Synthesis of Black g-C₃N₄ and Exploration of the Mechanism Underlying the Enhancement of Photocatalytic CO₂ Reduction

Abstract

The use of solar energy to convert CO₂ into value-added chemicals is a promising sustainable development strategy. In this study, a black graphitic carbon nitride (CN-B) photocatalyst was fabricated through a single-step calcination process, employing phloxine B and urea as the precursor materials. The catalysts were characterized using TEM, XRD, FTIR, XPS and so on. The amount of prepolymer phloxine B was 25 mg, 35 mg and 45 mg, respectively, and the obtained samples were CN-B-0.025, CN-B-0.035 and CN-B-0.045. All samples were used for visible-catalyzed CO₂ reduction. The experimental findings indicate that the CO evolution rate of the optimal photocatalyst CN-B-0.035 reaches 27.56 μmol g_cat.⁻¹ h⁻¹. This value is nine-fold higher than that of pure CN, which has a CO evolution rate of 3.22 μmol g_cat.⁻¹ h⁻¹. The excellent photocatalytic reduction performance is due to the following factors: Firstly, the exceedingly thin nanosheet structure of the catalyst enhances the velocity of the charge transfer, and transmission electron microscopy (TEM) analysis shows that the nanosheet thickness of the catalyst CN-B is significantly thinner. Secondly, the light absorption capacity of the catalyst is enhanced. The absorbance of CN-B increases significantly in the ultraviolet region and extends to the near-infrared region, as shown with UV diffuse reflection spectroscopy. Finally, the photothermal effect of CN-B causes the catalyst temperature to rise rapidly from 20 °C to 131 °C within 120 s, which further promotes photogenerated carrier separation. This research offers a novel approach to the development of photocatalysts aimed at the photothermal-assisted photocatalytic conversion of CO₂.

1. Introduction

Rapid population growth and industrialization have led to a dramatic increase in global energy demand. Simultaneously, the combustion of fossil fuels releases significant quantities of carbon dioxide (CO₂), a greenhouse gas that profoundly affects the global ecosystem. The effective utilization of CO₂ has become a hot topic in current research [1,2]. Due to the abundance of solar energy resources, photocatalysis presents a promising application prospect [3]. Therefore, photocatalytic CO₂ reduction holds significant potential for future applications. Through the photocatalytic reduction of carbon dioxide, the concentration of carbon dioxide in the atmosphere can be reduced, which helps to slow down the trend of global warming and has positive significance for environmental protection and coping with climate change.

Graphitic carbon nitride (g-C₃N₄) is a common chip semiconductor photocatalyst [4,5,6,7,8,9]. Due to its compatible energy gap (Eg = 2.7 eV), excellent physical and chemical stability, low cost, non-toxic properties and easy-to-adjust synthesis methods, it has attracted great attention in photocatalytic CO₂ reduction applications. However, challenges faced by g-C₃N₄ include poor visible light response, the continuous recombination of excitons and short specific surface area. All these hinder g-C₃N₄ to obtain the best photocatalytic performance [10]. Despite extensive research on photocatalytic CO₂ reduction, achieving high selectivity and activity in CN photocatalysts often requires the use of sacrificial agents or cocatalysts. This is mainly due to their high electron-hole recombination rate, narrow light absorption range and low CO₂ adsorption capacity, which limit their performance [11,12,13]. To tackle these challenges, multiple approaches have been employed to improve the photocatalytic performance of pristine g-C₃N₄, including morphology regulation, elemental doping and the creation of heterostructures [14,15,16]. Within these approaches, the separation of photoinduced electron-hole pairs is regarded as one of the most efficient ways to enhance photocatalytic performance [17]. For instance, Hu and his team fabricated a BaCN-C₃N₄ photocatalyst. This photocatalyst facilitated partial ring opening by acting on the heptazine ring. The presence of Ba²⁺ not only enhanced photocatalytic activity but also improved visible light absorption [18]. Nevertheless, because of the low temperature in the surrounding environment and the restricted utilization of photons, the efficiency of photocatalytic reduction achieved by the photocatalyst remains insufficient.

As luck would have it, the development of a photocatalytic system synergized with photothermal function can increase the temperature within the reaction space. This increase in temperature accelerates the chemical reaction kinetics on the photocatalyst’s surface, promotes the transfer of photogenerated charges and ultimately improves the photocatalytic degradation efficiency. A less explored aspect in developing CO₂ photo reduction systems involves photo-to-thermal conversion, a process in which materials absorb light and convert it into heat. Enhanced photo-to-thermal conversion could improve photocatalytic efficiency by supplying the energy needed to overcome activation barriers and enable reaction pathways [19]. For example, Zhang et al. developed an enhanced NdO/TiO photothermal catalyst using the sol–gel method. Due to the Z-type heterojunction formed at the interface between NdO and TiO and the photothermal effect that promoted charge separation, it showed excellent CO₂ reduction performance under light [20]. However, much of the research on photothermal-assisted photocatalysis involves the integration of photothermal materials, which often face challenges such as inconsistent preparation stability, complex processes and limited CO₂ adsorption capacity [21,22]. Therefore, the design of an ultra-thin g-C₃N₄ nanosheet photocatalytic material with its own photothermal effect is of great significance for the photothermal-assisted photocatalytic reduction of CO₂.

In this study, the red dye phloxine B was added during the preparation of carbon nitride, and the black carbon nitride (CN-B) photocatalyst was synthesized through a single-step calcination process. Our results show that the constructed CN-B ultra-thin nanosheet structures shorten the path length of charge transport. Moreover, the addition of phloxine B significantly enhanced the absorbance of the catalyst. As a result, the black substance showed a potent photothermal property. The temperature of the reaction surroundings increased, which in turn advanced the efficiency of photocatalytic reactions.

2. Results and Discussion

The morphology of the photocatalyst was tested. Transmission electron microscopy (TEM) images showed that there was little difference in the morphology of CN and CN-0.035 photocatalysts, both of which exhibited coiled, layered and stacked nanostructures (Figure 1a,b). TEM was employed to characterize both the microstructure and elemental composition of CN-B-0.035. Through TEM analysis, the morphological features and elemental makeup of CN-B-0.035 were determined. TEM was utilized to conduct an in-depth analysis of CN-B-0.035’s microstructure and elemental constituents. As seen in Figure 1c,d, a large sheet grid structure is evident. Upon enlarging the image, as shown in Figure 1e,f, the large sheet structure remains visible. From the above analysis, it is clear that the thickness of CN-B-0.035 is much smaller than that of CN, indicating that the addition of phloxine B not only maintains the original layer structure of CN but also reduces the charge transfer distance due to the ultra-thin thickness of nanoparticles and their sheet structure. This reduction in the charge transfer distance facilitates the transport of photogenerated charge carriers, thereby improving photocatalytic activity. Furthermore, from the AFM test image in Figure S1 (Supplementary Materials), it can be clearly seen that after adding phloxine B, the sample thickness decreases from 1.762 nm to 1.423 nm. Additionally, as shown in Figure 1g–j, the element mapping images confirm that the C, N and O elements are distributed on the surface of CN-B nanosheets.

The crystal structure of the sample was analyzed with XRD, as shown in Figure 2a. The diffraction peaks at 27.2° can be assigned to the (002) planes of the sample CN, respectively [23]. The initial characteristic peak uniquely corresponds to the organization of the triazine units in the CN network [24,25]. Conversely, the next characteristic peak indicates the ordered aggregation of carbon nitride nanosheets in the C-axis [26,27]. Moreover, when the XRD patterns of CN and CN-B photocatalysts are contrasted, it is evident that as the amount of phloxine B increases, the intensity of the (002) characteristic peak of the CN-B photocatalyst steadily declines. This decline suggests that the periodicity of the catalyst’s planar structure is disrupted [28]. Furthermore, due to the enhanced interfacial interaction between CN-B nanosheets and the reduced layer spacing, the (001) crystal plane diffraction peak of photocatalyst CN-B shows a slight blue shift [29,30]. The functional groups of the synthesized samples were further analyzed using Fourier transform infrared spectroscopy (FTIR). The peaks located between the 810 cm⁻¹ and 900–1800 cm⁻¹ regions correspond to the out-of-plane breathing vibrations of the triazine unit and the stretching vibrations of N–C=N/N–C₃, respectively. This indicates that the characteristic structure of CN is retained in both samples CN-B and CN [31]. These findings demonstrated that following the introduction of the precursor phloxine B, there was no substantial alteration in the functional-group configuration of CN [32,33].

To examine the chemical and elemental states of the photocatalyst samples, X-ray photoelectron spectroscopy (XPS) was utilized to characterize the electronic structure of both CN and CN-B. In Figure 3a, the peaks of C 1s, N 1s and O 1s can be clearly observed, which is consistent with the results of the element distribution mapping graphs. Moreover, the presence of a trace amount of element O stems from water adsorbed on the surface of the catalyst. When contrasted with CN, CN-B-0.35 exhibits a greater C atom ratio. Also, because phloxine B is added during preparation, its C/N atom ratio is elevated. Figure 3b illustrates that the C 1s spectrum for the CN-B-0.35 photocatalyst displays three distinct peaks. At 288.2 eV, this peak corresponds to N–C=N bonds, the 285.9 eV peak aligns with C–N–H bonds and the 284.8 eV peak is associated with C–C bonds [34,35]. In Figure 3c, regarding the N 1s spectrum of CN-B-0.035, three peaks stand out. The peak at 400.7 eV is assignable to the N–H bond. The 399.7 eV peak correlates with the triazine ring featuring C–N=C and N₂C structures. Meanwhile, the 398.89 eV peak corresponds to nitrogen N-(C)₃ groups [36]. In the O 1s spectrum of CN-B-0.035, two prominent characteristic peaks are seen at 532.1 and 533.5 eV, as shown in Figure 3d. The 532.1 eV peak corresponds to C=O bonds, with the 533.5 eV peak assignable to C–O–H bonds [37]. In summary, compared with CN, the positions of the peaks in the C 1s, N 1s and O 1s spectra of the CN-B-0.035 photocatalyst were slightly shifted, indicating that the addition of phloxine B may alter the internal structural environment of CN [38,39]. This alteration can affect the electron transfer within the catalyst network to a certain extent.

The light absorption capacity of the photocatalyst was studied using UV diffuse reflection (DRS) spectroscopy. As shown in Figure 4a, CN and CN-B exhibit different light absorption ranges and capacities. CN-B demonstrates a strong light absorption capacity from 450 nm to 800 nm, showing better light absorption in the visible light and near-infrared regions, thereby enhancing the utilization of light [40,41]. Additionally, based on the Tauc function (αh)² = A(hν-Eg) [42,43], the band gaps (Eg) of CN and CN-B are determined to be 2.75 eV and 2.62 eV, respectively (Figure 4b). The results indicate that the band gap value of CN-B (2.62 eV) is narrower than that of CN (2.75 eV), suggesting that more electrons can be generated under visible light. According to the results of XPS-valence band spectroscopy, the addition of prepolymer phloxine B significantly changes the valence band edge position of the sample. Figure 4c shows that the CN and CN-B-0.035 photocatalysts possess valence band (VB) values of 1.73 eV and 1.55 eV in sequence. This shift to a lower binding energy enhances the reducing power of the semiconductor. Using the position of the flat band potential and the equation E_VB = E_CB + Eg, the conduction band (CB) positions of CN and CN-B-0.035 are calculated to be −1.02 eV and −1.07 eV, respectively. This indicates that CN-B has a more negative conduction band position [44].

To assess the photocatalytic CO₂ reduction efficiency of the synthesized photocatalyst, a 300 W xenon lamp was used to test the photocatalytic CO₂ reduction of the sample without adding sacrificial agents and photosensitizers. As shown in Figure 5a, CN-B samples were prepared at different temperatures of 480 °C, 500 °C, 520 °C, 540 °C and 560 °C, and their carbon dioxide reduction properties were also tested. It is evident that the samples calcined at 540 °C exhibit the best photocatalytic CO₂ reduction performance. As shown in Figure 5b, the CO formation rate of CN within 4 h is only 3.12 μmol g_cat.⁻¹ h⁻¹, while the CO formation rates of CN-B-0.025, CN-B-0.035 and CN-B-0.045 are 13.78, 27.56 and 15.63 μmol g_cat.⁻¹ h⁻¹, respectively. The CO formation rate of the photocatalyst CN-B-0.035 is 8.8 times that of CN. In conclusion, an appropriate amount of phloxine B can enhance the migration of photogenerated electrons and boost the photocatalytic degradation performance of CN. However, an excessive amount of phloxine B may result in the formation of new recombination centers for photogenerated carriers, thereby decreasing the generation of active species. Additionally, photocatalyst stability is a key factor in practical applications. Thus, we conducted six cycles of photocatalytic carbon dioxide reduction experiments. Notably, after six cycles of experiments, the CN-B-0.035 photocatalyst still maintained a high CO yield (Figure 5c), indicating that the CN-B-0.035 photocatalyst has superior stability.

To reveal the contribution of phloxine B precursor addition to enhanced photocatalytic CO₂ reduction, CN and CN-B-0.035 photocatalyst powders were irradiated with a 300 mW cm⁻² Xenon lamp for 120 s, and the surface temperatures at 40 s intervals were recorded with infrared thermography, as shown in Figure 5g,h. Figure 5f illustrates the temperature variation of the CN and CN-B-0.035 photocatalysts. The temperature of CN-B-0.035 jumps swiftly from 26.0 °C to 131.4 °C and then stays steady. This final temperature significantly exceeds that of pure CN, which reaches only 47.7 °C. The temperature rise is attributed to the addition of the phloxine B precursor. This addition darkens the color of CN, enhancing the photothermal performance of CN-B. As a consequence, there is an elevation in temperature. Based on the above analysis, the CO₂ reduction properties of pure CN and CN-B-0.035 photocatalysts were tested at different temperatures by controlling the reaction temperature in a circulating condensing unit system and an oil bath. As seen in Figure 5d, both original CN and CN-B-0.035 exhibit some degree of temperature responsiveness, with an upward trend in temperature, and there is a corresponding increase in the CO production rate, indicating that heat can improve their performance. Additionally, as shown in Figure 5d,e, the CO generation rate of the CN-B-0.035 photocatalyst with a light source is 18.3 μmol g_cat.⁻¹ h⁻¹, compared to 1.3 μmol g_cat.⁻¹ h⁻¹ without a light source, demonstrating that light can significantly enhance its photocatalytic reduction performance. These findings suggest that by introducing the phloxine B precursor, the photocatalytic CO₂ reduction capabilities of CN are strengthened. This occurs via the combined action of the photothermal effect, enabling CN to more effectively convert CO₂ during photocatalysis.

Accordingly, the photoelectrochemical characterization of different samples was carried out to study the effect of the photothermal effect on the catalyst carrier kinetics. As shown in Figure 6a, CN-B-0.035 exhibits a lower PL signal than pure CN due to the introduction of a cyanogen defect that increases the photon-generated carrier separation rate of CN-B-0.035. In Figure 6b, CN-B-0.035 shows the highest photocurrent response intensity, which demonstrates that the modified photocatalyst has a strong electron transfer ability [45,46]. Furthermore, in the electrochemical impedance spectroscopy (EIS) diagram shown in Figure 6c, the CN-B-0.035 sample has the smallest arc radius, indicating lower charge transfer resistance [47].

In order to understand the conversion mechanism of CO₂ molecules on CN-B-0.035, FT-IR spectroscopy was used to study the intermediate products in the photoreduction process. As shown in Figure 7a, the CO₂ stretching vibration (2350–2250 cm⁻¹), H–O–H bending vibration band (1700–1600 cm⁻¹) and H–O–H stretching vibration band (3700–3200 cm⁻¹) are observed. This indicates the reduction of CO₂ and the dissociation of H₂O. Figure 7b shows that the CO₂ reduction signal of CN-B is enhanced. Evidently, introducing phloxine B boosts its reduction effectiveness. It can be observed that the two vibration absorption peaks of H₂O have different intensities. This difference may be due to the O–H bond stretching vibration in water, which involves a large relative displacement of oxygen and hydrogen atoms, resulting in a significant change in charge distribution, strong infrared absorption and high absorption peak intensity. Conversely, the dipole moment of the flexural vibration molecule changes little. Similar to the bending vibration of water molecules, the displacement of hydrogen atoms relative to oxygen atoms is small, the change in charge distribution is minimal, the infrared absorption is weak and the absorption peak intensity is low [48,49,50]. It is generally believed that *COOH is a key intermediate in the conversion of CO₂ to CO [50,51,52,53]. Based on the above analysis, the preliminary reaction mechanism of the CN-B photocatalytic reduction of CO₂ is proposed: First, CO₂ is adsorbed on the catalyst surface (CO₂ → *CO₂), followed by the reaction of adsorbed activated *CO₂ with H⁺ and e⁻ to produce the important intermediate COOH*. Then, *COOH is converted into H₂O and CO* through the proto-electron transfer reduction process. Finally, CO* desorbs to release CO (CO* → CO). Figure 7c illustrates the mechanism of the CO₂ reduction process.

$$\mathrm{CN}\text{-}\mathrm{B}\stackrel{\mathrm{h}\mathsf{\upsilon }}{\to }{\mathrm{e}}^{-}+{\mathrm{h}}^{+}$$

(1)

$$\mathrm{C}{\mathrm{O}}_2\to \mathrm{C}{\mathrm{O}}_2*$$

(2)

$${\mathrm{H}}_2\mathrm{O}+{\mathrm{h}}^{+}\to \mathrm{OH}*+{\mathrm{H}}^{+}$$

(3)

$$\mathrm{OH}*+{\mathrm{h}}^{+}\to 1/2{\mathrm{O}}_2\left(\mathrm{g}\right)+{\mathrm{H}}^{+}$$

(4)

$$\mathrm{C}{\mathrm{O}}_2*{+\mathrm{e}}^{-}+{\mathrm{h}}^{+}\to \mathrm{COOH}*$$

(5)

$$\mathrm{COOH}*+{\mathrm{e}}^{-}+{\mathrm{H}}^{+}\to \mathrm{CO}*\left(\mathrm{g}\right)+{\mathrm{H}}_2\mathrm{O}$$

(6)

$$\mathrm{CO}*\to \mathrm{CO}$$

(7)

3. Experimental

3.1. Chemicals and Reagents

For specific details on the chemicals and reagents, please refer to Section S1.1 of the Supporting Information.

3.2. Synthesis and Characterizations

For the synthesis procedures of black g-C₃N₄ (CN-B) and pure CN, as well as detailed characterization, please refer to Section S1.2 of the Supplementary Materials.

3.3. CO₂ Photoreduction Experiments

CO₂ reduction experiments were conducted in a 200 mL quartz reaction vessel using a 350 W xenon lamp (1000 mW·cm⁻²) as a white light source. The catalyst, amounting to 0.01 g, was made to disperse in two milliliter unit volumes of water free of ions. The mixture was deposited on a 1 cm × 2 cm quartz glass and dried under an infrared lamp at 60 °C. The dehydrated catalyst was positioned at the base of the quartz reaction apparatus. H₂O steam was introduced into the reactor for 20 min along with the CO₂ stream to completely purify the air. The light intensity of 1000 mW·cm⁻² on the photocatalyst was measured with an automatic optical power meter (CEL-NP200010A, Zolix Instruments, Beijing, China). Following a 60 min irradiation period (using a 300 W xenon lamp at 1000 mW·cm⁻²), the gaseous products were analyzed using a gas chromatograph (GC-7920, Nanjing Kejie Analytical Instrument Co., Ltd., Nanjing, China) fitted with a hydrogen flame ionization detector (FID).

3.4. Photo-Electrochemical Evaluations

For detailed procedures of the Photoelectrochemical Measurements, please refer to Section S1.3 of the Supplementary Materials.

4. Conclusions

Overall, a highly efficient photothermal-aided photocatalytic system for CO₂reduction was engineered. CN-B came into being through an on-pot calcination approach. Phloxine B along with urea were deployed as precursors in the synthetic process. The experimental results show that the CO yield of the optimal sample, CN-B-0.035, is 27.56 μmol g_cat.⁻¹ h⁻¹ within 4 h under the irradiation of a 500 mW cm⁻² Xenon lamp. The excellent photocatalytic reduction performance is mainly attributed to the following three reasons: (1) the ultra-thin nanosheets of black CN-B-0.035 reduce the charge transfer distance; (2) the addition of prepolymers promotes the separation and migration of photogenerated carriers; (3) due to the high-efficiency photothermal conversion effect of the catalyst, the temperature of the reaction environment is increased, so the photocatalytic degradation performance is further improved. This investigation presents a viable method and systematic approach for the development of photothermal-assisted photocatalytic CO₂ reduction photocatalysts based on CN.