Modulation of Z-scheme photocatalysts for pharmaceuticals remediation and pathogen inactivation: Design devotion, concept examination, and developments.

The recent outbreak of Covid-19 guarantees overconsumption of different drugs as a necessity to reduce the symptoms caused by this pandemic. This triggers the proliferation of pharmaceuticals into drinking water systems. Is there any hope for access to safe drinking water? Photocatalytic degradation using artificial Z-scheme photocatalysts that has been employed for over a decade conveys a prospect for sustainable clean water supply. It is compelling to comprehensively summarise the state-of-the-art effects of Z-scheme photocatalytic systems towards the removal of pharmaceuticals in water. The principle of Z-scheme and the techniques used to validate the Z-scheme interfacial charge transfer are explored in detail. The application of the Z-scheme photocatalysts towards the degradation of antibiotics, NSAIDs, and bacterial/viral inactivation is deliberated. Conclusions and stimulating standpoints on the challenges of this emergent research direction are presented. The insights and up-to-date information will prompt the up-scaling of Z- scheme photocatalytic systems for commercialization.

Introduction

A consensus has been reached by different stakeholders that energy crisis and environmental pollution are primary challenges facing the world in the 21st century with water pollution considered as the major contributor of human adverse health effects and environmental deterioration. Water pollutants consist of heavy metals, pharmaceuticals, bacteria, and organic matter from domestic waste. It is estimated that water pathogens such as bacteria, viruses, fungi, protozoa, prions and parasites initiates about 80% of ailments in unindustrialized nations [1] and there is poor prophylaxis of waterborne epidemics globally [2]. Pharmaceuticals form a major constituent of priority pollutants in aquatic environments initiating frequent detection studies and possible remediation techniques being a substance of contemporary attention.

Photocatalysis is defined as a technique that results in production of oxidizing agents like hydroxyl radicals (•OH), superoxide anion radical (•O2 -), holes (h+), and singlet oxygen (1O2). In semiconductor photocatalysis, the initial step (step i) involves the absorption of light energy equal or greater than the bandgap of the semiconductor. The subsequent steps that follow photoexcited electrons and holes may take four pathways that are distinctively known as electrons and holes surface recombination (route A), electrons and holes bulk recombination (route B), surface oxidation reaction initiated by photoexcited holes (route C), and surface reduction initiated by photoexcited electrons (route D) [3], [4]. Routes A and B occur within a period of a few ps - ns [5]. The photodegradation activity of photocatalysts reliant on these four routes is determined by (1) the light harvesting ability with a requirement of wide light energy absorption range, (2) rapid migration of photogenerated charges and transportation to the surface of the semiconductor active/reaction site, and (3) strong reduction and oxidation reactions at the photocatalytic reaction sites [6]. However, the requirement of wide light energy absorption is a narrow bandgap energy of the photocatalyst as the Coulomb field equation, i.e., Fc = Kqeqh/r2 demonstrates that force (Fc) of attraction of photogenerated charge pairs (qe, qh) is inversely proportional to the reduction in bandgap energy (r) [7], K is proportionality constant. Ironically, a wide bandgap results in strong redox ability and less likeliness for recombination of excited electrons and holes, creating a challenge for a distinct semiconductor to possess all features that are significant for proficient photocatalytic reactions [8].

Diverse tactics investigated for possible improvement of the general efficiency of semiconductors for removal of pharmaceuticals include metal doping [9], non-metal doping [10], multi-elemental doping with metals and/or non-metals [11], use of supports [12], and formation of heterojunctions [13], [14]. The contribution of heterojunctions towards abatement of different classes of pharmaceuticals such as β- blockers, hormones, antibiotics, cholesterol-lowering drugs, oestrogens, anti-psychotics, non-steroidal anti-inflammatory drugs (NSAIDs) and analgesics has been studied recently [15], [16], [17], [18], [19], [20]. There are numerous heterojunctions that have been reported in literature for effective degradation of pollutants and they comprise p-n, traditional type II, R-scheme, C-scheme, p-n-p and Z-scheme heterojunctions, etc. [21], [22], [23], [24]. The Z-scheme heterojunction evolved from mimicking the natural photosynthesis process that plants employ for food generation [25]. The illustration of electron transfer process in natural photosynthesis forms a shape identical to the English letter “Z” initiating its name. In some recent reviews, the natural photosynthesis process has been described in detail for better understanding [25], [26], [27]. In summary, Z-scheme photosynthesis encompass garnering of solar energy through chlorophylls and their subsequent transfer to the reaction centre to initiate water oxidation on the donor side of photoreaction system II (PS II) while nicotinamide adenine dinucleotide phosphate (NADP) aided reduction of CO2 to carbohydrates occurs at the acceptor side of photoreaction system I (PS 1) [25]. Like photosynthesis, Z-scheme photocatalytic heterojunctions offers high charge separation quantum efficiency with presence of PS I and PS II promising to conquer inadequacies of strong redox ability and wide light absorption requirements.

Numerous reviews have been compiled on the topic of different properties and applications of Z-scheme photocatalytic systems [3], [6], [28]. Xue et al. emphasized elementary ideologies and construction of silver-based Z-scheme photocatalytic systems with clear role of metallic silver explored under numerous applications [29]. Wang et al. [27] systematically outlined different and recent redox mediated properties and applications of traditional Z-scheme photocatalytic systems for water splitting towards renewable H2 evolution focusing on reaction mechanisms and different performance of various elected acceptor and donor pairs. Important discussions exists to distinguish the different charge transfer mechanisms involved in type II, Z-scheme and Step-Scheme mechanisms with the aim of signalling misinterpretations that may stir up [30]. Charge transfer mechanisms of direct Z-scheme have been analysed to occur through internal electric field (IEF), interfacial defect states and selective facet manipulation [31]. However, no specific review channels its discussion to pharmaceuticals which, considering the current Covid-19 pandemic outbreak, is of urgent attention due to an upsurge of drugs usage. It is imperative to outline appropriate designs, analyse the concept of Z-scheme towards degradation of pharmaceuticals, and the possibility of employing Z-scheme photocatalytic systems for removal of antivirals and antibiotics that are employed to reduce the impact of Covid-19.

This review focusses on the applications of Z-scheme composites towards degradation of pharmaceutical NSAIDs, antibiotics, antiviral, and antimicrobial activity, the concept proof of Z-scheme mechanism, efficiency, and determined intermediates. Moreover, the applications of Z-scheme photocatalysts for potential laboratory scale wastewater treatment plants and their real industrial applications towards mineralization of pollutants and medical applications have been reviewed. Lastly, suggestions on future directions and recommendations are made towards realization of industrial applications of Z-scheme heterojunctions.

Verification of Z-Scheme

It has progressively become imperative to scientifically analyse the actual charge transfer that transpires at the interface of different semiconductors due to its spontaneity which birthed different heterojunctions. Here, brief, and clear discussions of the characterisation methods employed towards elucidation of Z-scheme photocatalysis are outlined with subdivisions of theoretical and experimental validation techniques.

Theoretical Proof

Theoretical investigations that have been used and accepted scientifically to provide enough evidence are simulation techniques that predict the behavioural interaction of different elemental materials input into a software. They can be employed towards structural, electronic, and interaction predictions. The parameters are important to understand different assumptions made towards establishment of the results. Density functional theory (DFT), frontier electron density calculation (FED), etc. are some of the theoretical methods that have been used for different applications [32], [33], [34]. Just recently, these theoretical investigations were employed towards validation of charge transfer in Z-scheme heterojunctions.

The DFT calculations have been the most commonly used theories for different applications with different software. The detailed inspection and comparison of software performance towards a particular application would be an interesting study to actually determine the best software package to use for determining the charge transfer at the interface of two different semiconductors. With each different software, minor and major parameters are reported as they may affect the outcome of the whole process under investigation. The parameters include the module, cut-off energy, quality of k-points, and size of vacuum layer [34], [35], [36]. After the parameters are set, the most important part is in the interpretation of the DFT data obtained towards understanding the interfacial charge transfer. Density of states, bandgap and work functions together with electron depletion and accumulation, and planar averaged electron density difference Δρ(z) along the z direction collaboratively determine electron transfer [37]. Wang et al. constructed a novel AgI/BiSbO4 Z-scheme heterojunction via a hydrothermal-precipitation method towards degradation of acid red G (ARG) and tetracycline (TC) with in-situ generated surface plasmon resonance (SPR) from Ag0 during degradation [37]. They employed DFT calculations to study the interfacial charge transfer at the interface of AgI and BiSbO4 (Fig 1 a-e)

Figure 1

TDOS and PDOS of (a) BiSbO4 and (b) AgI. (c) Charge difference distribution between AgI with (002) plane and BiSbO4 with (020) plane: charge accumulation is in yellow and depletion in cyan. (d) Planar averaged electron density difference Δρ (z) along the z direction of the interlayer. (e) Electronic location function (ELF) of the interlayer [37].

From Fig 1a and b, the density of states of BiSbO4 and AgI are conveyed and bandgaps of the respective semiconductors were determined as lower than the experimental values. This result is also similar to what was found by most studies that determines both experimental and theoretical bandgaps like the work of Du and co-workers [38]. Fig 1c and d shows that AgI semiconductor donate free electrons to BiSbO4 and BiSbO4 shows positive value indicating it is an electron accumulation side while the negative value exhibited by AgI forming an electron transfer route across the heterojunction. Generally, DFT calculations successfully portrayed formation of interlayer at the interface, and that the exhibited accumulation of electrons on BiSbO4 after equilibrium demonstrated charge transfer from VB of AgI to BiSbO4 CB. DFT calculations offer a clear insight into interfacial charge transfer that occurs between two semiconductors which can then be employed to affirm the nature of the heterojunction. More details of validation of Z-scheme charge transfer mechanism under different applications can be found in a review by Bao and co-workers on S-scheme photocatalytic systems [7].

Experimental validations

Despite theoretical investigations towards validation of Z-scheme heterojunctions formation which is normally done with DFT calculations, numerous scientific methods are employed to validate the formation of Z-scheme charge transfer mechanism in a formed heterojunction. These methods include metal nanoparticle photodeposition, Mott-Schottky (M-S) test, reactive oxygen species (ROS) generation test, photocatalytic CO2 reduction, in situ irradiated X-ray photoelectron spectroscopy (XPS), and just recently microscopic time resolved imaging technique [26], [39], [40]. The use of a single validation method may however not be sufficient and numerous methods may be required as complementary experimental or theoretical evidence.

In metal nanoparticle photodeposition, the principle adopted is the photodeposition of noble metal co-catalysts that are described as (Mz+ + ze-→M0), where M is a metal ion with charge z required for as its reduction to zero valence state [39]. The photoreduction is normally done on the surface of the semiconductor with the intended metal precursor with a tendency for the photodeposition to occur where there is accumulation of electrons. Since Z-scheme mechanism has distinctive reduction and oxidation sides, photodeposition will occur on the reductive or sometimes on the oxidative semiconductor [40], [41]. It is possible to predict the charge transfer mechanism under light irradiation in the presence of electron donor or acceptor sacrificial agents, based on the metal photodeposition as the photo-deposited metals can be seen under TEM and mapping will show on which semiconductor the photodeposition occurred.

In sacrificial reagent testing, sacrificial reagents are carefully selected to determine regions of electrons and holes accumulation in a Z-scheme heterojunction system, especially with enough redox power for water-splitting. The selected sacrificial reagents are used such that one reagent can allow HER to occur without O2 production while the other can allow OER without detection of H2 evolution [40]. This process is also time consuming as it requires that the individual semiconductor pairs in a Z-scheme to be loaded with another semiconductor to analyse the charge transfer like in an example of PS I and PS II being loaded with Co3O4 in a work by Zhu et al. [42]. The common sacrificial reagent pairs that have been used include EDTA and AgNO3. Lastly, both PS I and PS II use must allow HER and OER without use of sacrificial reagents for the test to be complete. Therefore, sacrificial reagent experiment can be employed successfully towards the elucidation of charge transfer mechanism in direct Z-scheme photocatalyst systems.

In radical trapping experiment, a chemical reagent is added to quench the effect of a particular set of expected radical species through capturing it [43]. Comparisons based on the activity with and without the radical trapping chemical reagent will give insights into existence and possible magnitude of the specific radical contribution [44], [45]. The understanding of the involved radicals aid towards prediction of charge transfer mechanism in a Z-scheme heterojunction [44]. Since the most common initiators of photocatalytic redox reactions stems from h+, hydroxyl radicals and superoxide anion radicals, radical scavengers for each one of them include TEOA, EDTA and ammonium oxalate (AO) for h+ [46], [47], tert-butyl alcohol (TBA) and isopropanol (IPA) for OH radicals [48], [49], and N2 gas and p-benzoquinone (BQ) as superoxide anion radical quenchers [46], [50]. In our previous work, we managed to induce a Z-scheme charge transfer mechanism between CO3O4 and BiOI towards direct Z-scheme charge transfer using just radical trapping tests and some theoretical investigations that included electronic properties of the materials [43]. Similar reports exist from other groups for successful use of radical trapping experiments to confirm charge transfer pathway in Z-scheme [51].

With ongoing attempts to understand charge transfer mechanism in Z-scheme, Ebihara et al. applied pattern-illumination time-resolved phase microscopy towards in depth investigation to probe photogenerated charge carrier dynamics using a Mo-doped BiVO4 and Rh- doped SrTiO3 with indium tin oxide as the electron mediator [52]. They did not only manage to map reactive sites and recombination centres, but they also fruitfully observed position- and structure-dependency of photoexcited charge route. Interestingly, Mott-Scotty plots have also been implemented to understand band structure of individual composites that can then be used to predict the charge transfer dynamics in Z-scheme heterojunctions [53]. However, Mott-Schottky plots are more clearly used to further elucidate and complement the radical trapping experiments. In our previous work, we showed the use of M-S plots to portray formation of direct Z-scheme heterojunction between ceria and CN [54]. The M-S plots were used to give the band structure while the radical trapping experiments gave information on most active radicals which were used together to show existence of direct Z-scheme charge transfer between CN and ceria. Comparison of possible heterojunctions with generated radicals inevitably rules out formation of some heterojunctions and thus give insights into the charge transfer route. Also some reports exist for complementary use of M-S plots to further describe charge transfer in a Z-scheme [55].

XPS is used for chemical composition and surface oxidation states analysis and it has also proved to be a vital technique to study the charge transfer pathway in a heterojunction by exploring the relative shift of binding energy of individual elements in the composite [56]. When a foreign material is added to a semiconductor, the binding energy shift occurs subsequent to existence of electron transfer between them, where a positive shift in binding energy validates decrease in electron density of the element while a negative shift affirms increase in electron density. Therefore, a shift in binding energy in XPS spectra can be used to confirm electron transfer mechanism at the interface of a Z-scheme heterostructure. Jo et al. reported that the decrease in the intensity and shift to higher binding energy of elements in high XPS spectra of composites compared to pristine semiconductors confirmed Z-scheme charge transfer between g-C3N4/TiO2 [57].

In CO2 photoreduction reaction, photoexcited electrons play a major role based on their conduction band edge potential energy after a heterojunction is formed. The product of a photoreduction reaction is used to predict the charge transfer mechanism when the band edge positions of the semiconductors are determined (experimentally or theoretically) such that if a traditional heterojunction is formed, the reduction in the redox capability of the semiconductor reaches a level that is undesirable for photoreduction of CO2 to CO (-0.42 eV vs. NHE), CH3OH (-0.38 eV vs. NHE) or CH4 (eV vs. NHE) [58], [59], [60], [61]. Therefore, CO2 photoreduction gives different products based on the reduction potential [40]. Its application towards confirmation of Z-scheme transfer mechanism is portrayed in proposed Z-scheme BiOI/g-C3N4 towards CO2 photoreduction under visible light by Wang and co-workers [59]. The use of CO2 experiment towards confirmation of Z-scheme mechanism was employed after CB and VB potentials of BiOI were 2.42 and 0.67 eV, and that of g-C3N4 were determined to be 2.17 and −0.52 eV, respectively. If the photocatalytic system obeys a traditional double charge transfer mechanism, the accumulated electrons on the conduction band of BiOI would be more positive than the CO2 reduction potentials leaving the Z-scheme as the possible charge transfer mechanism. Studies towards confirmation of Z-scheme charge transfer mechanisms are influenced by (1) the reaction products obtained and (2) the band edge potentials of the individual semiconductors.

Electronic paramagnetic resonance (EPR) is a spectroscopic technique commonly employed towards systematic investigation of charge transfer mechanism at the interface of semiconductors. Its application in the determination of Z-scheme charge transfer mechanism has largely increased over the last decade and it is based on the determination of •OH and •O2 − indirectly through use of 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a trapping agent to generate EPR measurable DMPO-•OH and DMPO-•O2 − signals [60], [62], [63]. The increase of the signals intensity with increase in radiation time signals generation of OH radicals and superoxide anion radicals. From the present generated radicals and the band alignment of the individual semiconductors, the Z-scheme mechanism can be assured. In a study by Xu et al. [64], a Z-scheme mechanism was confirmed to occur during sulfadiazine (SDZ) oxidation and Ni(II) reduction by a Cu2O/Bi/Bi2MoO6 composite ruling out the possibility of p-n heterojunction formation due to determination of •OH radicals signal in EPR that would not occur when holes concentrate on the VB of Cu2O in traditional p-n heterojunction.

Another common technique that is employed towards systematic analysis of charge transfer mechanism in Z-scheme is UPS spectra. In short, linear approximation of the UPS spectra gives the work function of the catalysts that can be used to determine their Fermi levels and valence and conduction bands. This can then be easily employed to determine the Z-scheme charge transfer at the interface and the details of this method is given by Aguirre et al. [65]. The surface potentials of semiconductors can be measured by spectroscopic techniques such as Kelvin probe force microscopy (KPFM), surface photovoltage spectroscopy (SPS) and transient absorption spectroscopy (TAS) to affirm Z-scheme charge transfer route [40], [66], [67], [68]. Characterization of Z-scheme photocatalytic systems promotes understanding and optimisation of charge transfer and separation efficiency.

Z-Scheme photocatalysis

On pollutant degradation, general removal of all classes of organic pollutants has been inspected. However, given the current global pandemic outbreak of the Covid-19, it is of paramount importance to examine the applications of both direct and indirect Z-scheme charge transfer composites towards remediation of different classes of pharmaceuticals and pathogens in water bodies.

Pharmaceuticals degradation

Pharmaceuticals are classified as organic pollutants of concern by environmentalists while pharmaceuticals production and consumption has steadily escalated due to increased number of patients in need of medical attention due to sicknesses and symptoms concomitant with Covid-19 pandemic, and emotional traumas triggered by effects of the Covid-19 on the economy and freedom. Inevitable use of antiviral and antibiotic medications and their prescription has ensued their secretion into drinking water systems which is very precarious in different perspectives: (1) the pollutants bio-accumulate and their concentrations increase in drinking waters and can become deadly under acute or chronic exposures, (2) they can form extremely lethal metabolites when interacting with the ecosystem, (3) and they may result in another pandemic of drug resistance as elongated ingestion of pharmaceutical polluted water may lead to emergence of drug resistant strains of pathogens. In this section, the degradation of pharmaceuticals with exhaustive analysis focused on antibiotics and non-steroidal anti-inflammatory drugs (NSAIDs) is explored.

Antibiotics

Generally, antibiotics are classified as persistent organic pollutants, they are non-biodegradable and can persist in the environment causing harmful effects to human health. In recent studies on photocatalysis, emerging pollutants such as pharmaceuticals are being used as model water pollutants with some reviews focusing on the degradation of these materials using specific semiconductors such as Bi2WO6 based photocatalysts grouping the antibiotic pharmaceuticals into three different classes of sulfonamide antibiotics, tetracycline antibiotics [69] and fluoroquinolones antibiotics [70] which is adopted in this review for systematic and in depth analysis of Z-scheme heterojunctions towards degradation of antibiotics. These antibiotics are widely employed for deterrence and cures of photogenic diseases that have immensely escalated during Covid-19 pandemic ensuing their predominant determinations in water bodies.

Fluoroquinolones (FQ) antibiotics are subclass of quinolones and they have characteristic fluorine atom in the central ring system. Fluoroquinolones are widely used against a range of bacterial infections such as typhoid fever, and sexually transmitted diseases. The examples of detected fluoroquinolones in water systems that have been studied for possible degradation with Z-scheme heterojunctions include norfloxacin, levofloxacin, ciprofloxacin, and ofloxacin. Ofloxacin is a widespread FQ antibiotic used for the treatment of skin, urinary and respiratory infections, gonorrhoea, pneumonia, and bronchitis. Kumar et al. degraded ofloxacin under visible light irradiation with dual Z-scheme g-C3N4/Bi4Ti3O12/Bi4O5I2 heterojunction which possessed I3 −/I− and Bi3+/Bi5+ and redox mediators synthesised through the hydrothermal route [71]. M-S plots were used to determine the conduction band positions of the semiconductors towards proposal of charge transfer mechanism at their interfaces, and to show that all the semiconductors were p-type semiconductors as their slopes on the M-S plot is positive (Fig 2 a). The photocatalytic experiments were conducted in a gas circulation and pumping system connected Pyrex reactor with autosampler. Anaerobic conditions were induced through pumping dissolved oxygen with vacuum before the experiment began. A 500 W Xe lamp (240 mWcm−2) with 420 nm cut-off filter placed irradiated a 100 ml solution of 10 mg L-1 with 45 mg of photocatalyst during degradation process. 87.1% ofloxacin (OFL) removal efficiency was obtained in 80 minutes which was indexed to electron separation and formation of Z-scheme mechanism as shown by EIS Nyquist plots (Fig 2b), transient photo-current response (Fig 2c), and photoluminescence spectra (Fig 2d). The degradation pathway of OFL was proposed to result in CO2, H2O, and NH3 upon complete mineralization. M-S plots, ESR, and radical scavenging experiments were used to confirm the formation of Z-scheme mechanism in this work with numerous effects of different parameters explored on the degradation of OFL. Under aerobic conditions, dissolved oxygen reacts with photogenerated electrons forming •O2 - with the rate determining step assigned as the electron transfer to O2.

Figure 2

(a) Mott-Schottky plots, (b) EIS Nyquist plots, (c) transient photo-current response, (d) photoluminescence spectra of synthesised dual Z-scheme g-C3N4/Bi4Ti3O12/Bi4O5I2 heterojunction [71], (e) levofloxacin degradation rate of the synthesised S-scheme heterojunction Bi2O3/P-C3N4 photocatalyst [72], and (f) the luminescence inhibition (INH) of the bioluminescence of V. fischeri at different reaction time [73]

The degradation of other fluoroquinolones has been reported in literature recently. An efficiency of 89.2% during levofloxacin degradation was achieved in 75 minutes under simulated solar radiation on Bi2O3/P-C3N4 direct Z-scheme nanocomposite (Fig 2e) with improved hydrophilicity and BET surface area [72]. Density functional theory, (DFT), ESR, and radical trapping experiments were used collaboratively to affirm the charge transfer mechanism at the interface of the semiconductors to be Z-scheme. Z-scheme degradation of levofloxacin has also been reported with an efficiency of 90.8% in 70 minutes of irradiation on Sm6WO12/g-C3N4 nanocomposite through •O2 − and •OH as the main ROSs [74]. Moreover, ciprofloxacin degradation was accomplished with an indirect Z-scheme Bi2MoO6/g-C3N4 with nitrogen doped carbon dots as the electron mediator [75]. The Z-scheme mechanism was proved with ESR and radical trapping experiments while the effect of solution pH, initial concentration of CIP, catalyst dosage, and inorganic anions and organic matters were studied. The pH 8 of the solution, 1 mgL−1 initial solution concentration, and 1.0 gL−1 of catalyst dosage gave the highest photocatalyst performance. Similarly, to dissolution of organic constituents, inorganic anions broadly exist in water and were reported to inhibit degradation performance of the catalyst, and the main given reason was completion of active sites and also scavenging of radicals by anions where new radicals with less oxidising power can be formed (Eq. 1 - 4).

The degradation of norfloxacin was also examined with direct dual [28] and indirect Z-scheme mechanisms [73] with BiFeO3/CuBi2O4/BaTiO3 and phosphate-doped BiVO4/graphene quantum dots/P-doped g-C3N4 (BVP/GQDs/PCN) under visible light irradiation with efficiencies of 93.2% and 86.3% within 60 and 120 minutes, respectively. Evidently, the degradation time incurred during photodegradation of this class of antibiotics reach a maximum of 2 hours towards high mineralization efficiencies that are enticing for industrial applications of Z-scheme heterojunctions towards large-scale applications of antibiotics degradation in water. The incomplete degradation was studied with bioluminescence inhibition assay and portrayed that highly toxic intermediates were formed (Fig 2f) during the use of the Z-scheme heterojunction.

Tetracycline antibiotics are cheap and widely used antibiotic pharmaceutical agents which is not surprising that they are widely studied for their photocatalytic degradation in environmental waters including with Z-scheme photocatalysts [76]. The Mott–Schottky (M–S) plot was employed for examination of band structure towards confirmation of Z-scheme heterostructure in conjunction with radical trapping experiments for both BiOCl and Bi2O3 samples also providing information on whether the semiconductors are p-type or n-type [77]. The negative slopes indicated that the semiconductors are n-type and it was further employed to determine band edge positions of the semiconductors to compare with radical experiments to substantiate the interfacial charge transfer to follow indirect Z-scheme with Bi as the electron mediator between BiOCl and Bi2O3. There are numerous photodegradation experiments that have been conducted with different photocatalytic Z-scheme heterojunctions towards degradation of single [10], [78], [79] and multiple [80] tetracycline antibiotics in water. Dual direct Z-scheme MoS2/g-C3N4/Bi24O31Cl10 ternary heterojunction was evaluated towards degradation of TC with 97.5% degradation efficiency in 50 minutes of visible-light irradiation by Kang and co-workers [62]. PL spectra, time-resolved PL spectra, photocurrent response and EIS spectra (Fig 3 a-d) synchronously proved the charge separation in the composite compared to pristine semiconductors with reduced PL intensity, increase in average lifetime of photogenerated charges, highest photocurrent intensity, and small charge transfer resistivity, respectively. EPR and radical quenching experiments confirmed the formation of Z-scheme charge transfer mechanism.

Figure 3

(a) Photocurrent response, (b) EIS spectra, (c) photoluminescence spectra excited at 360 nm and (d) time-resolved PL spectra of dual Z-scheme MoS2/g-C3N4/Bi24O31Cl10 ternary heterojunction photocatalysts [62], (e) Effect of different inorganic anions on SMZ degradation by direct Z-Scheme AgI/Bi4V2O11 photocatalyst ([anions concentration] =10mM; [Catalyst dosage] =0.1 g L−1; [SMZ] =10mg L−1); and (f) Comparison of the survey XPS spectra of the used and fresh catalyst after 60 minutes of visible light degradation [81].

Sulfonamide antibiotics are characterized by presence of amide of sulfonic acid and they are normally employed as antibacterial, anticonvulsant, anti-inflammatory and diuretic humans or livestock drugs. Their determinations in water matrix are gaining momentum yearly due to their demand and their detections in the environment are increasing due to their resistance to biodegradation and their bio-accumulative nature, imperilling aqualife, human health and the ecosystem. Z-scheme AgI/Bi4V2O11 heterojunction was synthesized by Wen and co-workers towards visible light degradation of sulfamethazine (SMZ) with an efficiency of 91.47% in 60 minutes of irradiation [18]. SMZ degradation in presence of Cl−, SO4 2−, NO3 −, and HCO3 − portrayed inhibitory effect on the efficiency of SMZ degradation from 93.17% to 80.44% and 54.22% during 60 minutes in the presence of Cl− and HCO3 −, respectively (Fig 3e). The minor efficiency decrease in the presence of chloride ions could be mainly due to competition of active sites between chloride ions and pollutant on the surface of the catalyst while HCO3 − works as a scavenger to trap and convert active ROSs into less oxidative or reductive radicals lowering the overall efficiency of the system.

The systematic study of the re-used products showed that the crystal structure and the elemental composition of the composites did not change as analyzed by XPS (Fig 3f) with further analysis of the Ag+ peak showing that no Ag0 was formed. Similar degradation of SMZ together with cloxacillin (CLX) were performed with dual direct double Z-scheme Ag3PO4/Bi2S3/Bi2O3 under visible light radiation with 98.06% and 90.26% efficiency in 90 minutes, respectively [82].

Xu et al fabricated a direct Z-scheme a 0D/1D AgI/MoO3 photocatalyst towards visible light driven degradation of sulfamethoxazole (SMX) with 97.6% efficiency in 20 minutes [83]. ESR and radical trapping experiments confirmed the formation of direct Z-scheme heterostructure which achieved mineralization of SMX as proved by 3D excitation-emission matrix (EEM) spectra and HPLC-MS spectra. From the EEM spectra (Fig 4 a-d) as irradiation progressed, the peak intensity began to fade away to signal successful degradation of SMX until the weakest intensity at 20 minutes showing almost complete degradation of SMX in agreement with 97.6% calculated efficiency with the optimum composite.

Figure 4

the 3D EEM spectra of the SMX solution in AgI/MoO3 visible system: (a) 0 min, (b) 5 min, (c) 15 min, (d) 20 min [83], (e) XRD patterns and (f) FT-IR spectra of as-synthesized samples (g) ESR spectra of composites, and (h) schematic illustration of a possible charge transfer mechanism for BiOCl/CuBi2O4 S-scheme heterojunction [91].

A dual direct Z-scheme Ag2S/Bi2S3/g-C3N4 heterojunction was investigated for visible light and solar light irradiation of SMX with degradation efficiency of 97.4% in 90 minutes initiated by hydroxyl and superoxide anion radicals with in-situ generation of Ag as electron shuttle as proved by XPS [84]. The degradation pathway was investigated to with LC–MS analysis after 30 minutes and 90 minutes of SMX degradation with the optimum performing catalyst under visible light showing all the intermediates that were used to predict the degradation reaction pathway, showing that no aliphatic compounds were detected. However, the TOC results proved high mineralisation efficiency of 67.4% and 85.1% at 30 °C under solar and visible light irradiation after 120 minutes.

Other classes of pharmaceuticals have also been investigated for removal from water environment with Z-scheme heterojunctions. Carbamazepine is one of the pharmaceutical and personal care products (PPCPs) that is classified as an anticonvulsants/antiepileptics [85] capable of causing detrimental effects to human health and the environment despite some of its important medicinal uses in treating trigeminal neuralgia, epilepsy or other diseases [86]. In our group, we constructed a dual Z-scheme Co3O4/CuBi2O4/SmVO4 photocatalyst towards visible light driven degradation of carbamazepine (CBZ) with attained efficiency of 76.1% within 300 minutes of irradiation [87]. The bandgap energies attained from UV-Vis and the CB edge potentials estimated by XPS were used to draw up the energy diagram with band positions for the involved semiconductors. The radical scavenging tests affirmed involvement of holes, •O2 −, and •OH towards degradation of CMZ which was then used to show that in a traditional type II interfacial charge transfer, it would be impossible to form all the ROSs leaving only the probable mechanism as the dual Z-scheme. The changes in binding energies of the system was also observed when comparing high-resolution XPS spectra of pristine semiconductors to that of the composite which depicted electron acceptor and donor semiconductors from decrease and increase in binding energy respectively. The change in chemical environment is the phenomena behind this and the direction of electron transfer was obtained after observed increase and decrease in the binding energy of elements.

NSAIDs

The most common non-steroidal anti-inflammatory drugs found in water media are ibuprofen (IBU) and diclofenac (DFC) with expected escalated uses due to their low cost and high functionality towards treatment of health-related issues. DFC is a NSAID engaged for inflammation treatment and most frequently administered for pain relief in humans and animals with expected usage escalations in the course of the Covid-19 pandemic [88], [89]. IBU is a non-steroidal anti-inflammatory drug (NSAID), entrusted with pain and fever relief, arthritis, and to treat inflammation of minor injuries since its first introduction in 1969, being recently selected as the safest NSAID by spontaneous consumption which has surpassed 200 tons per year [90].

The existence and detections of both DFC and IBU in drinking water systems have influenced their removal from water bodies to provide clean and safe drinking water to humans. Wu et al. constructed a BiOCl/CuBi2O4 S-scheme heterojunction for visible light driven removal of diclofenac through the solvothermal method with 90% efficiency in 60 minutes [91]. XRD results (Fig 4e) and FTIR spectra (Fig 4f) confirmed the formation of BiOCl/CuBi2O4 composites with strong interfacial contact that was proved by decrease in peak intensity of tetragonal CuBi2O4 and increase in peak intensity of tetragonal BiOCl in XRD spectra. Electron spin resonance (ESR) spectroscopy was used to investigate the formation of oxygen vacancies in the samples and the results confirmed formation of oxygen vacancies in BiOCl and BiOCl/CuBi2O4 with the intensity higher (meaning high concentration of oxygen vacancies) in the composite than pristine BiOCl (Fig 4g). Similarly, ESR was employed to further determine the charge transfer mechanism at the interface of the semiconductors heterostructure in combination with photoluminescence (PL), electrochemical impedance spectroscopy (EIS), radical scavenger tests, transient photocurrent response, and time-resolved fluorescence (TRPL) measurements. From the affirmed S-scheme heterojunction in this study (Fig 4h), no distinctive difference exists with other direct Z-scheme reported mechanisms.

Ibuprofen degradation with direct Z-scheme Co3O4/BiOI was performed by our group together with a mixture of trimethoprim under visible light with efficiency of 97.02% in 100 minutes while the degradation of only IBU was achieved in 60 minutes with 99.98% efficiency. Self-assembled Co3O4 spherical rings were engulfed by flower like BiOI towards strong interaction and efficient separation of charge carriers studied with PL analysis [43]. Another direct Z-scheme photocatalyst was inspected for IBU degradation using Bi5O7I-MoO3 photocatalyst with 89.2% degradation efficiency in 120 minutes under visible light irradiation [92]. Analysis of different water matrixes incurred low reductions in photocatalytic activity of the Z-scheme composites showing that the formed Z-scheme heterojunction activity is suppressed by complex compositional contents of the real wastewater samples under investigation (Fig 5 a). However, the catalyst was still stable to some extend with stability tests confirming that it can be reused five times with less than 5% decrease in efficiency (Fig 5b). Some traces of adsorbed IBU after degradation and some reduction in and shifting of IR peaks (Fig 5c) and XRD peaks (Fig 5d) bands confirmed that adsorbed pollutant was present on the catalyst surface after degradation, proving that the catalyst can be inactivated after prolonged usage.

Figure 5

(a) Influence of water matrices on the photocatalytic IBU degradation by Bi5O7I-MoO3 catalyst, (b) stability test experiments in IBU degradation on Bi5O7I-MoO3 composite (c) FTIR spectra of IBU, fresh and reused Bi5O7I-MoO3 catalyst, and (d) XRD patterns of the fresh and reused catalyst [92].

Wei et al. designed dual direct Z-scheme α-SnWO4/UiO-66(NH2)/g-C3N4 with visible light induced photodegradation of IBU attaining 95.5% degradation efficiency within 2 hours [93]. The W, O and Sn binding energies shifted to higher values in α-SnWO4/UiO-66(NH2)/g-C3N4 compared with SnWO4 (α-SW), suggesting a strong interaction that proved the formation of a heterojunction. The kind of formed heterojunction was determined first by calculation of the bandgaps of the materials using UV-Vis data and the results were 2.62, 2.51, and 1.78 eV for CN, UiO-66(NH2) and α-SW, respectively. This showed that all the individual semiconductors absorb light in the visible range of the solar spectrum. Then M-S plots were used to classify CN, UiO-66(NH2), and α-SW as p-, p-, and n-type semiconductors due to negative and positive slopes, the values of which were used to determine the flat band (EFB) potentials and then from the Nernst’s equation (ENHE = EAg/AgCl + 0.21). The values of -0.91 and -0.5 eV (vs. NHE) were calculated as conduction bands for CN and UiO-66(NH2) while the valence band of α-SW gave +2.66 eV (vs. NHE). The relationship of (EVB = ECB + Eg) was employed to determine the corresponding valence and conduction band edge potentials. Similar steps have been followed with other Z-scheme semiconductors to calculate respective band edge positions experimentally using M-S plots [94]. Moreover, scavenger tests were conducted with BQ, IPA and OA for •O2 -, •OH and h+, respectively to confirm generation of •O2 - and •OH ROSs in the heterostructure and their involvement thereof in the photodegradation of IBU. ESR results also corroborated the radical scavenger experiments signalling that no radicals formed in the dark while peak intensities of •O2 - and •OH radicals increased proportionally to increase in irradiation time inducing the proposed dual Z-scheme mechanism. Both the inorganic CN and UiO-66(NH2) semiconductors acted as reduction semiconductors while the inorganic α-SW acted as an oxidative semiconductor ensuring provision of distinctive reaction sites for reduction and oxidation reactions respectively. Despite the increased charge transfer and separation enhancement, the redox ability of the heterojunction was higher than it would be if the traditional p-n-p heterojunction formed between the composites. The electrons would all accumulate on the conduction band of α-SW (+0.88 eV) while the holes would accumulate on the valence band of CN (+1.71 eV). Evidently, if this was to happen, the formation of the detected ROSs in this wok would not be possible due to inappropriate energies of electrons and holes towards respective formation of •O2 - and •OH radicals. It can be inferred that based on numerous discussions in this review, Z-scheme catalyst are promising heterostructures towards general photocatalytic applications in respect to quantum efficiency, charge separation efficiency, and photoactivity.

Activation of persulfate

Application of sulfate radical (SO4 •-, E = 2.5–3.1 eV, half-life: 30–40 μs) based advanced oxidation processes (SR-AOPs) has been regarded as a hotspot of innovative research towards environmental pollution remediation [95]. The production of sulfate radicals is achieved through activation of peroxymonosulfate (HSO5 −, PMS) or peroxydisulfate (PDS, S2O8 2−) by either homogeneous or heterogeneous photocatalysts [96], [97]. Most recently, Z-scheme photocatalysts are used for heterogeneous activation of persulfate in photocatalytic removal of various water pollutants [95], [98], [99]. In heterogeneous activation of PMS/PDS, the other component if not both should be a metal or metals (Co, Mn, Fe, Cu, etc.) that can effectively break the O-O peroxo bond of PMS/PDS via redox mechanism to generate sulfate radicals. Since this bond is highly stable at room temperature, breaking it is only realized when high energy and chemical activators are used [96], [100], [101]. Some researchers employed heat, ultraviolet irradiation, and ultrasound system as source of energy, whereas chemical activators include sulfides, transitional metal ions and their metal oxides together with their complexes. However, the choice of chemical activators should be investigated properly as they introduce secondary pollution from toxic metal ions. The common utilised precursors for sulfate radicals generation possess different properties and that lead to difference in activation and catalytic power. The PDS has higher potential (E° (S2O8 2−/SO4 •−) = 2.01 VNHE) than PMS (E° (HSO5 -/SO4 •−) = 1.75 VNHE) and its cost is estimated at 0.74 USD/kg (0.18 USD/mol) while 2.2 USD/kg (1.36 USD/mol) is the cost for PMS [102], [103]. Furthermore, PDS has a longer O-O peroxo bond distance (1.497 Å) as opposed to PMS (1.460 Å) and this makes it easier to be activated. For the purpose of this review activation of both precursors by Z-scheme photocatalysts will be discussed for degradation of pharmaceutical water pollutants. It should be noted that the higher degradation efficiencies of PMS/PDS based photocatalytic systems emanates from the production of reactive radical (SO4 •-, •OH, and •O2 -∙) and non-radical (1O2) species.

Ji et al. [100] prepared a perylene diimide (PMI)/MiL-101(Cr) (PM) Z-scheme heterojunction using one-pot method and used it for activation of PS under visible light for degradation of iohexol (IOH) drug. The authors achieved degradation efficiency of ∼100% in an illumination time and PS concentration of 35 min and 3.0 mM, respectively. As shown in Fig 6 a, the IOH degradation mechanism portrayed the production of reactive species for initiation of both radical and non-radical based reactions. Electron spin resonance (ESR) tests (Fig 6b-c) indicated that participation of Z-scheme was vital as the signals for SO4 •--, •OH, •O2 -, and singlet oxygen (1O2) was produced in the presence of visible light. The presence of oxygen containing radicals and holes was also confirmed by scavenger experiments which showed that the holes and •OH were dominant radicals that contributed to effective IOH degradation. The results proved the concept of high charge separation and utilization of visible light irradiation for effective activation of PS. Likewise, a 2D/2D CoAl-LDH/BiOBr Z-scheme photocatalyst was prepared via hydrothermal method and its application revealed 96% removal of ciprofloxacin (CIP) on 8 wt% LDH/BiOBr/PMS/Vis system within 30 min at optimum catalyst and PMS doses of 40 and 100 mg, respectively. The CIP degradation was initiated by reactive radicals as shown in Fig 6e. The authors allied the high CIP elimination to several factors such as strong intimacy interaction between the materials, high specific surface area (91.38 m2 g-1) which was higher than that of CoAl-LHD (50.80 m2 g-1), and BiOBr (30.60 m2 g-1), and enhanced charge carrier separation and migration as demonstrated by EIS, transient photocurrent response and PL results (Fig 6f-i). The 8 wt% LDH/BiOBr/PMS/Vis system was further extended to TC (98%), ENR (88%), NOR (88%), and RhB (100%), respectively. It was discussed that the high degradation efficiencies obtained confirmed the synergistic effect between photocatalysis and PMS activation in the LDH/BiOBr/PMS/Vis system. The scavenger and EPR experiments indicated the prevalence of radical (•O2 -) and non-radical (1O2) degradation of CIP [99]. Similar studies on Z-scheme mechanism activation of PMS/PDS have been explored by other researchers [95], [98], [104], [105], [106], [107], [108] for pharmaceuticals and other organic pollutant elimination.

Figure 6

The degradation mechanism of iohexol drug on PM/PS/Vis system. The EPR signals of (b) DMPO-•OH and DMPO-SO4•−, (c) DMPO-•O2− and (d) TEMP-1O2 in the PM-7/PS, PM-7/Vis and PM-7/PS/Vis systems [100], (e) mechanism of charge generation for CIP degradation in the 8 wt% LDH/BiOBr/PMS/vis system, (f) Nitrogen adsorption-desorption isotherms, (g) EIS plots, (h) Transient photocurrent responses, and (i) PL spectra of BiOBr and its composites [99].

Recently, studies have shown that direct Z-scheme heterojunctions are preferred over all solid state/mediated Z-scheme heterojunctions for activation of PMS/PDS [98], [100], [105]. The mediated Z-scheme has been plagued with notable drawbacks such as the cost of noble metals normally used, absorption of photons and electrons by mediators which tend to limit the performance of the process while direct Z-scheme is intimately designed to operate as a sieve for the used photo-electron-hole pairs possessing low redox potential, thus minimising the light shielding effect and uplifting the redox capabilities [103]. The intimate contact in a direct Z-scheme heterojunction also induces formation of electric field which is vital for redox ability needed for redistribution of carrier charges. Afterwards, the Z-scheme photocatalytic system results in more positive valence band and negative conduction band which are vital for both PMS/PDS activation and absorption of more solar light throughout the entire solar spectrum.

Pathogens inactivation

Numerous micro-organisms or disease-causing pathogens have been determined in surface water making their removal very important. This work focusses on bacteria and virus inactivation only. Viruses have been evaluated directly for their photocatalytic removal with foodborne, airborne and waterborne viruses of supreme prominence [109]. The photocatalytic virus inactivation was first reported on TiO2 photocatalyst by Sierka and Sjogren in 1994 leading to widespread application of photocatalysis for viral decomposition [110]. The mechanism (like in other pathogen degradation) involves generation of ROS like •OH and H2O2 that directly attacks and destroys the virus capsid or cell wall and the cytoplasm releasing contents inside the virus like genetic makeup and proteins [111]. The mechanisms of inactivation can be subdivided into metal ion toxicity(Fig 7 a) , physical damage (Fig 7b) and chemical oxidation (Fig 7c) as suggested by Zhang et al. [1]. Briefly, chemical oxidation involves the use of semiconductors under illumination by light of appropriate wavelength for initial generation of electrons and holes which are transformed through redox reactions towards formation of ROSs (•OH, h+, 1O2, H2O2, and •O2 -). It is noteworthy that interaction of catalyst and virus surface, like in photocatalytic pollutant degradation, will enhance efficiency of the process.

Figure 7

Three main mechanisms of viral/bacterial inactivation induced by heterogeneous photocatalysts (a) metal ion toxicity, (b) physical damage (c) chemical oxidation [111], (d) photocatalytic antibacterial activity of Z-scheme composite against E. coli at different irradiation times under LED light [112], (e) general schematic presentation of the step by step processes that occur throughout the inactivation process of viruses or bacteria with Z-scheme heterojunctions.

The most commonly studied inactivation of micro-organisms involves bacterial investigations. The antimicrobial evaluation of photocatalytic mechanism involves studies on different bacterial species. Bacterial contamination is a serious concern with numerous bacterial species determined to cause diseases that may result in death from severe diarrhoea making their inactivation a priority in surface and drinking water systems. Liu et al. studied Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) inactivation with direct Z-scheme BiO2-x/BiOBr heterojunction synthesised in-situ with oxygen vacancies with bactericidal efficiency of 100% in 20 minutes [112]. The increase in irradiation time supervenes decrease in the number of viable bacteria on the culture dish (Fig 7d). In a recent review by Wang and co-workers [1], different g-C3N4 modifications like doping and heterojunction formation were summarised towards photocatalytic inactivation of different classes of bacteria with specific conditions and efficiency highlighted. There are numerous traditional, all solid state and direct Z-scheme heterojunctions associated studies that principally emphasise only bacterial inactivation or bacterial inactivation plus an additional application under visible light irradiation [113], [114], [115]. Xia et al. hydrothermally synthesised a Z-scheme g-C3N4/m-Bi2O4 towards visible light driven inactivation of 6 log10 cfu/mL of E. coli K-12 in 1.5 hours [115]. Fluorescence microscopy images after staining with fluorescent dye portrayed an increase of green cells and a decrease in red cells attributed to membrane rapture. SEM images affirmed destruction and distortion of the bacterial structure due to leaking out of important contents resulting in instantaneously inactivation of bacteria by metabolic arrest. Fig 7e shows a schematic presentation of the step-by-step process that occurs during the inactivation of viruses or bacteria with Z-scheme heterojunctions. Firstly, there should be an interaction between the catalyst and the pathogen, under visible light irradiation, the catalyst attached on the cell wall of bacteria or virus generate ROSs that rapture the cell membrane. The attack leads to destruction of the membrane and the continued release of the ROSs would ensure that contents inside the cell membrane are also destroyed.

The popularity of bacterial inactivation with different heterojunctions resulted in fabrication of handy prototypes towards real environmental applications of bacterial inactivation, giving hope for large scale applications of these systems. In medical applications, thermodynamic therapy involves the production of radicals to block electron transport to the tumor cells for energy and nutrients deficiency induced cell death [116]. In wound healing, the ROSs induced by light illumination inhibit bacterial growth resulting in healing of wounds. For example, an indirect Z-scheme ZnO/C-dots/g-C3N4 heterojunction fabricated through ultrasonication method was evaluated for wound healing by Xiang et al with antibacterial efficacy of 99.97% and 99.99% for S. aureus and E. coli after 15 min of visible light irradiation, indexed to synergistic contributions of ROSs generation and hyperthermia [117]. In their study, both gram-negative and gram-positive bacteria was selected for visible light driven inactivation on a spread plate method. The results of E. coli in 15 minutes showed no reduction of bacterial count with the control, little reduction with CN and complete removal with a Z-scheme ZnO/C-dots/g-C3N4 ternary heterojunction (Fig 8 a). The SEM analysis of the bacteria structure shows more distortion with holes when the Z-scheme ZnO/C-dots/g-C3N4 ternary heterojunction was employed compared to no changes and little change for control and CN respectively (Fig 8b). In-vivo wound healing was organized with rats wounds (20 mm diameter) infested with 20 µL of S. aureus (1 × 108CFU/mL) to assess the wound therapeutic ability of the catalyst. The images of the wounds confirmed faster healing in 12 days when the Z-scheme catalyst is used compared to the control experiment after initial S. aureus infection after 2 days in both cases (Fig 8c). Further inspection of the hematoxylin and eosin (H&E) was performed to stain wound tissues. After 2 days, the composite had less inflammatory cells (GRAN, neutrophil granulocyte, red arrows) than the control group, yet after 12 days, necrosis of epidermal cells and epidermal crevices were witnessed in the control group while restoration curative outcome, and unharmed hypodermal tissue arrangement was observed with Z-scheme composite group(Fig 8d).Fig 10.

Figure 8

(a) Spread plate results of E. coli grown on different samples, (b) Surface morphologies of E. coli treated with different groups with or without 15 min visible light irradiation, scale bars: 1 µm, in vivo assessment of the ZCCN with antibacterial effects and wound healing capability: (c) Photographs of the S. aureus-infected wounds treated with different dressings at time points of 0, 2, 6 and 12 days. Scale bars, 2 mm. and (d) H&E stained images showing the degree of infection of the skin tissue after 2 and 12 days. Scale bars are 50 µm [117].

Figure 10

(a) Example of S-scheme mechanism charge transfer heterostructure [91] and (b) Schematic diagram of a PCO fluidized bed reactor [133], and (c) experimental setup of a Z-scheme twin reactor for overall water splitting reaction.

The direct degradation of viruses has been studied and it is an emergent area of research with numerous articles focusing on the topic of their possible removal with advanced oxidation processes in water, air and food [111], [118], [119], [120]. Z-scheme heterojunction inactivation of viruses is no exception, but only a handful of research outputs exist to this regard similarly to Z-scheme remediation of antiviral drugs in water systems. Zhang et al. used a facile solvothermal-hydrothermal method to fabricate a direct Z-scheme heterostructure of oxygen-doped graphitic carbon nitride microspheres (O-g-C3N4) and carbonation carbon (HTCC) for visible light inactivation of human adenovirus type 2 (HAdV-2) with virucidal efficiency of -5log[C/C0] in 120 minutes [121]. The increase of temperature to 37°C further ensued complete removal efficiency at pH 5. The viral particle damage was studied with TEM where after 120 minutes of irradiation, HAdV-2 structure was ruptured and only virus debris was observed indicating the severity and lethality of the all-organic O-g-C3N4/HTCC-2 Z-scheme heterostructure under visible light irradiation. The fabricated nanoparticles cytotoxicity was evaluated through the XTT assay with human A549 cells portraying that within photocatalytic inactivation dosages, the nanoparticles toxicity is negligible with 85% remaining even at higher concentrations of 150 μg/mL. This validates the biocompatibility and the negligible cytotoxicity indexed to stability and chemical composition of the composites. Cheng and co-workers investigated Ag3PO4/g-C3N4 heterojunction with Z-scheme charge transfer at the interface synthesised by hydrothermal method towards complete inactivation of bacteriophage f2 with concentration of 3 x 106 PFU/mL under visible light irradiation in 80 minutes [122]. Scavenger tests were performed to validate charge transfer mechanism of Z-scheme and humic acid retarded the virucidal activity of the composites. As with bacterial inactivation, increase with irradiation time decreased the cultured viral count with none observed after 80 minutes of irradiation. Numerous investigations into Z-scheme heterojunctions have propelled knowledge broadening proved by recent increase of publications in this regard with numerous novel discoveries like in-situ generation of electron shuttles in direct Z-scheme composites.

Advancements in Z-scheme Photocatalysts

Evolution of S-scheme

After numerous investigations of the direct Z-scheme, another mechanism referred to as the S-scheme was introduced in 2019 by Fu et al [123]. There is no difference in the direct Z-scheme and the reported S-scheme mechanism (Fig 9a) based on materials and thermodynamic perspectives, and the pre-requisite conditions are the same with the same properties at the interfaces that results in their formation as evident from already reported S-scheme heterojunctions [36], [72], [91] in comparison with some direct Z-scheme heterostructures [77], [124], [125]. For example, as correctly pointed out by Liao and co-workers in their review based on Z-scheme catalytic systems, both S-scheme and direct Z-scheme mechanisms require a unique principle of high work function of oxidative semiconductor (PS II) compared to that of the reductive semiconductor (PS I) [126]. The build-in electric field is reported as the only means of interface charge recombination driving force in Z-scheme while coulomb attraction and bend bending also act as driving forces for interface recombination of charges in S-scheme system. Equivocally, some reports on Z-scheme mechanisms exist with Xu et al. reporting band bending induced by In-O-Cd bond as a driving force for interface charge recombination enhancement in ZnIn2S4/CdS material fabricated via cation exchange reaction towards efficient photoelectrochemical water splitting [127]. Therefore, the coulomb force, band bending and internal-build in electric field contributions for recombination of electrons and holes at the interface of semiconductors can be considered a good discovery for enhancement of knowledge and understanding in direct Z-scheme heterostructures. A strong recommendation is made that the word S-scheme can be regarded as another name for direct Z-scheme heterojunction by scientific community. It is reported that the S-scheme comprises mainly of two n-type photocatalysts in which one is oxidation photocatalyst and another reduction photocatalyst. The migration and movement of electrons follows a “step” (macroscopic viewpoint) or “N” (microscopic viewpoint) type for effective separation of carrier charges [30]. However, this kind of heterojunction is at its infancy stage and need to be explored more in the future and all-encompassing assessment with analogous scientific procedures maybe directed to determine a reasonable alteration with direct Z-scheme systems to evade confusion and misconception of forthcoming works.

Laboratory Up-scaling

The possible applicability of Z-scheme photocatalytic heterojunctions for real wastewater applications needs laboratory validations mimicking real wastewater compositions for proper analysis of actual expected behavioural activities and the minimum required time towards achieving meaningful efficiencies. In this regard, Wang et al. used a hydrothermal method to synthesise a 0D/2D/2D ZnFe2O4/Bi2O2CO3/BiOBr double Z-scheme heterojunctions for visible light degradation of three tetracycline antibiotics degradation in tetracycline (TC), oxytetracycline (OTC), and doxycycline (DOX) with 93%, 90.1%, and 89.4% degradation efficiencies under sunlight and PMS system for 20 minutes [80]. Direct double Z-scheme mechanism was established with quenching experiments, DFT calculations, EPR measurements, and XPS valence band spectra with two oxidative semiconductors (Bi2O2CO3 and BiOBr) and one reductive semiconductor (ZnFe2O4) forming the arrow-up heterojunction. From this study, despite the study of the effects of different anions, the analysis of different secondary effluents from a sewage treatment plant prepared with 20 ppm solution of TC OTC, and DOX gave 85.6%, 82.8%, and 82.5% degradation efficiencies in 20 minutes respectively. The small decrease of less than 10% gave extraordinary confidence towards real wastewater applications with oxidants (PMS) enhanced dual Z-scheme photocatalytic systems in real wastewater matrixes for remediation of pharmaceuticals in polluted water.

Different reactor designs have been investigated to address the challenge of expensive filters required to dodge nanomaterials discharges into the environment. This dispersion which results in high quantum efficiency due to good pollutant and catalyst interaction is common despite causing secondary pollution [128]. The widely employed fluidized bed reactor (Fig 9b) contains a system for easy recovery of catalysts after application. The fluidized-bed photoreactors are designed such that the light source can either be tubular UV lams or LED lights placed at the centre, externally or wrapping the interior of the reactor walls with a barrier between them and the solution. The reviews on Z-scheme photocatalysis for degradation of pollutants do not have a discussion on possible application of Z-scheme photocatalytic reactors at laboratory scale [129], [130]. Therefore, it is important to analyse this aspect with a view to encourage more studies to be tailored towards pilot scale testing of Z-scheme heterojunctions towards water pollution remediation for drinking water security and safety. Most reported Z-scheme photoreactors have been reported towards water splitting for HER and OER. Chandran et al. [131] assessed a micro-scale photocatalyst suspension in aqueous solution with a traditional Z-scheme charge transfer mechanism on a tandem particle-suspension reactor design for solar water splitting. Expedition of redox species conveyance between hydrogen and oxygen evolution reaction compartments to forfend gas crossover is achieved through a porous separator while its superior performance is attributed to proton-coupled electron mediators (para-benzoquinone/ hydroquinone and iodide/iodate) with zero visible light absorption capacity. A 1 cm high reaction stall containing 2.2 mg of BiVO4 and SrTiO3: Rh particles was predicted to sustain reactor operational efficiency at 1% with fast diffusive species transportation. Su et al. designed a mediator free Z-scheme twin reactor for isolated HER and OER during photocatalytic water splitting reaction [132]. A single Nafion membrane divided reactor was segmented into two distinctive chambers of CuFeO2 and Bi20TiO32 as HER and OER powder catalysts, respectively (Fig 9c). Reduced graphene oxides were used as electron mediators with branched copper wires employed to further enhance the electron collection in a Z-scheme interfacial charge transfer mechanism for water splitting. HER and OER efficiency was 2.23 and 1.14 in this reactor showing a stoichiometric ration of 2:1 with advantages such as hydrogen purification cost avoidance, and avoidance of potential explosion. It was suggested that the system can potentially be employed for wider range of applications like water pollution remediation. However, alarming scarcity of industrial systems in water purification sums up the lack of interest and trust from investors and thus limiting research aimed at large scale application of Z-scheme heterostructures and advanced oxidation processes in general.

Conclusion and Perspectives

With the current Covid-19 pandemic inducing muddle amongst scientists and the public, the time to exercise the supremacy of scientific technology in unravelling tangible and current problems on a global scale is now. Z-scheme photocatalysts posture high charge separation efficiency with separate PS I and PS II ensuring strong redox ability making them more efficient than other heterojunctions. The addition of oxidants like PMS/PDS have been established to heighten catalytic performance while reactor design is envisaged as an essential aspect towards industrialization of Z-scheme photocatalytic composites for water treatment. Despite tremendous work and discoveries associated with Z-scheme photocatalytic applications in pharmaceuticals degradation and pathogens inactivation, lack of practical competency appeals for imminent considerations.

Z-scheme photocatalytic degradation of pharmaceuticals or inactivation of pathogens should be explored largely with profuse sunlight energy with appropriate reactors designed and optimised for this purpose using experimental findings and modelling techniques such as DFT for efficient performance. This requires careful selection of oxidative and reductive semiconductors with strong visible light absorption index, optimised charge transfer at the interface of the semiconductors with in-depth study of more than one technique to complementarily prove Z-scheme mechanism existence, ease separation of the photocatalytic materials from the treated water with options like use of supports, magnets, etc., evaluation and assurance of the photocatalytic composites safety to human health and proper evaluation of their permissible limits in the water environment, and optimization of fabricated prototypes for economical viablity, readiness of investors into this new technology, and its possible integration into existing wastewater treatment technologies for low installation, operation and maintenance costs. The recent reports on S-scheme heterojunctions needs to be explored in detail to reach a scientific consensus with evidence of their difference with direct Z-scheme heterojunctions.

Considering the multitudes of investigations performed on elimination of bacteria with Z-scheme heterostructured composites, it is imperative that more work needs to be done to completely understand the mechanism of their destruction with systematic evaluations of the fate of the contents inside the membranes after destruction of the cell wall. Furthermore, the lack of evaluations of Z-scheme heterostructures for viral inactivation is alarming, and considering the current Covid-19, more work needs to be pioneered towards laboratory scale investigations of SARS-CoV-1, MERS-CoV and Covid-19 virucidal activities of Z-scheme charge transfer composites. Lastly, the applications of external oxidants like PMS/PDS, hydrogen peroxide, and ozone should be explored with Z-scheme photocatalytic composites as viral disinfection materials in water systems with explorations of near real environmental compositions as they are appeasing photocatalytic performance enhancers.

The issues above can be considered both challenges and opportunities to environmental, materials and chemical engineers, investors, and researchers in water disinfection and photocatalysis. The onus of this work is to stimulate knowledge in Z-scheme photocatalytic materials, draw more attention to researchers and investors in this field, and encourage its large-scale applicability to solve current problems for socio-economic sustainability and development.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.