Thermal treatment impact on the evolution of active phases in layered double hydroxide-based ZnCr photocatalysts: Photodegradation and antibacterial performance

: This study investigated the in ﬂ uence of thermal treatment on the photocatalytic performance of ZnCr layered double hydroxide-based mixed metal oxides in the degradation of methylene blue and brilliant cresyl blue organic dyes under simulated solar light irradiation. The photocatalysts were synthesized using a simple coprecipitation method and subjected to thermal treatment at temperatures ranging from 100°C to 900°C. Additionally, the study explored the anti-bacterial activity against Escherichia coli and Staphylococcus aureus using a novel antibacterial experimental setup. It not only involved the introduction of ZnCr samples into BioPeptone/prepared cell suspension to enhance photocatalyst – bacteria cell contact but also included research on antibacterial activity induced by solar irradiation and also in the absence of light, providing crucial insights into photocatalytic antibacterial activity of ZnCr photocatalysts. Despite satisfactory e ﬃ ciencies observed for all thermally treated ZnCr samples (removal e ﬃ ciency ranging from 40% to 90%), ZnCr 900 (thermally treated at 900°C) exhibited exceptional performance, achieving nearly 100% removal e ﬃ ciency and complete growth inhibition for both bacteria. Integrating these ﬁ ndings with structural and textural characterization data, as well as kinetic studies, our comprehensive analysis enhances the understanding of structure-dependent photocatalytic activities. These insights open possibilities for the application of ZnCr photocatalysts


Introduction
Over the past few decades, the large-scale discharge of various organic and inorganic pollutants into the wastewater has raised significant concerns due to their adverse impact on water eco-systems [1,2].The development of modern textile industry has contributed to the pollution of around 15% of water streams with effluent dyes from this industry [3].Removal of organic dyes emerged as a matter of high importance due to their toxic and carcinogenic properties, which pose a serious threat to water ecosystems [4][5][6].Azo dyes, which include compounds such as methylene blue (MB) and brilliant cresyl blue (BCB), constitute the largest group of textile dyes, comprising over 60% of the total dye pollutants released into water ecosystems [7][8][9].Azo dyes have toxic, cancerogenic, and bio-accumulative nature and are extremely harmful to aquatic eco-systems and human health if discharged into the environment without previous treatment [7,10,11].Furthermore, bacterial pollution of wastewater is another emerging environmental problem that represents a serious hazard to human health, as pathogenic bacteria not only exhibit toxicity to humans but also show resistance to antibiotics, requiring their removal from wastewater [12][13][14].
In the pursuit of the most effective method for wastewater purification, several factors should be considered: efficiency of the pollutant treatment process, microbial removal efficiency, simplicity, and long-term costs.Due to high stability and persistency of dye compounds, various purification methods for wastewater treatment have been researched over the past few decades, including adsorption, biodegradation, heterogeneous photocatalysis, ion-exchange, etc. [15][16][17].Compared to other purification methods that merely concentrate pollutants (adsorption, ion-exchange), environmentally friendly photocatalytic processes have advantages since they are activated by free, abundant solar radiation and lead to complete demineralization of pollutants [18,19].Advanced photocatalytic oxidation processes based on metal oxides as photocatalysts, coupled with the use of highly reactive radicals induced by solar light in situ, enable the development of cost-effective processes capable of decomposing resistant water pollutants [20,21].Moreover, recent studies on photocatalytic processes have shown great potential for antimicrobial wastewater treatment [12,13,22].
Efficiency studies of several well-known semiconductors, such as TiO 2 , Bi 2 WO 6 and SrTiO 3 , have been excessive.Additionally, metal oxides like ZnO [23][24][25], CuO [26], and Cr 2 O 3 [27] have shown promising additional functional properties, which include photoinduced antibacterial activity, enhancing their potential for use in water purification treatments [20,[28][29][30][31]. Currently, there is ongoing research focused on developing novel materials capable of overcoming the limitations of the aforementioned semiconductors, such as their relatively low charge separation and narrow light absorption spectra.For instance, Kumar et al. [32] demonstrated that Bi 2 WO 6 -TiO 2 (20% mass)-Ti 3 C 2 (5 mass%) nanocomposite degraded 98.5% of methyl green after 40 min of sunlight irradiation, outperforming TiO 2 , which achieved 74% efficiency after 80 min.Additionally, Sharma et al. [33] found that ciprofloxacin could be efficiently removed (77% in 160 min) under solar irradiation with novel ternary nanocomposites obtained from coupling Bi 2 WO 6 , ZnO, and Ti 3 C 2 nanosheets.Lately, noble metal nanoparticles have emerged as potential doping agents for photocatalysts due to their unique properties.Mohanty et al. [34] reported that the heterojunction effect created after Au decoration of SrTiO 3 and combined with Ti 3 C 2 into a ternary interfacial heterostructure led to improved photocatalytic removal of ciprofloxacin (63% in 120 min).Interestingly, Choudhary et al. [35] determined that heterojunctions formed after combining CoFe 2 O 4 with ZnO notably increased photocatalytic efficiency in MB removal (90.8% in 28 min) compared to pristine CoFe 2 O 4 nanostructures (65% in 28 min).Also, Choudhary et al. [36] reported that optimal doping (0.5%) of Ce on ZnO nanowires significantly enhanced light utilization capability in visible spectral region due to reduced band gap of ZnO, which led to an increase of 20% in photocatalytic MB removal.
Layered double hydroxide (LDH) materials have emerged as promising photocatalysts for wastewater purification, given their reported ability to degrade various dyes [2], pesticides [37], and antibiotics [38].Furthermore, these materials have exhibited antibacterial activity, which has been reported in numerous studies.Balcik et al. [39] reported that nanocomposite membranes incorporated with ZnFeCe LDH improved the bactericidal effect on Escherichia coli when compared to membranes without LDH.Additionally, Cardinale et al. [40] reported that a 150 mg disk of ZnAl-SO 4 LDH had a 1.95 cm inhibition zone against E. coli.
LDHs have the following general formula: , where M 2+ represents divalent metal cations, M 3+ represents trivalent metal cations, A n− represents anions that reside in the interlayer region, x = M 3+ /(M 2+ + M 3+ ), and m corresponds to the interlayer water amount [41,42,[44][45][46].The layered structure of LDHs consists of brucite-like (M (OH) 2 ) layers, where a portion of M 2+ cations is substituted by M 3+ cations, and the excess charge is compensated with interlayered anions.The optimal range for the synthesis of a single LDH phase has been reported when the value of x is between 0.2 and 0.4, and exceeding this range may lead to the formation of additional hydroxides or other compounds [45,47,48].The nature of the interlayer anion, A n− , determines the height of the LDH crystal unit cell, and it is possible to synthesize LDHs with different interlayer anions, such as inorganic anions (F − , Cl − , Br − , I − , (NO 3 ) − , (ClO 3 ) − , (IO 3 ) − , OH − , (SO 4 ) 2− ), heteropolyacids ((Pmo 12 O 40 ) 3− , (PW 12 O 40 ) 3− ), and organic acids (adipic, oxalic, succinic, malonic) [42,43].Nevertheless, carbonate anions have the highest affinity for incorporation during the synthesis, and it is difficult to obtain well-crystallized materials with an anion other than carbonate since carbonates are easily absorbed from atmospheric carbon dioxide into the reaction solution [43].It has been observed that an appropriate M 2+ /M 3+ ratio and careful selection of metal cations during synthesis enable the formation of favourable LDH properties [49,50].Thermal treatment of LDHs, also called thermal activation, leads to the formation of nonstoichiometric mixed oxides (MO) with numerous advantageous properties, such as strong oxidation ability, photocatalytic activity, and coupling with semiconductors [17,22].Various metals have been used to induce photocatalytic properties of LDHs and derived MOs, including Zn, Ni, Fe, Cr, Al, Co, and Cu [17,[50][51][52][53].Among these metals, zinc oxide stands out for its photocatalytic and microbial properties, being non-toxic and stable, with the ability to tailor properties by coupling with various semiconductors with a narrow energy band gap to improve its photocatalytic and antimicrobial efficiency [23,53].Furthermore, due to the favourable properties of LDHs and their growing application in different fields, several studies have been conducted to assess the toxicity and potentially hazardous effects of LDHs in living organisms.Torbati et al. [54] reported that a high concentration of ZnFe-SO 4 LDH causes stress and inhibits superoxide dismutase and peroxidase enzymes in microalgae (T.obliquus) at a concentration of 100 mg•L −1 .
Various studies on ZnCr LDH and its derived MOs have been conducted in the last decade.Mohaptra and Parida [2] demonstrated that ZnCr LDH successfully photodegraded 100% of rhodamine B and rhodamine 6G after 120 min of visible light irradiation.El Mersly et al. [55] studied the effect of different anions (CO 3 2− , Cl − , and SO 4 2− ) in the interlayer of ZnCr LDHs on photocatalytic degradation efficiency of AO7 dye (21%, 54%, and 66%, respectively).Bencherif et al. [50] concluded that after calcination of ZnCr LDH at 500°C, the photodegraded amount of crystal violet was notably higher (87.8%) than the amount degraded by the LDH sample (∼45%) due to the formation of ZnO and ZnCr 2 O 4 phases.Pausova et al. [56] also reported increased efficiency in photocatalytic removal of Orange II dye after calcination of ZnCr LDH at temperatures higher than 600°C (from 34% to almost 100%), attributing this improvement to highly crystalline ZnO and ZnCr 2 O 4 phases.Sadeghi Rad et al. [57] reported that ZnCr LDH exhibited higher efficiency in the removal of rifampicin when a combination of different advanced oxidation processes was used (sonication and photocatalysis).This study also demonstrated that ZnCr-LDH reduced the viability of S. aureus by 55.4%.
In light of specific requirements for inducing both photocatalytic and antibacterial properties, in this study, zinc (Zn) and chromium (Cr) were selected as constituent metals for the synthesis of LDHs.A simple and inexpensive low supersaturation coprecipitation method with carbonates as interlayer anions was selected for the synthesis of LDHs since it facilitates the formation of precipitates with a large number of low crystalline particles that, after thermal treatment, enables the formation of active MO phases.This research aims to investigate the influence of thermal treatment on ZnCr LDH-based materials and their photocatalytic behaviour, particularly in the degradation of selected organic dyes (MB and BCB).Moreover, standard Gram-negative (E.coli) and Gram-positive (S. aureus) model bacteria representatives were chosen for preliminary evaluation of antibacterial activity.While the existing literature has explored various aspects of BCB dye photodegradation [9,58,59], this research presents a novel approach by introducing ZnCr LDH-based mixed metal oxides as potential photocatalysts for degrading brilliant cresyl blue.Furthermore, the study introduces a novel antibacterial experimental setup aimed at not only enhancing photocatalyst-bacteria cell contact but also includes research on antibacterial activity induced by solar irradiation and also in the absence of light, thereby providing valuable insights into the photocatalytic antibacterial mechanism.By addressing these specific gaps in the literature, this study offers additional contributions to the field of photocatalysis and antibacterial materials.
2 Materials and methods

Sample preparation
Taking into consideration our previous experience and literature references [15,17,21,22,41,49,[60][61][62]64], the parameters of the low supersaturation coprecipitation method at constant pH for the synthesis of ZnCr LDHs were selected.The precursor solution (70 mol% of Zn(NO 3 ) 2 •6H 2 O and 30 mol% of Cr (NO 3 ) 3 •6H 2 O) was added continuously (4 cm 3 •min −1 ) and stirred vigorously, along with the base solution (0.67 M Na 2 CO 3 ; 2.25 M NaOH) that was used to maintain constant pH (9.4) at constant temperature (40°C).The molar ratio of constituent metals (Zn and Cr) was selected in order to enable the formation of photocatalytic active phases (value of x = Cr 3+ /(Zn 2++ Cr 3+ ) = 0.3 in the optimal range for the synthesis of single ZnCr LDH phase) that, after thermal treatment, facilitates the formation of active ZnO phase and the spinel phase, thereby avoiding the formation of additional unfavourable phases, such as Cr 2 O 3 phase and Zn(OH) 2 , which are not photocatalytically active [50].
Precipitation products were aged 12 h under the same conditions, washed with distilled water until pH = 7, and dried (24 h; 100°C in the air) and thermally treated at different temperatures (300°C, 500°C, 700°C, 900°C) for 5 h in air.To induce the formation of active phases, the temperatures for sample thermal treatment were selected considering the temperatures of the following thermal decomposition steps: (1) dehydration of LDHs (loss of the physisorbed and interlayer water) in the temperature range of 100-250°C; (2) dihydroxylation (loss of hydroxyl groups from the brucite-like layer) in the temperature range of 350-450°C; (3) decarbonation (collapse of the layered structure and formation of nonstoichiometric MO phases) in the temperature range of 420-470°C; and (4) formation of stoichiometric spinel oxides and bivalent oxides in the temperature range of 600-800°C [62,63].The obtained samples were denoted as follows: ZnCr 100, ZnCr 300, ZnCr 500, ZnCr 700, and ZnCr 900.

Characterization
Crystalline phases were identified by X-ray powder diffraction (XRD) using Rigaku MiniFlex 600 diffractometer (CuKα radiation, λ = 0.15406 nm; 2θ 10-70°; scan rate = 0.02 s −1 ).Scherrer formula (Eq. 1) and full width at half-maximum (FWHM) of intense diffraction peaks were used in order to calculate the crystallite size (CS): where D is the CS (nm), k is the shape function (0.9), λ is the X-ray wavelength, θ is the angle of diffraction, and β is the FWHM of the considered peak.The lattice parameters were calculated for the dominant phase detected by XRD analysis.The unit cell parameters a (width of the crystal unit cellcation-cation distance in Brucite-like sheets) (Eq.2), c (height of the crystal unit cell) (Eq. 3) and c′ (brucite-like sheet thickness) (Eq.4) for the LDH phase were calculated from the positions of ( 110) and (003) reflections: For the MO, a hexagonal zincite structure of ZnO, and the unit cell parameters a (width of the crystal unit cell) (Eq.5) and c (height of the crystal unit cell) (Eq.6) were calculated as [65,66]: The texture of all samples was analysed by low-temperature nitrogen adsorption at -196°C (Microtrac Belsorp Max II).The specific surface area was calculated by the Brunauer-Emmer-Teller (BET) method.The pore size distribution and cumulative pore volume were determined by the Brunauer-Joyner-Hallenda (BJH) method applied to the desorption branch of the isotherm.The absence of micropores in samples was confirmed using the t-plot method.
Zeta-potential measurements were performed by dynamic light scattering using a Malvern Instrument, model Nano ZS.The refraction index of the investigated dispersions for ZnCr-LDHs was n = 1.22 and that for MB and BCB solutions was 1.34, while the refraction index of the dispersant (demineralized water) was n 0 = 1.33.The light absorption α was 0.2.The mean zeta potential values were obtained as the average value after performing 12 scans.

Photocatalytic experiments
All photocatalytic experiments were performed in an open Pyrex vessel using Osram Ultra Vitalux 300 W lamp (I (VIS) = 20.52W•m −2 ; I (UV A ) = 17.6 W•m −2 ) with the emission spectrum that simulates solar light positioned 45 cm above the top surface of each dye solution.The photocatalytic efficiency of all prepared samples was analysed by monitoring the photodegradation of two dye pollutants: MB and BCB.The suspensions were exposed to air without additional aeration throughout the duration of the ongoing experiment.Prior to each photocatalytic test, the reaction mixtures containing 50 mg of photocatalysts and 100 mL of dye solution (C MB = 10 mg•L −1 and C BCB = 10 mg•L −1 ) were stirred in the dark for 30 min in order to establish the required adsorption/desorption equilibrium between the dye and the catalyst surface.When the equilibrium was achieved, reaction mixtures were irradiated by light, and aliquots were taken and analyzed at defined time intervals using a UV-VIS spectrophotometer (EVOLUTION 600 spectrophotometer) at maximum absorption wavelengths (λ MB = 664 nm and λ BCB = 622 nm).
The MB/BCB photodegradation efficiency, E ff (%), was calculated as follows (Eq.7): where C 0 (mg•L −1 ) is the initial dye concentration and C t (mg•L −1 ) is the concentration at the defined time, t [67].
In order to exclude the influence of catalyst dosage and pollutant concentration on the photocatalytic degradation efficiency, degradation turnover (dTON) was calculated as follows [68,69]: where C 0 and C f (μmol•L −1 ) are the initial and final concentrations of the dye, t (h) is the time, and M (g•L −1 ) is the catalyst dosage.

Kinetics of the photocatalytic reaction and artificial neural network (ANN)
The kinetics of organic dye photodegradation were calculated using the pseudo-first-order reaction rate (Eq.8) that followed the Langmuir-Hinshelwood kinetic model: where k app (min −1 ) is the apparent pseudo-first-order reaction constant, which was obtained from the slope of the ln (C 0 /C) vs time linear function.
The reaction half-time (Eq.9), which represented the time required for the initial concentration of dye to reduce by half, was calculated as follows: The electrical energy per order, E EO (kW•h•L −1 •order −1 ), is a scale-up parameter used for the comparison of different photocatalytic water systems.This parameter provides information on the necessary electrical energy for the 90% order-of-magnitude pollutant degradation in a studied volume of polluted water [69,70]: where P (kW) represents the electric lamp power and V (L) is the volume of the reaction mixture [69].
A predictive model for the photocatalytic behaviour of thermally treated ZnCr samples in the photodegradation of two organic dye pollutants was proposed.The created ANN for the photodegradation rate constant consisted of four inputs from experimental variables: the thermal treatment temperature (TTT), organic dye type (ODT), CS, and pore diameter of the most present pores (MPPD).A limitedmemory Broyden-Fletcher-Goldfarb-Shanno optimization algorithm was used in the solver function of the neural network, with the number of iterations limited to 100.In order to optimize the network training speed, the early stopping function was enabled after ten iterations with no improvements in the validation of data sets.For this type of data modelling, it is crucial to determine the right number of neurons in the hidden layer, and the trial and error method showed that the number of ten neurons in the hidden layer works best in this ANN.It was estimated that the optimal function for connecting input neurons to hidden neurons was hyperbolic tangent sigmoid (tansig).The ANN predictive model simulations were carried out using Python 3.

Antibacterial experiments
In order to assess the antibacterial activity of the samples, microbiological experiments were performed on both Gramnegative (E. coli ATCC 25922) and Gram-positive (S. aureus ATCC 25923) bacteria [30,71].Prior to microbiological analysis, both cultures were grown on plate count agar at 37°C for 24 h.Then, the cultures were suspended in 9 cm 3 of BioPeptone (HiMedia, Mumbai, India), corresponding to McFarland density number 5 (approximately 1.5 × 10 9 CFU cm −3 ).The prepared ZnCr samples (50 mg) were suspended in 100 cm 3 of BioPeptone, and 1 cm 3 of freshly prepared cell suspensions was added to obtain the final cell concentration of 6-7 log CFU•mL −1 .In order to pinpoint and distinguish solely photocatalytic antibacterial activity of photocatalysts, an additional novel setup was introduced.The antibacterial effect of each sample was tested not only for solar irradiation (Osram Ultra Vitalux 300 W lamp) but also in the dark, without light irradiation.For better dispersion of photocatalysts, all investigated samples were stirred vigorously in the suspension.Control samples containing only BioPeptone with 1 cm 3 of the same cell suspension, without photocatalysts, were subjected to the same conditions.In order to track the antimicrobial effect of ZnCr samples, an aliquot of 1 cm 3 was sampled at defined time intervals (0, 3, and 6 h), and serial dilution was prepared and transferred onto adequate selective solid media: tryptone bile X-glucuronide agar for E. coli and Baird-Parker agar for S. aureus.After the incubation period (37°C, 24 h), the grown colonies were counted.
These characteristic reflections of the dried ZnCr 100 sample confirmed the layered structure with R3m rhombohedral symmetry, which is expected for LDHs synthesized with nitrate precursors [74].The structure of the synthesized sample and the lattice parameters of ZnCr 100 were defined for the dominant LDH phase and the evaluated unit cell parameters are presented in Table 1.
The calculated values of the structural lattice parameters a and c were consistent with those reported in the literature [50,51].The c′ value (d 003 ), which described the thickness of one layer consisting of a Brucite-like sheet and one interlayer, was also calculated and compared with d 006 .Considering that d 003 (0.74 nm) was twice as high as d 006 (0.37 nm), favourable stacking of layers can be suggested for the ZnCr 100 sample [15,75].
The structural analysis results indicated that the chosen molar ratio initiated the formation of a layered structure for the sample ZnCr 100 [78].Furthermore, after thermal treatment, the formation of ZnO and ZnCr 2 O 4 phases was observed, aimed at enhancing photocatalytic efficiency through synergistic and heterojunction effects [50].

Textural characterization
The adsorption isotherms and pore size distribution, as well as textural parameters of all studied samples (specific surface area (S BET ), pore volume (V p ), and the most present pore diameter obtained from BJH pore size distribution (d p )), are presented in Figure 2 and Table 1.Notably, all  samples, except ZnCr 100, exhibited similar adsorption isotherms.ZnCr 100 displayed a type IV adsorption isotherm (Figure 2a) with an H2 hysteresis loop type, indicating a mesoporous structure with ink bottle-shaped pores [79].Furthermore, this sample showed a monomodal pore size distribution (Figure 2b) with one intense peak at ∼4 nm, suggesting a higher presence of smaller mesopores, developed BET surface area (Table 1), and a small value of the calculated average pore diameter.Upon calcination at temperatures of 300°C and higher, samples ZnCr 300, ZnCr 500, ZnCr 700, and ZnCr 900 showed a type II isotherm.Type II isotherm is characteristic of the non-porous or macroporous materials that exhibit unrestricted monolayer/multilayer adsorption.The isotherms of thermally treated samples displayed an H3 hysteresis loop type, indicating a mesoporous structure with wedge-shaped pores formed by the loose stacking of flaky particles [79].However, the hysteresis loop for ZnCr 300 began forming at a relative pressure of 0.4, suggesting non-uniformed pores in this sample [17].
Pore size distribution for ZnCr 300 revealed a low-intensity peak at around 4 nm (similar to the ZnCr 100 sample) and a broad peak at larger pore diameters (20-60 nm).These results suggest incomplete LDH thermal decomposition at the lowest TTT of 300°C.As the TTT increased, smaller mesopores completely disappeared, and the pore size distribution shifted towards larger values of pore diameters, consistent with the decrease in the BET surface area of these samples.
The difference in the adsorption/desorption isotherm of the sample ZnCr 100 compared to thermally treated samples can be attributed to its layered structure, which disappeared at higher temperatures of thermal treatment.The collapse of the layered structure caused changes in the pore type (transformation from ink-bottle pores to wedgeshaped pores) and size (disappearance of smaller pores), affecting N 2 gas adsorption [56].During thermal treatment at 300°C, dihydroxylation occurred, and the spinel MO phase was formed, as confirmed by XRD analysis.With increasing temperature, the crystallinity of the spinel phase also increased, leading to a decrease in the BET surface area [80].
The results obtained from the t-plot are also provided in Table 1, offering better insight into the pore structure [81].The calculated external surface area from the t-plot (S tplot ) and the calculated BET surface area were similar, indicating that the surface area originated solely from the presence of mesopores.These results also confirmed the absence of micropores in all samples.

Photocatalytic experiments
Photocatalytic experiments were conducted using MB and BCB dye pollutants.In order to rule out possible direct photolysis of dyes when irradiated, test solutions of both pollutants (MB and BCB) without a photocatalyst were treated under the same conditions as photocatalytic mixtures, and their concentration was measured with time.After 120 min, no photolysis was observed, suggesting that the organic dye pollutants used in these experiments were photochemically stable and the photocatalytic reaction was solely responsible for the dye degradation [55].It was observed that the photocatalytic efficiency for the MB removal (Figure 3a) increased with the temperature of thermal treatment.After 9 h of irradiation, the removal efficiency for ZnCr 300 and ZnCr 500 was approximately 40%, while a high removal efficiency (85%) was detected for the ZnCr 700 sample.However, complete MB decolourization was achieved after only 4 h for the ZnCr 900 sample.The superior efficiency of the ZnCr 900 sample could be attributed to its favourable phase composition, particularly the amount and heterojunctions between photocatalytic active phases (ZnO and ZnCr 2 O 4 ).In contrast, ZnCr 100 exhibited negligible MB photodegradation efficiency, likely due to its unsuitable phase composition (low crystalline LDH phase).
Subsequent investigation of BCB photodegradation efficiency (Figure 3b) revealed high removal efficiency ranging from 70% to 90% for all thermally treated samples, with ZnCr 900 exhibiting the highest efficiency.As expected, ZnCr 100 showed moderate BCB removal efficiency (10%).The favourable phase composition, as explained for MB removal, also contributed to the enhanced BCB photodegradation of thermally treated samples.
All of the thermally treated samples exhibited significant capability for the removal of both dye pollutants, with efficiencies ranging from approximately 40% to 100%.It can be concluded that the temperature of the thermal treatment had a significant impact on photocatalytic degradation efficiency by triggering the formation of photocatalytic active phases responsible for the dye removal reactions.
The differences in the photodegradation efficiency between BCB and MB removal could be attributed to a combination of factors related to the chemical structure of the dyes, impacting adsorption affinity, interaction with photocatalysts, and formation of intermediates during the photocatalytic reaction.Although both MB and BCB are azo dyes, BCB possesses a complex heterocyclic structure containing a central thiazine ring with various substituents [59,82], whereas MB has a simpler structure of a central thiazine ring with methyl groups as substituents [83].
Considering all the above factors, an overall potential photocatalytic degradation mechanism can be proposed regarding both dyes [69,84] When in contact with the surface of the ZnCr photocatalysts, MB and BCB molecules were adsorbed due to the interactions between dye molecules and active sites on the photocatalytic surface.Upon solar irradiation, the photocatalysts generated electron-hole pairs (Eq.12) that participated in the redox reactions: (i) photogenerated hole (h + ) reacted with the chemisorbed dye molecules, leading to the oxidation of dyes and forming cationic radical cations (Eq.13), and other intermediate products, and (ii) photogenerated electrons (e − ) reduced oxygen (Eq.14) or other remaining species producing reactive oxygen species (ROS) [83,69].Intermediate products, formed during oxidation and reduction, underwent further degradation, producing simpler and less harmful products (Eqs.( 15)-( 17)).In summary, the overall photocatalytic efficiency depends on a complex relationship of various factors: the chemical nature of the dyes, properties of the obtained photocatalysts, solar absorption, and charge separation [84].
A comparative table of results from numerous studies on photocatalytic dye degradation using various semiconductor photocatalysts is presented in Table 2.When compared to similar work in recent years, the results presented in our studies reveal that the application of ZnCr-LDHbased catalysts positively influences the photocatalytic performance in the photodegradation process.
The differences in the photodegradation efficiency of ZnCr 900 between MB and BCB removal could be elucidated by the surface charge of both the dye molecules and the photocatalyst.The pH-dependent changes in surface charge can influence the adsorption affinity of MB and BCB onto the photocatalyst surface.The surface charge of the photocatalyst can vary with pH due to alterations in protonation and deprotonation of the surface functional groups.The pK a and pH values of MB and BCB dye solutions and reaction mixtures are provided in Table 3.The pH value (6.08) of MB exceeded its pK a value (3.8), rendering the solution more alkaline than the solute's pK a .Consequently, the MB solute contains basic functional groups capable of accepting protons, resulting in proton loss (H⁺ ion) and increased negative charge.On the contrary, the pH value of the BCB solution (4.6) was lower than its pK a (6), resulting in a positively charged surface due to acidic functional groups capable of donating protons.The interaction between the photocatalysts and the dye solution resulted in an increase in pH values for both dyes.Introducing the catalyst into the MB solution amplified the difference between the pK a value and pH value, enhancing the negative surface charge.In the case of positively charged BCB dye molecules, the introduction of the photocatalysts neutralized the positive charge, as the difference between the pK a value (6) and pH value (6.03) was negligible.The photocatalytic reaction initiated a significant elevation of pH values in both reaction mixtures, reaching an alkaline pH after the reaction.It can be concluded that a more alkaline pH of the MB solution (8.1) enhanced and strengthened the electrostatic attraction between MB and  ZnCr 900, leading to stronger adsorption and higher photocatalytic activity compared to BCB, which had almost a neutral pH value (7.5).Furthermore, the zeta potential (Table 3), used to measure the surface charge, additionally explained the magnitude of the electrostatic catalyst-dye interactions, considering that higher positive or negative zeta potential magnitude points to stronger repulsion or attraction [91].The measured zeta potential of ZnCr 900 was positive (12.2 mV), indicating a slightly positive surface charge under neutral pH conditions.The positive zeta potential of the ZnCr photocatalyst, combined with the negative zeta potential of both MB and BCB dyes, demonstrated strong electrostatic attraction between the photocatalyst and dye molecules, which is favourable for higher adsorption and potentially higher photocatalytic activity [92].Comparatively, a higher photocatalyst/MB zeta potential magnitude could better explain the photocatalytic behaviour in MB photodegradation.Overall, the combination of pH-dependent changes in surface charge and the positive zeta potential of ZnCr 900 induced higher adsorption affinity, formation of surface reactive sites, and, consequently, higher photocatalytic MB degradation.A higher value of the photodegradation process turnover parameter, dTON, for MB (15.5 μmol•h −1 •g −1 ), compared to BCB (6.7 μmol•h −1 •g −1 ), additionally supports the previous observations regarding the better photocatalytic efficiency of the ZnCr 900 photocatalyst in the MB degradation reaction.
Determining heavy metal leaching from the obtained photocatalysts is an important step in providing information on any potential secondary pollution during water treatment.For the most efficient photocatalysts (ZnCr 900), the amount of metal leaching into the MB and CBC solution during the photocatalytic degradation was measured using the following instruments: ICP-OES Thermo Scientific icap 6500 series (for Zn 2+ leaching detection) and Perkin Elmer spectrophotometer (for Cr 6+ leaching detection).The concentration of Zn in the MB solution was 1.39 mg•L −1, and in the BCB solution was 2.09 mg•L −1 , which is lower than the WHO drinking water standards [69].Furthermore, results showed that Cr 6+ leaching was not detected (below the detection limit of the instrument), which was expected considering that only Cr 3+ was used in the synthesis process and is known to be non-toxic and non-carcinogenic [93].Therefore, the leaching study suggests that the ZnCr photocatalysts used do not initiate secondary pollution during water treatment.

Kinetic studies and predictive model of the photocatalytic reaction
The results of the kinetic study revealed that the experimental data obtained from photocatalytic reaction experiments for both investigated dyes were best fitted with linear regression and followed pseudo-first-order kinetics (Figure 5).The kinetic parameters obtained from the pseudo-first-order kinetic model are presented in Table 4, where the linear coefficients for determination, R 2 , for all photocatalytic active samples ranged between 0.83 and 0.98, indicating a good correlation with the suggested model.The kinetic study of azo dye photodegradation confirmed that the mechanism of both photocatalytic reactions followed the Langmuir-Hinshelwood kinetic model, which was expected since photocatalytic reactions on single substrates typically adhere to this kinetic model [55].The most efficient photocatalyst ZnCr 900 exhibited the highest value of the apparent rate constant and the shortest reaction half-time.
To compare the photodegradation efficiency of ZnCr 900 for different water treatment systems/dyes, a scaleup parameter, the electrical energy per order (E EO ), was calculated.This parameter represents the electrical energy required for the 90% pollutant degradation order-of-magnitude in a particular volume of contaminated water [69].The lower value of E EO for the MB removal (8.6 kW•h•L −1 •order −1 ), compared to BCB removal (13.4 kW•h•L −1 •order −1 ), indicated a lower amount of energy/lower cost needed for the pollutant removal treatment.
The results of the ANN approach for photodegradation rate constant showed that the accuracy of networks could be improved by adding additional hidden neurons, as evidenced by the increasing trend of correlation coefficient values with the decrease of mean squared values [94].ANNs exhibit significant advantages compared to traditional mathematical modelling methods and are widely used in chemical engineering for process optimization and various predictions [94,95].The correlation coefficient R 2 and mean squared error (MSE) for 1 neuron were 0.814 and 0.0115 (Figure 4a and b), respectively, while the best results, without over-fitting, were achieved after adding ten neurons to the hidden layer, and the correlation coefficient and MSE were 0.979 and 0.0039, with tansig activation function.
The ANN trained for the prediction of important valuables for the photodegradation rate constant (Figure 4c) showed the highest relative influence of TTT (35.1%), followed by the CS, ODT, and pore diameter of the most present pores (MPPD).

Effect of pollutant concentration and catalyst dosage on photocatalytic performance
The effect of MB dye concentration on photodegradation efficiency was analysed, and the results are presented in Figure 5a.ZnCr 900 was selected for this test since it exhibited the best removal efficiency in previous photocatalytic tests.It was observed that as the dye concentration increased, the efficiency improved, but this trend was not pronounced for higher concentration values.Consistent with previous reports [67,96,97], the removal efficiency increased with the initial dye concentration up to a specific limit, followed by a decrease in photodegradation efficiency.This decrease can be attributed to the light absorption of dye molecules, which inhibits the activation of photocatalytic active sites on the catalyst surface.The dosage of photocatalyst and its influence on the photocatalytic efficiency of MB removal was also studied (Figure 5b).The results indicate that the increase of photocatalyst mass did not improve photodegradation, probably due to the decrease in light penetration throughout the reaction mixture caused by light scattering on photocatalyst particles [2].

Stability tests
The stability of photocatalysts is an important factor that should be investigated in order to provide valuable insights into the catalysts' properties.Therefore, the stability of ZnCr 900 for MB photodegradation under solar light irradiation was assessed after consecutive cycles of use without any additional treatment between cycles (Figure 6).The obtained results revealed a gradual decrease in photodegradation efficiency after every cycle.This behaviour could be attributed to the limited availability of active sites, hindering the mass transfer of reactive species.Since no treatment of the photocatalyst was conducted between cycles, such as washing and drying, further experimental studies  are needed to elucidate the reasons for the deactivation of the photocatalyst and to enhance its photocatalytic performance.Nevertheless, the results indicated a moderate stability of the ZnCr 900 photocatalyst.

Antibacterial activity
Antibacterial tests were conducted to evaluate the antibacterial properties of all prepared samples by determining the total cell number of Gram-negative E. coli and Grampositive S. aureus bacteria (as indicator strains) over time during contact with the ZnCr photocatalyst under solar irradiation (Figures 7 and 8).
It was observed that, in the dark, the variation of the number of cells for E. coli and S. aureus did not exceed ±1 log unit, leading to the conclusion that ZnCr photocatalysts did not exhibit antibacterial properties under the defined conditions of these experiments, regardless of the applied thermal treatment during sample preparation (Figures 7a  and 8a).
The consistent number of microorganisms in the dark throughout the experiment duration suggests that the environmental factors were convenient for the selected microorganisms and did not adversely affect microbial growth.Moreover, the constant number of microbial cells in the control samples exposed to the solar light indicated that the light source did not influence the number reduction of both tested microorganisms (Figures 7b and 8b).
Under solar irradiation, a decrease in the cell number of E. coli was observed for all ZnCr photocatalysts with the exception of the ZnCr 100 samples (Figure 6b).The lack of antibacterial activity of the ZnCr 100 sample could be attributed to its phase composition, with only the LDH phase present, which could not induce the inhibition of cell growth.Thermally treated ZnCr samples gradually reduced the number of cells, with the ZnCr 900 sample exhibiting the highest antibacterial activity, inhibiting almost complete cell growth after only 3 h of contact.ZnCr 500 and ZnCr 700 achieved similar cell count reductions after 6 h of irradiation.The variation in antibacterial activity among samples could be explained by the amount of ZnO and ZnCr 2 O 4 phases detected by XRD analysis, wherein an increase in TTT corresponded to higher intensity and sharpness of peaks, indicating a higher amount and higher crystallinity of the active phases.
A different antibacterial behaviour was observed when investigating the growth inhibition of S. aureus under solar light irradiation (Figure 8b).Samples thermally treated at temperatures higher than 500°C (ZnCr 500, 700, and 900) induced antibacterial activity under solar light irradiation,  achieving practically complete growth inhibition after only 6 h of contact, whereas samples thermally treated at 100°C and 300°C did not reduce the initial number of cells during the experiment.As previously mentioned, the absence of antibacterial activity in samples is strongly connected with their phase composition and the amount of active phases.
Crystallinity and phase composition can significantly influence the photocatalytic properties of samples, including the ability to generate ROS under solar irradiation [22,98], which plays a crucial role in antibacterial activity [99].The variations in thermal treatment most probably influenced ROS generation, impacting the antibacterial effectiveness against both bacteria.Therefore, higher amounts of active phases (ZnO and ZnCr 2 O 4 ) resulted in more efficient antibacterial properties.Consequently, ZnCr 900 achieved the highest antibacterial activity for both bacteria due to the higher concentration of these active phases.
Based on the results and literature obtained, an antibacterial mechanism can be proposed for ZnCr photocatalysts, involving several key processes triggered by their photocatalytic activity under light irradiation.These processes lead to the generation of ROS and other effects that directly attack bacterial cells, resulting in their inactivation or death through oxidative damage to the cells.Additionally, the ROS can target the lipid bilayer of bacterial cell membranes, causing lipid peroxidation, compromising the membrane integrity, increasing permeability, and ultimately leading to cell lysis [100,101].While the primary antibacterial action of ZnCr photocatalysts is attributed to ROS generation, the presence of zinc (Zn) and chromium (Cr) ions may also contribute  to their antibacterial activity.Zn ions, in particular, are known for their antimicrobial properties, which can include disrupting membrane integrity and interfering with nutrient uptake and enzyme function within bacterial cells [102,103].
Furthermore, the investigation revealed differences in antibacterial activity between investigated E. coli and S. aureus, with the overall greater activity observed against E. coli.This disparity can be attributed to variations in bacterial cell wall structures and overall susceptibility to photocatalytic activity [104].While Gram-negative bacteria (E.coli) have a more complex cell wall structure, they also have a more permeable outer membrane compared to Gram-positive bacteria (S. aureus), allowing better penetration of antibacterial agents, including ROS generated by the ZnCr photocatalysts, into the bacterial cell [57,104].Also, E. coli possesses lipopolysaccharides (LPS) in the outer membrane that can be sensitive to oxidative stress [105,106].Hence, it can be suggested that the solar light irradiation initiated ROS generation that targeted and disrupted LPS integrity, contributing to the more effective antibacterial activity against Gram-negative bacteria than Gram-positive bacteria [107].The sensitivity of Gram-negative bacteria towards ROS generated by photocatalysts, such as singlet oxygen and superoxide radicals, can have a more pronounced effect on the cellular components [104,108] and agrees with our findings.Thereby, from the results obtained and from the literature, it could be suggested that the penetration of the generated ROS inside S. aureus required a longer time due to a thicker membrane [109].Additionally, Gram-negative bacteria typically have a larger surface areato-volume ratio due to their smaller size and complex shapes, providing more sites for interaction with the photocatalyst, and consequently, increasing the antibacterial activity [85].In conclusion, despite the more complex cell wall structure of E. coli, the combination of factors and unique properties such as membrane permeability, sensitivity to ROS, and specific interactions with active phases of the photocatalyst can result in more efficient antibacterial activity against E. coli compared to S. aureus.
Different catalyst materials and the main variables that influence both photocatalytic and antibacterial activity are presented in Table 5.Each study had different experimental setups, making comparison of the results challenging.Hallak et al. [110] reported that TiO 2 , ZnO, and Au/ZnO showed a progressive increase in antibacterial activity against E. coli with gold-doped ZnO exhibiting the highest log reduction after 2 h under UVA light, indicating that Au doping enhances photocatalytic activity.ZnO samples presented high antibacterial efficiency against S. aureus and E. coli, likely due to the catalyst's modification and the intense xenon lamp light source used for the experiment [107].Also, BiVO 4 demonstrated broad-spectrum antibacterial properties across various strains, including MRSA, under visible light, with effectiveness indicated by the diameter of the inhibition zones [111].Additionally, TiO 2 loading onto ZnAl LDH facilitated the formation of favourable interactions among active phases, improving the photocatalytic and antibacterial efficiency [22].However, the results presented in our study demonstrate that the application of our suggested antibacterial experimental setup, along with enhanced photocatalyst-bacteria cell contact, achieved high photoinduced antibacterial efficiency even with significantly lower amounts of used photocatalysts, compared to other studies.

Conclusions
The study of the impact of thermal treatment on LDHbased ZnCr mixed metal oxides revealed enhanced photocatalytic efficiency in the removal of MB and BCB dyes.Furthermore, the antibacterial activity against E. coli and S. aureus demonstrated slight variations among photocatalysts, with ZnCr 900 exhibiting exceptional high performance under solar irradiation.The formation of active phases during the thermal treatment played a pivotal role in the observed high efficiency in both antibacterial and photodegradation processes.Structural and textural analysis unveiled the active phases, specific surface area, and pore size distribution of all samples.Structural analysis indicated the formation of the layered structure for the sample ZnCr 100 and the development of ZnO and ZnCr 2 O 4 phases after thermal treatment, enhancing not only the photocatalytic efficiency but also antibacterial efficiency through synergistic and heterojunction effects.Moreover, kinetic studies using Langmuir-Hinshelwood kinetic modelling confirmed that the experimental data obtained from photocatalytic reaction experiments for both investigated dyes best fit the linear fit, following pseudo-first-order dynamics.Future studies regarding a detailed investigation of photocatalytic and antibacterial mechanisms should provide guidelines for the modification of LDH-based ZnCr photocatalysts that will further improve efficiency and stability.This study contributes valuable insights into the structure-dependent photocatalytic behaviour of ZnCr photocatalysts, paving the way for environmentally friendly applications of ZnCr photocatalysts in water purification and environmental remediation.

Figure 4 :
Figure 4: ANN results: relationship between (a) MSE and (b) correlation coefficient (R 2 ) and the number of hidden layer neurons for the degradation rate constant and (c) relative importance of input variables: TTT, ODT, CS, and MPPD.

Figure 6 :
Figure 6: Stability tests of ZnCr 900 for MB photodegradation induced by solar light.

Figure 7 :
Figure 7: Reduction of E. coli cell number under antibacterial influence of the synthesized samples (a) in absence of light/in dark and (b) under solar light (experimental conditions: I [VIS] = 20.52W•m −2 ; I [UV A ] = 17.6 W•m −2 ).

Figure 8 :
Figure 8: Reduction of S. aureus cell number under antibacterial influence of the synthesized samples in (a) in the absence of light/in dark and (b) under solar light (experimental conditions: I [VIS] = 20.52W m −2 ; I [UV A ] = 17.6 W m −2 ).

Table 1 :
Structural and textural parameters of the obtained samples :

Table 2 :
Comparison of the photocatalytic dye degradation efficiency of ZnCr 900 with the other published resultsCatalyst

Table 3 :
Zeta potential, pK a , pH values of MB and BCB dye solutions and reaction mixtures

Table 4 :
Kinetic parameters for MB and CBC photodegradation reaction

Table 5 :
Comparison of photoinduced antibacterial efficiency of ZnCr 900 with the other published results a Inhibition zones.