Effect of the synthesis method on the MnCo₂O₄ towards the photocatalytic production of H₂

2.1 Pechini synthesis method

The material obtained by this method is denoted as MCOP. The synthesis of the MCOP by Pechini consisted in a modification of the method used in the reference literature [23,34,56]. According to the basis of Pechini method [52], citric acid (CA) and ethylene glycol (EG; Sigma-Aldrich) were used as chelating agents for the formation of the polymeric network. First, the calculated amount of EG was heated at 70°C until evaporation was noticed and then the required amount of CA was added. It was kept under constant stirring until a translucent gel-like solution was achieved and allowed to cool for 20 min at room temperature. Then, the necessary amount of Mn(NO₃)₂·6H₂O and Co(NO₃)₂·6H₂O (Sigma-Aldrich) was added, in stoichiometric ratio of 1:2, considering that spinel structure is AB₂O₄. It was kept under constant stirring at a temperature of 80°C until a gel was obtained. It was left to dry until it hardened forming a polymeric resin with the metal ions. Once dried, a thermogravimetric analysis of the organic resin was carried out to thermally decompose it and obtain the desired oxide.

2.2 Hydrothermal synthesis method

The methodology for the hydrothermal consisted in the modification of the methods used in the literature [33,58]. EG and 0.8 g NaOH as nucleating agent were used. First, stirring of 20 mL of EG was carried out at 70°C. Then, stoichiometric amounts Mn(NO₃)₂·6H₂O and Co(NO₃)₂·6H₂O were added and mixed by constant stirring with the EG. NaOH was then added dropwise. The solution was then transferred to a 40 mL Teflon autoclave and left at 120°C for 12 h. Subsequently, it was placed in a centrifuge and then the material was filtered and washed with a water-ethanol solution (1:1) several times. It was left to dry at 60°C for 5 h and finally it was exposed to a thermal treatment of 500°C for 4 h. The material obtained is denoted as MCOH.

The hydrothermal method was modified by using Mn and Co metal acetates as cation precursors, and the use of urea as reported in literature [57,59,60]. 20 mL of EG was stirred at room temperature and the calculated amounts of Mn(CH₃COO)₂·4H₂O and Co(CH₃COO)₂·4H₂O were added. It was kept under constant stirring for 20 min and then urea was added as precipitant, maintaining the stirring for another 30 min. It was then transferred to a 40 mL Teflon autoclave, and left at 150°C for 10 h. It was then filtered and washed with water-ethanol solution. It was left to dry at 80°C for 5 h and then heat treated at 350°C for 4 h. This material is denoted as MCOHA.

3.1 Electrochemical characterization

An electrochemical characterization of the material was carried out by means of Cyclic Voltammetry (CV) in a bipotentiostat model AFCBP1 Pine Instruments. The system consisted of a 3-electrode cell configuration: a 0.5 cm diameter (0.196 cm²) glassy carbon working electrode on which the materials were deposited, a counter electrode using a 4.5 cm² platinum foil (9 cm² on both sides), and an Ag/AgCl reference electrode with KCl solution at 3 M concentration (0.210 V vs normal hydrogen electrode [NHE]). The solution used for the electrochemical cells was 0.1 M potassium ferrocyanide trihydrate (K₄Fe(CN)₆·3H₂O) at approximately pH 7. The solution was bubbled with nitrogen for approximately 5 min to remove possible dilute oxygen content before each measurement. Preparation of the working electrode consisted of dispersing 30 mg of the material in 80 μL of isopropanol and 0.5 μL of Nafion^®, then the suspension was sonicated for a period of 10 min, and a drop of the mixture was taken and placed on the surface of the glassy carbon electrode. The isopropanol was evaporated at a temperature of 60°C for 5 min, obtaining a membrane of the material to be characterized.

3.2 Photocatalytic evaluation

The evaluation as a photocatalyst was carried out by measuring the evolution of hydrogen using the gas chromatography technique in a Clarus 500 PerkinElmer Chromatograph equipped with a thermal conductivity detector) using a specific method for measuring hydrogen and employing nitrogen as a carrier gas. The system used to carry out the separation of the water molecule consisted of a reactor with a quartz tube of 20 cm long and 4 cm in diameter with aluminum caps on the ends and an NPT sample inlet on one of them. The reactor is sealed by means of steel rods that press membranes on the ends of the tube allowing a partially leak-free seal. A magnetic stirrer, 200 mL of distilled water with 4% methanol as sacrificial reagent, and 0.2 g sample for each evaluation are placed inside the reactor. The reactor is kept in constant agitation for a period of 8 h, sampling every hour with a gas syringe while the system is irradiated by a Philips metal halide lamp of 250 W power that emits light in the visible range.

4.1 TGA

Figure 1 shows the thermogravimetric analysis obtained from the MCO synthesized by different synthesis methods, where the thermal behavior of the material is observed. This test was used to determine the minimum calcination temperatures. Figure 1a shows the TGA obtained from the material synthesized by the Pechini method. It shows a 70% weight loss at 370°C attributed to the loss of the organics used during the synthesis: EG and CA. Figure 1b shows the behavior of the material obtained by the hydrothermal method with NaOH. In this figure, it is observed that at a temperature of 500°C there is a loss of 10% by weight, which is lower compared to the previous figure since fewer organics are used. Finally, Figure 1c shows the analysis of the material obtained by the modified hydrothermal method. In this figure, it is observed that at the temperature of 350°C, the highest loss in weight is obtained, being 40%. This loss is due to the EG, and the urea used in the method. From Figure 1 it can be observed that the loss of weight % starts at 300°C due to the elimination of water molecules and organics present in the samples. The literature presents a varied range of temperatures used to obtain the material by different synthesis methods, from 200 to 1,000°C [33,61,62]. In the case of the MCOP, different calcination temperatures were used, but at 900°C, the desired crystalline phase of the MCO was obtained (being the MCO reported in this work). In the MCOH, the temperature of 500°C was chosen according to the results presented in the TGA diagram (Figure 1b), where above this temperature no weight loss is observed meaning that a thermally stable material can be obtained. Finally, the sample synthesized by modifying the hydrothermal method (MCOHA) used a temperature of 350°C as shown in the TGA diagram in Figure 1c. Again, temperatures higher than 350°C did not present a weight change, thus obtaining a thermal stable material. For all samples, a heat treatment time of 4 h was employed.

Figure 1

Thermograms of the MCO obtained by the method: (a) Pechini, (b) hydrothermal with NaOH, and (c) hydrothermal with acetates.

The reduction of the calcinated MCOs was carried out by TGA using a flow with 5% H₂ in balanced Ar atmosphere to evaluate the presence of oxygen vacancies, which are related to the amount of O₂ present within each sample [21]. These vacancies are defects in the crystalline lattice and the surface of the material which play a key role in the performance of a photocatalyst. The weight loss due to oxygen release in each MCO is shown in Figure 2. Theoretical amount of O₂ in the complete reduction of the MCO is 20%, while in the MCOP, it is about 19%, 19.5% in MCOH, and 18.4% in the MCOHA, which is slightly lesser than the others and could be associated with more oxygen vacancies.

Figure 2

Thermograms of the weight loss of each MCO due to the oxygen released in the reduction with 5% H₂/Ar. (a) MCOP, (b) MCOH and (c) MCOHA.

4.2 XRD

Figure 3 shows the diffractograms obtained from the cobaltite synthesized by different methods. This figure shows the comparison of the peaks obtained with the signals of the ICDD pattern corresponding to the spinel phase MCO. The samples show the characteristic peaks at 30.56, 36.03, 37.67, 43.8, 54.39, 57.96, and 63.68° well indexed to the (220), (311), (222), (400), (422), (511), (440), and (531) crystal planes of the spinel-type MnCo₂O₄ (Fd-3m, JCPDS #00-023-1237). This indicates that the material obtained in each diffractogram coincides with the consulted card, in addition to not presenting any other signal that indicates the presence of impurities. Figure 3a shows the diffractogram obtained by the Pechini method. As mentioned above, the calcination of the MCOP was carried out at different temperatures, 900°C being the one that performed the best, so it is the one presented in this work. In the diffractogram, very defined and elongated peaks are observed, which indicates that it is possible to obtain a very large crystallite size, in addition to sintering, due to the high calcination temperature used. Figure 3b shows the diffraction pattern of the MCOH calcined at 500°C with peaks of lower intensity, but with a greater width than the previous ones. This indicates the possibility of obtaining small particle sizes due to smaller crystallite size than obtained in the previous MCO. The last pattern (Figure 3c) shows peaks with lower intensity and less definition than the previous ones, due to the calcination temperature of 350°C and the use of the hydrothermal method for this MCOHA. In the Figure, a shift in the peaks of the MCOH and the MCOHA can be observed. This behavior can be explained by the observations of Stangeland et al. [63] who attributed this to the incorporation of manganese into the face-centered cubic structure of spinel type.

Figure 3

XRD patterns of the MCO synthetized by (a) Pechini method at 900°C, (b) hydrothermal methos with NaOH at 500°C, and (c) hydrothermal method with acetates at 350°C.

To evaluate the impact in the crystalline structure of the MCO samples due to the temperature used in the synthesis methods, their crystallinity index (CI) was calculated using the following equation and procedure as reported in the literature [64]:

where Sc represents the area of the crystalline domain and St is the area of the total domain in the diffractogram. Furthermore, the average crystallite sizes from each diffractogram were calculated with the Scherrer equation [65]. For the MCOP, the crystallinity obtained was 90% with an average crystallite size of 53 nm, as expected due to the higher temperature used. A CI of 43% and crystallite size of 20 nm was obtained for MCOH, and finally, 36% of crystallinity with crystallite size of 12 nm was obtained for MCOHA. From the above, it can be summarized that as the temperature increases, higher degrees of crystallinity are obtained but larger crystallite sizes are produced.

4.3 XPS

XPS characterization was used to determine the chemical composition and the valence state of the elements in the MCO samples. Table 1 summarizes the binding energy details of all the elements in the spectra of the MCO. As shown in Figure 4, the spectrum only represents elemental peaks of Mn, Co, and O in the three MCOs. The inevitable existence of the C peak is derived from adventitious carbon species as reported in literature [42,51,66]. The peak deconvolution were carried out by a Gaussian fitting method based on the Shirley background correction with CasaXPS software [67]. In the high-resolution spectrum of Co 2p of the 3 MCO, 2 main peaks are observed at 795 and 780 eV with spin−orbit splitting of ∼15 eV attributed to the multiplets Co 2p_1/2 and Co 2p_3/2, respectively. The evident shake up satellite peaks (denoted as “sat.” in the figure), characteristic of Co²⁺ and Co³⁺ cations [66], also can be seen at ∼803 and ∼786 eV. Co 2p_1/2 signal deconvoluted into subpeaks centered at 797 and 795 eV, while Co 2p_3/2 at 782 and 780 eV approx. Subpeaks at 795.50 and 780.10 eV are ascribed to Co²⁺, while subpeaks at 797.26 and 781.88 eV are ascribed to Co³⁺. In XPS of Mn 2p of the MCO, 2 main peaks at 653.7 and 642 eV are observed for Mn 2p_3/2 and Mn 2p_1/2, respectively. These 2 signal deconvoluted into 4 subpeaks, 2 centered at 653.5 and 641.5 eV assigned to the binding energy of Mn²⁺, and the other 2 to Mn³⁺ at 654.2 and 643.8 eV. The O 1s spectrum is often divided by 3 peaks at 530, 531, and 533 eV which belong to the lattice oxygen (metal–oxygen bond) in the spinel structure, the number of defect sites with low oxygen coordination and/or presence of hydroxyl groups and the third is atribuited to the physisorbed or chemisorbed H₂O. Using the area under the signal curve at ∼531 eV of the O 1s spectre, the concentration of vacancies of the 3 MCOs were calculated as reported [68], being the MCOHA with a relative concentration of 34% oxygen vacancies, more than the others (26% MCOP and 19% MCOH) and these results are according to the reduction obtained in Section 4.1 (MCOHA > MCOP > MCOH). The peak values obtained with the XPS are agreeing to that reported in the literature [42,51,58,66,69–71]. The fitted peaks and the corresponding satellites are well-indexed to the presence of Co²⁺/Co³⁺ and Mn²⁺/Mn³⁺ mixed valances revealing the coexistence of both states in the 3 MCOs. XPS also allow us to determinate the cation distribution between octahedral and tetrahedral sites to determinate the type of spinel-cobaltite obtained. If a normal spinel is formed, all the Co³⁺ ions occupy octahedral sites, and all the Mn²⁺ ions occupy tetrahedral sites. For an inverse spinel, half of the Co³⁺ ions are occupying the octahedral sites (Oh), and the other half occupies the tetrahedral sites (Th) along with all the Mn²⁺ ions. In 2015, Aghavnian et al. [72] and Rodríguez-Rodríguez et al. [73] in 2019, calculated these sites with the fitting of the peaks Co 2p_3/2 and Mn 2p_3/2 since it is more precise due to their higher intensity signal than the 2p_1/2. The resulting fitting gives the proportion of the Oh and the Th sites occupied by the cations. The Co(Oh) obtained was 53.04% for MCOP, 68.61% for MCOH, and 47.64% for MCOHA, hence almost half of the Co³⁺ are in octahedral sites and therefore the spinel obtained is an inversal spinel type.

Figure 4

XPS spectra of the MCO samples: The survey spectrum and high-resolution spectra of Co 2p, Mn 2p, and O 1s.

4.4 Scattering transmission electron microscopy (STEM)

Figure 5 presents the images obtained by STEM of the MCO synthesized by the different synthesis methods with a unimodal particle size distribution. In the upper part (Figure 5a) the MCOP is shown, being evident the degree of sintering present in the particles due to the 900°C temperature used for calcination. The morphology is also affected, observing particles with irregular polygon shape and mode values with size of 478 nm. Figure 5b shows the MCOH. In this image, irregular particles with mode size of 30 nm are observed, much smaller compared to the MCOP; this is attributed to the hydrothermal method with the NAOH since it allows to have a better control in the particle size in addition to a lower use of calcination temperature [53,54]. This is also the case in Figure 5c, where MCOHA is observed with a mode particle size of 10 nm with irregular morphology. This size can be attributed to the modification of the hydrothermal method by using acetates as precursors favoring an increase in surface area and small particle size as mentioned by Kurajica et al. [60]. The size obtained in MCOHA is much smaller compared to MCOP and MCOH, even after using the same synthesis method. The results of the particle size were according to the expected crystallite size obtained in Section 4.2, since the temperature used was higher, the obtained particle size was larger.

Figure 5

STEM images of the MCO synthetized by different methods: (a) pechini at 900°C, (b) hydrothermal with NaOH at 500°C, and (c) hydrothermal with acetates at 350°C.

4.5 BET area surface

For the analysis of the surface area, the BET method [74] was used, obtaining the nitrogen adsorption–desorption isotherms in the MCO shown in Figure 6. In the case of MCOP and MCOH, isotherms show a normal behavior on the physisorption, and a significant presence of hysteresis loop is not observed, therefore the surface area obtained in these two is attributed only to the particles. In the MCOHA isotherm, where this loop is observed, this type is classified as type IV according to IUPAC [75]. This is characteristic of a presumably mesoporous material, hence the MCOHA obtained is mesoporous, which is possible as reported by Li et al. [57], where the cobaltite obtained by hydrothermal using nitrates showed mesoporosity. Furthermore, there are reports of other cobaltites presenting the same hysteresis behavior [76–78].

Figure 6

BET isotherms of the MCO synthetized by the method: (left) Pechini at 900°C, (middle) hydrothermal with NaOH at 500°C, and (right) Hydrothermal with acetates at 350°C.

Table 2 shows the surface area values obtained from the analysis of the isotherms, which vary from 5 to 155 m²·g⁻¹. This remarkable change in the areas is due to the temperatures and the method used in each synthesis, since in the MCOP the 900°C temperature had a great impact on both the particle size and the surface area of the material, while in the hydrothermal method, better results were obtained with lower temperatures. Li et al. [59] reported having obtained an area of 80 m²·g⁻¹ and reported that it is higher than those found in other MCO materials, but this value is lower than that obtained in the MCOHA in this work. Table 2 shows this comparison.

4.6 UV-Vis spectroscopy

A UV-Vis spectroscopy analysis was performed on the powders to determine the region of the light spectrum absorbed by the materials. From the diffuse reflectance spectra, the band gap of each material was determined employing the Kubelka–Munk method. By means of the reflectance R obtained from each phase and substituted into the Kubelka–Munk equation (2) [3,80]:

Then, using the Tauc relation: (αhν) ⁿ vs E, the plots in Figure 7 were obtained. For semiconductors with direct band gap, n = 2 and for indirect, n = 1/2. In the graphs, a tangent line is drawn at the point with the steepest slope of the graph extending to the x-axis estimating the band gap in eV [3]. In these Tauc plots the band gap of each MCO was estimated, obtaining an approximate value of 1.30 eV in all three. Table 3 shows the comparison of the band gap of each of the MCOs with what is reported in the literature. The table shows that the values obtained in this work are different from those reported in the literature, but even among the same authors there is a difference in the values. Therefore, a characterization by CV was performed in order to verify the values obtained in this work.

Figure 7

Absorption spectra and Tauc plots for the estimation of the direct band gap for the MCOP, MCOH, and MCOHA.

4.7 CV

Since the band gap value obtained is just the necessary potential, it is important to determine that the CB level of the photocatalyst is sufficiently negative compared to the H⁺/H₂ redox potential. Therefore, an electrochemical characterization of each MCO was performed by CV. The electrochemical data obtained from the voltammetry plots is very valuable information since it allows to have a relative estimation of the position of the redox potentials of the material. These values in turn are approximations to the ionization potential of the material between its valence and conduction bands [83–85]. The voltammograms of the synthesized MCO are shown in Figure 8. In the −0.5 V potential region is the reduction potential of each of the MCOs. The value is very similar in all, being slightly different only by the synthesis method, stoichiometry, doping, etc., as mentioned in the literature [86]. Table 4 shows the data of each of the MCO compared with the optical band gap calculated in this work and with that reported by Zheng and Lei [51]. They report the characterization obtained by the Mott–Schottky method, which, although it is not the same characterization method used, also makes use of the basic principles of electrochemistry so the values are very close [87]. The table also shows the correction of the potential values obtained, since these values were obtained against the Ag/AgCl electrode, and it is necessary to convert them against the NHE to know if it is sufficiently negative.

Figure 8

Voltammograms of the MCO synthetized by the methods: (top left) Pechini, (top right) hydrothermal with NaOH, and (bottom) hydrothermal with acetates.

According to the electrochemical analysis, the synthesized MCOs presented an approximate negative reduction potential value of −0.3 eV, higher than the H⁺/H₂ redox potential. Based on these results, it is possible to draw a representation of the photogenerated charge carriers of the MCOs and this is presented in Figure 9. Here it can be observed that this value (−0.3 eV) is barely sufficient since, when calculating the electrochemical band gap of each one, a value within the range of 1.44 eV is achieved. By comparing this value against the needed 1.23 eV and considering that it has been previously reported that it is required to reach a value of 1.6 eV to be able to perform the photocatalytic phenomenon without problems, this value may become an obstacle to obtain a good performance of the photocatalyst in the production of hydrogen. According to the research carried out by Admassie et al. [88], the values obtained for the electrochemical gap and the optical gap can differ by ±0.15 eV, generally with the optical gap being slightly smaller. This, according to Inamdar et al. [89], is due to the Coulombic interaction of the electron–hole pair, i.e., slightly more energy is required in the redox process than in the excitation of charge carriers from the conduction band to the valence band.

Figure 9

Schematic of the photogenerated charge carriers of the MCOs for the visible-light photocatalytic H₂ production from H₂O.

4.8 Hall effect

To obtain two of the main electrical properties from synthesized MCOs such as mobility and charge carrier concentration (CCC), the Hall effect with the configuration of Van de Pauw [90] was measured obtaining the results as shown in Table 5. From this table, the MCOP presented a CCC of −2.6 × 10¹⁶ cm⁻³ with the highest mobility of −48.6 cm²·V⁻¹·s⁻¹. MCOH shows a CCC of −4.2 × 10¹⁶ cm⁻³ with −25.9 cm²·V⁻¹·s⁻¹. Finally, the MCOHA shows better CCC with −13.0 × 10¹⁶ cm⁻³ and a mobility of −3.7 cm²·V⁻¹·s⁻¹. On one hand, the Hall effect measurements showed that samples belong to a n-type conductive nature, as reported in the literature [91]. On the other hand, results from CCC indicate that MCOP and MCOH exhibit a similar behavior between 3−4 × 10¹⁶ cm⁻³. Whereas MCOHA shows a higher CCC in the order of magnitude which can be attributed to the greater amount of oxygen vacancies, as mentioned previously in the XPS section and in agreement with that reported by Selim et al. [92], who associated the increase in CCC with a higher concentration of oxygen vacancies present in the material.

The order of the materials in terms of mobility is as follows: MCOP > MCOH > MCOHA. This order is presumably related to the crystallinity index as reported in the XRD section. The higher crystallinity of MCOP result in higher mobility, agreeing to that reported in the literature that charge typically moves faster in crystalline regions [93]. In the case of the MCOH, its crystallinity is slightly higher than the MCOHA, obtaining greater mobility than this last one, but with less CCC.

Since there is scarce knowledge about these properties of the MCO, the obtained results provide new information for these types of materials, which can be used in further investigations.

4.9 Charge separation efficiency

The method of the PL spectroscopy is a common technique used to study the separation efficiency of the photogenerated charge carriers, as well as their migration and transference in a photocatalyst [91]. Figure 10 represents the PL spectra of the MCOP, MCOH, and MCOHA. In this figure is evident a strong PL signal for the three samples when are excited by a wavelength of 470 nm, suggesting that the higher the intensity of the peak the faster the annihilation of the photoexcited electron–hole pairs as reported in the literature [15,16,94].

Figure 10

PL spectra of the MCO samples.

In general, a low PL intensity shows a better efficiency of charge separation. Likewise, these efficiencies can be directly related to the CCC [95] and as stated before this is due to the oxygen vacancies of the materials that hinder charge carrier recombination. Analysis of PL results indicate that the most efficient charge separation is presented by MCOH with the highest CCC, followed closely by MCOHA, and the least efficient with the lowest CCC is shown by MCOP.

4.10 Hydrogen production

Figure 11 shows the monitoring of the photocatalytic hydrogen production by evaluating the different MCOs, in a total time of 8 h. According to the figure, the MCOHA is the one that shows the best performance as a photocatalyst at the end of the evaluation time. Table 6 shows the results obtained in the evaluation of the materials as photocatalysts. The MCOP presented a production of 9,200 nmol of H₂/g_cat, the MCOH of 5,074 nmol of H₂/g_cat and, finally, the MCOHA with a production of 12,050 nmol of H₂/g_cat. As mentioned in Section 2.1, the first synthesis method used to obtain the MCO was the Pechini method, and it was found that calcination at 900°C resulted in the best production, sacrificing a good surface area and particle size (5 m²·g⁻¹ and 500 nm). Later, the hydrothermal method with the use of NaOH allowed improving these properties, since the excess of OH⁻ accelerates nucleation and the EG control the growth [33] obtaining an increase in surface area and small particle size (58 m²·g⁻¹ and 40 nm), affecting hydrogen production. By modifying the hydrothermal method using acetates, urea, and EG as precursors, it was possible to obtain smaller particle sizes and increase the surface area significantly, due to the fact that the formation of a metal alkoxide is favored and when it decomposes it forms the desired oxide with very high areas with the possibility of larger active sites [59,60,96]. In addition, a better photocatalytic performance was obtained in terms of time with the MCOHA but when comparing the production in terms of surface area, MCOP is positioned as the best. This is due to the fact that a higher temperature favors higher crystallinity; higher crystallinity leads to a decrease in defects, as vacancies of oxygen, which in turn improve the recombination between charge carriers [3,39]. In the XPS analysis of O 1s shown in Section 4.3, the MCOHA have 34% oxygen vacancies, more than the MCOP, and these vacancies can act as doping elements introducing defect and preventing recombination, one of the main obstacles in photocatalysis [97,98]. The high calcination temperature (900°C) of the MCOP results in better crystalline quality but specific surface area lower than the MCOHA (350°C): 90% crystalline with 5 m²·g⁻¹ vs 36% and 155 m²·g⁻¹, respectively. This result leads to high surface area and more active sites and therefore a better performance, even with the disadvantage of possible defects. If the calcination temperature is high, the crystalline quality is good, but the specific surface area is low and could result in poor photocatalytic activity due to the lack of active sites for hydrogen evolution [9], so it is important to get a balance between these properties to obtain a good photocatalyst. Also, PL and Hall Effect measurements confirmed findings discussed above. These results showed that the MCOP presented the highest mobility and fastest recombination of the charge carriers, possibly due to its high crystallinity, but with the lowest CCC, which influence in its intermediate photocatalytic performance. Whereas the properties of MCOHA are superior, since its highest CCC, slowest recombination, and high surface allow more photocatalytic interaction with the water molecules; even with the lowest mobility, its photocatalytic performance was the best.

Figure 11

Photocatalytic evaluation of the MCOs synthetized for the hydrogen production.

The MCOHA was the sample with highest evolution rate of 1.506 μmol·g·h⁻¹, but when compared with the 730 μmol·g·h⁻¹ of SrFe₂O₄ reported by our group [23], the result is not as promising as expected. Recently, heterojunctions and precious metals as dopants were used to improve the performance of the photocatalyst as reported by Fang et al. [10], enhancing the rate 2.7 times of the pure g-C₃N₄ preparing the Pt/g-C₃N₄/BiOBr photocatalyst (361 μmol·g⁻¹·h⁻¹). Lv et al. [18] using the NaY zeolite with Cds and Pt, showed a H₂ production rate of 686.5 μmol·g⁻¹·h⁻¹. Yet, the development of TiO₂ remains under investigation as reported recently by Chen et al. [19] in 2022, loading Pd nanoparticles and doping ions in TiO₂ nanosheets, obtaining 76.6 μmol·g⁻¹·h⁻¹ with Pd/0.2%K⁺ -TiO₂.

Although it was possible to improve the properties of the material without affecting its performance, the production obtained is low compared to that reported in the literature. This may be mainly since the band gap of 1.3 eV and reduction potential (−0.3 eV) presented by the material are just adequate for this application; therefore, it does produce hydrogen; however, it could be improved as discussed above with further development to achieve the performance that would be expected from a good active photocatalyst under visible light.