Reduction potentials of some species that may be involved in photocatalytic systems (adapted from ).
The unique physicochemical properties of carbon quantum dot-(CQD)-based photocatalysts, notably their exceptionally good light absorption in the UV and near-visible region, tunable photoluminescence, extraordinary upconversion photoluminescence, outstanding electron affinity, and photoinduced electron transfer, and electron mobility, have attracted considerable attention in different photocatalytic applications. In this review, we summarized the fundamental mechanism and thermodynamics of heterogeneous photocatalysis of aqueous pollutants and the fundamental multifaceted roles of CQDs in photoredox process. Furthermore, we discussed the recent developments in the use of CQD-based materials as visible-light active photocatalysts in water purification. Finally, the challenges and future direction of CQD-based materials as photocatalytic materials for environmental decontamination were highlighted.
Carbon quantum dots (CQDs) are a recently discovered class of carbon-based nanomaterials that are typically discrete, quasi-spherical, and less than 10 nm in size  (although CQDs of sizes >10 nm have been reported [2, 3, 4, 5]). They are composed of -hybridized carbon atoms, have various surface functional groups, and possess composition-dependent fluorescence. This relatively new material has attracted huge interest in many fields including, but not limited to, electrocatalysis , biosensing [7, 8, 9, 10], bioimaging [11, 12, 13, 14], chemical sensing , and nanomedicine , due to their unique tunable photoluminescence (PL) properties, chemical inertness, high water solubility, ease and low cost of fabrication and, more importantly, low toxicity. Additionally, CQDs have also attracted considerable interest in various photocatalytic applications—environmental remediation [17, 18, 19, 20, 21], water splitting to produce production [22, 23, 24, 25, 26, 27], conversion [28, 29, 30], and synthesis of chemicals [28, 29, 30, 31, 32, 33]—because when coupled with a semiconductor photocatalyst, CQDs can provide with several advantages, including improved light harvesting ability, efficient usage of the full spectrum of sunlight, efficient charge carrier separation, stability, and hinder charge recombination. Figure 1 illustrates the various applications of CQDs in photocatalysis (left) and their competitive optical and structural properties during each photocatalytic procedure (right).
2. Principles of heterogeneous photocatalysis
The fundamentals of heterogeneous photocatalysis underlying the degradation of a pollutant employing a semiconductor catalyst have been intensively reported in many literatures [35, 36, 37, 38, 39, 40, 41].
A semiconductor (SC) is a material that has a valence band (VB) and a nearly empty conduction band (CB) that are separated from one another by a band gap . In the ground state (i.e., at ), all the electrons are found in the VB of the SC. When illuminated by light of energy greater than or equal to the band gap () of a semiconductor, electrons in the VB are excited to the CB, leaving behind equal number of voids (or holes) in the valence band. The generation of electron-hole pairs is the first step in the heterogeneous photocatalysis of organic and inorganic compounds by semiconductors. This step is illustrated by the enlarged section of Figure 2. Once generated, the fate of the electron and hole can follow several de-excitation pathways as shown in Figure 2.
First, the spatially separated charge carriers can migrate to the semiconductor surface where they can induce a redox reaction. At the surface, the electrons can be transferred to reduce an electron acceptor (A) (pathway 1). On the other hand, a donor species (D) can donate an electron to the surface hole and gets oxidized in the process (pathway 2). The probability and rate of the charge transfer processes for electrons and holes depends upon the respective positions of the band edges for the conduction and valence bands and the redox potential levels of the adsorbate species [40, 41, 42].
In competition with charge transfer to adsorbed species is electron and hole recombination. Recombination of the separated electron and hole can occur on the surface (pathway 3) or in the volume of the semiconductor particle (pathway 4). The electron-hole recombination can occur nonradioactively or radioactively to dissipate the input energy either as heat (E) or as a photon () [44, 45], respectively. The widely postulated steps of heterogeneous photocatalysts may be summarized in a simplified way according to Eqs. (1)–(5).
A thermodynamic requirement of spontaneous reaction is a negative change in Gibb’s energy, (i.e., G < 0) . In a photocatalytic reaction, this condition is met when the quasi-Fermi (usually taken as the flat band potentials) energy levels of the photogenerated electrons and holes of the photocatalyst straddle the reduction/oxidation potentials of the substrate. In other words, the bottom of the photocatalyst’s conduction bands () must be located at a more negative potential than the reduction potential () of the substrate, while the top of its valence bands () must be positioned more positively than the oxidation potential () of the substrate [43, 47] as illustrated in Figure 3. Therefore, the spontaneity or otherwise the charge carrier-induced redox reactions, generically depicted by Eq. (5), is determined by the potential of the band edges of the semiconductor photocatalyst [48, 49].
As shown in Figure 4, the redox potential of the VB and the CB for different semiconductors varies between +4.0 and −1.5 V vs. normal hydrogen electrode (NHE), respectively. Many organic compounds have more negative oxidation potentials than the valence potential of most semiconductor photocatalysts. Accordingly, it is thermodynamically possible for the organic pollutants to be oxidized by photocatalysts. In contrast, fewer organic compounds can be reduced by photocatalyst since only a smaller number of them have a potential below that of the conduction band of most photocatalysts .
While choosing a photocatalyst for application in environmental remediation, it is also essential to take into account the reduction potentials (E) of the substrate, as well as those of the intermediates that are formed during the photocatalytic reaction(s). However, the reduction potentials for different organic or inorganic compounds are usually dependent on the reaction conditions such as pH and the employed electrolyte. The reduction potential of the couples refers to the reactions described by Eqs (7) and (8):
These reactions refer to one-electron reduction reactions—the potentials of which are given versus the standard hydrogen electrode (SHE). Table 1 includes the reduction potentials of the most important inorganic species that may be present in photocatalytic systems.
|Reductiom reaction||Ea, (V)||Remark|
For example, over a photocatalyst in pH 7 solution, reaction Eq. (9) is feasible because the redox potential of the vs. SHE is sufficiently more negative than the reduction potential (V vs. SHE) of the superoxide radicals. Reaction (10) occurs because the redox potential of the V vs. SHE is sufficiently more positive than the oxidation potential (V vs. SHE) of the hydroxyl radical.
Moreover, it is in fact critical to provide an overpotential for each process, to initiate and subsequently drive the electron transfer process. Without an overpotential, even a good photocatalyst cannot ensure sufficiently a high rate of reaction.
3. Beneficial roles of CQD materials in photocatalysis
3.1. Extends the optical absorption range of photocatalysts
The absorption of large portion of incident solar radiation by a photocatalyst is one of the key factors for achieving useful efficiencies in photocatalysis . For wide bandgap semiconductors such as , a common strategy for achieving this is through the use of photosensitizers. A photosensitizer is excited by low-energy photons to generate electrons that are subsequently injected into the conduction band of the wide bandgap photocatalyst. The common photosensitizers are usually characterized as expensive [53, 54, 55], toxic , unstable , and polluting . These drawbacks limit the practical and large-scale application of conventional photosensitizers. Accordingly, it is essential to find an alternative photosensitizer material(s) that is free from these drawbacks and one that can harvest larger portion of the solar radiation. Carbon quantum dot appears to be such a material. It is easy to produce, cheap, nontoxic, green, stable, and abundant [1, 58, 59]. Additionally, it displays strong blue photoluminescence and good optical absorption in the UV and near-visible region . Futhermore, CQDs exhibit upconversion photoluminescence properties [19, 61, 62, 63] and exhibit promising electron transfer properties [64, 65]. These optical characteristics of CQDs render them promising candidates as photosensitizers for photocatalytic applications. Three possible photsensitization mechanisms are proposed for CQDs in the literature. Firstly, CQD can be excited by a low-energy radiation to generate holes () and electrons (). Under favorable thermodynamic condition, the photogenerated is injected into the conduction band of a wide bandgap photocatalyst (such as ) to initiate a reaction as shown in Figure 5(a) . Secondly, the CQDs with upconversion PL properties can convert longer wavelength light to the short wavelength light, which in turn can excite a wide bandgap photocatalyst (ZnO) to form pairs (Figure 5(b) ). Finally, addition of CQDs may lead to bandgap narrowing of the semiconductor owing to the chemical bonding between semiconductor and CQD, which results in the extended light absorption range as shown in Figure 5(c) .
Recently, several reports have been published that show CQDs to dramatically extend the optical absorption range of wide and narrow bandgap photocatalysts to the entire visible-light range and beyond [18, 19, 20, 21, 58, 60, 62, 68, 69, 70, 71]. For example, in a recent publication, Ye et al.  reported using CQDs to extend the light absorption range of to the entire visible range. The resulting CQD/composite photocatalyst achieved an absolute photocurrent density of as shown in Figure 6(a). A photoanode fashioned from the composite (NiOOH/FeOOH/CQDs/) achieved a photocurrent density of at 1.23 V vs. RHE under AM 1.5G in aqueous solution without a hole scavenger (pH 7) and record a high applied bias photon-to-current efficiency of 2.29% at 0.6 V vs. RHE. Ren et al.  reported that coupling of CQDs with resulted in a composite, CQD/photocatalyst that is active over the entire range of the solar spectrum (Figure 6(b)).
Carbon quantum dots have also been applied as visible-light photosensitizer for non-noble metal -evolution catalyst [24, 60, 68]. Martindale et al. used CQDs as a photosensitizer to energize a nickel-based molecular catalyst, Ni-bis-(diphosphine) (NiP) to produce under visible-light irradiation [24, 60]. A proposed scheme of production in the homogeneous CQD-NiP system is presented in Figure 6(c). Irradiation of photoluminescent CQDs with visible-light results in the direct transfer of photoexcited electrons to the catalyst NiP with subsequent reduction of aqueous protons to . The electron donor EDTA  or TCEP/EDTA  quenches the photogenerated holes in the CQDs. In a similar work, McCormick and co-workers  used a PVP-coated CQD as a photosensitizer for a nickel nanoparticle (NiNP) catalyst (Figure 6(d)) to produce at a much higher quantum yield of 6%. In this report, it was observed that the fluorescence quantum yield of CQDs increases with increasing PVP coating. However, production decreased when the PVP coating of CQD is greater than 20%. A total of 330 /g CQD was collected from a 20% PVP-coated CDQ/Ni nanoparticle system at a pH of 4.51 after five irradiation with a 470-LED light source.
Upconversion photoluminescence property (UCPL) of CQDs was used to improve the visible-light and NIR response of narrow bandgap composite photocatalysts. For example, Liu and coworkers  used the upconversion property of nitrogen-doped CQDs to improve the visible-light and near infra-red response of graphitic carbon nitride (g-). Similarly, Mao et al.  improved the visible-light and NIR response of by coupling it with CQDs. The resulting composite photocatalyst exhibited improved photoactivity under NIR irradiation. Figure 6(e) shows the proposed CQD sensitization mechanism.
3.2. Enhances charge separation and electron transfer
One of the main processes that limit the quantum efficiency of a photocatalytic system is the fast recombination of photogenerated charge carriers . Therefore, to enhance the performance of a photocatalyst, it is essential to improve charge carrier separation and minimize the rate of their recombination. Several approaches have been devised to achieve this goal. These included strategies such as surface modification of the semiconductor particles with noble metals [73, 74], coupling of two semiconductor particles with different electronic levels [75, 76, 77], and using sacrificial reagents to scavenge for photogenerated electrons or holes [78, 79, 80]. Another strategy is to couple carbon-based materials with a photocatalyst because of their high charge storage capacity and electrical conductivity [81, 82, 83]. In particular, many reports have been recently published that show the beneficial effect of coupling of CQDs with photocatalysts [18, 19, 20, 21, 63, 71, 84, 85, 86]. The intrinsic band gap and strong electron affinity [59, 65, 87] of CQDs give them the ability to readily accept photogenerated electrons from an electron donor such as a semiconductor with a more negative conduction band minimum. The transferred electrons are then shuttled freely along the conducting paths of the CQDs allowing for effective charge separation, stabilization, and prevention of charge recombination. The longer-lived charge carriers have greater probability to induce transformations, thus accounting for the much improved quantum efficiency of the CQD-based photocatalysts. Figure 7 depicts a proposed photoinduced electron transfer mechanism on a layered composite of CQD/CdS photocatalyst .
Results of time-resolved (TR) PL and photocurrent measurements corroborated the slow recombination rates and enhanced the separation of electron-hole pairs observed in CQD-based photocatalysts. Figure 7(b) and (d) showed the time-resolved photoluminescence (TRPL) measurements on pure and CdSe and their CQD-based counterparts with different amounts of CQD. Because the TRPL plots of the CQD composite photocatalysts exhibited lower PL intensities, it implies that the CQDs’ modification effectively inhibits the recombination of photogenerated electron-hole-electron pairs in the hybrid photocatalysts. Figure 7(c) and (e) showed the corresponding photocurrent responses. The higher photocurrent signals recorded for the CQD-based photocatalysts compared to the pure photocatalysts demonstrated that much more photogenerated charge carriers were produced and the electron-hole pairs could separate more efficiently [62, 89].
3.3. Provides additional surface for adsorption and reaction
Another key role of CQD, which promotes photocatalytic activity, is its capacity to provide additional surface for adsorption and reaction of substrates. As a nanomaterial, CQDs possess a larger surface area to volume ratio, and thus, its composite with other nanomaterials potentially acquires much enhanced surface area, thereby increasing the adsorption capacity. Like other carbon materials, adsorption capacity of CQD originates from the flexible -bonded carbon structure and the large surface area. Because the surface of CQDs contains a plethora of oxygenated functional groups , its adsorption capability depends on its interaction with the adsorbate. The aromatic regions of CQDs can form a -stacking interaction with organic adsorbates containing aromatic structures to enhance their adsorption. On the other hand, the presence of functional groups such as carbonyl, epoxy, hydroxyl, and amino groups on the surface of CQDs promotes the adsorption of a wide variety of molecules and metal ions. The increased adsorption of target reactants on the surface of photocatalyst increases their chance of reacting with photogenerated reactive oxidative species, thus enhancing photocatalytic activity of the composite CQDs-based photocatalyst [90, 91, 92].
4. CQD-based semiconductor composites as visible-light active photocatalysts for water treatment
Recently, a large number of studies [17, 18, 19, 20, 21, 85, 86, 93, 94, 95] have reported on the application of CQD/semiconductor nanocomposites as photocatalysts for water purification owing to the exceptional properties exhibited by the composites. Carbon quantum dots have been coupled with variety of metal-oxide semiconductors such as , , , , , and others and have found to be an excellent photocatalyst for the degradation of pollutants. Table 2 summarizes some of the CQD-based composites that have been investigated as photocatalysts for the degradation of aqueous pollutants.
|CQD composite||Targeted pollutant||Visible-light source||Experiment conditions||, %||Reference|
|Methylene blue (MB)||300 W Xe lamp, 400-nm filter||100 mg cat., MB (100 mL, 10 mg/L), 30 C, 120 min||—|||
|Methylene blue||250 W Xe lamp, 420-nm filter||50 mg cat., MB (100 mL, 10 mg/L), 150 min||100|||
|Methylene blue||500 W Xe lamp, 420-nm filter||MB (15 mL, 5 mg/L), 100 min||35|||
|Methylene blue||> 420 nm light||MB (50 mg/L), 25 min||100|||
|Methylene blue||> 420 nm light||120 min||90|||
|Methylene blue||> 420 nm light||MB (50 mg/L), 15 min||100|||
|Methylene blue||400 W halogen, 400-nm filter||50 mg cat., MB (100 mL, 20 mg/L), +35% , 90 min||97.3|||
|Methylene blue||400 W halogen, 400-nm filter||30 mg cat., MB (100 mL, 50 mg/L), 120 min||88|||
|N-CQD||Methylene blue||> 420 nm light||10 cat., MB (3 mL, ),, 6.5 min||100|||
|Ciprofloxacin (CIP)||300 W Xe lamp with 400-nm filter||100 mg cat., CIP (100 mL, 10 mg/L), 30, 120 min||88|||
|Ciprofloxacin||300 W Xe lamp, 400-nm filter||50 mg cat., CIP (100 mL, 10 mg/L), 120 min||87|||
|CQD/BiOBr||Ciprofloxacin||300 W Xe lamp, 400 nm filter||30 mg cat., CIP (100 mL, 10 mg/L), 4 h||44.3|||
|Bisphenol A (BPA)||300 W Xe lamp, 400-nm filter||100 mg cat., BPA (100 mL, 10 mg/L), 30, 120 min||54|||
|CQDs/BiOI||Bisphenol A||300 W Xe lamp, 420-nm filter||30 mg, 50 mg, 70 mg Cat., BPA(100 mL, 10 mg/L), 120 min||99|||
|Bisphenol A||300 W Xe lamp, 400-nm filter||50 mg cat., BPA (100 mL, 10 mg/L), 120 min||—|||
|CQD/BiOBr||Bisphenol A||300 W Xe lamp, 400-nm filter||50 mg cat., BPA (100 mL, 10 mg/L),||66|||
|Tetracycline chloride (TC)||300 W Xe lamp, 400-nm filter||100 mg cat., TC (100 mL, 20 mg/L), 30, 120 min||—|||
|CQDs/BiOI||Tetracycline chloride||300 W Xe lamp, 420-nm filter||30 mg, 50 mg, 70 mg Cat., TC(100 mL, 20 mg/L), 120 min||68|||
|Tetracycline chloride||300 W Xe lamp, 400-nm filter||50 mg cat., TC (100 mL, 20 mg/L), 120 min||—|||
|Cr^6+||> 420 nm light||180 min||99|||
|Methyl orange (MO)||150 W Xe lamp, 420-nm filter||200 mg cat., MO (100 mL, 66 ppm), 25 min||100|||
|Rhodamine B (RhB)||500 W Xe lamp, 420-nm filter||100 mg cat., MB (100 mL, 10 ppm), 60 min||95.4|||
|CQDs/BiOI||Rhodamine B||300 W Xe lamp, 420-nm filter||30 mg, 50 mg, 70 mg Cat., RhB(100 mL, 10 mg/L), 30 min||61|||
|CQD/BiOBr||Rhodamine B||300 W Xe lamp, 400-nm filter||20 mg cat., RhB (100 mL, 10 mg/L), 30 min||100|||
|Rhodamine B||150 W NIR lamp, 700-nm filter||50 mg cat., RhB (100 mL, 10 mg/L), 150 min||48|||
|Rhodamine B||300 W Xe lamp, 400-nm filter||50 mg cat., RhB (100 mL, 10 mg/L), 120 min||—|||
|CQDs/BiOI||Rhodamine B||300 W Xe lamp, 400-nm filter||30 mg cat., RhB (100 mL, 10 mg/L), 120 min||94.9|||
|CQDs/CdS||Nitro-benzene||3 W LED light, 420-nm filter||30 mg cat., RhB (15 mL, 10 mg/L), 120 min,||—|||
Chen et al.  reported the fabrication of nitrogen doped (i.e., NCQD/) complex photocatalyst with improved activity under visible-light irradiation. As shown in Figure 8(a), the activities of the hybrid NCQD/materials are significantly enhanced for the degradation of methyl orange (MO) compared with the pure under visible light (> 420 nm). A composite of NCQD/prepared from a 3 mL solution of 0.1 g CQD/L of distilled water denoted 3-NCQD/exhibited the highest activity. The 3-NCQD/sample eliminated 98% of the MO present in 18 min. In another report, Ren et al.  reported improved activity for the reduction of Cr(VI) ion over CQD/composite photocatalyst compared to the activity over pure (Figure 8(b)). Similarly, Xie et al. also reported on the beneficial effects of complexing CQDs with photocatalysts. Visible light active to improve the rate of charge transfer and reducing the rate of recombination in photocatalysts. The resulting CQD/composites exhibited improved photocatalytic activities as shown in Figure 8(c).
Li et al.  first reported the exploitation of the unique optical properties of CQDs in photocatalysis. The team synthesized /CQDs and /CQDs nanocomposite systems (Figure 9a) from CQDs with upconversion photoluminescence (PL) properties (Figure 9b) and used them to investigate the degradation of aqueous methylene blue (MB) under visible light. The as-synthesized CQDs-based photocatalysts exhibited enhanced ability to promote the degradation of the dye compared to their precursor components (CDQs, , or ) as shown in Figure 9c. Under the given experimental conditions (50 MB solution, suspended /CQDs or /CQDs, 300 W halogen lamp irradiation) used, the MB was almost completely degraded in 15 and 25 min over /CQDs and /CQDs, respectively.
Li and co-workers attributed the enhanced photocatalytic activities of the nanocomposites under visible light to upconversion and efficient charge carrier separation in the nanocomposite systems. The former allows CQDs to emit short-wavelength photons after absorbing visible light. The emitted short-wavelength UV light then induces photoexcitation of the and to generate the electron/hole () redox pairs that go to initiate the process of MB degradation via the generation reactive oxygen species such as and and others. The appropriate position of the conduction band edge of the CQDs allows electron transfer from the surface of the semiconductors, thereby enhancing charge carrier separation and inhibiting recombination resulting in the increased number of and available for photoreactions.
Yu et al.  reported the synthesis of CQDs/complex photocatalysts via a facile solvent-thermal process in an aqueous solution (Figure 10a). The photocatalytic activity of photocatalyst was evaluated by the degradation of MB as a model contaminant under visible-light irradiation. The results (Figure 10b) showed that coupling of CQDs with greatly enhances the decomposition of MB under visible-light irradiation. The authors attributed the enhanced activity of the photocatalyst to the mesoporous architecture and large surface area of the -, which provides more abundant active sites for the adsorption and degradation of MB to occur and also for the effective electron transfer from the conduction band of the particle to the conducting networks of the loaded CQDs as shown in Figure 10c.
5. Conclusions and outlook
In summary, various types of carbon quantum dots/semiconductor composites have been investigated for the design of novel materials for water remediation via heterogenous photocatalysis. The coupling of CQDs with semiconductors synergistically enhances the activity of the photocatalysts for the degradation of various waterborne pollutants. The unique properties of CQDs, in conjunction with size-dependent properties of nanomaterials, induce further functionalities to the composites such as high adsorption capacity, extended light absorption range, and improved charge separation properties along with high stability. Despite the confirmation of the excellent activity of CQDs-based photocatalystsic aforementioned reports, the applicability of the heterogeneous photocatalytic technology in large-scale water treatment is constrained by several key technical issues that have to be addressed.
The main constraints associated with the field scale application of CQD-based photocatalysts essentially lies in the design of (i) a catalyst that can utilize large portion of the solar spectra, (ii) catalyst immobilization strategy to provide a cost-effective solid-liquid separation, (iii) improvement in the photocatalytic operation for wider pH range and to minimize the addition of oxidant additives, (iv) new integrated or coupling system for enhanced photomineralization or photodisinfection kinetics, and (v) effective design of photocatalytic reactor system or parabolic solar collector for higher utilization of solar energy to reduce the electricity costs.
The authors gratefully acknowledge the financial support of the Tertiary Education Trust Fund of the Nigeria Ministry of Education.