Background

In recent decades, wastewater containing heavy metal ions and organic compounds brings serious damages for environment and human beings. As one of common heavy metal ions, hexavalent chromium (Cr(VI)) derived from electroplating, leather tanning, and printing poses a serious threat for our health owing to its high toxicity [1]. On the other hand, most of organic pollutants (such as dyes) are also toxic and non-biodegradable, which destroy our living environment [2]. Up to now, many techniques have been developed to eliminate organic pollutants and reduce Cr(VI) to Cr(III) [3,4,5]. Among these methods, photocatalytic and photo-Fenton-like catalytic techniques are regarded to be the promising methods for efficient degradation of organic contaminants and Cr(VI) reduction in wastewater because of their inexpensive cost, non-selectivity, and simplicity of operation [6,7,8,9]. The basic steps involved in a photocatalytic degradation process can be described as follows: excitation of photocatalysts, separation and migration of the photogenerated charges, generation of active species on the surface of catalysts, and decomposition of organic compound as well as reduction of Cr(VI) caused by the redox reaction of active species and photo-induced charges [10, 11]. The photo-Fenton-like catalytic reaction is based on the synergistic effects of the Fenton reaction and photocatalytic process. The generation of active species during the Fenton reaction process can be promoted after the introduction of suitable light irradiation, which leads to improved catalytic activity [12, 13]. However, the wide application of photocatalytic and photo-Fenton-like catalytic techniques is limited due to the large bandgap of photocatalysts only responding to UV light (which accounts for ~ 5% of sunlight energy) and their low charge separation efficiency [14]. Generally, it is known that the visible light and near-infrared (NIR) light occupy ~ 45% and ~ 46% of solar energy, respectively, and their application has received a great deal of interest [15, 16]. As a result, the development of broad spectrum (UV-vis-NIR) active catalysts with efficient separation of photogenerated charges is very important for their practical applications [17,18,19,20]. Up to now, the iron-contained catalysts with narrow bandgap are considered as ideal candidates in the photocatalytic and photo-Fenton-like catalytic applications [21,22,23,24,25].

As one of typical iron-contained catalysts, BiFeO3 with perovskite-type structure is known to be an interesting visible light-driven photocatalytic and photo-Fenton-like catalytic material for the degradation of dyes [26,27,28,29,30,31,32,33,34]. Nevertheless, its catalytic activity is not so strong to meet the application requirements owing to the high recombination rate of photogenerated charges. Moreover, the light response range of BiFeO3 needs to be further extended to NIR light region for effective utilization of sunlight energy. Therefore, many strategies have been used to overcome these shortcomings [35,36,37,38,39,40].

Carbon quantum dots (CQDs), as an important class of zero-dimensional nanocarbon material, have attracted considerable attentions due to its distinct properties, such as large surface area, low toxicity, high biocompatibility, good water solubility, high chemical stability, good electrical conductivity, and excellent optical properties [41,42,43,44]. These prominent properties make it a promising candidate for the practical application in different fields [41,42,43,44]. More importantly, the photoexcited CQDs are demonstrated to be an excellent electron donors and acceptors to promote the separation of photogenerated charges in photocatalysts [45]. On the other hand, CQDs are found to be an unique up-converted photoluminescence material, which allows the generation of short-wavelength emission light (from 450 to 750 nm) by the excitation of long-wavelength light (NIR light, from 700 to 1000 nm) [42, 44]. The up-converted emission light can be employed as the excitation light for the production of photogenerated charges in the semiconductors, which extends their light response region [45]. As a result, incorporation of CQDs with photocatalysts is demonstrated to be a promising way to form excellent hybrid composite photocatalysts [46,47,48,49,50,51,52]. Chen et al. prepared CQD/BiFeO3 nanocomposites and found their enhanced visible light photocatalytic activity for the dye degradation [53]. To the best of our knowledge, however, there is no work devoted to the photo-Fenton dye degradation and photocatalytic Cr(VI) reduction performances of CQD/BiFeO3 composite photocatalysts under visible or NIR light irradiation.

In this work, the CQD/BiFeO3 composite photocatalysts were prepared by a hydrothermal route. Their photocatalytic and photo-Fenton-like catalytic performance for acid orange 7 (AO7) degradation as well as photocatalytic Cr(VI) reduction activity under visible and NIR light irradiation were systematically investigated. The corresponding catalytic mechanism was proposed.

Methods

Preparation of CQDs

The CQDs were prepared by a hydrothermal method [54]. Glucose (1 g) was added into distilled water (80 ml) under magnetic stirring and ultrasonic treatment to obtain a homogeneous solution. Subsequently, this solution was transferred into a 100-mL Teflon-lined stainless steel autoclave and heated at180 °C for 4 h. After the reaction, the resultant solution was filtered by filter paper twice, and then, the reddish-brown CQDs suspension was obtained.

Fabrication of CQD/BiFeO3 Composites

BiFeO3 nanoparticles were prepared through a polyacrylamide gel route as reported in the literature [55]. The CQD/BiFeO3 composites were fabricated as follows (Fig. 1): BiFeO3 nanoparticles (0.1 g) were introduced into distilled water (70 ml), followed by ultrasonic treatment for 0.5 h to obtain uniform suspension. After that, a certain amount of CQD suspension was added drop by drop into the BiFeO3 suspension under magnetic stirring. The mixture was moved into the Teflon-lined stainless steel autoclave (100 ml) and heated at130 °C for 4 h. Finally, the product was collected by centrifugation, washed with deionized water, and dried at 60 °C for 8 h. To explore the impact of the CQDs content on the catalytic actives of the composites, a series of CQD/BiFeO3 composites with different mass contents of CQDs were prepared by adding different volumes of CQDs suspension (3, 6, 12, and 24 ml). These composites were correspondingly named as 3C/BFO, 6C/BFO, 12C/BFO, and 24C/BFO.

Fig. 1
figure 1

The schematic illustration of preparation process for CQDs/BiFeO3 composite

Full size image

Photo-Fenton Catalytic and Photocatalytic Degradation of Dye

The photo-Fenton catalytic performance of the as-prepared CQD/BiFeO3 composites was investigated toward the degradation of AO7 separately irradiated by visible light (300-W xenon lamp with a 420-nm cutoff filter) and NIR light (300-W xenon lamp with a 800-nm cutoff filter). In a typical experiment, the photocatalyst (0.1 g) was placed into AO7 solution (200 ml, 5 mg/L), and magnetically stirred in dark for 0.5 h to achieve a adsorption-desorption equilibrium between the photocatalyst and AO7 molecules. Subsequently, a certain amount of H2O2 solution was added into the suspension, and the xenon lamp was turned on to start the catalytic reaction. In the catalytic process, a small amount of the reaction solution (2 ml) was taken and centrifuged to eliminate the catalyst. The absorbance of the supernatant was measured by a UV-vis spectrophotometer at 484 nm to obtain the AO7 concentration. On the other hand, the photocatalytic degradation of AO7 over the samples was performed to evaluate their photocatalytic activities under the same conditions in the absence of H2O2.

The recycling catalytic experiments were carried out to test the catalytic reusability of the samples. After the first catalytic experiment, the catalyst was separated from the solution by centrifugation, washed with deionized water, and dried. The collected catalyst was added into the new dye solution for the next catalytic reaction with the same condition.

To confirm the reactive species involved in the photocatalytic and photo-Fenton catalytic degradation processes, the active species trapping experiments were performed by adding several scavengers under the same conditions as mentioned above. Ethanol (10% by volume) and ammonium oxalate (AO, 2 mM) were used as the scavengers of hydroxyl (·OH) and photogenerated holes (h+), respectively [56]. N2 purging can expel the dissolved O2 in the solution, leading to the inhibition of superoxide (·O2−) generation.

Photocatalytic Reduction of Cr(VI)

Cr(VI) was employed as another model pollutants to measure photocatalytic activity of the samples. The photocatalytic reduction process of Cr(VI) to Cr(III) was similar to that of the dye degradation. The initial concentration of Cr(VI) was 10 mg/l and the photocatalyst dosage was 0.2 g in 200 ml Cr(VI) solution (i.e., 1 g/l). The initial pH value of the Cr(VI) solution was adjusted by H2SO4 to 2~3. The residual concentration of Cr(VI) solution was detected by UV-vis spectrophotometer using the diphenylcarbazide (DPC) method [57].

Hydroxyl Radical Detections

Fluorimetry was employed to detect the ·OH radicals generated on the irradiated samples by using terephthalic acid (TA) as a probe molecule. Generally, the ·OH will react with TA to generate highly fluorescent compound, 2-hydroxyterephthalic acid (TAOH). The information of ·OH can be detected through measuring the photoluminescence (PL) intensity of TAOH with the excitation wavelength of ~ 315 nm. Typically, the TA was introduced into NaOH solution (1.0 mmol l−1) to obtain TA solution (0.25 mmol l−1). The catalyst (60 mg) was placed into TA solution (100 ml) under magnetically stirring for several minutes. After that, a certain amount of H2O2 was dissolved into above mixture, which was irradiated by visible light (300-W xenon lamp with a 420-nm cutoff filter) or NIR light (300-W xenon lamp with a 800-nm cutoff filter). At given intervals of irradiation, 3 ml of the reaction solution was sampled and centrifuged to remove the catalyst. The PL spectra of the supernatant were determined by fluorescence spectrophotometer. On the other hand, the generation of ·OH in the photocatalytic reaction was also measured under the same conditions without the addition of H2O2.

Characterization

The phase purity of the samples was examined by X-ray powder diffraction (XRD) and Fourier-transform infrared spectroscopy (FTIR). The morphology and microstructure of the samples were observed by field-emission transmission electron microscopy (TEM). The chemical states of the surface elements on the samples were detected by X-ray photoelectron spectroscopy (XPS). The ultraviolet-visible (UV-vis) diffuse reflectance spectra of the samples were recorded through a TU-1901 double beam UV-vis spectrophotometer. The PL spectra of the samples were determined by a fluorescence spectrophotometer. The transient photocurrent response and electrochemical impedance spectroscopy (EIS) measurements were carried out on an electrochemical workstation with a three-electrode system. The working electrode fabrication and test procedures were similar to those previously reported [56]. Particularly, the photocurrent response measurement was performed under visible light (300-W xenon lamp with a 420-nm cutoff filter) irradiation.

Results and Discussion

XRD Analysis

Figure 2 presents the XRD patterns of BiFeO3, CQDs, and 24C/BFO. The BiFeO3 and 24C/BFO sample show similar diffraction patterns, which can be readily indexed to the rhombohedral BiFeO3 phase (JCPD file no: 74-2016). No trace of impurities, such as Fe2O3 and Bi2O3, is found. The results indicate that the high-purity BiFeO3 is obtained and the introduction of CQDs and hydrothermal treatment do not obviously change the crystal structure of BiFeO3. From the XRD pattern of CQDs, one can see that a broad diffraction peak is observed at ~ 23.5°, which is mainly attributed to the amorphous structure of CQDs. Notably, for the composite, no characteristic diffraction peaks of CQDs are detected owing to the low content of CQDs in the 24C/BFO sample. To confirm the existence of CQDs in the composite, the FTIR characterization is performed.

Fig. 2
figure 2

XRD patterns of BiFeO3, CQD, and the 24C/BFO composites

Full size image

FTIR Analysis

Figure 3 shows the FTIR spectra of BiFeO3, CQD, and 12C/BFO composites. In the case of bare BiFeO3, the peaks at ~ 440 cm−1 and ~ 560 cm−1 are assigned to the stretching and bending vibrations of Fe–O, which is consistent with the reported result [55]. For the CQDs, the deformation vibration for C–H at ~ 638 cm−1, the stretching vibration for C–C at ~ 1630 cm−1, and C–OH stretching at ~ 1120 cm−1 are found [58]. In addition, the characteristic peaks of BiFeO3 and CQDs are detected in the spectrum of 12C/BFO composite. The results suggest the existence of CQDs and BiFeO3 in the composite. Moreover, the peak located at ~ 1380 cm−1 is attributed to the stretching vibration of O–H from the absorbed H2O [59].

Fig. 3
figure 3

FTIR spectra of BiFeO3, CQD, and the 12C/BFO composites

Full size image

Optical Absorption Property

It is well established that the optical absorption property of nanomaterials has an important effect on their performance [60, 61]. The optical absorption property of BiFeO3, CQD, and CQDs/BiFeO3 composites were investigated by UV-vis diffuse reflectance spectra, as shown in Fig. 4a. Compared with BiFeO3, the CQD/BiFeO3 composites exhibit obviously enhanced optical absorption capability in the entire UV-vis light region. It is worth noting that the optical absorption intensity of the composites gradually increases with increasing the content of CQDs. This phenomenon can be attributed to the strong light absorption of CQDs in the UV-vis light region. To obtain the light absorption edge of the samples, the first derivative curves of the UV-vis diffuse reflectance spectra are carried out (Fig. 4b), in which the peak wavelength is considered to be the absorption edge of the samples [62]. It is found that absorption edges of BiFeO3 and CQD/BiFeO3 composites are located at ~ 588 nm, suggesting that the decoration of CQDs does not change the bandgap energy of BiFeO3.

Fig. 4
figure 4

a UV-vis diffuse reflectance spectra of BiFeO3, CQD, and CQDs/BiFeO3 composites. b The corresponding first derivative of the diffuse reflectance spectra

Full size image

XPS Analysis

The chemical states of elements in the 12C/BFO sample were monitored by XPS and the results are presented in Fig. 5. On the Bi 4f XPS spectrum (Fig. 5a), the observed two strong peaks at 164.1 (Bi 4f5/2) and 158.8 eV (Bi 4f7/2) demonstrate the existence of Bi3+ in the composite [63]. In Fig. 5b, the Fe 2p XPS spectrum indicates two obvious peaks at 723.6 and 709.6 eV, which are attributed to Fe 2p1/2 and Fe 2p3/2. Notably, the broad peak of Fe 2p3/2 can be divided into two peaks at 712.0 and 709.6 eV, corresponding to Fe3+ and Fe2+, respectively [40]. In addition, it is seen that the satellite peak of Fe 2p3/2 is found at 717.8 eV. As shown in the XPS spectrum of O 1s (Fig. 5c), the obvious peak located at 529.6 eV is attributed to the lattice oxygen and the shoulder peak at 531.3 eV belongs to the chemisorbed oxygen of surface vacancies [64]. For the XPS spectrum of C 1s (Fig. 5d), the signal of C 1s can be divided into two distinct peaks. The major peak at ~ 284.9 eV is ascribed to the C–C bond with sp2 orbital, whereas the peak at 287.7 eV is caused by the oxygenated carbon. The results further demonstrate the coexistence of CQDs and BiFeO3 in the composite [65].

Fig. 5
figure 5

High-resolution XPS spectra of a Bi 4f, b Fe 2p, c O 1s, and d C 1s for the 12C/BFO composite

Full size image

Morphology Observation

The TEM and high-resolution TEM (HRTEM) images of BiFeO3 nanoparticles are shown in Fig. 6a and b, respectively. It is seen that the bare BiFeO3 possesses a sphere-like shape and smooth surface with an average diameter of ~ 120 nm. The lattice spacing of 0.288 nm belongs to the (110) spacing of BiFeO3. The TEM image in Fig. 6c indicates that the CQDs are composed of spherical-like particles with an average particle size of ~ 15 nm. From the TEM image of the CQD/BiFeO3 composites (Fig. 6d–g), one can see that the CQDs are decorated on the surface of BiFeO3 nanoparticles. The HRTEM image of the 12C/BiFeO3 sample (Fig. 6h) reveals the interplanar distance of 0.389 nm corresponding to the (012) plane of BiFeO3. Alongside of BiFeO3, the decorated CQDs exhibit amorphous characteristic. This result suggests the formation of hybrid composite structure between BiFeO3 and CQDs.

Fig. 6
figure 6

a and b TEM and HRTEM images of bare BiFeO3 nanoparticles, respectively; c TEM image of CQDs; d–g TEM images of 3C/BFO, 6C/BFO 12C/BFO, and 24C/BFO respectively; h HRTEM image of 12C/BFO

Full size image

The dark-field scanning TEM (DF-STEM) image and the corresponding elemental mappings of the 12C/BFO sample are shown in Fig. 7a–e, respectively. The results reveal that the sample presents not only uniform distribution of the Bi/Fe/O elements but also uniform distribution of the C element. This confirms that CQDs are uniformly assembled on the surface of BiFeO3 nanoparticles.

Fig. 7
figure 7

a DF-STEM image of the 12C/BFO composite. b–e The corresponding energy dispersive X-ray elemental mapping images

Full size image

Photo-Fenton Catalytic and Photocatalytic Performance

The photocatalytic performance of the samples was first assessed by the degradation of AO7 under visible light irradiation, and the result is shown in Fig. 8a. Prior to the photocatalytic reaction, the adsorption (in the dark) and blank (without catalyst) experiments were carried out. A small amount of AO7 (~ 5%) is degraded after 3-h irradiation without catalyst, indicating that the self-degradation of the dye can be neglected. In the photocatalytic reaction, the photodegradation ability of pure BiFeO3 is weak and only ~ 33% of AO7 is observed to be decomposed after 3-h exposure. When BiFeO3 nanoparticles are decorated by CQDs, the CQD/BiFeO3 composites exhibit obviously enhanced photocatalytic activity. Moreover, it is found that the catalytic activities of the composites are highly related to the content of CQDs. Among these composites, the 12C/BiFeO3 composite displays the optimal degradation percentage of ~ 73% after 3-h irradiation, which is 2.2 times higher than that of bare BiFeO3. However, with further increase of the CQD content (e.g., 24C/BFO), excessive CQDs decorated on the surface of BiFeO3 nanoparticles may shield BiFeO3 from absorbing visible light, which leads to the decrease of the photocatalytic activity.

Fig. 8
figure 8

a Photocatalytic degradation of AO7, b photocatalytic reduction of Cr(VI), and c photo-Fenton catalytic degradation of AO7 over BiFeO3 and CQD/BiFeO3 composites under visible light irradiation. d Recyclability of the 12C/BFO composite for photocatalytic degradation of AO7, photocatalytic reduction of Cr(VI), and photo-Fenton catalytic degradation of AO7 under visible light irradiation

Full size image

In this work, the photocatalytic ability of the samples for the reduction of Cr(VI) under visible light irradiation was also studied, as shown in Fig. 8b. The blank experiment indicates that the reduction of Cr(VI) after 3-h illumination in the absence of catalysts is negligible. It is seen that the CQD/BiFeO3 composites possess much higher photocatalytic reduction ability than pure BiFeO3. The reduction efficiency of Cr(VI) over the samples increases in the order: BiFeO3 < 3C/BFO < 6C/BFO < 24C/BFO < 12C/BFO. The result demonstrates the visible light-driven photocatalytic reduction property of BiFeO3, which can be obviously improved by the decoration of CQDs.

Besides the photocatalytic activity, it is demonstrated that BiFeO3 also displays promising photo-Fenton-like catalysis ability. Figure 8c shows the photo-Fenton degradation of AO7 over the samples under visible light irradiation with the addition of H2O2, from which one can see that the degradation percentage of AO7 in the photo-Fenton-like catalytic process is much higher than that in bare photocatalytic reaction. For example, about 96% of AO7 is photo-Fenton catalytically degraded over 12C/BFO sample under 3-h irradiation, which has a ~ 23% enhancement compared with the photocatalytic degradation of AO7 (~ 73%). In addition, it is found that the photo-Fenton catalytic activities between the samples have an order same to the photocatalytic activities between the samples. This suggests that the CQD/BiFeO3 composites can be used as effective photo-Fenton catalysts for the degradation of dyes.

Generally, the reusability of catalysts is regarded as an important parameter for their practical application. According to above catalytic results, the 12C/BFO sample was chosen as the catalyst for the investigation of photocatalytic and photo-Fenton catalytic stabilities. Figure 8d presents the catalytic activities of the 12C/BFO sample during three successive visible light-driven photocatalytic and photo-Fenton catalytic processes. After three consecutive cycles, the catalytic activities of the 12C/BFO sample do not undergo obvious decrease. This indicates that the CQD/BiFeO3 composite exhibits good catalytic reusability under visible light irradiation.

In this work, the NIR light-driven photocatalytic and photo-Fenton catalytic activities of BiFeO3 and 12C/BFO were investigated. Figure 9a–c display the time-dependent photocatalytic degradation of AO7, photocatalytic reduction of Cr(VI), and photo-Fenton catalytic degradation of AO7 over BiFeO3 and 12C/BiFeO3 under NIR light irradiation, respectively. It can be seen that bare BiFeO3 exhibits almost no NIR light photocatalytic activity because it cannot respond to NIR light, while about 22% of AO7 is degraded by BiFeO3 during the photo-Fenton catalytic reaction. In contrast, the 12C/BFO sample displays obvious NIR light-driven catalytic activities. After 3-h NIR light irradiation, the photocatalytic degradation of AO7, photocatalytic reduction of Cr(VI), and photo-Fenton degradation of AO7 over the 12C/BFO sample reach ~ 35%, ~ 63%, and ~ 49%, respectively. The result indicates that the introduction of CQDs onto the surface of BiFeO3 plays an important role in the enhancement of its NIR light-driven catalytic activity. The NIR light catalytic stabilities of the 12C/BFO sample were also studied by recycling catalytic experiments, as shown in Fig. 9d. It is found that the CQD/BiFeO3 composite also has steady NIR light-driven catalytic activity.

Fig. 9
figure 9

a Photocatalytic degradation of AO7, b photocatalytic reduction of Cr(VI), and c photo-Fenton catalytic degradation of AO7 over BiFeO3 and CQD/BiFeO3 composites under NIR light irradiation. d Recyclability of the 12C/BFO composite for photocatalytic degradation of AO7, photocatalytic reduction of Cr(VI), and photo-Fenton catalytic degradation of AO7 under NIR light irradiation

Full size image

Active Species Trapping

To explore the effect of active species on the catalytic degradation reaction, reactive species trapping experiments were carried out. Figure 10a and b show the photocatalytic and photo-Fenton catalytic degradation of AO7 using the 12C/BFO sample with the addition of quenchers under visible light illumination, respectively. From Fig. 10a, the introduction of ethanol and AO leads to relatively small inhibition on the AO7 degradation. In contrast, the photocatalytic degradation of AO7 is dramatically suppressed with N2 purging. This suggests that the ·O2− is the primary reactive species, whereas ·OH and h+ are the secondary reactive species responsible for the dye degradation. As shown in Fig. 10b, the degradation percentage of AO7 decreases from 96% (without scavengers) separately to ~ 60% (N2 purging), ~ 71% (adding AO), and ~ 45% (adding ethanol). This reveals that ·O2−, h+, and ·OH participate in the visible light-driven photo-Fenton catalytic reaction, and ·OH plays a relatively large role in this process. Figure 10c and d present the photocatalytic and photo-Fenton catalytic degradation of AO7 over the 12C/BFO sample in the presence of scavengers with the irradiation of NIR light, respectively. It can be seen that in the both catalytic processes, the dye degradation depends on ·O2−, h+, and ·OH. Particularly, ·O2− is demonstrated to be the main active species in the NIR light-driven photocatalytic process, whereas ·OH exhibits a key duty in the NIR light photo-Fenton catalytic reaction.

Fig. 10
figure 10

a and b Effects of ethanol, N2 purging, and AO on the photocatalytic and photo-Fenton catalytic degradation of AO7 over 12C/BFO under visible light irradiation, respectively. c and d Effects of ethanol, N2 purging, and AO on the photocatalytic and photo-Fenton catalytic degradation of AO7 over 12C/BFO under NIR light irradiation, respectively

Full size image

Figure 11 displays the time-dependent PL spectra of the TPA solution using the 12C/BFO sample as the catalyst in the photocatalytic and photo-Fenton catalytic reaction under visible and NIR light illumination. It is seen that, in all cases of the catalytic processes, the PL emission peak located at ~ 429 nm becomes intense gradually with the increase of the illumination time, indicating the generation of ·OH radicals. Based on the PL signal intensity, it is concluded that more ·OH radicals are generated in the photo-Fenton process than in the photocatalytic process, and the visible light irradiation leads to the increased generation of ·OH radicals when compared with the NIR light irradiation.

Fig. 11
figure 11

a and b PL spectra of the TA solution as a function of visible light irradiation time over the 12C/BFO sample in the photocatalytic and photo-Fenton catalytic reactions, respectively. c and d PL spectra of the TA solution as a function of NIR light irradiation time over the 12C/BFO sample in the photocatalytic and photo-Fenton catalytic reactions, respectively

Full size image

Photogenerated Charges Performance

Photoelectrochemical measurement is very useful for the investigation of the migration and recombination performance of photogenerated charges. The transient photoresponse currents of BiFeO3 and 12C/BFO under visible light irradiation with several on/off cycles are shown in Fig. 12a. One can see that the photocurrent density of 12C/BFO is much higher than that of bare BiFeO3, indicating the effective separation of photogenerated charges in the CQDs/BiFeO3 composite. Figure 12b displays the EIS curves of BiFeO3 and 12C/BFO. It is well known that the semicircle in the Nyquist plot at the high-frequency region reflects the interfacial charge-transfer process and a smaller diameter of semicircle means a lower charge-transfer resistance [66]. The 12C/BFO sample exhibits a smaller semicircle diameter compared with bare BiFeO3, suggesting that the migration of photogenerated charges can be promoted in the CQD/BiFeO3 composites.

Fig. 12
figure 12

a Photocurrent response plots of BiFeO3 and 12C/BFO under visible light irradiation. b EIS spectra of BiFeO3 and 12C/BFO

Full size image

Catalytic Mechanism

A possible visible light-driven photocatalytic mechanism of CQDs/BiFeO3 for the dye degradation and Cr(VI) reduction is proposed, as shown in Fig. 13a. When the CQD/BiFeO3 composite is irradiated by visible light, the BiFeO3 nanoparticles will be excited to generate photogenerated electrons and holes. On the other hand, the electrons in the CQDs can be also excited from their π orbital or σ orbital to the lowest unoccupied molecular orbital (LUMO) to obtain photoexcited electrons. It has been demonstrated that the excited CQDs can act as excellent electron donors and electron acceptors. Therefore, the photogenerated electrons in the conduction band (CB) of BiFeO3 nanoparticles will easily migrate to the π orbital or σ orbital of CQDs, while the photoexcited electrons of CQDs will transfer to the CB of BiFeO3. During the above converse electron migration process, the separation of photogenerated charges in BiFeO3 can be promoted, as revealed by photoelectrochemical measurement (see Fig. 12a). Thus, more photogenerated charges are available for participating in the photocatalytic reaction, leading to the improvement of photocatalytic activity.

Fig. 13
figure 13

a Photocatalytic degradation mechanism of AO7 over the CQDs/BiFeO3 composite photocatalysts under visible light irradiation. b Up-converted PL spectra of CQDs under different excitation wavelengths. c Photo-Fenton catalytic degradation mechanism of AO7 over the CQD/BiFeO3 composites under visible light irradiation. d NIR light photocatalytic mechanism of the CQDs/BiFeO3 composites. e NIR light photo-Fenton mechanism of the CQD/BiFeO3 composites

Full size image

More importantly, the up-converted PL property of CQDs also plays an important role in the enhancement of photocatalytic activity. Figure 13b presents the up-converted PL spectra of CQDs with the excitation wavelength from 810 to 890 nm, from which one can see that the up-converted emission peaks are centered at shorter wavelengths in the range of 400–680 nm. Because the light absorption edge of the as-prepared BiFeO3 nanoparticles is located at ~ 588 nm (see Fig. 4), the up-converted emission light (400–588 nm) of CQDs can be used to excite BiFeO3 nanoparticles to produce photogenerated electrons and holes, which provides additional photogenerated charges for the photocatalytic reaction. This also contributes to the enhancement of photocatalytic activity for BiFeO3 nanoparticles.

Besides the yield of photogenerated charges, the redox ability of photogenerated charges is considered to be another important factor for understanding the catalytic mechanism of catalysts. In our previous work, the CB and VB potentials of prepared BiFeO3 nanoparticles are calculated to be + 0.4 and + 2.47 V vs. NHE, respectively [55]. From a thermodynamic point of view, the generation of ·OH will be smoothly achieved because the VB potential of BiFeO3 is more positive than the redox potential of OH−/·OH (+ 1.99 V vs. NHE) [67]. Compared with the redox potential of Cr(VI)/Cr(III) (+ 0.51 V vs. NHE) [57], the photogenerated electrons in the CB of BiFeO3 is negative enough to reduce Cr(VI) to Cr(III). Another active species ·O2− can be obtained from the reaction between the photoexcited electrons of CQDs and O2 [68].

Figure 13c presents the visible light-driven photo-Fenton catalytic degradation mechanism of the dye over the CQD/BiFeO3 composites. In this case, the photocatalytic and Fenton reactions will simultaneously happen. When H2O2 is introduced into visible light-driven photocatalytic system, the H2O2 can react with Fe2+ on the surface of BiFeO3 to obtain additional ·OH along with the generation of Fe3+. Simultaneously, the Fe3+ will be reduced to Fe2+ by the photogenerated electrons of BiFeO3 and CQDs [69]. During this cycle reaction, more ·OH is produced, which is beneficial for the enhancement of catalytic efficiency.

Figure 13d and e display the photocatalytic and photo-Fenton catalytic mechanism of the CQDs/BiFeO3 composite under NIR light irradiation. It is known that the BiFeO3 do not response to NIR light (> 800 nm). As a result, only CQDs can be excited under NIR light irradiation in the two catalytic processes. The photogenerated charges migration and up-converted excitation of CQDs are similar to those as depicted in Fig. 13a and b. Because the BiFeO3 cannot be directly excited by NIR light, NIR light-excited CQD/BiFeO3 composite has a relatively lower yield of photogenerated charges compared with the visible light-excited composite. This is why photocatalytic and photo-Fenton catalytic activities of the CQD/BiFeO3 composites under NIR light irradiation are weaker than those under visible light irradiation.

Conclusions

The CQDs were successfully decorated on the surface of BiFeO3 nanoparticles through a hydrothermal route to obtain CQD/BiFeO3 composites. Under visible and NIR light irradiation, these composites manifest remarkably enhanced photocatalytic degradation of AO7, photocatalytic reduction of Cr(VI), and photo-Fenton catalytic degradation of AO7 compared with bare BiFeO3 nanoparticles. They can be reused without obvious decrease of catalytic activities. It is found that the introduction of CQDs leads to the efficient separation of photogenerated charges in the composites. The improved catalytic activities of CQD/BiFeO3 composites can be ascribed to the two factors: the excellent up-converted photoluminescence property and photogenerated electron transfer ability of CQDs.