle and rod morphologies by transmission electron microscopy (TEM) in bright field mode are illustrated in Fig. 3. Can be observed in Fig. 3(a), BFO included zero-dimensional nanoparticles with an average size of ~ 48 nm, which exhibit a distribution with noticeable agglomeration, likely due to a shortage of surfactants. Also, adjacent spherical particles are joined together in a layered arrangement during the hydrothermal synthesis by KOH (Fig. 3a). The reduction of surface energy of nanoparticles through the adsorption of hydroxyl ions on the particle surfaces leads to spherical nanoparticles with the least surface energy³⁷.

Fig. 3.

TEM images of (a) nanoparticle, (b) nanorod of BFO.

Figure 3(b) illustrates the 1D-nanorod structure. It seems that the nanorods were generated through the assembly of various irregularly shaped nanoparticles, as shown in Fig. 3b(ii). The growth of nucleation in a one-dimensional state and anisotropic in the presence of surfactants contributes to the formation of rod-shaped structures²¹. Furthermore, based on the Gibbs-Thomson law, the growth environment has a major impact on determining the shape of bismuth ferrite particles by dissolution and crystallization. Firstly, Crystals grow into bigger particles in an Alkaline environment. Subsequently, based on the concentration of alkaline, other nanocrystals are adsorbed onto the surface of the particles, which leads to anisotropic growth²⁰.

Figure 4 illustrates that the assessment of the physical characteristics of the synthesized BFO nano-catalysts can be conducted through both BET and BJH techniques. The isotherms of the nitrogen (N₂) adsorption-desorption for BFO nanoparticle and nanorod in diagrams correspond to type IV, featuring an H-type loop (H₃-type hysteresis loop). In addition, the pore size distribution for the synthesized samples concentrated around 4–12 nm, which classifies them as mesoporous material²⁵. As presented in Table 2, the area of the surface for the BFO nanoparticle was 5.2 m²/g, where as it increased to 12.8 m²/g for the BFO nanorod after morphological changes, and the total pore volume for the nanoparticle was 0.0213 nm, which was increased to 0.0411 nm for the nanorod. These outcomes imply that nanorods could have greater photocatalytic effectiveness than nanoparticles. Similarly, other photocatalytic studies have shown high degradation rates (100%), despite a low value of BET (13m²/g)⁴³.

Fig. 4.

(a) N₂ adsorption-desorption isotherms of BFO, nanoparticle and nanorod morphologies (b) BJH pore size distribution for BFO samples.

Table 2.

Surface area, total pore volume, and mean pore diameter of BFO.

Catalyst	BET SSA (m2/g)	Mean Pore Diameter (nm)	Total Pore Volume (m 3/g)
BFO nanoparticle	5.230	11.4	0.0213
BFO nanorod	12.836	21.007	0.0411

Optical properties

An investigation of the optical properties was carried out using DRS and PL analyses for BFO with various morphologies. The prepared photocatalysts’ UV-Vis DRS spectra (Reflectance), as illustrated in Fig. 5(a), both BFO nanorod and nanoparticle, illustrate the light reflectance within the visible light spectrum. However, the reflectance intensity for the nanorod structure was decreased compared to nanoparticles, demonstrating the enhancement in light absorption¹. Compared to the BFO nanoparticle (absorption edge around 415 nm), the BFO nanorod shows a red shift (absorption edge around 520 nm), indicating higher photocatalytic performance in the region of visible light⁸. Figure 5(b) demonstrates that, from the Tauc plots, the band gap values for the nanorod and nanoparticle samples are determined at 2.10 eV and 2.12 eV, respectively. Although there was no significant change in the band gap, the BET analysis confirmed that the area of the surface of the nanorods has increased by more than twofold concerning the nanoparticles. Numerous studies indicate that the form of semiconductors has an impact on both the absorption of light and the energy of the band gap^23,44. The slight variation in the band gap is likely attributed to the minor differences in the nanostructure of the produced BFO particles. Consequently, particles of the BFO may be a beneficial choice as a photocatalytic material for removal tasks that require the utilization of a wider UV–visible light range²⁰.

Fig. 5.

(a) UV–vis absorption spectrum, (b) Tauc plots, (c) Mott-Schottky plot of the BFO nanorod, (d) nanoparticle.

Also, the energy corresponding to the band gaps of the BFO nanostructures was determined by applying the equation (Eq. (8):

 

8

9

Here, α, hυ, , A, and represent the absorption coefficient, the frequency of the incident light, the band gap energy, and a constant parameter¹. Additionally, the quantity of n is contingent upon the type of optical transition occurring in a semiconductor, and for BFO, it is identified as 2. The quantity of the band gap for the produced particles is calculated by graphing (αhυ)ⁿ against hυ²⁴. Furthermore, the energy of the band gap () is 2.10 eV for the BFO nanorod and 2.12 eV for the BFO nanoparticle, as determined by the DRS analysis results, which is less than the formerly recorded quantities of 2.27 eV and 2.25 eV⁴⁵. The positive slopes of the Mott-Schottky curves for BFO in Fig. 5(c, d) confirm that it is an n-type semiconductor⁴⁶. The conduction band potential (CB) in n-type semiconductors is 0.1–0.3 eV less than the potential of the flat band (E_f). Following the information from Mott-Schottky diagrams, the flat-band potentials (E_f) of BFO nanorod and nanoparticle are − 0.74 and − 0.59 V (vs. Ag/AgCl, pH = 7). The flat-band potential (E_f) is convertible to the NHE, which was calculated as follows⁴⁶:

10

Hence, for BFO nanorod and nanoparticle, the conduction band energy (E_CB) was determined approximately − 0.60 and − 0.40 eV (vs. NHE). Subsequently, using Eq. (9), the V_B potential of BFO nanorod and nanoparticle can be determined, resulting in values of 1.5 and 1.72 eV⁴.

To analyze the efficacy of charge carrier capture, transport, migration, and separation in semiconductor materials, photoluminescence (PL) emission spectra are employed. PL occurs due to electron-hole pairs (free charge carriers) recombination in the valence band (VB). Consequently, a great photoluminescence (PL) intensity in a semiconductor indicates increased recombination of excited charge carriers^47–49. Figure 6 displays a strong emission peak at 605 nm is observed for BFO nanoparticles, while the intensity of this peak decreases for the nanorods. Compared to the nanoparticles, a notable reduction in charge carrier recombination occurs for BFO nanorods. This implies that the BFO nano-catalyst morphology can influence the process of charge-carrier recombination.

Fig. 6.

PL spectra showing the emission of BFO nanostructures.

Statistical assessment of the RSM procedure

The RSM was utilized to assess the effect of response factors on the MB degradation rate. An experimental design was developed using CCD to achieve optimal efficiency for MB degradation in aqueous solutions through the adjustment of reaction parameters. As presented in Table 3, an experimental design consisting of 50 runs was created through the application of CCD^50–52. The expected yield of MB photodegradation can be determined by the following Eq. (11) via the RSM approach, based on the key actual factors.

Table 3.

Design of experiment based on RSM.

Run	Lamp (watt)	Time (min)	Dose Catalyst (mg)	pH	MB Concentration (mg/L)	Result (%)
1	100	40	10	4	10	12.67
2	200	40	10	4	10	40.84
3	100	80	10	4	10	38.02
4	200	80	10	4	10	69.01
5	100	40	20	4	10	16.9
6	200	40	20	4	10	43.66
7	100	80	20	4	10	33.8
8	200	80	20	4	10	66.19
9	100	40	10	8	10	59.45
10	200	40	10	8	10	93.24
11	100	80	10	8	10	95.94
12	200	80	10	8	10	93.24
13	100	40	20	8	10	59.45
14	200	40	20	8	10	89.18
15	100	80	20	8	10	74.32
16	200	80	20	8	10	89.19
17	100	40	10	4	20	23.27
18	200	40	10	4	20	54.71
19	100	80	10	4	20	43.39
20	200	80	10	4	20	81.13
21	100	40	20	4	20	13.8
22	200	40	20	4	20	51.57
23	100	80	20	4	20	35.22
24	200	80	20	4	20	74.8
25	100	40	10	8	20	59.66
26	200	40	10	8	20	91.4
27	100	80	10	8	20	85.7
28	200	80	10	8	20	96.09
29	100	40	20	8	20	30.7
30	200	40	20	8	20	73.94
31	100	80	20	8	20	54.33
32	200	80	20	8	20	86.5
33	50	60	15	6	15	18.86
34	250	60	15	6	15	56
35	150	20	15	6	15	26.95
36	150	100	15	6	15	74.7
37	150	60	5	6	15	64.7
38	150	60	25	6	15	49.68
39	150	60	15	2	15	0.64
40	150	60	15	10	15	96.66
41	150	60	15	6	5	90
42	150	60	15	6	25	59.5
43	150	60	15	6	15	66.08
44	150	60	15	6	15	64.7
45	150	60	15	6	15	66.6
46	150	60	15	6	15	67
47	150	60	15	6	15	67.3
48	150	60	15	6	15	67.3
49	150	60	15	6	15	58.13
50	150	60	15	6	15	54.065

MB degradation (%) = 188.54 0.86 A 1.64 1.831.93.340.002 0.006 0.020.008 0.01 0.05 0.0040.280.10.410.002² 0.0036² 0.0058

11

The parameters are represented as: () LED lamp, (Irradiation time, () Dose catalyst, () pH, () Initial MB concentration.

As illustrated in Table 4, an Analysis of Variance (ANOVA) was conducted to validate the influence of significant parameters within the model. F-value of 20.08, low P-value (< 0.0001), and a high sum of square values of 28,454 demonstrated that the quadratic model is the most appropriate choice. The software indicated that the high regression coefficients (R² = 0.93, R²_adj = 0.87) between the variables and responses reflect a high level of accuracy in estimating the experimental data using the quadratic model. Additionally, a coefficient of variation (CV) of 14.12% suggests that the model is statistically significant. The value of the R-squared, anticipated to be 0.75, aligns well with the Adjusted R-squared of 0.87. Adequate Precision, which assesses the signal-to-noise ratio, ideally should exceed 4. In this case, an Adequate precision of 17.50 for the degradation of MB indicates a strong response⁵¹.

Table 4.

The analysis of variance ANOVA.

Source	Sum of Squares	Df	Mean Square	F Value	p-value Prob > F	Status
Model	28454.06607	20	1422.703304	20.08521557	< 0.0001	significant
A-Lamp	7084.913063	1	7084.913063	100.0222645	< 0.0001
B-Time	3958.707123	1	3958.707123	55.88760898	< 0.0001
C-Dose catalyst	759.0765625	1	759.0765625	10.71637098	0.0027
D-pH	13154.7663	1	13154.7663	185.7142781	< 0.0001
E-Concentration	159.5603025	1	159.5603025	2.252615189	0.1442
AB	141.1620031	1	141.1620031	1.99287459	0.1687
AC	94.35945312	1	94.35945312	1.332132956	0.2578
AD	160.2497531	1	160.2497531	2.262348606	0.1434
AE	153.4314031	1	153.4314031	2.166089584	0.1519
BC	32.26052813	1	32.26052813	0.455442579	0.5051
BD	135.5069531	1	135.5069531	1.91303862	0.1772
BE	5.942628125	1	5.942628125	0.083895895	0.7741
CD	253.1812531	1	253.1812531	3.574322231	0.0687
CE	224.5610281	1	224.5610281	3.170272148	0.0855
DE	548.5500031	1	548.5500031	7.74423243	0.0094
A2	735.2653781	1	735.2653781	10.3802132	0.0031
B2	66.78790312	1	66.78790312	0.94288769	0.3396
C2	0.687378125	1	0.687378125	0.009704158	0.9222
D2	126.5242781	1	126.5242781	1.786224432	0.1918
E2	658.5727781	1	658.5727781	9.297494552	0.0049
Residual	2054.167438	29	70.83335994
Lack of Fit	1878.734991	22	85.39704506	3.407461539	0.0507	not significant
Pure Error	175.4324469	7	25.06177813
Cor Total	30508.23351	49

R² = 0.93, Adj R² = 0.87, Pred. R² = 0.75, Std. Dev. = 8.416, CV (%) = 14.12, Adj precision = 17.50.

The influence of photocatalytic variables on the MB degradation

To demonstrate the interplay among the five parameters, the RSM was employed using Design- Expert software. In Fig. 7(a), the impact of MB concentration (5–25 mg/L) and catalyst amount (5–25 mg) on the degradation effectiveness during the photocatalytic activity is presented. The factor of catalyst dose has a significant impact on the economic efficiency, so a lower dose could be more appropriate. As the photocatalyst dosage is increased from 5 to 25 mg, the effectiveness of degradation exhibits trend of decline, from 87.86 to 58.71%. A decline in the effectiveness of degradation might result in the agglomeration of catalyst particles, which reduces the active site count on the surface available for the absorption of photons. Further, this phenomenon diminishes the intensity of light that reaches the photocatalyst⁵³. An increase in the MB concentration molecules prevents the adsorption of irradiated photons on the catalyst particles, which leads to a decrease in.

Fig. 7.

Design- Expert software plot; (a-c) 3D plots of the parameters, d) Diagrams of normal probability from MB removal using BFO.

photocatalytic yield. Figure 7(b) indicates that the MB degradation improves when the time of exposure to irradiation is raised from 20 to 100 min. When the time of illumination is short, the light may not produce species of reactivity on the surface of the photocatalyst, leading to a low photodegradation efficiency of MB dye. Prolonged light exposure enhances the absorption of photons, resulting in the production of species of reactive species. The degradation process and overall efficiency are enhanced by the effective interaction between MB molecules and reactive species⁵⁴. Additionally, Fig. 7(b) illustrates the impact of light intensity by LED lamps on the MB degradation. An increase in the power of the LED lamp (50 to 250 W) improved the percentage of MB degradation. The highest efficiency of MB degradation, recorded at 64.26%, was observed at a light power of 250 W.The rise in intensity of irradiation improves electron excitation, resulting in the production of a greater number of active radicals³⁶. Moreover, pH is a pivotal factor in the process of photodegradation. According to Fig. 7(c), a rise in pH (from 2 to 10) leads to an improvement in the effectiveness of MB degradation. The highest efficiency of MB degradation was recorded, 0.64%, 67.5%, and 96.6%, at pH = 2, 6, and 10. According to the relevant figure, under alkaline pH, at the presence of hydroxide ions () photocatalyst surface attracted cationic MB dye. In contrast, under acidic conditions, the positive charge () on the surface causes repulsion of the MB. As a result, it is expected that under alkaline environments, the adsorption of positive charge pigment molecules on the catalyst increases, and the degradation effectiveness will be enhanced⁵⁵.

In Fig. 7(d), the normal probability plotsof the residuals are displayed.The graph illustrates a compact scatter of points. This correlation illustrates the model’s usefulness and a High level of precision. Based on the statistical analysis, the experimental data did not exhibit notable deviations from the predicted values⁵⁶.

Optimization of the photodegradation process

The optimal conditions for each operational parameter are demonstrated in Fig. 8. The optimal degradation of MB was attained by following the reaction conditions. Under optimum conditions, pH: 8.97, MB concentration: 5.02 mg/L, dosage of catalyst: 19.52 mg/L, time of irradiation: 20 min, and lamp power: 195.18 W, the expected optimal removal rate is 100%. Additionally, the effectiveness of 98.2% was recorded under experimental conditions. This highest efficiency was achieved in the shortest optimal time.

Fig. 8.

Optimum condition of operational factors.

Photocatalytic performance

The effectiveness of BFO as a photocatalyst for MB degradation was evaluated under visible light (home LED lamp) and sunlight over time. In Fig. 9(a, b), the UV–vis absorption spectrum of MB dye demonstrates a gradual reduction in the peak intensity at approximately 664 nm for 60 min, which implies that the photocatalyst is causing the breakdown of MB dye. The control experiment (conducted without a photocatalyst) revealed that the concentration of MB dye remained virtually steady over time, implying the stability of the MB dye and the absence of a self-photolysis process⁴³. Moreover, the required time to reach equilibrium of adsorption-desorption between the photocatalyst and the MB solution is 30 min in a dark environment. Also, during the dark time, very nominal decolorization was detected. It is clear that in Fig. 9(c), only 76% of MB dye was removed in 20 min when BFO nanoparticles were present, whereas the percentage of degraded MB dye rose to 98.2% in the presence of BFO nanorods. Moreover, as depicted in Fig. 9(d), the degradation of MB dye under natural sunlight with BFO nanorods achieved a rate of 91.3% (0.0277 min⁻¹), following 90 min. Furthermore, in comparison to BFO nanoparticles (0.0715 min⁻¹), BFO nanorods (0.2021 min⁻¹) exhibited a greater rate of MB dye degradation when exposed to visible light (LED lamp), as depicted in Fig. 9(e, f) and Table 5. These conclusions imply that the high removal efficiency is influenced by the morphology of the nano-catalysts, as the BFO nanorods exhibited greater dye degradation under visible light and sunlight in comparison to the BFO nanoparticles.

Fig. 9.

(a) MB absorption spectra by BFO nanorod visible light (LED lamp), (b) sunlight exposure, (c) MB degradation via photocatalytic processes under visible light and (d) sunlight irradiation, (e) degradation kinetics under visible light (f) sunlight, (19.52 mg BFO, 100 ml MB, pH = 8.97, 5.02 mg/L, 195.18 W).

Table 5.

The photocatalytic performance of the MB degradation.

No.	Photocatalyst	Degradation (%)	R Square	Rate constant K (min−1)	Time (min)
1	BFO nanoparticle (LED lamp)	76	0.9572	0.0715	20
2	BFO nanorod (LED lamp)	98.2	0.99113	0.2021	20
3	BFO nanorod (sunlight)	91.3	0.97907	0.0277	90

Photodegradation mechanism

To determine the mechanism of MB dye photodegradation by BFO nanorods, active species trapping tests of photocatalytic, along with calculations of conduction band and valence band potentials (based on DRS analysis results), were conducted. Scavengers, including BQ, IPA, and EDTA, were utilized to capture ·, ·, and correspondingly. As presented in Fig. 10(a), the effectiveness of dye degradation decreased dramatically from 98.2% under no scavenger situation to 20.1% when BQ was used as · scavenger. Adding the BQ (a scavenger for ·) decreased the dye degradation greatly, indicating that the charge carriers generated by light contribute significantly to this photocatalytic activity. Conversely, the incorporation of the IPA scavenger for · had no significant impact on the photocatalytic dye degradation activity. However, adding EDTA (a scavenger for ) slightly impedes the degradation of dye, suggesting that has a minimal impact on the destruction rate. Consequently, under visible light irradiation · had a most significant impact on the degradation of dye by BFO nanorods, afterwards and · locations in the second and third, respectively ·>⁺> ·.

Fig. 10.

Influence of scavengers (a) and ions (b) on the photodegradation of MB using BFO nanorods.

According to the conclusions obtained from scavenger experiments (active species), DRS, and Mott-Schottky analyses (the potentials of VB and CB), the proposed photocatalytic mechanism in the visible/BFO system for the MB photodegradation through the use of the BFO has been presented in Fig. 11. According to Eqs. (12–16), when the BFO photocatalyst is subjected to visible light, it is capable of absorbing photons that have energy identical to or greater than its energy of band gap (≥2.1). This process leads to the excitation of a significant number of electron-hole pairs, and the generated charges () move from the valence band (VB) to the conduction band (CB). Consequently, the generated electrons are capable of reacting with the O₂ present on the surface to produce superoxide reactive radicals (·⁻), subsequently facilitating the photodegradation of pollutants. Moreover, because the values of potential ·/ and /H₂O (+ 1.99 eV and + 2.27 eV, respectively) are greater than the BFO nanorod valence band energy in the Mott-Schottky diagram, the holes in the valence band ) of the photocatalyst cannot interact with water (H₂O) and hydroxyl ions () to generate hydroxyl radicals (·). Consequently, methylene blue dye molecules are oxidized by the holes present in the valence band of the BFO nanorod. Ultimately, the reactive species that significantly contribute to the decolorization of methylene blue are superoxide radicals (·⁻) and holes (⁺)²⁴.

12

13

14

15

16

Fig. 11.

Suggested schematic depicting the transfer of charge carriers in BFO nanorods under visible light.

The impact of anions

The effect of three common anions () on the MB photodegradation by BFO was investigated^57,58. Anions were added to the MB solution at concentrations of 0.1 mM, and the relevant findings are illustrated in Fig. 10(b). After 20 min under visible light, the removal of the BFO catalyst decreased from 98.2 to 97%, 93%, and 90% in the presence of , SO₄²⁻, and Cl⁻, respectively. Adding ions to the solution had no notable influence on MB degradation. In addition, the catalyst’s activity was minimally affected by sulfate ions [58]. It is possible that ions functioned as scavengers for · radicals and holes via the reaction [Eqs. (17,18)]:

17

18

The reaction of with · and can produce the formation of radicals () that are less reactive. It is clear that chloride ions have a more significant effect on the removal than other ions. This can be explained through the trapping of hydroxyl radicals by the Eq. (19):

19

The lower reactivity of hydroxyl radicals compared to ions of chloride, adding had a negative effect on the degradation. For the BFO catalyst, as discussed in the scavenger section, and · were not the dominant active species in the MB degradation. Generally, organic material degradation is not notably influenced by the mentioned ions⁵⁹.

Reusability of BFO photocatalytic

Reusability tests were performed to evaluate the stability of BFO nanorods. A stability assessment of the photocatalytic sample was conducted over five consecutive cycles for the photodegradation of MB. Following each usage period, the BFO nanorod catalyst was reactivated, and its performance was assessed. Figure 12 depicts the BFO photocatalyst stable experiment in MB dye photodegradation. The figures imply that the BFO nanorod sample maintained significant photocatalytic activity after undergoing five consecutive cycles of MB dye photodegradation. The analysis of the first and last cycles of the stability test shows that the photocatalytic performance of the sample diminished by less than 20%, suggesting that the photocatalyst demonstrated commendable stability throughout the test. The decrease in process effectiveness is caused by the diminished amount of the catalyst during both its recovery and photodegradation. Furthermore, the agglomeration of particles after centrifugation, which causes a reduction in specific surface, can hinder photocatalysis^53,60.

Fig. 12.

Reusability of BFO nanorod in the degradation process of pollutants.

Table 6 displays the outcomes of pollutant photodegradation for different BFO composite photocatalysts, highlighting that nanorod (1D)-shaped BFO exhibits enhanced photocatalytic effectiveness at a lower time relative to earlier research findings.

Table 6.

Comparative BFO composites photocatalyst for the photodegradation of pollutants.

Catalyst (Amount, mg)	Conditions	Efficiency (%)	Ref.
BiFeO3/MOF (30 mg)	80 ml (MB 20 mg/l) - visible light/100 min	93	61
CuS/BFO (25 mg)	100 ml (alachlor 5 mg/l)- visible light/60 min	95	62
BiFeO3/GdFeO3 (10 mg)	10 ml (MB 10 mg/l) - Natural sunlight/180 min	56	63
BiFeO3/g-C3 N4 (50 mg)	25 ml (Rh B 40 mg/l) - natural sunlight/40 min	96	64
BFO (20 mg)	100 ml (MB 5 mg/l)- visible light/20 min	98.2	present study
BFO (20 mg)	100 ml (MB 5 mg/l)- natural sunlight/90 min	91.3	present study

Conclusions

Two different morphologies of bismuth ferrite perovskite, including nanorod and nanoparticle, were prepared using a hydrothermal technique. The as-synthesized photocatalysts were characterized using various analyses such as TEM, XRD, BET, DRS, Mott-Schottky, and PL. The TEM analysis demonstrated the successful synthesis of nanorod and nanoparticle morphologies. Following the BET analysis, the ratio of the surface area of nanorods (12.83 m²/g) to nanoparticles (5.23 m²/g) was more than doubled. Additionally, the total pore volume of the samples was 0.04 cm³/g for the nanorod and 0.02 cm³/g for the nanoparticle. The DRS analysis revealed that both morphologies of BFO are capable of responding to visible light for photocatalytic reactions. However, nanorod morphology displayed red shift (within the visible light range) and decreased reflectance intensity compared to nanoparticles, indicating higher photocatalytic activity. The band-gap values of the nanorods and nanoparticles were determined at 2.10 eV and 2.12 eV. The synthesized nanorods displayed reduced PL in comparison with the nanoparticles, suggesting efficient charge transfer within the bands and a delayed recombination electron-hole process. Performance of the photocatalytic samples was compared via the MB decolorization process. The optimization was applied through the RSM-CCD technique for variables: (pH: 8.97, MB concentration: 5.02 mg/L, catalyst dosage: 19.52 mg/L, time: 20 min, Lamp power: 195.18 W) under the noted optimal conditions. The optimal photodegradation yield for BFO nanorods and nanoparticles was obtained at 98.2% and 76% in 20 min, respectively. Also, the BFO nanorod removed 91.3% of the MB dye in 90 min under natural sunlight. The superoxide radicals and positive holes were vital for the photodegradation of MB by BFO nanorods under visible light. Further, the BFO nanorods illustrated excellent photo-stability after 5 cycles of the MB degradation process.

Author contributions

Material preparation, data collection and analysis were performed by F.A., M.M., and M.T. All authors contributed to the study conception and design. M.M. was the senior supervisor of the study. All authors are fully aware of this manuscript and have permission to submit the manuscript for possible publication.

Funding

M. M. thanks the University of Ilam, for the award research fellowship.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.