Photocatalytic degradation of organic pollutants using green oil palm frond-derived carbon quantum dots/titanium dioxide as multifunctional photocatalysts under visible light radiation

Zeng Wei Heng ,Woon Chan Chong,2,*,Yean Ling Pang,2 ,Lan Ching Sim ,Chai Hoon Koo

1 Lee Kong Chian Faculty of Engineering and Science,Universiti Tunku Abdul Rahman,Sungai Long Campus,Jalan Sungai Long,Cheras,43000 Kajang,Selangor,Malaysia

2 Centre for Photonics and Advanced Materials Research,Universiti Tunku Abdul Rahman,Sungai Long Campus,Jalan Sungai Long,Cheras,43000 Kajang,Selangor,Malaysia

Keywords:Oil palm frond biomass N-doped carbon quantum dots Titanium-dioxide In-situ hydrothermal Visible light photocatalysis Methylene blue

ABSTRACT The present work suggested the use of waste oil palm frond as an alternative precursor for nitrogendoped carbon quantum dots (NCQDs) and proposed a straightforward in-situ hydrothermal method for the preparation of NCQDs/TiO2 nanocomposites.The elemental composition,morphological,structural and optical characteristics of NCQDs/TiO2 nanocomposites have been comprehensively investigated.The successful grafting of NCQDs on TiO2 matrix was confirmed by the formation of Ti-O-C bond and the electronic coupling between the π-states of NCQDs and the conduction band of TiO2.For the first time,the oil palm frond-derived NCQDs/TiO2 was adopted in the photodegradation of methylene blue(MB)under visible-light irradiation.As a result,the photocatalytic efficiency of NCQDs/TiO2 nanocomposites (86.16%) was 2.85 times higher than its counterpart TiO2 (30.18%).The enhanced performance of nanocomposites was attributed to the pivotal roles of NCQDs serving as electron mediator and visiblelight harvester.Besides,the optimal NCQDs loading was determined at 4 ml while the removal efficiency of NCQDs/TiO2-4 was the highest at a catalyst dosage of 1 g·L-1 under alkaline condition.This research work is important as it proposed a new insight to the preparation of biomass-based NCQDs/TiO2 using a facile synthetic method,which offers a green and sustainable water remediation technology.

1.Introduction

Since industrial revolution,clean and accessible water supply could not keep pace with the massive development of industrial economy,particularly in the textile industry.Each year,280,000 tonnes of organic dye is being introduced into the environment,which accounts for 17%-20% of industrial water pollution [1].Azo dyes that composed of complex aromatic amine structure are toxic and mutagenic in nature.It is therefore of utmost importance to develop advanced wastewater treatment and remediation technologies to treat the carcinogenic effluent and mitigate widespread pollution in aquatic ecosystem.However,the treatment of azo dyes in conventional wastewater treatment systems such as adsorption,coagulation-flocculation,membrane separation and biological treatment is difficult since they are structurally persistent and thereby condemned as one of the detrimental pollutants in wastewater [2].In this regard,advanced oxidation processes(AOPs),particularly photocatalysis is proposed as a powerful and environmental friendly approach in water remediation.Photocatalysis adopted high oxidative reactive oxygen species (ROS) to destruct complex structures of organic contaminants without generating secondary waste [2,3].

Titanium dioxide (TiO2) nanoparticle is deemed as one of the promising candidates in wastewater remediation in view of its versatile properties such as unique electronic structure,long-term chemical inertness,strong oxidizing capability,nontoxicity and low cost[4,5].Despite all the excellent physiochemical properties,TiO2-based photocatalysts encountered a bottleneck in visible light absorption.The major setback is associated with its intrinsic wide band gap especially anatase TiO2(～3.2 eV),signifying that TiO2could only be excited under high energy ultraviolet light(＜387 nm) [6].However,UV light which only accounted for 5% of the sunlight irradiated on the earth surface has greatly limited the application of TiO2in photocatalytic degradation[7].Moreover,the performance of TiO2is restricted by the rapid recombination of photogenerated electron-hole pairs.Therefore,intensive attention has been devoted to modify the band structure of TiO2by assembling it on a variety of carbon materials.Carbon materials such as activated carbon,fullerene,graphene and carbon nanotubes were incorporated to promote visible-light utilization and charge separation [8].Nevertheless,the practical applicability of these abovementioned carbon materials is still limited due to the use of expensive and non-renewable precursors such as carbon nanotubes(CNTs),graphite and its derivatives.Moreover,the reaction also required tedious pre-treatment steps and harsh condition such as high temperature and pressure [9].

In the past decades,fluorescent-based QDs have grasp substantial attention in this domain owing to their tuneable optoelectronic features [8].The newly emerging QDs can be classified into polymer dots (PDs),graphene quantum dots (GQDs) and carbon quantum dots (CQDs) [10].Among all,CQDs has appeared as an ideal substitute to traditional quantum dots due to the use of nontoxic carbon sources and facile synthetic method,which compensate for the environmental deficiency of traditional quantum dots.CQDs not only inherit the superior up-converted photoluminescence and photostability of traditional quantum dots,but it also possesses good chemical stability and water solubility [9,11].The outstanding electronic properties and high biocompatibility of CQDs have been widely adopted in various application fields such as bioimaging,chemical sensing,optoelectronics,and photocatalysis [12].In recent years,the incorporation of CQDs with various semiconductors have been broadly studied including carbon nitride [13,14],bismuth vanadate [15],zinc oxide [16],zinc stannate [17],cadmium sulphide [18] and titanium dioxide [19].Although numerous works on the modification of TiO2by CQDs have been conducted,most reported chemical-derived CQDs are still consuming chemicals that are harmful to the environment despite their excellent photocatalytic performance [9].In view of this,green synthesis of CQDs using eco-friendly natural resources such as rice husk [20],lemon peel [21],coal tar pitch [6],peach juice [22] and sugarcane juice [23] have drawn much attention.Despite the fascinating properties of CQDs,some researchers disclosed that bare CQDs without modification showed poor photosensitization and spectra efficiency [23].For these issues,doping of heretoatom such as nitrogen,sulphur and phosphorus on CQDs has been implemented to modify the electronic structure of CQDs,hence enhancing the photoluminescence properties of CQDs[24,25].

Herein,we propose the use of oil palm frond as a renewable carbon source for NCQDs fabrication.To the best of our knowledge,the application of oil palm frond-derived NCQDs/TiO2in the field of photocatalytic degradation has not been reported yet.In this work,the formation of TiO2and deposition of NCQDs on TiO2were conducted simultaneouslyviaa facile and effectivein-situhydrothermal method.The NCQDs content in the modified nanocomposites was varied to determine the optimal NCQDs concentration with the best photocatalytic performance.Additionally,the effect of initial solution pH and catalyst dosage on the removal efficiency of photocatalysts were studied in-depth to determine the optimal photocatalytic condition.Lastly,a possible mechanism of NCQDs/TiO2for the enhancement of visible-light utilization was proposed.

2.Materials and Methods

2.1.Materials

In this study,isopropanol (IPA,C3H8O,99%),ethylenediamine(EDA,C2H8N2,99%) and ethanol (C2H5OH,95%) were purchased from Merck while titanium (IV) isopropoxide (TTIP,Ti(OCH(CH3)2)4,97%) was obtained from Sigma-Aldrich.All chemicals were used as received without further purification.Dated oil palm frond collected from an oil palm estate in Tanjung Tualang,Perak was utilized as the green precursor for NCQDs.

2.2.Preparation of NCQDs

Typically,NCQDs were synthesized through the hydrothermal technique.Briefly,2.5 g of oil palm frond powder was dispersed in a mixture of IPA (70 ml) and distilled water (25 ml) together with 1 ml of EDA as nitrogen doping agent.Then,the solution was transferred into a Teflon-lined stainless steel autoclave,sealed and heated at 180 °C for 12 h.After cooling down naturally,a brown dispersion was observed.The residual weight precipitate was removed by centrifugation at 10000 r·min-1for 10 min followed by vacuum filtration to remove smaller insoluble matter.The supernatant containing NCQDs was kept in the dark before use.

2.3.Preparation of NCQDs/TiO2 nanocomposites

Nanocomposites NCQDs/TiO2were synthesized throughin-situhydrothermal method.Initially,5 ml of TTIP precursor was added into 50 ml of isopropanol solution.This was followed by dropwise addition of distilled water to form a white suspension.Then,different amount of synthesized NCQDs solution was added and stirred for 2 h at room temperature.The resulting solution was transferred into a Teflon-lined stainless steel autoclave and heated at 180 °C for 12 h.The autoclave was cooled down naturally and the mixture was centrifuged at 8000 r·min-1for 10 min.The particles obtained were rinsed and washed with deionized water and ethanol for several times,followed by drying at 80°C for 12 h.Hereafter,the products were designated as NCQDs/TiO2-xwherexindicated the volume of NCQDs added which were 1 ml,4 ml,6 ml and 10 ml.The overall process for the synthesis of NCQDs/TiO2nanocomposite is illustrated in Fig.1.

2.4.Characterization of photocatalysts

The morphological structure and surface elemental composition of the prepared photocatalysts were characterized using a field emission scanning electron microscopy equipped with energy dispersive X-ray (FESEM-EDX,JEOL,JSM-6701F),transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM,Tecnai,G2F20).X-ray diffraction (XRD,Shidmazu,XRD-6000) was conducted using Cu-Kα radiation(λ=0.15406 nm) with diffraction angles (2θ) ranging from 5° to 80° at a scanning speed of 2 (°)·min-1.Besides,Raman spectrometer (Horiba Labram HR Evolution) was used to study the crystal phase of the photocatalysts with a 514 nm laser beam.Fourier transform infrared (FTIR,Nicolet,IS10) spectra were obtained to determine the presence of diverse functional groups on the prepared samples.UV-vis diffuse reflectance spectra (UV-vis DRS)of photocatalysts over the range of 220-1400 nm wavelength were obtained using UV-vis DRS spectrophotometer (Shimadzu,UV-2600) with an integrating sphere accessory and BaSO4as the reflectance standard.Photoluminescence (PL) spectra were recorded by fluorescence spectrophotometer (Perkin Elmer,LS55)with an excitation wavelength of 320 nm.Zeta potential of the photocatalyst was determined using dynamic light scattering equipment (DLS,Horiba,SZ-100).Some of the characterization studies were conducted using the photocatalyst with optimum performance.

Fig.1.Illustration of the synthesis process of NCQDs and NCQDs/TiO2 nanocomposite.

2.5.Adsorption and photocatalytic activity study

The photocatalytic performance of the prepared nanocomposites and bare TiO2were evaluated by the degradation of MB under visible light irradiation.The photocatalytic experiment was conducted by dispersing 100 mg of the prepared photocatalysts into 100 ml of 10 μg·L-1MB solution.Prior to light irradiation,the suspension was placed in the dark under stirring for 1 h to establish the adsorption-desorption equilibrium between the photocatalyst and MB solution.After equilibration,the concentration of the MB was taken as the initial concentration (C0) to discount the adsorption contribution in the dark phase.Then,the MB suspension was illuminated by an 80 W LED lamp as a visible light source.The distance of the lamp from the reactor was fixed at a distance of 10 cm.The MB solution was withdrawn using a syringe filter at an interval of 30 min.Then the collected samples were sent for analysis using UV-vis spectrophotometer (PG Instruments T60) at 663 nm to determine the MB’s concentration.

Control experiments were also conducted using commercial P25 and blank solution for comparison purposes.Besides,different initial solution pH and catalyst dosage were studied to determine the optimal reaction condition.To study the kinetics of MB degradation over the prepared photocatalysts,the corresponding kinetic curves of prepared samples were plotted following a pseudo-firstorder model expressed in Eq.(1):

whereC0represents the initial MB concentration,Ctis the MB concentration at different irradiation times andkis the apparent rate constant.

3.Results and Discussion

3.1.Characterization of the photocatalysts

3.1.1.X-ray diffraction (XRD)

XRD spectra in Fig.2 demonstrate the phase structures of the bare TiO2and NCQDs/TiO2nanocomposites.The XRD pattern of the bare TiO2shows distinctive peaks at 25.3°,37.9°,48.3°,54.1°,55.1°,62.8°,69.2°,70.3° and 75.4°,which were perfectly indexed to the reflection of (101),(004),(200),(105),(211),(204),(116),(220)and(215)planes of anatase TiO2,respectively[26].The single anatase structure of TiO2is beneficial to the photocatalytic reaction.This is owing to the modestly larger band gap of anatase(3.2 eV) than rutile phase (3.03 eV),hence exhibits better charge separation capacity[4].All NCQDs/TiO2nanocomposites exhibited similar characteristic peaks of bare TiO2with no significant shift.Besides,no other obvious peaks were detected,implying that the introduction of NCQDs maintained the high purity and crystalline structure of TiO2.The characteristic signal of NCQDs was hardly noticed in all modified nanocomposites due to the poor crystallinity,high dispersion and low loading of NCQDs in the heterostructure [27].It is worth mentioning that as the NCQDs content increased,the anatase peaks became slightly broader and the peak intensity reduced notably as compared to that of bare TiO2.The result is in accordance with the study by Choiet al.[28].This phenomenon is presumably due to the intercalation of NCQDs atoms within the TiO2matrix,thus resulting in disordered lattice structure and lower degree of crystallinity of the nanocomposites [29].

Fig.2.XRD patterns of pure TiO2 and NCQDs/TiO2 nanocomposites with different NCQDs contents.

3.1.2.Surface morphologic structure

The FESEM micrographs of the pure TiO2and NCQDs/TiO2-4 nanocomposite at a magnification of ×105are illustrated in Fig.3.In Fig.3(a),a hierarchical microsphere structure was observed for bare TiO2with an approximate diameter of 18 nm,which is similar to the structure reported by Zhanget al.[30]and Liuet al.[31].The microspheres particles were uniformly distributed and each particle showed smooth and dense outer surface.After the deposition of NCQDs,the microstructure of TiO2was preserved,yet the surface morphologies of the NCQDs/TiO2-4 particles became irregular as shown in Fig.3(b).Also,it can be noticed that the individual microspheres formed larger aggregates after the incorporation of tiny NCQDs onto the host matrix,which is in good agreement with the findings of Kumaret al.[32].

Fig.3.FESEM images of(a)pure TiO2 and(b)NCQDs/TiO2-4 nanocomposite.(c)TEM image and(d)HRTEM image of NCQDs.(e)TEM image and(f)HRTEM image of NCQDs/TiO2-4 nanocomposite.

Since the expected nanosized NCQDs were too small to be discerned under the reported FESEM magnification,TEM and HRTEM were further conducted to validate the presence of NCQDs and examine their distribution over the TiO2microspheres.The TEM image of pure NCQDs (Fig.3(c)) shows homogeneous dispersion of quasi-spherical nanoparticles with an average diameter of 3.05 nm.In order to evaluate the lattice spacing of NCQDs,HRTEM(Fig.3(d)) was conducted and a clear lattice fringe of 0.24 nm was obtained,which correlated with the (100) facets of graphitic NCQDs [27].The corresponding fast Fourier transform (FFT) pattern of NCQDs in Fig.S1 shows diffused rings,indicating the amorphous phase of NCQDs nanoparticles [33].

From the TEM image of NCQDs/TiO2-4 nanocomposite depicted in Fig.3(e),an abundant amount of dark dots were found to have dispersed uniformly among the clusters of microspheres.Another notable feature is the overlapping of nanoparticles over each other,signifying that intimate interface contact has been established between NCQDs and TiO2,which would be beneficial for electrons migration during the photocatalytic reaction.In Fig.3(f),the smaller particle with lattice distance similar to that of Fig.3(d) was identified as NCQDs,while the relatively larger particle with lattice spacing of 0.35 nm was thereby determined as TiO2.The interplanar spacing agreed well with the (101) crystallographic plane of anatase titania as discussed in the XRD results [6].

EDX analysis was conducted to identify the elements and their respective composition present in the prepared samples.As can be observed in Table S1,pure TiO2was mainly composed of Ti and O elements,with a trace amount of C atoms.After NCQDs were coupled with TiO2,obvious Ti,O,C and N signals were observed in the nanocomposite sample.The weight percentage of C and N atoms increased along with the NCQDs content.This result in a way proved that the formation of TiO2and the deposition of NCQDs can be achieved simultaneously usingin-situhydrothermal technique.

3.1.3.Fourier-transform infrared spectroscopy (FTIR)

The infrared spectra (IR) depicted in Fig.4 were analysed to determine the presence of various functional groups on the catalyst surface.In the FTIR spectrum of NCQDs,a broad vibrational band found between 3000 to 3700 cm-1was ascribed to the stretching of O-H (～3100-3700 cm-1) and N-H (～3300-3500 cm-1) bonds [34].The presence of these functional groups on NCQDs surfaces is conducive to the attraction and transportation of excited electrons during photocatalytic reaction[35].Meanwhile,the absorption peaks at 2970 cm-1,2920 cm-1,2870 cm-1and 1370 cm-1were correlated with the aliphatic,antisymmetric and symmetric stretching mode as well as the bending vibration mode of C-H bond,respectively [36].

To a certain extent,the doping of nitrogen hereatoms on CQDs was evidenced by the stretching vibration of C=O in amide and carboxylic groups located at 1640 cm-1[37].Some of the carboxylic groups present might probably due to the conversion of nitrogen-containing groups during the reaction [38].In addition,the characteristic peaks at 1410-1470 cm-1were in accordance with the skeleton stretching of C=C bond [39].Several adsorption peaks that fell in the range of 1040 to 1300 cm-1were associated with the C-O and C-N stretching (～1020-1250 cm-1) [22].Besides,the bands at 947 cm-1and 814 cm-1were originated from the epoxy groups (C-O-C) [36].It can be concluded that the surface of NCQDs was modulated with abundant moieties,which made NCQDs a highly water-dispersible material [36].Moreover,the functional groups-rich surface of NCQDs could serve as a series of surface emissive sites,which are responsible for the luminescence properties of NCQDs [35].

On the other hand,the FTIR spectrum of bare TiO2presents a broad absorption band centred at 3340 cm-1and a slight peak at 1640 cm-1,which were attributed to the stretching and bending of O-H groups,respectively [40].Moreover,the strong and broad band below 1000 cm-1corresponded to the Ti-O-Ti stretching vibration [28,40].After the decoration of NCQDs on TiO2,all nanocomposites exhibited a comparable spectrum to bare TiO2.The addition of NCQDs characteristic peaks at 1090 cm-1(C-N),1410 cm-1(C=C) and 2920 cm-1(C-H) implied the successful loading of NCQDs on the TiO2surface.When NCQDs dosage increased,the C=O vibration peak became more intense while a wider band below 1000 cm-1was observed concurrently.Besides,the gradual shift of the wide band towards higher wavenumber can be designated to the synergistic combination of Ti-O-Ti and Ti-O-C stretching vibration [6,28].In this regard,the carbon framework of NCQDs and the TiO2structure were most likely linkedviaTi-O-C bonds during the grafting process.

The XPS analysis of NCQDs/TiO2-4 nanocomposites was previously reported in a study published[41].The chemical composition and bonding environment of the NCQDs/TiO2-4 photocatalysts deposited on a membrane support were proven by the wide XPS spectrum which displays signals of C 1s,O 1s,Ti 2p,and weak signal of N 1s.The signals detected were mainly derived from the NCQDs/TiO2coating due to the thin penetration depth of XPS.Furthermore,the narrow scan of Ti 2p and O 1s indicates the presence of NCQDs and TiO2in the composite which can be ascribed to the lattice oxygen atom in TiO2and the formation of Ti-O-C bonds between NCQDs and TiO2.The detailed information can be found in the prior study.The FTIR results are in accordance with the XPS results and the presence of NCQDs in the composite is further proved.

Fig.4.FTIR spectra of NCQDs,pure TiO2 and NCQDs/TiO2 nanocomposites with different NCQDs contents.

3.1.4.Raman analysis

As compared to XRD,Raman scattering is more sensitive to the presence of carbon state or structure of NCQDs[42].To further verify the loading of NCQDs on TiO2,Raman analysis of pure NCQDs and NCQDs/TiO2-4 nanocomposite was conducted and the corresponding vibrational spectroscopy spectra are illustrated in Fig.5.From Fig.5,the Raman spectra of both pure NCQDs and NCQDs/TiO2-4 nanocomposite showed the characteristic band of NCQDs at 1380 cm-1(D band) and a shoulder band at 1590 cm-1(G band),implying the successful decoration of NCQDs on the anatase TiO2[43].As seen,the shoulder G band was similar to the observation reported by Wanget al.[44].The less obvious G and D bands of CQDs were due to the low carbon-lattice-structure content of CQDs.Typically,the D band is assigned to the disordered carbon structure originated from the vibration of sp3hybridized carbon atoms.On the other hand,the G-band represents the ordered,crystalline plane of graphite lattice resulted from the vibration of sp2-bonded carbon atoms [22,45].The existence of both D and G bands signified the successful construction of carbon dots with sp2core and sp3hybridized matrix through hydrothermal method.

Fig.5.Raman spectra of pure NCQDs and NCQDs/TiO2-4 nanocomposite.

Besides,the Raman spectrum of NCQDs/TiO2-4 nanocomposite displayed four distinctive peaks at 146 cm-1,399 cm-1,514 cm-1and 639 cm-1,which were well-matched with theEg,B1g,A1gandEgvibrational mode of anatase TiO2,respectively[45].To gain a profound understanding of the structural variety of NCQDs,the degree of disorder or graphitic order of NCQDs can be represented by the intensity ratio of D to G band(ID/IG).In general,CQDs with a high D/G ratio are amorphous in nature whereas high degree of crystalline GQDs is indicated by a relatively low D/G ratio[42].In this case,the D/G ratio of the synthesized NCQDs was calculated to be 3.6.The high D/G intensity ratio and the obvious intensity of D-band had ascertained the amorphous nature of NCQDs.

3.1.5.UV-Visible diffuse reflectance spectrophotometer (UV-vis DRS)

The optical properties of NCQDs solution and the nanocomposite photocatalysts were characterized by UV-vis spectroscopy and UV-vis DRS techniques,respectively.Fig.6(a) illustrates the UVvis absorption spectrum of NCQDs,showing two typical peaks at 278 nm and 357 nm.The peak at 278 nm correlated with the ππ* transition of the sp2-hybridized conjugated C=C bond,while the shoulder peak at 357 nm was assigned to the n-π* transition of C=O or C=N bonds in the NCQDs structure[21,22,28].Inset picture of Fig.6(a) (top right corner) presents the highly dispersed NCQDs nanoparticles in isopropanol solution under ambient and UV light.The yellowish-brown NCQDs emitted a bright green fluorescence after being excited by UV illumination.The luminescence properties proved the successful fabrication of NCQDs from waste oil palm frond biomass.

To explore the impact of NCQDs on the absorption properties of TiO2,UV-vis DRS of bare TiO2and NCQDs/TiO2nanocomposites were analysed.As presented in Fig.6(b),the bare TiO2exhibited a strong absorption band at UV region with an absorption edge at approximately 383 nm,which agreed well with the intrinsic property of bulk anatase (Eg=3.2 eV,corresponding to λ=387 nm) [46].Meanwhile,bare TiO2had no photo-response in the visible region ranging from 400 to 700 nm.An obvious red-shift of absorption edge to visible region (from 383 nm to 427 nm) was noticed after the coupling of NCQDs.The extended photo-response of NCQDs/TiO2nanocomposites is related to the electronic coupling between the π-states of NCQDs and the conduction band of TiO2[6].With the increment of NCQDs loading,the absorption intensity over the visible range increased remarkably.The enhancement in visible-light absorption was in good agreement with the colour transition of nanocomposite powder from white to brownish as depicted in Fig.S2.

The band gap value of bare TiO2and NCQDs/TiO2nanocomposites were calculated using Kubelka-Munk plot [6] as presented in Fig.6(c).Notably,the band gap energy of nanocomposites was significantly reduced to 2.77 eV as compared to that of bare TiO2(3.2 eV).The band gap narrowing can be better explained by the introduction of new carbon energy state into the band structure of TiO2,thus reducing the overall band gap of nanocomposites[6,32].The reduced optical band gap energy is expected to enhance the light-harvesting ability of heterostructures during photocatalytic reaction [28].

Fig.6.(a) UV-vis absorption spectrum of NCQDs.(b) UV-vis diffuse reflectance spectra of pure TiO2 and NCQDs/TiO2 nanocomposites.(c) Kubelka-Munk plot of pure TiO2 and NCQDs/TiO2 nanocomposites.

3.1.6.Photoluminescence spectroscopy (PL)

Despite the broad spectra response range of NCQDs/TiO2nanocomposites,the photocatalytic efficiency of nanoparticles might still be affected if the charge recombination process occurs rapidly.Therefore,PL study was conducted to evaluate the charge separation efficiency within pure TiO2and NCQDs/TiO2nanocomposites using an excitation wavelength of 320 nm.In general,a lower PL spectra intensity corresponds to a lower recombination rate of electron-hole.[42].As shown in Fig.7(a),the PL intensity of nanocomposites was remarkably quenched by the addition of NCQDs,signifying the delayed recombination rate of charge carriers.The accelerated charge transfer was due to the close interface of NCQDs-TiO2contactedviaTi-O-C bond [6].Among all the nanocomposites,NCQDs/TiO2-4 exhibited the weakest PL intensity,which did not correlated well with the UV-vis DRS results.This is because nanocomposites with high coverage of NCQDs create trap states on the NCQDs/TiO2surface,thus decelerating the interfacial charge transfer [23].

Fig.7(b)and(c)demonstrate the excitation-dependent PL emission spectra of NCQDs under varying excitation wavelengths.As the excitation wavelength increased from 300 nm to 500 nm,the corresponding emission peaks were red-shifted to longer wavelengths from 390 nm to 500 nm.The intensity of the downconverted PL peaks gradually increased with the excitation wavelength and showed a maximum emission peak centred at 390 nm when the excitation wavelength was 350 nm.The PL intensity then showed a decreasing trend as the excitation wavelengths further increased to 500 nm.The obvious excitation-dependent emission property of NCQDs revealed the tunable emission nature of NCQDs,which is consistent with the results reported in previous literature [27,42].

The broad emission spectra range is generally attributed to the presence of different surface emissive traps on NCQDs surface[22,32].Hence,we can conclude that the surface passivation or functionalization may be responsible for the PL behaviour of NCQDs.As observed in Fig.7(b) and (c),no emission peak was detected at excitation wavelength of 500 nm and above,indicating the NCQDs did not exhibit up conversion fluorescence property.This result is consistent with the work reported by Simet al.[23].Therefore in this study,the compounding of TiO2with NCQDs extended the life span of photo-induced electrons which potentially lead to the enhancement in photocatalytic decomposition of MB.

3.2.Photocatalytic ability of NCQDs/TiO2 nanocomposites

The photocatalytic performance of the prepared photocatalysts under visible light irradiation was assessed by selecting MB as the model pollutant.To reasonably study the photocatalytic activity of the modified photocatalysts,blank experiment and adsorption study were conducted to discount the effect of MB selfdegradation and catalyst adsorption of MB.As depicted in Fig.8,the reduction of MB concentration in the absence of photocatalyst was found negligible,thereby the self-degradation of MB by photolysis can be ignored [47].For the adsorption study,the pure TiO2and all the nanocomposites exhibited considerably low adsorption capability as compared to their catalytic activity,hence surface adsorption did not play decisive role in the removal of MB.Therefore,the degradation efficiency of NCQDs/TiO2nanocomposites can be evaluated in comparison to the pure TiO2and commercial P25.As seen,the sudden dip observed att=-30 was due to the rapid adsorption of dye molecules on the abundance active sites when the photocatalysts were first dumped into the reaction medium.This resulted in the drastic drop in MB concentration during the first 30 min of contact time.Beyond that point,the adsorption process slowed down while desorption started to take place until an equilibrium point was reached.

Fig.7.(a) PL spectra of pure TiO2 and all NCQDs/TiO2 nanocomposites;(b) and (c) The PL emission spectra of NCQDs at different excitation wavelengths.

Fig.8.Photodegradation of MB under visible-light irradiation by all photocatalysts.

As can be seen in Fig.8,the pure TiO2and commercial P25 decomposed 30.18%and 26.29%of MB after 4 h of visible light irradiation.The low degradation rate was attributed to the poor TiO2response to visible light irradiation,as expected by the wide band gap energy observed in the Kubelka-Munk plot.In contrast,all NCQDs/TiO2nanocomposites showed enhanced photodegradation activity after 4 h of visible light illumination.The photocatalytic performance of nanocomposites was significantly improved from 30.18% to 71.01%,86.16%,47.23% and 40.08% when the NCQDs loading increased from 0 to 1 ml,4 ml,6 ml and 10 ml,respectively.Obviously,NCQDs/TiO2-4 exhibited the highest removal efficiency of MB which increased by 56% as compared to that of pure TiO2.The outstanding photocatalytic performance of NCQDs/TiO2nanocomposites was mainly attributed to the cooperative interaction between NCQDs and TiO2.The multifunctionality of NCQDs enhanced the interfacial charge separation and lightharvesting capacity of TiO2,as validated in the PL and UV-vis DRS analysis above [42].

Though an increase in NCQDs loading is beneficial for the photocatalytic activity,however,the further addition of NCQDs beyond 4 ml had rendered these photocatalysts with low degradation rate.This is well consistent with the PL results which explained that the redundant amount of NCQDs deposited on the TiO2surface served as recombination centers for photo-induced charge carriers [15].Besides,the surface coverage of TiO2by the thick NCQDs layer may inhibit light penetration to TiO2particles [6,43].Moreover,the excessive NCQDs nanoparticles will compete with TiO2for photon absorption,thus reducing the light absorption efficiency of TiO2[6,27,43].Furthermore,the agglomeration of nanoparticles will minimize the surface active sites available for interfacial adhesion of pollutant molecules.The reduction in contacting surface might reduce the photocatalytic activity since the attraction of MB molecules towards photocatalyst is mainly facilitated by the aromatic bond and the oxygen-containing functional groups on the NCQDs surface[27].On this account,appropriate NCQDs loading is an important factor to be considered in order to realize the full potential of NCQDs during the photocatalytic degradation of MB.

The kinetic rate constants of photocatalysts were calculated to understand the photodegradation activity of different photocatalysts.Fig.S3 demonstrates that the experimental data was wellfitted to the pseudo-first-order kinetic model with regression coefficients,R2greater than 0.97 (Table S2).The degradation rate constants were obtained from the slope of the linear lines and displayed in Fig.9.Notably,NCQDs/TiO2-4 showed the highest degradation rate constant of 0.0081 min-1,which was around 6-fold faster than that of pure TiO2(0.0014 min-1).Table 1 summarizes and compares the degradation performance of NCQDs/TiO2-4 with that of other CQDs modified photocatalysts.It can be observed that the prepared NCQDs/TiO2in this work exhibited a considerably high degradation rate among the other biomassderived CQDs-TiO2in previous literature.On the other hand,it can be noticed that the photocatalytic performances of biomassderived CQD/TiO2are much lower than that of CQDs/TiO2prepared using chemical precursors.As can be noticed from Table 1,the degradation efficiency of the chemically-derived CQDs/TiO2prepared using citric acid were above 90%and required shorter degradation time.Despite the numerous advantages of biomass carbon sources,the application of green CQDs in photocatalysis is still not well-established as compared to that of chemical-derived CQDs.Hence,additional exploration is required to develop environmentally-benign CQDs with remarkable optical and catalytic properties.

Table 1Comparison of kinetic rate constant of different CQDs/TiO2 under visible light irradiation

Fig.9.The corresponding kinetic rate constant of all photocatalysts.

The reusability of the optimal NCQDs/TiO2-4 nanocomposites deposited on a membrane support has been previously reported[41].The cycling run for the photodegradation of MB was studied by repeating the degradation test for 3 cycles.The results showed that about 70% of MB could be removed after three consecutive experiments.The minor decrement in the degradation activity may be attributed to the adhesion of small organic species that was not removed completely in the previous run,thereby retarding the degradation activity in the following cycles.The steady degradation behaviour has verified the stability and recyclability of the NCQDs/TiO2nanocomposites which is feasible for the practical application in water purification.

3.3.Influence of the reaction conditions

3.3.1.Effect of initial pH of solution

As observed in Fig.10,the photocatalytic activity of NCQDs/TiO2-4 was greatly dependant on the initial pH of MB solution since the surface charges on the photocatalysts could affect the adsorption and dissociation of organic molecules [48].In Fig.10,it was observed that the reduction of MB concentration was slow when the pH value fell in the acid range of pH 3 to 5.On the contrary,the removal rate of MB increased remarkably when the pH was raised from 5 to 12.The high adsorption efficacy of the photocatalyst under basic condition can be explained by the strong electrostatic interaction between the photocatalyst surface and the targeted MB molecules.To better understand this phenomenon,zeta potential of NCQDs/TiO2-4 was measured in the pH range of 2-10.As shown in Fig.S4,the point zero charge (pHpzc) of NCQDs/TiO2-4 was discovered at pH 6.8,which is very similar to the results described in the previous literature [37,44].The negatively charged nature of NCQDs modified nanocomposite(-10 mV,pH=7.6) was favourable for the adsorption of cationic MB dye on the photocatalyst surface [49].

Generally,when the pH value is less than pHpzc,the outer surfaces of nanoparticles are protonated and become positively charged.When the pH is beyond the isoelectric point,the nanoparticles become predominantly negatively charged and exhibit high affinity with cationic reactant.Good adsorption is deemed as the prerequisite for enhanced degradation efficiency since it facilitates the direct contact of MB molecules with photocatalysts.In an alkaline environment,the high availability of OH ions gave rise to a more negatively charged surface of NCQDs/TiO2-4,hence resulting in strong electrostatic attraction to the cationic MB molecules.Meanwhile,the large amount of OH ions in the basic medium could promote the photogeneration of·OH radicals which contribute to the high degradation efficiency of NCQDs/TiO2-4 [50].

In an acidic medium,electrostatic repulsion force was dominant since the like charges inhibited the migration of MB molecules towards the nanoparticles.Besides,the degradation of MB was retarded by the intense competition ofh+ions with the cationic MB molecules for adsorption sites,thereby reduced the uptake of cationic organic molecules.Other than the formation of·OH,low reactivity radicals such aswere also generated and participated in the degradation process [51].The competition among the abundant amount of radicals generated would lead to a scavenging effect on·OH,as reported in the study by Ghasemiet al.[51].Furthermore,the·OH radicals generated might be deactivated rapidly under elevated pH [48].

Fig.10.Effects of initial solution pH on the photocatalytic activity of NCQDs/TiO2-4 nanocomposite.

Though there was a tremendous reduction in MB concentration at pH 10 and 12,however,almost 80% of the MB reduction was adsorbed rather than degraded.The degradation rate of MB was insignificant and was even slower than that occurred under acidic conditions.In summary,the optimal photodegradation performance was found at pH 8 since a good balance between the adsorption and the decomposition of MB molecules had been achieved under this pH condition.

Fig.11.Effects of catalyst dosage on the photocatalytic activity of NCQDs/TiO2-4 nanocomposite.

3.3.2.Effect of catalyst dosage

As shown in Fig.11,the effect of catalyst loading on the degradation of MB was investigated by varying the catalyst dosage(0.5 g·L-1,1 g·L-1,1.5 g·L-1,2 g·L-1).From Fig.11,the maximum MB degradation of NCQDs/TiO2-4 was found at 1 g·L-1.Initially,the improvement in degradation performance with increased catalyst dosage was due to the increasing availability of active sites,which allowed more interaction between the photocatalysts and MB molecules[48,49].Meanwhile,a greater amount of·OH radicals can be produced by an increased dose of photocatalysts[26].However,when the catalyst dosage was raised to 1.5 g·L-1and 2 g·L-1,the degradation rate drop from 86.16% to 74.41% and 68.67%,respectively.The decreasing trend revealed that high catalyst dosage did not necessarily bring a positive impact to the photocatalytic activity.

When the concentration of solid catalyst was in excess,the nanoparticles tend to agglomerate and form large aggregates,hence minimized the effective active sites available and hampered the photocatalyst from light irradiation [52].In addition to the above,excessive catalyst multiplied the collision frequency between activated molecules with other ground-state molecules,ultimately leading to deactivation of photoactivated molecules[51,53].Moreover,the bulk catalyst increased the turbidity of the reaction medium which blocked the light from penetrating through the suspension [52].Also,light scattering effect might become pronounced and thus,producing lesser photoinduced electrons for MB destruction[52,53].In this case,the optimum amount of catalyst dosage was found at 1 g·L-1which able to attain efficient light absorption and good adsorption of MB molecules during the photocatalytic reaction.

3.4.Mechanism of photocatalytic degradation

3.4.1.Free radicals scavenging test

To illustrate the potential degradation mechanism of NCQDs/TiO2,scavenging test was conducted to determine the dominant reactive oxygen species (ROS) involved in the photocatalytic system.The three primary scavengers used to study the inhibition effects of h+,·OH andradicals were EDTA,IPA and BQ,respectively.As shown in Fig.12,the inhibition effect of the active radical species followed the order of h+＞·OH ＞The holes were identified as the key reactive species since the degradation rate was significantly quenched from 86.16%to 37.48%by the EDTA scavenger.Besides,the photodegradation of MB was also restricted to some extent in the existence of IPA and BQ scavengers,which was 55.77% and 73.81% respectively.The slight inhibition exerted by IPA and BQ scavengers on the MB degradation process revealed that both·OH andradicals played a secondary role in the photocatalytic reaction.

3.4.2.Band structure analysis

To elucidate the reactive species generated in the photocatalytic system,it is necessary to determine the band-edge potentials of NCQDs/TiO2nanocomposites.The energy levels of the conduction band (ECB) and valence band (EVB) were deduced theoretically using Butler and Ginley approach [54] as shown by the following relation:

Fig.12.Effects of different scavengers on the removal of MB by NCQDs/TiO2-4.

whereXis the absolute electronegativity of TiO2(5.9 eV)[55],Eeis the energy of free electrons on the hydrogen scale (ca.4.5 eVvsNHE) [23] andEgis the band gap energy of TiO2(3.2 eV).The VB and CB of TiO2were calculated at 3.0 eV and -0.2 eV,respectively.According to the band-edge potentials obtained,the CB energy level of TiO2(-0.2 eV)was more positive than the standard redox potential of(-0.33 eV),implying that the excited electrons of TiO2were unable to reduce oxygen molecules intoradicals [44].However,this is in opposition to the scavenging result which indicated the involvement ofspecies in the degradation process.Considering the abovementioned facts,NCQDs most probably act as a photosensitizer in the NCQDs/TiO2photocatalytic system[44,56,57].

Fig.13.Effects of superoxide scavenger for pure TiO2 on MB removal.

To validate this assumption,a trapping experiment over pure TiO2photocatalyst was performed as a comparative study.In Fig.13,the introduction of BQ can hardly exert any inhibitory effect on the photocatalytic degradation of MB,implying that noradical was generated due to the lower CB position of TiO2.In contrast,the slight inhibition rate observed in Fig.12 proved that NCQDs can generateradicals during the photocatalytic process,hence serving as a co-catalyst in the photocatalytic system[56].While for the band structure of NCQDs,the CB and VB band edges are established as -0.73 eV and 1.80 eV as obtained from the work by Zhanget al.[6]who synthesised green CQDs from biomass,which was very similar to this work.

Based on the edge potentials obtained,the band structure diagram of NCQDs/TiO2nanocomposite is illustrated in Fig.14.Considering the band gap alignment diagram,there are two possible mechanisms for the charge transfer to take place,which are conventional type-II and direct Z-scheme heterojunction system[56].In a type-II heterojunction (Fig.14(a)),the excited electrons would transfer from the CB of NCQDs to the CB of TiO2whereas the holes would transfer in the opposite direction.Although the charge separation is effective,however,the electrons and holes that accumulate on the CB of TiO2and the VB of NCQDs,respectively are unable to generateand·OH radicals due to the low activation ability of the band potentials.[44].On top of that,theand·OH radicals identified in the scavenging test suggested that this type-II mechanism cannot be applied in the NCQDs/TiO2heterostructure.

While in the direct Z-scheme mechanism (Fig.14(b)),the electrons on CB of TiO2would quench with the holes in the VB of NCQDs,thus leading to the reservation of electrons and holes on the CB of NCQDs and VB of TiO2,respectively [57].Consequently,bothand·OH could be generated since the CB of NCQDs(-0.73 eV)and VB of TiO2(3.0 eV)possess stronger redox potential with high activation ability[44].Together with the scavenging test and the band gap analysis,it is reasonable to propose that the NCQDs/TiO2nanocomposite followed the direct Z-scheme heterojunction mechanism and had achieved an effective spatial charge separation,hence resulting in enhanced photocatalytic performance.

3.4.3.Mechanisms insights

The pivotal roles of NCQDs in enhancing the photocatalytic activity of NCQDs/TiO2nanocomposites are summarized as below.Firstly,NCQDs can direct the preferential flow of photo-induced electrons across the heterojunction interfacesviaTi-O-C bonding.The rapid electron migration may efficiently suppress the recombination of charges in the nanocomposites,as validated by the PL analysis[27].Secondly,the new carbon state of NCQDs that introduced within the band structure of NCQDs/TiO2nanocomposites can tune the band gap of TiO2and extended the absorption edge to visible light region,as evidenced by UV-vis DRS.Thirdly,the photosensitizing effect of π-conjugated NCQDs contributes to the enhanced photocatalytic activity of NCQDs/TiO2by generating more electrons for the degradation of MB [23].Upon visible light irradiation,NCQDs will absorb low energy photon and excite electrons from highest occupied molecule orbital (HOMO) to lowest un-occupied molecular orbital (LUMO) of NCQDs [58].Although the NCQDs in this study did not display UCPL properties,however,the reduced overall band gap can harvest visible light and generate more electrons for the degradation of MB [23].

Based on the above analysis,a plausible mechanism for the enhanced photocatalytic performance of NCQDs/TiO2nanocomposite was explored in detail using the schematic diagram illustrated in Fig.15.Upon light irradiation,the π-conjugated NCQDs absorb light in visible and NIR regions to generate electron-hole pairs for the sensitization of TiO2particles [59].Meanwhile,the reduced band gap of TiO2nanoparticles will be activated by visible light followed by the excitation of electrons from the VB to CB of TiO2[26].In the Z-scheme heterojunction system,the electrons from CB of TiO2will migrate and then combine with the holes at the VB band of NCQDs.The close interfacial contact aroused from the strong chemical bonding and the strong electron affinity of NCQDs will accelerate the migration of photo-induced electrons across the hetero-interface from CB of TiO2to NCQDs surface,thus prohibiting the photoexcited electrons from leaping back to the lower state [38].As a result,the high charge density and the prolonged life span of charge carriers will subsequently participate in a series of redox reactions by generating reactive oxygen species(ROS) [43].

Fig.14.The proposed charge-transfer mechanisms of (a) type-II and (b) direct Z-scheme heterojunctions for NCQDs/TiO2 photocatalysts.

Fig.15.Schematic mechanism of NCQDs/TiO2 nanocomposite.

The electrons that accumulate on the NCQDs surface will reduce the adsorbed oxygen molecules to yield superoxide radicals ().The unstablethen further react with holes (h+) to produce hydroxyl radicals (·OH) [34].Meanwhile,the photo-excited holes that remain on the valence band of TiO2will oxidize the surface adsorbed water molecules or hydroxyl groups(OH-)to form highly reactive·OH radicals[31].Bothand·OH radicals exhibit strong oxidizing power to decompose the MB molecules into non-harmful substances such as carbon dioxide and water [21].Besides,the conjugated structure of NCQDs would form π-π stacking interaction with the aromatic ring of organic pollutants [40].In addition,the presence of oxygenated surface groups on NCQDs will engage and interact with the MB molecules.The close interfacial contact of MB pollutants with NCQDs surface speed up the radicals attack and destruct the organic structure.The synergistic effect of the multi roles of NCQDs eventually leads to the outstanding degradation performance of NCQDs/TiO2nanocomposites.

4.Conclusions

In summary,a direct Z-scheme NCQDs/TiO2nanocomposite has been successfully synthesizedviaa facilein-situhydrothermal treatment using waste oil palm frond as an alternative precursor for NCQDs.The incorporation of NCQDs had significantly promoted the utilization of visible light and hindered the recombination of electron-hole pairs in NCQDs/TiO2nanocomposite.The photocatalytic activity of NCQDs/TiO2-4 towards MB degradation was found to be the highest,proving that the degradation efficiency was highly dependent on the dosage of NCQDs.Moreover,NCQDs/TiO2-4 nanocomposite exhibited the best performance at a catalyst dosage of 1 g·L-1under the alkaline condition.The reaction rate constant of NCQDs/TiO2-4 was 5.8-fold faster than that of pure TiO2.The present study provided a renewable and sustainable strategy in the design of visible-light responsive photocatalyst by turning waste biomass into higher-value products.

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.

Acknowledgements

The authors would like to express their gratitude towards the funding provided by Universiti Tunku Abdul Rahman Research fund (IPSR/RMC/UTARRF/2020-C2/C06) and Centre for Photonics and Advanced Materials Research(CPAMR,UTAR)for their support.

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2021.10.021.