Photocatalytic degradation of textile dye utilizing biocl: synthesis, optimization, characterization and toxicity assessments

Abstract

In this study, BiOCl was synthesized hydrothermally to optimize effective synthesis parameters, which had previously not been addressed systemically. In this regard, optimal synthesis parameters such as temperature (157 °C), pH (8.97), residence time (23.96 h) and concentration of mannitol (0.458 mol/L) were specified utilizing central composite design (CCD) methodology. In order to investigate characterizations of the photocatalyst, various techniques such as FTIR, XRD, DRS, FESEM, EDX, EIS and mapping were employed. Moreover, the influence of various operational parameters such as pH (5, 7, 9), catalyst dosage (0.1, 0.5, 1 g/L) and pollutant concentration (10, 20, 50 ppm) were assessed. The obtained results revealed that at optimal operational conditions (pH = 5, catalyst dosage = 0.5 g/L and pollutant concentration = 20 ppm), the dye is removed completely within 90 min. Furthermore, in order to investigate the degradation pathway due to the generated intermediates, the secondary wastewater was analyzed by GC-MS. Finally, toxicity evaluation of generated species was assessed by T.E.S.T. The obtained results demonstrated that the toxicity of generated intermediates is about AO7 or lower, which, proves that this photocatalyst demonstrates effective performance.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-25451-0.

Introduction

Serious worldwide environmental contamination has occurred due to excessive application of organic materials^1,2, resulting in significant harm to human health and ecosystems^3,4. Numerous methods and technologies have been employed to deal with these organic materials⁵; however, these methods have to face the challenge of high stability of these organic pollutants^2,6,7. Certain limitations such as low efficiency, inadequate treatment and elevated costs are commonly exhibited by the current pollution removal techniques². To tackle environmental pollution, photocatalysis processes can be considered as most effective methods due to being cost-effective, eco-friendly, and efficient^8–11. To investigate the degradation of organic pollutants, common photocatalysts, notably TiO₂ and ZnO, are extensively utilized^12–14. Due to the wide band gap of most of the common photocatalysts, the irradiation adsorption of them is limited to the Ultra Violet region (UV)^8,9,15. Furthermore, other reasons such as low activity of photocatalysts and slow reaction speed lead to some challenges to degrade environmental pollutions^10,13,16,17. Over the years, numerous photocatalysts have been synthesized by various researchers to address the issue of environmental contaminations^10,13,17. To modify photocatalysts, two different strategies are suggested. The first approach is to reduce the band gap of photocatalysts, and the second approach is to decrease recombination (migration of electrons and holes) rates^18–20. Additionally, another approach to improve catalytic efficiency that has been suggested by numerous studies is to increase the specific surface area of photocatalysts. Among different photocatalysts, bismuth oxychloride (BiOCl) has attracted a lot of attentions among researchers based on its favorable attributes such as unique layered structure (21–23), narrow band gap, suitable photocatalytic activity and chemical stability, which provide suitable photoresponse and excellent physiochemical properties^24,25. BiOCl can degrade organic pollutants under UV irradiation^26,27, and numerous studies have demonstrated that BiOCl can degrade dyes and organic pollutants under significant modifications²⁸. Another issue which can impact on large-scale applications and fabrication efficiency of photocatalysts is the complexity of photocatalytic modification techniques^24,26,29. Although these modification techniques result in extending the optical attributes of BiOCl into a visible region, the rate of pollutant degradation changes to a low amount, and in some cases, it takes several hours to achieve 80% removal of the pollutant. This issue limits utilization of this powder in addressing environmental contaminations^24,29–31 (For more information about feasibility of large-scale application, refer to Supporting Information).

Recently, publications have presented papers focused on modifications of BiOCl’s structure to improve photocatalyst performance^26,32. These studies investigated the effect of different crystallinity structures such as flowerlike, spheroidal, lamellar, and filamentous^31,33,34 on the efficiency of photocatalyst. The obtained results revealed that these modifications led to visible light adsorption improvement and acceptable degradation performance. The researchers believe that improving the hierarchical structure of synthesized BiOCl leads to improvement in light reflection, light utilization, generation of electrons and holes, and increases specific surface area and active sites, thus improving photocatalytic degradation performance under sunlight^30,31,33,34. The separation and transfer of charge carried inside the photocatalyst can be facilitated by (110)/(001) the crystal surface of BiOCl¹⁵.

Our literature review has revealed that in order to employ BiOCl as an effective photocatalyst, there are various research gaps that should be addressed. So, some important highlights of the present study are listed below:

• Impacts of synthesis parameters such as temperature, residence time, pH and additive concentration on photocatalyst performance have been investigated.

• Performance of photocatalyst in dye removal was maximized by employing response surface method (RSM), particularly the central composite design (CCD).

• Performance of the optimum sample was assessed under various operational conditions such as dosage of photocatalyst, pollutant concentration and pH of wastewater.

• Secondary wastewater was analyzed toxically by T.E.S.T and @Microtox.

Methods and materials

Materials

Bi(NO₃)₃.5H₂O (99≥, Merck), KCl (99≥, Merck), D-Mannitol (99≥, Merck), NH₃ solution (28≥, Merck), HNO₃ (65≥, Merck), N-Methyl-2-pyrrolidone or NMP (99≥, Samchun), PVDF (Arkema), carbon black (Dr. Mojallali) and Acid Orange 7 (99≥, Merck) were purchased from local shops. All the chemicals are analytically pure and were utilized without further purification.

Photocatalyst synthesis

Our current approach to synthesise BiOCl is in line with the literature³⁵. Moreover, our study of the literature showed that there is hardly any evidence of optimizing the synthesis conditions of BiOCl. Using the central composite design (CCD), the current work sought to optimize the synthesis parameters including additive concentration (D-Mannitol), temperature, residence time (t_R), pH with respect to pollutant degradation response.

The synthesis of bismuth oxychloride entails the production of an initial solution of 60 mL of deionized water that includes 0.1 mol/L of mannitol and 1 mmol of potassium chloride. Treating the solution with 0.1 mol/L ammonia/nitric acid solution changes its pH to match the recorded values in Table 1. Afterwards, 1 mmol of bismuth nitrate pentahydrate is added to the solution, and the mixture is stirred by ultrasonic and magnetic devices for twenty minutes. Then, the solution is transferred to autoclave and the autoclave is placed in an electric oven at a specified temperature for a specific time period (t_R) as shown in Table 1. After the autoclave cools naturally, the generated powder is washed with distilled water and subsequently dried under an electric heater set at between 60 and 70 °C (For more information about the optimization procedure, refer to SI file).

Table 1.

Design experiments using central composite design and response.

STD	Run	Temperature (°C)	Mannitol Conc.(Mol/L)	pH	tR (h)	Response
28	1	160	0.55	7	13.5	72.74
22	2	160	0.55	9	13.5	86.21
17	3	120	0.55	7	13.5	63.8
7	4	140	0.775	8	8.25	66.58
29	5	160	0.55	7	13.5	64.70
27	6	160	0.55	7	13.5	63.95
1	7	140	0.325	6	8.25	57.52
5	8	140	0.325	8	8.25	70.62
2	9	180	0.325	6	8.25	73.8
6	10	180	0.325	8	8.25	73.80
23	11	160	0.55	7	3	76.83
30	12	160	0.55	7	13.5	61.24
18	13	200	0.55	7	13.5	56.89
13	14	140	0.325	8	18.75	76.60
15	15	140	0.775	8	18.75	85.94
25	16	160	0.55	7	13.5	68.3504
3	17	140	0.775	6	8.25	77.03
19	18	160	0.1	7	13.5	47.38
8	19	180	0.775	8	8.25	62.51
21	20	160	0.55	5	13.5	64.25
12	21	180	0.775	6	18.75	65.92
20	22	160	1	7	13.5	75.84
24	23	160	0.55	7	24	70.84
9	24	140	0.325	6	18.75	42.39
10	25	180	0.325	6	18.75	52.10
14	26	180	0.325	8	18.75	64.06
16	27	180	0.775	8	18.75	79.21
26	28	160	0.55	7	13.5	62.92
4	29	180	0.775	6	8.25	70.79
11	30	140	0.775	6	18.75	63.56

Characterization

FTIR (Fourier Transform Infrared Spectroscopy), XRD (X-ray Diffraction), and DRS (Diffuse Reflectance Spectroscopy), FESEM (Field Emission Scanning Electron Microscope), EDS (Dispersive X-ray Spectroscopy) and mapping analyses were used to probe and define the produced bismuth oxychloride powder. Furthermore, in order to identify degradation intermediates, GC-MS analysis was utilized. Moreover, EIS (Electrochemical Impedance Spectroscopy) analysis was implemented in order to investigate the coupled mechanisms of dye removal.

Degradation experiments

To assess the photocatalytic activity of the synthesized powders, a specific experimental procedure was designed. Initially, 200 ml of an Acid Orange 7 dye solution with a defined concentration (typically 20 ppm) was prepared. A measured quantity of the photocatalyst (usually 0.1 g) was then added to the solution. The mixture was stirred magnetically in a dark environment for 30 min to achieve equilibrium. After this, a portion of the solution was sampled, centrifuged at 10,000 rpm for 10 min, and its absorbance was measured using a spectrophotometer. The photocatalytic degradation process was then conducted under UVC irradiation for 30 min (t_D= time of physical adsorption + time of photocatalytic degradation), after which another sample was collected, centrifuged, and its absorbance was measured again. Furthermore, the color removal, considered as the response level is calculated by Eq. 1:

1

In this equation C₀ and C_t refer to the initial concentration and concentration of dye at time t, respectively. Additionally, the diagram of the experimental setup used in this study is presented in Fig. 1.

Fig. 1.

Schematic display of applied photoreactors.

One of the most important parameters affecting photocatalytic processes is pH. To investigate this parameter in this study, some solutions containing 20 ppm AO7 dye were prepared and their pH were adjusted to 5, 7 and 9 by utilizing ammonia/nitric acid (0.1 mol/L). Then, 0.1 g of optimized synthesized photocatalyst was added to 200 ml of the solution and stirred for about 30 min to achieve equilibrium. In this regard, the solution was irradiated with UVC light for 60 min and the absorbance was recorded after each 10 min. To investigate the validation of the experiments, each one of them was repeated at least two times.

Another parameter that can be considered a critical parameter is the dosage of photocatalyst. To investigate its effect, at first, different solutions containing 200 ml of AO7 dye were prepared. Different dosages of photocatalysts (0.1, 0.5 and 1 g/L) were introduced to them. The degradation process was performed based on the previous procedure.

Another parameter that can affect photocatalyst performance is the initial concentration of contaminant. To examine this parameter, different solutions containing 200 ml of AO7 solutions (10, 20 and 50 ppm) were prepared. Next, the same degradation processes were followed as mentioned above.

To investigate the durability of electrodes, a solution (200 ml) containing AO7 (20 ppm) was prepared. Next, 0.1 g of synthesized BiOCl (Opt6) was added to the reservoir and the absorption in the reservoir was measured after physical adsorption and photocatalyst degradation. Additionally, residual powder was washed with distilled water, alcohol and acetone respectively. Furthermore, the powder was dried in an oven over night and utilized for another run. This process was repeated 7 times.

Results and discussion

Experiment design

The results of degradation experiments are demonstrated in Table 1:

Based on the response level, the Design Expert software suggested the following models in both actual (real levels base) and coded (coded levels base) coefficients as demonstrated in Eqs. 2 and 3, respectively. In Eq. 2 T, C, t and pH refer to temperature, mannitol concentration, residence time (t_R) and pH, respectively. Additionally, in Eq. 3 A, B, C and D refere to temperature, mannitol concentration, pH and residence time (t_R), respectively.

2

3

Table 2 compares the adaptability of different models with experimental results:

Table 2.

Comparison of different proposed models.

Source	Sum of Squares	df	Mean Square	F-value	p-value
Mean vs. Total	2.729E + 05	1	2.729E + 05
Linear vs. Mean	2512.55	4	628.14	10.06	< 0.0001
2FI vs. Linear	1475.74	6	245.96	6.16	< 0.0001
Quadratic vs. 2FI	947.16	4	236.79	10.55	< 0.0001	Suggested
Cubic vs. Quadratic	368.17	8	46.02	2.65	0.0209	Aliased
Residual	642.16	37	17.36
Total	2.789E + 05	60	4647.65

As shown in Table 2, different models were investigated and based on p-value (< 0.0001) and F-value (10.55), the quadratic model was suggested by the software. Furthermore, the suitability of quadratic model was investigated by its lack of fit statistical merit as demonstrated in Table 3.

Table 3.

Comparison of lack of fit index for the proposed models.

Source	Sum of Squares	df	Mean Square	F-value	p-value
Linear	2810.14	20	140.51	7.89	< 0.0001
2FI	1334.40	14	95.31	5.35	< 0.0001
Quadratic	387.23	10	38.72	2.9	0.0539	Suggested
Cubic	19.06	2	9.53	0.5354	0.5901	Aliased
Pure Error	623.10	35	17.80

As demonstrated in Table 3, the p-value of lack of fit merit for the quadratic model is reported above 0.05, indicating lack of fit index is not significant statistically. Consequently, the model can be utilized for predictive purposes. In this vein, the analysis of variance (ANOVA) for the suggested quadratic model is presented in Table 4.

Table 4.

Analysis of variance (ANOVA) of suggested (quadratic) model.

Source	Sum of Squares	df	Mean Square	F-value	p-value
Model	4935.46	14	352.53	15.70	< 0.0001	significant
A-Temp	5.44	1	5.44	0.2423	0.6249
B-mannitol Conc	1281.89	1	1281.89	57.10	< 0.0001
C-pH	1140.72	1	1140.72	50.81	< 0.0001
D-Time	84.49	1	84.49	3.76	0.0587
AB	46.64	1	46.64	2.08	0.1564
AC	113.55	1	113.55	5.06	0.0295
AD	97.72	1	97.72	4.35	0.0427
BC	237.42	1	237.42	10.57	0.0022
BD	212.23	1	212.23	9.45	0.0036
CD	768.18	1	768.18	34.21	< 0.0001
A²	457.78	1	457.78	20.39	< 0.0001
B²	74.94	1	74.94	3.34	0.0743
C²	214.28	1	214.28	9.54	0.0034
D²	69.83	1	69.83	3.11	0.0846
Residual	1010.33	45	22.45
Lack of Fit	387.23	10	38.72	2.9	0.0539	Not-significant
Pure Error	623.10	35	17.80
Cor Total	5945.79	59

As shown in Table 4, the p-value index is reported less than 0.0001 for the quadratic model. Due to this fact that p-value is less than 0.05, it can be concluded that the quadratic model is significant and there is an acceptable fit between the experimental data and proposed model. Additionally, the F-value parameter for the proposed model is 15.7, which can be considered as significant. This amount for F-value suggests that there is only 0.01% chance that the F-value was controlled by noise. Furthermore, due to the p-value of different parameters of the model, it can be concluded that terms B, C, AC, AD, BC, BD, CD, A², and C² have more influence on model’s response compared to other terms of the equation. In this regard, the results related to adaptability of the model are presented in Table 5.

Table 5.

Coefficient of determination for quadratic model.

Std. Dev.	4.17	R²	0.8920
Mean	67.44	Adjusted R²	0.8278
C.V. %	6.18	Predicted R²	0.7533
		Adeq Precision	15.4932

As demonstrated in Table 5, the R² value of 0.8920 exhibits a suitable correlation between the model and experimental data which indicates an acceptable accuracy for the model. In addition, there is a disparity of about 0.06 between adjusted and predicted R²s. Due to the fact that the disparity is less than 0.2, it can be concluded that there is a robust predictive capability for the proposed model. Moreover, the calculated Adeq precision value that is reflecting signal-to-noise ratio is 15.5. Adhering to the fact that this value is higher than 4, it means the model demonstrates a high level of reliability and precision in predicting response levels (for further statistical investigations refer to Supporting Information file). Based on Design Expert calculations, the proposed optimum samples are listed in Table 6.

Table 6.

Suggested synthesis conditions for maximum AO7 removal.

Name	Temperature(°C)	Mannitol Conc(mol/L)	pH	Time(h)	Response(%)
Opt1	127	1	9	24	62.83
Opt2	155	0.54	8.9	20.6	85.73
Opt3	152	0.55	9	17.7	82.91
Opt4	143	0.62	8.49	22.8	76.07
Opt5	176	0.69	5.45	200	84.64
Opt6	157	0.458	8.97	23.96	88.35

As can be seen in Table 6, the maximum degradation occurs at a temperature of 157 °C, 0.458 mol/L of mannitol concentration, pH of 8.97 and 23.96 h of residence time (t_R) in the electric oven. The degradation process that was operated under the above-mentioned conditions led to 88.35% removal of AO7 dye in 60 min (t_D = 30 min for physical adsorption and 30 min for photocatalytic degradation). Obviously, since the R² of the model is about 0.9, to achieve the maximum degradation, different proposed optimum samples were tested, and Opt6 demonstrated the maximum degradation efficiency. In the following, Opt6 is used for characterizations and investigation of operational factors.

Characterization

FTIR

To investigate formed vibrations and bonds FTIR analysis was conducted. The result of Opt6 powder analysis by FTIR in the range of 400 to 4000 cm^− 1 is illustrated in Fig. 2:

Fig. 2.

FTIR spectra of Opt6 powder.

As shown in Fig. 2, the observed peak around 1561 cm^{− 1} is responsible for bending vibration of O-H groups. Certainly, the peak’s absence around 1600–2500 cm^{− 1} range leads to water’s absence in BiOCl molecule and confirms the high purity of synthesized BiOCl³⁶. Additionally, the stretching vibration of Bi-O bonds can be concluded from the peak in the vicinity of 547 cm^{− 137}. Further evidence of effective synthesis of BiOCl is provided by the peaks between 1035 and 1461 cm^{− 1}, which are ascribed to Bi-Cl bonds.

XRD

To investigate the crystallography of the synthesized powder XRD was conducted³⁸ and the results of analysis are displayed in Fig. 3.

Fig. 3.

XRD spectra of Opt6 powder.

The peaks appearing at 10.4°, 11.9°, 23.7°, 25.8°, 31.1°, 32.5°, 33.1°, 34.6°, 35.9°, 40.7°, 46.7°, 49.0°, 54.1°, 58.3°, 60.1°, 68.2°, 69.4°, 74.2°, 77.7°, and 82.4° can all be considered as characteristic peaks for XRD analysis. To investigate the crystal structure of BiOCl, the above- mentioned peaks were compared with standard Jade ASCII patterns. The results revealed that the synthesized material corresponds to tetragonal phase of bismuth oxychloride. The lack of additional characteristic peak demonstrates high purity of synthesized BiOCl powder. Furthermore, calculated lattice parameters are presented as follows: a = b = 3.8910 nm, c = 7.3690 nm and α = β = γ = 90°. The same results were mentioned in reference³⁵.

UV-visible DRS

One of the most important factors which plays a major role in investigating photocatalytic performance, is band gap energy. To characterize light response and analyzing the optical properties of samples, UV-Visible DRS analysis was performed and the obtained results are presented in Fig. 4:

Fig. 4.

(a) UV-Visible diffuse reflection spectra, (b) (αhυ)² versus hυ for Opt6 powder.

As demonstrated in Fig. 4a, the maximum photon adsorption occurred at 295 nm wavelength, which belongs to UVC region. Additionally, the band gap of Opt6 powder was examined at about 3.3 eV. These findings indicate that optimizing the synthesis parameters by RSM method can decrease band gap of BiOCl powder from 3.5 eV to 3.3 eV as mentioned in reference³⁹.

FESEM

To investigate the geometrical structure of synthesized powder, FESEM analysis was performed and the obtained results are shown in Fig. 5.

Fig. 5.

(a–d) FESEM images at various zoom values (starting from the top left image, a rectangular area is specified in the image, and that area is sequentially zoomed in the subsequent images for a larger view of that area).

As demonstrated in Fig. 5a, it can be concluded that the synthesized powder has a plate-like morphology. To examine further, different specified parts of the sample were zoomed and all the results from Fig. 5b and d demonstrate this fact. Plate-like shape morphology in photocatalyst causes higher surface area to adsorb UVC irradiation and pollutant molecules. The same findings were reported in reference⁴⁰. By comparing the results of the present study with the Nonthing study⁴¹, it can be concluded that the systematic adjustment of synthesis parameters will result in more effective morphology for photocatalytic application. This has become the case, because in the present study, we have achieved plate-like morphology, whereas in the other study, they had synthesized a microsphere structure, which has lower surface area and active sites in comparison with a plate-like structure. Furthermore, to confirm this interpretation, BET test was employed and the results were illustrated in SI file. The obtained results demonstrated, that surface area of Opt6 sample is about 27.9 m²/g, and for the same study by Nonthing, it is about 22.4 m²/g. As a result, there were more active sites, and that bigger number of reactive species can be released during photocatalytic experiments.

EDS

To identify and quantify the elements present in the sample, EDX analysis was performed and the results are shown in Fig. 6.

Fig. 6.

Obtained results of EDX and elemental analysis of Opt6 powder.

As demonstrated in Fig. 6, the elements bismuth, oxygen and chlorine are presented by observable and specified peaks. Furthermore, other observable peaks are related to the gold element, which are added due to gold coating of the sample. Additionally, gold related peaks have not been considered in quantitative analysis. Moreover, atomic and weight percentages of the sample are reported in Fig. 6. The results are in agreement with the results of reference⁴².

Mapping

To investigate the distribution of elements of the sample, mapping analysis was conducted and the results are shown in Fig. 7, which indicate that Bi, Cl and oxygen elements were distributed homogenously.

Fig. 7.

EDX elemental mapping of (a): Bismuth, (b) Chlorine, (c) Oxygen and (d) mixture in Opt6.

EIS

In order to obtain further insights about mechanisms governing the dye removal process, electrochemical impedance spectroscopy (EIS) test was performed. In this regard, a solution containing Opt6, PVDF and carbon black (2:1:1) in NMP was pasted on a titanium plate and the experiment was conducted in an AO7 electrolyte (20 ppm.). The results are shown in Fig. 8:

Fig. 8.

Electrochemical impedance spectroscopy results of Opt6.

As shown in Fig. 8, Nyquist plot comprises of two major parts, a semicircle in the high- frequency zone and a linear segment in the low-frequency zone. The semicircle part refers to electron transfer process, which leads to occurrence of a reaction between BiOCl and electrolyte. Due to photoreactions (mentioned in subheading 3.3.2), this analysis can be considered reasonable. Moreover, the linear region corresponds to ion diffusion within the electrolyte towards the electrode surface, which seems reasonable with respect to the physical adsorption experiments. Consequently, dye removal process in present study is a combination of physical adsorption and chemical degradation⁴³.

Degradation experiments

Kinetic investigation

In this study, three different kinetic models such as pseudo first-order (), pseudo second-order () and Langmuir-Hinshelwood () were utilized to investigate the kinetic of degradation experiments (excluding physical adsorption data). The results including coefficient of determination and parameters of the models are demonstrated in Table 7:

Table 7.

Adaptability of various kinetic models.

Model	K	K’		R 2
Pseudo first-order	0.0138			0.96
Pseudo second-order	0.0076			0.78
Langmuir-Hinshelwood	0.61523	6.7075		0.0055

As shown in Table 7, the performance of pseudo first-order model in adaptability to experiment data can be considered suitable. Low pollutant concentration can account for high R² value for pseudo first-order model as mentioned in reference⁴⁴. In the following, the performance of photocatalytic degradation in different situations is examined by pseudo first-order coefficient.

Effect of pH on photocatalyst performance

As shown in Fig. 9, the decrease or increase in pH levels compared to neutral state improves the performance of photocatalytic process. Another point that should be considered is that the acidic condition in pollutant reservoir demonstrates superior performance in comparison with the basic condition. In quantitative view, the pseudo first-order kinetic constant in acidic condition is approximately 35% higher than in basic condition. In the photocatalyst process, the photocatalyst material is initially exposed to UVC irradiation. Some reactive species such as superoxide radicals and oxygen ions are generated when electrons transfer from valence band (VB) to conduction band (CB) after receiving energy from photons as shown in Reactions 1 to 4.

Fig. 9.

Effect of solution pH on (a) dye decrease and, (b) pseudo first-order coefficient.

(Reaction 1).

(Reaction 2).

(Reaction 3).

(Reaction 4).

(Reaction 5).

Furthermore, another reactive species like hydroxyl radical is generated when the hole generated in valence band combines with hydroxide ions as demonstrated in Reaction 5. According to the outlined mechanism, the number of protons (H⁺) increases when the levels of pH decrease. Due to Coulombic forces that form between protons and electrons on the surface of photocatalyst, the performance of photocatalyst and rate of photoreactions improve. Moreover, the concentration of hydroxide ions increases when pH shifts from neutral to basic condition. Consequently, reaction rate and photocatalytic performance improve compared to the neutral state. Another notable point is the superior performance of photocatalyst in acidic condition compared to basic condition. This phenomenon can be explained as follows: based on Coulombic forces between protons and electrons in acidic situation, the hole and electron separation (Reaction 1) occurs more effectively as mentioned in reference⁴⁵. The obtained results demonstrate that only the systematic optimization of photocatalyst performance promotes the efficiency of BiOCl compared to studies^{46] and [47} which have utilized other modification techniques.

Effect of dosage on photocatalyst performance

Another operational parameter affecting the degradation process can be considered the dosage of photocatalyst. The obtained results are shown in Fig. 10:

Fig. 10.

Effect of catalyst dosage on (a) dye decrease and, (b) pseudo first-order coefficient.

As demonstrated in Fig. 10a, increasing photocatalyst dosage from 10ppm to 50ppm leads to an improvement in the rate of pollutant degradation. Furthermore, the increase in dosage levels causes increasing of pseudo first-order coefficient as shown in Fig. 10b. The explanation for this phenomenon is that an increase in the amount of photocatalyst leads to more photon adsorption in the photo-reactor. Consequently, more reactive species such as hydroxyl radicals are generated and more pollutants degrade within a certain time interval. This reasonable trend was mentioned in various references such as⁴⁸.

Effect of pollutant concentration on photocatalyst performance

As shown in Fig. 11, increasing the initial concentration of pollutant leads to a decrease of pollutant removal rate and pseudo first-order constant. When initial concentration is increased from 10ppm to 50ppm, it leads to saturation of active sites in Opt6. Furthermore, the generated intermediates during degradation process result in a competitive adsorption to active sites. The findings demonstrate the same results referred to in reference⁴⁹. Table 8 is prepared to compare the results of this study with other ones:

Fig. 11.

Effect of initial dye concentration on (a) dye decrease and, (b) pseudo first-order coefficient.

Table 8.

Comparison of obtained results with other studies.

Name	Pollutant	Photocatalyst	Degradation time (min)	Lamp	%Removal
Farzadkia et al.50	Metronidazole	ZnO/PANI	180	UV	99
		TiO2/PANI	120	UV	98
		ZnO/PANI	180	Vis	46
		TiO2/PANI	120	Vis	43
Shiekhmohammadi et al.51	Methylen blue	TCN	120	Vis	40
		Ag/ZrO2			60
		Ag/ZrO2/TCN			< 100
	Methyl orange	TCN	120	Vis	50
		Ag/ZrO2			40
		Ag/ZrO2/TCN			> 90
Haghighi et al.46	Acid Orange 7	Mn3O4	120	UV	< 5
		BiOCl			40
		Mn3O4/BiOCl			70
Aghdam et al.47	Acid Orange 7	BiOCl/C3N4	140	Vis	76.8
		BiOI/C3N4			67.2
		BiOCl/BiOI/C3N4			97.4
This study	Acid Orange 7	BiOCl	90	UVC	100

Cycle ability of photocatalyst

The results of cycle ability of photocatalyst are shown in Fig. 12.

Fig. 12.

(a) Dye removal percentage and, (b) photocatalyst performance decrease during 5 cycles.

As shown in Fig. 12a, chemical stability of Opt6 photocatalyst remains at 88% for up to five runs. The obtained results reveal that the performance of Opt6 starts to decrease about 5% with the sixth run. The formation and adsorption of AO7 and degradation intermediates leads to occupation of active sites of photocatalyst, which leads to decreased performance of photocatalyst as mentioned in reference⁵² (For more information about cycle ability of photocatalyst, refer to SI file).

Scavenger test

To investigate reactive species, methanol (Me), p-benzoquinone (BQ) and isopropyl alcohol (IPA) were utilized as trapping agents of holes, superoxide anions and hydroxyl radicals, respectively⁵³. Photodegradation and adsorption efficiency were significantly changed by addition of trapping agents as demonstrated in Fig. 13. When no radical trapper was added, pollutant concentration in photoreactor reached to 1.6 ppm after 90 min. Furthermore, after addition of methanol pollutant, the concentration reached to 4 ppm. Additionally, the addition of p-benzoquinone resulted in concentration of AO7 dye to reach 12 ppm. Moreover, utilizing isopropyl alcohol in quench experiment resulted in 16 ppm of pollutant concentration after 90 min of degradation process. These findings demonstrate that hydroxyl radical plays a major role in Opt6 degradation. As well, superoxide anion and hole can also affect degradation and adsorption process, respectively.

Fig. 13.

Effect of trapping agents on (a) dye decrease and, (b) pseudo first-order coefficient.

Suggested degradation pathway

To investigate generated intermediates for degradation experiments conducted under optimum circumstances (pH = 5, t_D= 90 min, catalyst dosage = 0.5 g/L, C₀ = 50ppm), GC-MS analysis was conducted, and based on the obtained results a pathway is suggested as demonstrated in Fig. 14.

Fig. 14.

Proposed degradation pathway of AO7.

As mentioned in various references such as⁵⁴, degradation process of azo dyes starts with cleavage of N-N as demonstrated in Fig. 14. Going forward, generated active species such as hydroxyl radicals and superoxide ions orient toward chromophore of Acid Orange 7. This phenomenon leads to formation of identified species such as I₁ (3-Methyl-2, 3-epoxycyclohexan-1-one) and I₂ (Caryophyllene-(I1)). These generated species under oxidative ring opening reactions change to new species such as I₃ (Lauryl chloride), I₄ (Hexacosane), I₅ (Tetradecane, 1-chloro-), I₆ (Heneicosane), I₇ (Nonahexacontanoic acid) and I₈ (Tetratriacontane). Finally, these intermediates tend to mineralize and produce carbon dioxide and water. The existence of chlorine atoms in species I₃ and I₅ is noticeable. Due to chlorine ion release reaction in acidic circumstances⁵⁵, which was demonstrated in Reaction 6, the formation of the above mentioned species can be considered reasonable (For more information about chlorine ions, refer to SI).

(Reaction 6).

Toxicity evaluation

In this study, the evaluation of environmental damage due to the generated intermediates during the degradation process plays a crucial role. As a result, the toxicity assessment of AO7 potential generated species was conducted by Toxicity Estimation Software Tool (T.E.S.T). To obtain insights on the bioaccumulation factor and fathead minnow’s 50% lethal dose (LC50-96 h), consensus method was employed. The results are demonstrated in Fig. 15.

Fig. 15.

T.E.S.T results, (a) fathead minnow’s 50% lethal dose and, (b) bioaccumulation factor.

As illustrated in Fig. 15, the data related to I₇ was not reported due to poor stability of the mentioned intermediate. As shown in Fig. 15a, the majority of generated intermediates such as I₄ to I₈, reveal values lower than AO7, which leads to increase in toxicity, however, values are near each other as well as AO7. Furthermore, as revealed in Fig. 15b, the same trend was observed, and the bioaccumulation factor (BAF) increased in intermediates such as I₃ to I₈. The noticeable point is that all reported BAFs are lower or around 100, which demonstrate low toxicity impact on the environment. Based on toxicity evaluation results, the species generated during photocatalyst degradation process have the same toxicity effects as AO7 and as a result, BiOCl can be considered a promising photocatalyst in wastewater treatment. Since practical assessment of toxicity of secondary wastewater is crucial, @Microtox experiments were employed. For further information about this test, refer to SI file.

Conclusion

In the present study, the central composite design (CCD) methodology was employed to achieve the most effective BiOCl photocatalyst by tuning hydrothermal synthesis parameters such as temperature, t_R, pH and manni_tol additive concen_tration. In order to characterize synthesized powder, physical, chemical and electrochemical characterizations were conducted. Moreover, the influence of operational parameters on performance and kinetic of degradation were investigated. Furthermore, the durability and degradation mechanism of photocatalytic which, is about generated reactive species were assessed. Finally, due to formation of intermediates, a degradation pathway was proposed and toxicity of generated species was assessed by T.E.S.T and @Microtox.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Amirhossein Karimirahnama (1) Data Gathering (2) Writing Original Manuscript (3) Software (4) Visualization (5) Conceptualization (6) Methodology. Mehrdad Mozaffarian (1) Funding (2) Supervision (3) Review and Editing Manuscript (4) Software (5) Validation.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.