EnvironmentalScienceNano

PAPER

Cite this: Environ. Sci.: Nano, 2019,

6, 948

Received 6th December 2018,Accepted 19th January 2019

DOI: 10.1039/c8en01375f

rsc.li/es-nano

Micro/nano-bubble assisted synthesis of Au/TiO2@CNTs composite photocatalyst forphotocatalytic degradation of gaseous styreneand its enhanced catalytic mechanism†

Weiping Zhang, Guiying Li, Hongli Liu, Jiangyao Chen,Shengtao Ma and Taicheng An *

A facile micro/nano-bubble method was firstly applied to synthesize an Au/TiO2@CNT composite photo-

catalyst for photocatalytic degradation of gaseous styrene. The morphologies, structures and compositions

of the photocatalysts were investigated by using a series of analytical techniques (scanning and transmis-

sion electron microscopy, FT-IR spectroscopy, Raman spectroscopy, X-ray diffraction, and X-ray photo-

electron spectroscopy). The micro/nano-bubbles can effectively facilitate the reaction of Au and TiO2 NPs

decorated onto CNTs, forming a stable ternary composite structure. The photocatalytic performance of

Au/TiO2@CNTs was investigated by the methodology of central composite design (CCD) in response sur-

face methodology (RSM). The photocatalytic degradation and mineralization of styrene over Au/

TiO2@CNTs drastically increased with the rise of reaction temperature due to the formation of a compact

structure. The analysis of EPR, UV-vis DRS, electrochemical properties and TPD-O2 further revealed the en-

hanced photocatalytic mechanism of Au/TiO2@CNTs. The further identification of free radicals showed that

the photocatalytic degradation and mineralization of styrene were closely related to oxidative radicals such

as hydroxyl radicals and superoxide radicals, which were mainly attributed to the synergistic effects of Au

NPs and CNTs to enhance the photocatalytic activity. This work will provide a new insight into designing

photocatalysts with high photocatalytic performance and stability for the degradation of hazardous VOCs.

1. Introduction

Photocatalytic oxidation has been extensively studied for airpurification, such as the removal of volatile organic com-

pounds (VOCs).1–4 Among various semiconductor materials,TiO2 is one of the most promising photocatalysts for environ-mental purification owing to its low cost, nontoxicity, goodstability, and excellent photocatalytic activity.5 However, itswide band gap of 3.2 eV, high recombination rate of photo-generated electron/hole pairs and low specific surface areaare obstacles for its industrial application. Many attemptshave been already taken to improve the photocatalytic proper-ties, and the main strategy is to fabricate a novel structureand improve the surface performance to obtain high-

948 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

Guangzhou Key Laboratory of Environmental Catalysis and Pollution Control,

Guangdong Key Laboratory of Environmental Catalysis and Health Risk Control,

School of Environmental Science and Engineering, Institute of Environmental

Health and Pollution Control, Guangdong University of Technology, Guangdong,

510006, China. E-mail: antc99@gdut.edu.cn

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8en01375f

Environmental significance

In the field, photocatalytic oxidation in the presence of TiO2 to give total or partial oxidation of gaseous contaminants to benign substances is one of themost promising environmentally friendly techniques for the abatement of volatile organic compounds (VOCs). However, TiO2 has a wide band gap, highrecombination rate of photogenerated electron/hole pairs and low specific surface area. In this work, physical micro/nano bubbles were used as a softtemplate to induce the synthesis of Au/TiO2 NPs assembled with CNTs. On the basis of the experimental results of central composite design (CCD) inresponse surface methodology (RSM), the intrinsic connections between the photocatalytic properties of the Au/TiO2@CNT composite and various factorssuch as reaction temperature, the mass fraction of CNTs, the amount of HAuCl4 and reaction time were systematically revealed. Furthermore, the influenceof the structure and composition on the photocatalytic properties, recombination of photogenerated electron/hole pairs, electrochemical properties andfree radical generation of the Au/TiO2@CNT composite was also discussed. Finally, a model of the photocatalyst system was carefully proposed to predictthe migration and transformation mechanisms of photogenerated electron/hole pairs according to the obtained results.

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

.

View Article OnlineView Journal | View Issue

Environ. Sci.: Nano, 2019, 6, 948–958 | 949This journal is © The Royal Society of Chemistry 2019

efficiency TiO2-based visible-light photocatalysts. For in-stance, doping with metals or nonmetals, as a better way tooptimize the electronic structure of TiO2, has been frequentlyused to enhance the photocatalytic performance by reducingthe excitation energy and increasing the surface active sitesof the photocatalyst.6–8

Recently, the heterogeneous structures of Au nano-crystal/TiO2 spheres have been attracting much attention for theirvisible-light photocatalytic activities due to surface plasmonresonance (SPR).9–12 As is reported, compared with single-phase TiO2 photocatalysts, Au nano-crystal/TiO2 spheres havea light-response range of 400–600 nm, which can significantlyenhance the utilization of photons for organics degrada-tion.13,14 However, Au nano-crystal/TiO2 spheres possess lowspecific surface areas and also provide few adsorption activesites.15 Generally, the photocatalytic reaction takes place atthe active sites of the photocatalyst defined as micro-reactorswith their finite oxidizing ability and mass-transfer ability. Inthat case, the corresponding micro-reactors suffer a heavierburden for photocatalytic degradation of organics, resultingin the excessive accumulation of intermediates on the photo-catalyst.16 As a result, the decreased contact probability be-tween the reactant and photocatalyst can lead to a lowerphotocatalytic activity of Au nano-crystal/TiO2. Throughoutthe process, the main decreased photocatalytic activity is at-tributed to the lower specific surface area and the mass trans-fer rates of Au/TiO2, resulting in difficulty in desorption of in-termediates and adsorption of the reactant.17–19 Thus, toobtain a high-efficiency photocatalyst with more active sites,it is necessary to conduct regulation of the surface structureand performance.

Carbon nanotubes (CNTs), as typical one-dimensionalnanostructures, have been attracting much attention in fabri-cating novel photocatalytic materials due to their unusualmechanical, electrical, and chemical properties.20–22 Owingto their large specific surface area, CNTs have been consid-ered to be promising co-adsorbents or supports for the ad-sorption and photocatalytic degradation of organiccontaminants.17,18,20–22 For example, CNTs could apparentlyimprove the photocatalytic activity of TiO2 due to the CNTsbeing electron acceptors that promote the separation ofphotogenerated electron/hole pairs.23–25 Hence, it is highlydesired to integrate the special characteristics of CNTs withAu/TiO2 to fabricate composite photocatalysts with efficientmass transfer and high utilization of photogenerated chargecarriers by an appropriate method.

Concerning the material combination of Au/TiO2 withCNTs, several self-assembly strategies have been developedrecently. Zeng et al.26,27 presented a self-assembly approachto fabricate Au/TiO2/CNT nanocomposites by photo-assistedgrowth of Au nanoparticles and using premade TiO2/CNTs asthe support, which was considered as an appropriate methodfor fabricating complex 1D systems of ternary nano-composites with better control of nanoparticles with activephases (Au and TiO2 nanoparticles). Nonetheless, thesemethods bear some negative implications, such as the utiliza-

tion of an organic solvent or additive which then suppressesthe photocatalytic activity due to residual organic matter.

A firm chemical interface is the key factor to obtain ter-nary nanocomposites or more complex nanocomposites withhigh photocatalytic activity and an in situ self-assembly strat-egy proves to be an excellent method for fabricating newtypes of nanocomposites. However, two things need specialattention when applying this strategy. First, appropriate inter-facial inducers should be selected to induce the formation ofchemical interfaces between CNTs and TiO2 nanoparticles toavoid overusing organic matter. Recently, chemical micro/nano-bubbles were used as a soft template through a self-assembly method to synthesize various nanomaterials due totheir typical properties of interfacial charge, enduring stabil-ity, and efficient interfacial mass-transfer.28–31 Second, owingto the special interfacial characteristics of micro/nano-bub-bles, they can be used as an interfacial inducer to facilitatethe interfacial adsorption, mass-transfer and chemical reac-tion for the formation of a specialized chemical interface.That is, the micro/nano bubbles can also show potential forfabricating multi-element composite materials, such as theternary composite material Au/TiO2@CNTs.

In this work, physical micro/nano bubbles were used as asoft template to induce the synthesis of Au/TiO2 NPs assem-bled with CNTs. Typical micro/nano-bubbles could effectivelycontrol the size and distribution of Au/TiO2 nanoparticles onCNTs, resulting in the formation of a firm chemical structurebetween Au/TiO2 and CNTs, which could significantly im-prove the photocatalytic performance and the stability in gas-eous styrene degradation. On the basis of the experimentalresults of central composite design (CCD) in response surfacemethodology (RSM), the intrinsic connections between thephotocatalytic properties of the Au/TiO2@CNT composite andvarious factors such as reaction temperature, the mass frac-tion of CNTs, the amount of HAuCl4 and reaction time weresystematically revealed. Furthermore, the influence of thestructure and composition on the photocatalytic properties,recombination of photogenerated electron/hole pairs, electro-chemical properties and free radical generation of the Au/TiO2@CNT composite was also discussed. Finally, a modelphotocatalyst system was carefully proposed to predict themigration and transformation mechanisms of the photo-generated electron/hole pairs according to the obtainedresults.

2. Experimental section2.1 Central composite design for preparation ofphotocatalysts

The design approach for photocatalyst preparation wasperformed by using a CCD with four independent variables,including reaction temperature, the mass fraction of CNTs,reaction time and the amount of HAuCl4. The determinedvalues of the variables were referenced to our preliminary ex-periments as presented in Table S1.† The experimental datawere analysed by Design-Expert software and a total of 30

Environmental Science: Nano Paper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

950 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

experimental runs were performed. The complete experimen-tal design results are shown in Table S2.†

2.2 Synthesis

The Au/TiO2@CNT composites were synthesized with amicro/nano-bubble method to induce the formation ofheterostructures based on electrostatic attraction of negativecharges on the surface of the micro/nano-bubbles. The func-tional CNTs were obtained through a modified oxidationmethod and the details can be seen in the ESI.† First, approx-imately 0.0832 g of functional CNTs, 2 mL of 2 mol L−1 TiCl4(99%, Aladdin Reagent Co., Ltd), 2 mL of glycerol (99%,Sinopharm Chemical Reagent Co., Ltd.) and 80 mL of micro/nano-bubble water were added into a 250 mL conical flask.Second, after 30 min of ultrasonic treatment, the conicalflask was placed in a water bath and 1.5 mL of ammonia(13.33 mol L−1, Guangzhou Chemical Reagent Factory) wasadded into the mixture dropwise under stirring and kept for13 h at room temperature. Third, a given amount of HAuCl4(Sinopharm Chemical Reagent Co., Ltd.) was slowly addedinto the conical flask and reduced with a 300 W xenon lamp(PLS-SXE300/300UV, Perfectlight Co., Ltd.) for 15 min. Then,the obtained mixture was kept at selected reaction tempera-ture for different durations. After the reaction, the productswere collected by centrifugation, washed with distilled waterand ethanol (Guangzhou Chemical Reagent Factory) thor-oughly, and then dried at 80 °C for 18 h. Finally, the obtainedAu/TiO2@CNT composite was calcined at 420 °C for 2 h withthe protection of N2 gas (flow rate: 30 mL min−1). The gasused to generate the micro- and nano-bubbles was the airsuctioned from the atmosphere, passing through a filter inthe gas inlet. The micro/nano-bubbles were produced by acommercial micro-bubble generator (Xiazhichun Environ-mental Protection and Technology Co. Ltd., China) composedof a magnetic gear pump, a pressurized tank and a micro/nano-nozzle in the outlet. Deionized water was used in allexperiments.

2.3 Characterization

The morphologies of the as-synthesized samples were ob-served using scanning electron microscopy (SEM, ZEISS Mer-lin, German) and transmission electron microscopy (TEM,Hitachi, Japan). The phases of the obtained samples werecharacterized on a Bruker D8 Advance X-ray diffractometer(XRD, German) with a Cu tube for generating Cu Kα radia-tion (λ = 1.5418 Å). The functional groups of the sampleswere analysed using a Nicolet6700 Fourier transform-infraredspectrometer (FT-IR, Thermofisher, America). The Brunauer–Emmett–Teller (BET) surface area and pore distribution werecharacterized using an ASAP 2020 PLUS HD88 surface areaand porosity analyzer (Micromeritics, America). The chemicalcompositions and molecular structures of the as-synthesizedsamples were analysed through UV laser Raman spectroscopy(HORIBA Jobin Yvon, France). The surface electronic stateswere analysed by using a Phi X-tool instrument (XPS, Ulvac-

Phi, Japan). Electron paramagnetic resonance (EPR) experi-ments were performed with a Bruker EMXPlus spectrometeroperating at the X-band (9.8 GHz), equipped with a cylindri-cal cavity operating at 100 kHz field modulation. The pres-ence and mobility of oxygen species on the photocatalystwere studied by using temperature-programmed desorptionof oxygen (O2-TPD) performed on an Autochem II 2920 chem-ical adsorption instrument (Micromeritics). The UV-vis dif-fuse reflectance spectra (DRS) were collected on an AgilentCary 300 spectrophotometer by using BaSO4 as the reflec-tance standard. The fluorescence spectra were collected byusing an F-7000 fluorescence spectrophotometer (Hitachi, Ja-pan) with an excitation wavelength of 321 nm at wavelengthsbetween 330–600 nm. The electrochemical properties wereanalysed by using an Autolab PGSTAT302N electrochemicalworkstation, and the details can be seen in the ESI.† The eval-uation of the photocatalytic activity of the Au/TiO2@CNTcomposite photocatalyst was conducted on a continuousphotocatalytic reaction system by using a 300 W xenon lampas the simulated solar light source, and the details are de-scribed in the ESI.†

3. Results and discussion3.1 Morphology, structure and composition analysis

The Au/TiO2@CNT ternary composite was typically synthe-sized using a micro/nano-bubble method and the morphol-ogies were observed by SEM (Fig. 1a1–c1) and TEM (Fig. 1a2–c2). The 1D CNTs serve as a novel support which is uniformlydecorated with Au/TiO2 NPs, and the Au/TiO2 NPs are closelycombined with the CNTs. Thus, the close neighborhood ofAu/TiO2 NPs and CNTs achieved by this method is believedto favor the vectorial transfer of photogenerated electronsfrom Au/TiO2 to CNTs, resulting in the enhancement ofcharge separation and photocatalytic efficiency.24,25

To investigate its interfacial structure, the resultant Au/TiO2@CNT composite was also subjected to high-resolutionTEM (HR-TEM) characterization. It is observed that the Au/TiO2 NPs are homogeneously decorated onto the CNTs(Fig. 2a and S1†) and there is no obvious separation bound-ary even under ultrasonic treatment for 2 h. This further con-firmed that chemical bonds could be formed among Au, TiO2

and CNTs. Furthermore, a selected area electron diffraction(SAED) pattern recorded from Au/TiO2@CNTs (the yellowsquare in Fig. 2a) shows a diffraction pattern correspondingto single-crystal anatase along the (101) zone axis (Fig. 2b).HR-TEM images (Fig. 2c and d) show that both the TiO2 andAu NPs are highly crystallized, as evidenced by the well-resolved Au (111) (0.248 nm) and TiO2 (101) (0.365 nm) crys-talline lattices. A close contact between them can facilitatethe electron transfer from the excited Au NPs or TiO2 toCNTs.25,32

The functional groups and crystal structure of the Au/TiO2@CNT composite were determined by FT-IR and Ramanspectroscopy to confirm the formation of chemical bondsamong Au, TiO2 and CNTs. As shown in Fig. 3a, the main

Environmental Science: NanoPaper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

Environ. Sci.: Nano, 2019, 6, 948–958 | 951This journal is © The Royal Society of Chemistry 2019

infrared features of oxidative CNTs are assigned to the CO(1714 cm−1) and C–OH (1215 cm−1) vibrations as comparedwith pure CNTs.33 The abundance of functional groups suchas CO and C–OH decorated on the CNTs is more favorablefor coupling with TiO2 NPs. The broad peaks at 500–900 cm−1

are ascribed to the Ti–O stretching vibration of TiO2 deco-rated on CNTs. Typically, the decreased peak intensity of

CO and C–OH indicates that the functional CNTs areclosely coupled with Au/TiO2 NPs due to possible chemicalbonds between CNTs and TiO2. Fig. 3b shows the Ramanspectra of TiO2, Au/TiO2 and Au/TiO2@CNTs. Au/TiO2@CNTsshows several characteristic bands at 155, 394, 509 and 628cm−1, which correspond to the Eg(1), B1gIJ1), A1g + B1gIJ2), andEgIJ2) modes of anatase TiO2, respectively.34,35 In particular,the observed EgIJ1), B1gIJ1), A1g + B1gIJ2), and EgIJ2) bands of Au/TiO2@CNTs are apparently shifted in comparison with theEgIJ1) (146 cm−1), B1gIJ1) (398 cm−1), A1g + B1gIJ2) (517 cm−1), andEgIJ2) (642 cm−1) bands of TiO2 and Au/TiO2 NPs, which are as-cribed to the bond distortion due to the possible chemicalinterface formed between Au/TiO2 NPs and CNTs.15 These re-sults are closely consistent with the morphology and FT-IRanalysis.

To investigate the crystal structure of the as-synthesizedphotocatalysts, the X-ray diffraction spectra of TiO2 and Au/TiO2@CNTs were carefully analysed (Fig. S2†). The anatasephase of TiO2 (PDF# 21-1272) is observed as the main crystalstructure in the Au/TiO2@CNT composite. The characteristicpeaks observed at 2θ = 38.2 and 44.3° are assigned to thecrystal faces (111) and (200) of Au NPs (PDF# 65-2870), re-spectively, which are consistent with the HR-TEM results.However, no obvious peaks of CNTs are observed from thespectra, which is possibly ascribed to the characteristic peakat 2θ = 26.0° of CNTs being covered by the main peak of ana-tase TiO2 at 2θ = 25.3°, although the highly crystallized struc-ture is beneficial for developing photocatalysts with highphotocatalytic activity.18

Furthermore, XPS was performed to confirm the composi-tion of the Au/TiO2@CNT composite. Fig. S3† shows the char-acteristic peaks of Au4f, Ti2p, O1s and C1s from the full XPSspectra of Au/TiO2@CNTs, further providing evidence for theformation of the Au/TiO2@CNT ternary composite. It can beseen from Fig. S4a† that the main sharp peaks of Ti 2p1/2 lo-cated at 465.2 eV and Ti 2p3/2 located at 459.4 eV wereassigned to Ti4+ in TiO2. The small peaks at 463.4 eV for Ti2p1/2 and 460.7 eV for Ti 2p3/2 clearly indicate the presenceof Ti3+.36,37 From the deconvolution plots (Fig. S4b†), the Oelement is mainly in the form of lattice oxygen in anataseTiO2 (530.7 eV), Ti–OH or OC–O (532.2 eV), C–O species ofCNTs (533.1 eV) and O–H of adsorbed H2O (534.5 eV), respec-tively.38,39 The peak of C1s is divided into four peaks, 284.6,285.9, 287.6 and 289.0 eV (Fig. S4c†), which are ascribed to

Fig. 1 SEM and TEM images of Au/TiO2@CNTs synthesized via amicro/nano-bubble method at different reaction temperatures (a1 anda2, 55 °C; b1 and b2, 70 °C; c1 and c2, 90 °C).

Fig. 2 TEM images of Au/TiO2@CNTs. a, TEM image of Au/TiO2

nanoparticles grown on CNTs; b, SAED pattern of Au/TiO2@CNTs; cand d, high-resolution TEM images of the Au/TiO2@CNT photocatalyst.

Fig. 3 Structure analysis of Au/TiO2@CNTs: FT-IR spectra of CNTs,oxidative CNTs and Au/TiO2@CNTs (a); Raman spectra of TiO2, Au/TiO2

and Au/TiO2@CNT composite (b).

Environmental Science: Nano Paper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

952 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

the sp2-hybridized carbon of CNTs, defect-containing sp2-hybridized carbon, C–O species and OC–O of CNTs, respec-tively.27,38,40 As shown in Fig. S4d,† the characteristic peaksof Au are observed at 84.3 eV (Au 4f7/2) and 87.8 eV (Au 4f5/2). The lower peak intensity of Au 4f is mainly ascribed to theexcessively low concentration.

3.2 CCD-RSM model establishment and analysis

To obtain a stable structure and high photocatalytic activity,the RSM was applied to optimize the experimental factorsand regulate the structure influencing the degradation andmineralization performance of the Au/TiO2@CNT composite.All experimental data in CCD for the degradation and miner-alization of gaseous styrene are listed in Table S2.† Based onthe experimental data, a semi-empirical expression with aquadratic polynomial model is obtained using the Design-Expert software and expressed as follows (equations in termsof coded factors):

YDeg = 23.28 + 9.83A + 0.51B + 0.47C − 4.00D − 4.56AB+ 3.62AC − 2.76AD + 3.96BC + 0.12BD + 2.64CD+ 1.02A2 − 0.69B2 − 0.12C2 + 2.99D2 (1)

YMiner = 17.56 + 5.60A − 4.41B − 1.55C − 5.75D − 5.36AB+ 1.33AC − 4.66AD + 6.16BC + 2.38BD + 1.19CD+ 2.19A2 + 4.08B2 − 5.34C2 + 1.82D2 (2)

where YDeg and YMiner are the response variable of degrada-tion and mineralization efficiency of styrene, respectively.

The given eqn (1) is used to predict the photocatalytic deg-radation efficiency of gaseous styrene catalyzed by Au/TiO2@CNTs. From Table S2,† it is found that the predicteddegradation efficiencies of gaseous styrene matched with theexperimental values very well. This is also observed from theplot of the predicted values against the experimental values(Fig. S5a†), with a high correlation coefficient (R2 = 0.9584).From the variance (ANOVA) analysis (Table S3†), the adjustedR2 = 0.9196 is very close to the corresponding R2 value,suggesting good consistency between the experimental andpredicted values of the degradation efficiencies of gaseousstyrene catalyzed by Au/TiO2@CNTs.

The mineralization performance of Au/TiO2@CNTs wasalso considered (eqn (2) and Table S2†), and a good correla-tion between experimental and predicted values was also ob-served with a high correlation coefficient (R2 = 0.9513) (Fig.S5b†), suggesting that the statistical model is appropriate toexplain the relationship of mineralization efficiencies andstructural performance. From Table S4,† the adjusted R2 =0.9092 based on the experimental results further confirmsthe adaptability of this model. Besides, the data residualsaccording to the analysis of the experimental and predictedvalues are given to evaluate the adequacy of the model (Fig.S6†). All data residuals showed obvious linearity with the deg-radation or mineralization efficiency, and there was no severenon-normality. Meanwhile, the scatter diagram of data resid-

uals against predicted values also demonstrates that the re-siduals appear to be a random scatter, suggesting that themodel is adequate to describe the correlation of the re-sponses and the four factors for the synthesis of the Au/TiO2@CNT composite.

The significance and adequacy can be tested based on var-iance analysis of the quadratic model. From the ANOVA re-sults (Tables S3 and S4†), the total variation is subdividedinto two components: variation related to the model and theexperimental error, to decide whether the variation from themodel is significant or not when compared with the residualerror.41–43 This comparison is performed using the F-value,which is the ratio of the mean square of the model to the re-sidual error. If the model is adequate, the F-value should begreater than the tabulated value of the F-distribution for acertain number of degrees of freedom in the model at a levelof significance. In this study, the F-values are obtained as24.69 and 21.75 for the degradation and mineralization effi-ciencies, respectively. These are clearly greater than the tabu-lated F-values (2.43 at 95% significance), indicating the ade-quacy of the model.

The Pareto analysis gives more significant information tointerpret the results, and the percentage effect of each vari-able on the response can be calculated according to the fol-lowing equation (eqn (3)):17,43

Pii

i

aa

i

2 2 100 0 (3)

where Pi represents the percentage of each variable and airepresents statistically significant coefficients in eqn (1) and2, respectively.

As shown in Fig. 4a, the reaction temperature (A, 51.46%)and amount of HAuCl4 (D, 8.52%) for Au/TiO2@CNT synthe-sis are the main contributors to its photocatalytic degrada-tion of gaseous styrene. As such, the reaction temperature (A,13.24%), mass fraction of CNTs (B, 8.21%) and HAuCl4amount (D, 13.96%) play important roles in the mineraliza-tion of gaseous styrene (Fig. 4b). Because the formation of acontact structure between Au/TiO2 and CNTs is closely relatedto the reaction temperature, it determines the interfacialcharge transfer and formation rate of radicals.24,25 Mean-while, the amount of HAuCl4 determines the amount of Au

Fig. 4 Pareto graphic analyses for the degradation (a) andmineralization (b) efficiencies of gaseous styrene catalyzed by the Au/TiO2@CNT composite.

Environmental Science: NanoPaper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

Environ. Sci.: Nano, 2019, 6, 948–958 | 953This journal is © The Royal Society of Chemistry 2019

NPs supported on TiO2, which is associated with the utiliza-tion of photons. Hence, the parameters of reaction tempera-ture and the amount of HAuCl4 are very beneficial for en-hancing the photocatalytic degradation of gaseous styrene.Note that the mineralization efficiency is also associated withthe mass fraction of CNTs aside from the above two impor-tant parameters because the CNT amount can also enhancethe specific surface area of the composites and promote theadsorption of styrene.

To reveal the interaction effects of the synthesized variatesof the Au/TiO2@CNT composite on the degradation and min-eralization efficiencies, three-dimensional response surfaceplots were constructed with the statistical software and areshown in Fig. 5. Fig. 5a1 shows the effect of the reaction tem-perature and mass fraction of CNTs on the degradation effi-ciency of styrene (amount of HAuCl4, 0.018 mmol; reactiontime, 12 h). Note that when the reaction temperature in-creases from 25 °C to 85 °C (mass fraction of CNTs, 26%),the degradation efficiency rapidly increases from 7.8% to47.0%, suggesting the important role of reaction temperaturein enhancing the degradation efficiency. Possible reasons canbe found from the characteristic results in Fig. 1–3b and 6a.The reaction temperature can significantly promote the for-mation of a contact interfacial structure among Au NPs, TiO2

NPs and CNTs, and the formed contact structure is very con-ducive to the separation of photogenerated electron/holepairs (Fig. 6a),24,25 and enhancement of the degradation effi-ciency of styrene. This regularity is also suitable for the varia-tion of mineralization efficiency. From Fig. 5b1, the minerali-zation efficiency of styrene increases from 15.1% to 37.4%when the reaction temperature rises from 25 °C to 85 °C(mass fraction of CNTs, 26%). These results showed that ahigher reaction temperature was not only conducive to sty-rene degradation but also its deep oxidation process.

In the case of the mass fraction of CNTs, the degradationefficiency slightly increases from 19.6% to 23.1% with the in-crease of mass fraction of CNTs from 4% to 37%. Meanwhile,the mineralization efficiency of styrene firstly decreased from42.6% to 16.2%, and then gradually increased to 17.2%.These results indicate that the high mass fraction of CNTs inAu/TiO2@CNTs can restrain the deep oxidation of gaseous

styrene. This is mainly due to the increase of the amount ofCNTs improving the adsorption capacity of styrene in a con-tinuous reaction system, which surely can facilitate the degra-dation of styrene.17,18 The fluorescence analysis also demon-strates that the increasing mass fraction of CNTs effectivelypromotes the separation of photogenerated electron/holepairs and increases the formation rate of radicals (Fig. 6c).However, with the accumulation of large amounts of interme-diates on the surface of the photocatalyst, the deep-oxidationprocess is possibly hindered due to ineffective mass transferand formation of interfacial radicals.44–46

Fig. 5a2 shows the effect of the reaction temperature andamount of HAuCl4 (mass fraction of CNTs, 26%; reactiontime, 12 h) on the degradation efficiency. The degradation ef-ficiency is found to slowly decrease with increasing amountof HAuCl4 from 0.004 to 0.032 mmol. Similar to the degrada-tion efficiency, the mineralization efficiency also slightly de-creases with the increase of the amount of HAuCl4. This isbecause the excessive load capacity of Au NPs can increasethe recombination of photogenerated electron/hole pairs asdescribed in Fig. 6b, inhibiting the degradation of styreneand the subsequent deep-oxidation process.43,47,48

As for the effect of reaction temperature and reaction time(mass fraction of CNTs, 26%; amount of HAuCl4, 0.018mmol) (Fig. 5a3), the degradation efficiency of styrene in-creases rapidly with the increase of the reaction temperature,but an increase of only 1.9% in degradation efficiency is ob-served when the reaction time was increased from 4 to 20 h.Obviously, the reaction time has a lower contribution to thedegradation efficiency. However, the relationship of the reac-tion time and mass fraction of CNTs has a significant effecton the mineralization efficiency as shown in Fig. 5b3 (reac-tion temperature, 55 °C; amount of HAuCl4, 0.018 mmol). A

Fig. 5 The response surface plots of the degradation (a1–a3) andmineralization (b1–b3) efficiencies of gaseous styrene catalyzed by Au/TiO2@CNT photocatalysts.

Fig. 6 Fluorescence spectra of Au/TiO2@CNT composites prepared atdifferent (a) reaction temperatures (mass fraction of CNTs, 26%;reaction time, 12 h; amount of HAuCl4, 0.018 mmol); (b) amounts ofHAuCl4 (mass fraction of CNTs, 26%; reaction time, 12 h; reactiontemperature, 55 °C); (c) mass fractions of CNTs (amount of HAuCl4,0.018 mmol; reaction time, 12 h; reaction temperature, 55 °C); (d)reaction times (amount of HAuCl4, 0.018 mmol; mass fraction of CNTs,26%; reaction temperature, 55 °C).

Environmental Science: Nano Paper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

954 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

notable decrease in mineralization efficiency is obtainedfrom 42% to 17.3% with the increase of the mass fractionfrom 4.0% to 37.0%, indicating the adverse impact on themineralization efficiency. The effects of reaction time showthat the mineralization efficiency gradually increased to17.7% when the reaction time increased from 4 to 11 h. How-ever, further increasing the reaction time would reduce themineralization efficiency. These results may be attributed toan increase in particle size of TiO2 over CNTs with the in-crease of reaction time, which can contribute to the recombi-nation of photogenerated electron/hole pairs (Fig. 6d).18

To sum up, the reaction temperature was found to be thekey parameter in adjusting the photocatalytic degradationand mineralization efficiencies of gaseous styrene on the Au/TiO2@CNT composite photocatalyst, based on the experimen-tal results. Although the CNTs are conducive to the enhance-ment of the photocatalytic degradation efficiency, excessivelyhigh proportions of CNTs in the composites lead to the accu-mulation of intermediates due to the excessive adsorption ofstyrene and will restrain the deep-oxidation process of sty-rene. The addition of Au NPs can effectively improve thephotocatalytic activity of Au/TiO2@CNTs, but the excessiveload capacity of Au NPs can increase the recombination ofphotogenerated electron/hole pairs. Herein, a verification ex-periment was carried out under the optimized conditionsbased on the numerical optimization results in the DesignExpert software. The highest photocatalytic degradation effi-ciency of 69.2% and mineralization efficiency of 52.5% forgaseous styrene are obtained with the optimum values of thevariables of 85 °C, 4.04%, 6.58 h and 0.004 mmol for the re-action temperature, mass fraction of CNTs, reaction time andamount of HAuCl4, respectively. For comparison, the photo-catalytic performance of this optimum sample was also mea-sured under visible-light (VL) irradiation. The results indicatethat the VL sample shows lower photocatalytic degradation(31.2%) and mineralization efficiencies (12.8%) comparedwith the sample under simulated solar light, suggesting thatthe excitation of TiO2 by UV light was also important to en-hance the photocatalytic performance of this compositephotocatalyst.

3.3 Enhanced photocatalytic mechanism of Au/TiO2@CNTs

According to the results of CCD-RSM analysis, Au/TiO2@CNTs possesses higher photocatalytic activity undersimulated solar light. For comparison, the photocatalytic per-formance curves of pure TiO2, TiO2/CNT and Au/TiO2 photo-catalysts are also given in Fig. S7.† As seen from Fig. S7 andTable S5,† the photocatalytic activity and stability of the Au/TiO2@CNT photocatalyst were much higher than those ofTiO2, TiO2/CNT and Au/TiO2 photocatalysts. The additionalXRD spectra shown in Fig. S8† reveal that there are nochanges in the crystal structure of Au/TiO2@CNTs after thephotocatalysis test, confirming the stability of the photo-catalyst. However, the essential reasons for the enhancedphotocatalytic mechanism of the Au/TiO2@CNT composite

are still not clear. To give a reasonable explanation for the en-hanced photocatalytic activity of Au/TiO2@CNTs, further evi-dence is given to further reveal the synergetic mechanism ofCNTs coupled with Au NPs for enhancing the photocatalyticperformance of TiO2 NPs.

To obtain information about the oxygen vacancies, EPRspectra were measured (Fig. S9†). For Au/TiO2, the sharp peakwith the corresponding gyy = 2.004 is assigned to oxygen vacan-cies49 and the highest intensity means that Au/TiO2 has thehighest level of oxygen vacancies compared with the other sam-ples. The EPR signals at gzz = 2.104 and gxx = 1.938 can be at-tributed to O2

− and Ti3+.50 From the EPR spectra, the character-istic peaks of CNTs in Au/TiO2@CNTs and TiO2/CNTs areobserved and exhibit three components including a wide com-ponent, narrow component and broad component comparedwith TiO2 and Au/TiO2. Among them, the wide component wasrelated to both the external magnetic field and the molecularfield; the narrow peak is assigned to either amorphous local-ized defects, such as vacancies, or dangling bonds; the broadpeak is assigned to the delocalized electrons over theconducting domains of carbon nanotubes.51 From the spectralanalysis and the photocatalytic experimental results, theobtained numbers of oxygen vacancies of TiO2 and Au/TiO2 arehigher than that of Au/TiO2@CNTs, but their photocatalyticperformances are lower than that of Au/TiO2@CNTs. That is,the enhanced photocatalytic performance of Au/TiO2@CNTs isrelated to not only the properties of Au/TiO2, but also the elec-tric conductivity and oxygen active sites of CNTs.20,21

It is known that the adsorption and mass transfer of VOCsonto the photocatalyst are not only closely related to the es-sential attributes of the specific surface area (SBET) but alsothe pore size of the photocatalyst. As shown in Fig. 7, the spe-cific surface area of Au/TiO2@CNTs was determined to be104.4 m2 g−1, which is higher than that of TiO2 (61.4 m2 g−1),TiO2/CNTs (78.7 m2 g−1) and Au/TiO2 (90.0 m2 g−1). Thehigher SBET of the Au/TiO2@CNT composite would allow it tohave more adsorption sites and active sites, which can obvi-ously promote the adsorption and degradation of gaseous sty-rene.52 Besides, Au/TiO2@CNTs possessed a larger pore sizeof 34.7 Å, which made it exhibit lower resistance of masstransfer during the degradation of gaseous styrene, resultingin a rapid mineralization of the intermediates from gaseousstyrene into CO2 and H2O.

Fig. 7 Nitrogen adsorption–desorption isotherms (a) and pore sizedistribution (b) of TiO2, TiO2/CNT, Au/TiO2 and Au/TiO2@CNTcomposite. The adsorption and desorption portions of the isotherms in(a) are marked with closed and open markers, respectively.

Environmental Science: NanoPaper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

Environ. Sci.: Nano, 2019, 6, 948–958 | 955This journal is © The Royal Society of Chemistry 2019

Optical characterization techniques such as UV-vis andfluorescence spectroscopy also provide information on opti-cal adsorption and electron transfer, which can indirectly re-flect the photocatalytic performance of photocatalysts.10 Asshown in Fig. 8a, it is noted that the absorption intensity ofAu/TiO2@CNTs is higher than that of TiO2 and Au/TiO2 dueto the excellent electrical conductivity of CNTs. For the Au/TiO2 photocatalyst, there is a large broad peak ranging from500 to 700 nm, suggesting the excellent visible-light responseof TiO2 decorated with Au NPs. In addition, the fluorescencespectrum also shows that the PL intensity of the Au/TiO2@CNT composite is much lower than that of TiO2 andAu/TiO2 (Fig. 8b), indicating that the Au/TiO2@CNT compos-ite possessed a higher separation efficiency of photo-generated electron/hole pairs.53 According to these results,the co-modification of CNTs and Au NPs will surely increasethe utilization of photons and depress the recombination ofphotogenerated electron/hole pairs of the TiO2 photocatalyst.

To deeply understand the migration and recombination ofphotoinduced electron/hole pairs, the photoelectrochemicalproperties of the photocatalysts were also carefully discussed.From Fig. 8c, the Au/TiO2@CNT photocatalyst shows a higherphotocurrent intensity and still possesses high photocurrentresponse after five cycles compared with pure TiO2 and Au/TiO2, indicating higher visible-light response and efficientelectron transfer. In addition, the EIS analysis gives the ap-propriate evidence to further explain the separation efficiencyof photogenerated electron/hole pairs. From the Nyquistcurve (Fig. 8d), the Au/TiO2@CNT composite shows a smallersemicircular diameter compared with TiO2 and Au/TiO2,where a smaller arc radius indicates a higher separation effi-ciency in charge transfer,54,55 suggesting that the Au/TiO2@CNT composite possessed higher separation and trans-fer efficiency of photo-generated electron/hole pairs. Throughthe analysis of the experimental results, it is speculated thatthe synergetic effect of Au NPs and CNTs can greatly promotethe separation and facilitate the transfer of photogeneratedelectron/hole pairs,24,25 which is very beneficial for the forma-

tion rate of interfacial radicals which then influences thephotocatalytic performance of the Au/TiO2@CNT composite.

It is well known that enhanced mobility is another crucialparameter for enhancing the catalytic activity due to the pro-motion of O2 adsorption and activation.56,57 Herein, tempera-ture programmed desorption of oxygen (O2-TPD) was alsoconducted to confirm the ability of oxygen activation over dif-ferent photocatalysts (Fig. S10†). Two O2 desorption peaksare observed over the Au/TiO2@CNT composite. The peak at75 °C is attributed to the chemisorption of O2 onto the sur-face, and another peak at 160 °C could be attributed to thedesorption of surface capping oxygen.58 As compared withpure TiO2 and Au/TiO2, enhanced desorption of O2 is ob-served for Au/TiO2@CNTs, indicating that more O2 can ad-sorb onto this ternary photocatalyst at 40 °C. Typically, thiseffect is obvious for Au/TiO2@CNTs, which is consistent withthe fact that Au/TiO2@CNTs shows the highest photocatalyticperformance under the same conditions.

The reactive oxidizing species (ROS) associated with theseparation of photogenerated electron/hole pairs are the keyfactors for the decomposition of gaseous styrene. Here, theEPR spectra were recorded by using DMPO as the trapping re-agent (see the ESI†). As shown in Fig. 9, strong EPR signalsof DMPO–˙OH adducts with a 1 : 2 : 2 : 1 intensity and DMPO–O2˙

− adducts with six characteristic peaks are observed in theAu/TiO2@CNT photocatalyst system under stimulated solarlight irradiation, indicating that ˙OH and O2˙

− are the impor-tant ROS in this system.59,60 Moreover, the EPR intensity forDMPO–˙OH and DMPO–O2˙

− of Au/TiO2@CNTs is muchhigher than that of TiO2 and Au/TiO2, demonstrating the syn-ergetic effects of Au NPs and CNTs on enhancing the photo-catalytic activity of TiO2.

Based on the above analytical results, a possible enhancedphotocatalytic mechanism of the Au/TiO2@CNT photocatalystunder solar light irradiation is proposed in Fig. 10. Generally,electrons and holes are produced by the excitation of TiO2 bya small amount of UV from solar light, and rapidly accumu-late on the conduction band (CB) and valence band (VB) ofTiO2, respectively.

61,62 Owing to the oxidability of holes, theycan oxidize H2O into ˙OH in the photocatalytic reaction. Inparticular, the CB potential of TiO2 is higher than those ofCNTs and Au NPs. Hence, there are two migration and trans-formation patterns for the photogenerated electrons: (1) thephotogenerated electrons transfer from TiO2 to the carbonchains of CNTs, and subsequently react with the adsorbed O2

Fig. 8 UV-vis (a), fluorescence (b), photocurrent response (c) andelectrochemical impedance (d) spectra of as-synthesized composites.

Fig. 9 EPR spectra of TiO2, Au/TiO2 and Au/TiO2@CNTs under solarlight irradiation: (a) DMPO–˙OH and (b) DMPO–O2˙

−.

Environmental Science: Nano Paper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

956 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

to form O2˙−;23,24 due to local surface plasmon resonance

(LSPR); the produced electrons on the surface of Au NPscan also transfer to the carbon chains of CNTs or react withthe adsorbed O2 to form O2˙

−.8,10,14,15 It is well known thatO2˙

− is an important ROS,63 which can also participate inthe photocatalytic degradation of styrene. Meanwhile, O2˙

−

can be transformed into ˙OH through a series of reactions,promoting styrene degradation. Due to the combined effectof h+, ˙OH and O2˙

−, the gaseous styrene could be subse-quently mineralized into CO2 and H2O.

64 This is becausefor a single TiO2 photocatalysis system, only ˙OH is themain ROS.61,65 Unlike the TiO2 photocatalysis system in thiswork, more charge carriers were transformed into oxidativeradicals on Au/TiO2@CNTs which participate in the photo-catalytic degradation of gaseous styrene. As described in thevisible-light photocatalytic tests, the Au/TiO2@CNT photo-catalyst still has photocatalytic activity under visible-light ir-radiation, which is mainly attributed to the contribution ofLSPR of Au NPs and electron transfer of CNTs for the for-mation of O2˙

− to degrade styrene. The Au/TiO2@CNT sys-tem with high SBET provides not only more O2 adsorptionsites for efficiently enhancing the formation rate of oxida-tive radicals, but also more adsorption and active sites dueto the addition of CNTs. In a word, the high decompositionof gaseous styrene is attributed to the intrinsically strongphoto-oxidation ability of Au/TiO2 and high adsorption ca-pacity and electron transfer of CNTs, resulting in efficientphotocatalytic degradation and mineralization of gaseousstyrene. But the formation of firm chemical interfaces be-tween Au/TiO2 and CNTs is confirmed to be very necessaryduring the synthesis of the Au/TiO2@CNT composite, facili-tating the formation of this ternary photocatalyst with highperformance and stability.

Conclusions

An Au/TiO2@CNT ternary composite was firstly synthesizedby a self-assembly method with micro/nano-bubbles as a softtemplate. Based on the CCD-RSM results, the reaction tem-perature was found to be the key factor for the enhancementof the degradation and mineralization efficiencies of gaseousstyrene. In particular, the enhanced mechanism of the photo-

catalytic performance of the Au/TiO2@CNT composite wascarefully discussed. It was noted that the photocatalytic per-formance and durability of the Au/TiO2@CNT composite areby far higher than those of TiO2, TiO2/CNT and Au/TiO2

photocatalysts. This was mainly because the synergetic effectsof Au NPs and CNTs could not only increase the migrationand transformation rates of photogenerated electron/holepairs, but also provide more adsorption and active sites forgaseous styrene. This work may provide deep insight into de-signing multiple-element photocatalysts with high efficiencyand stability for removal of typical VOCs, and typically revealsthe significant influence of the structure and composition onthe photocatalytic performance and durability of the as-synthesized photocatalysts.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural ScienceFoundation of China (21425015 and 41373102), the Local In-novative and Research Teams Project of Guangdong PearlRiver Talents Program (2017BT01Z032), the Innovation TeamProject of Guangdong Provincial Department of Education(2017KCXTD012), the China Postdoctoral Science Foundation(2017M622639 and 2018T110851), and the Guangdong SpecialBranch Plan of Science and Technology for Innovation lead-ing scientists (2016TX03Z094 and 2016TQ03Z291).

Notes and references

1 O. Debono, V. Hequet, L. Le Coq, N. Locoge and F.Thevenet, Appl. Catal., B, 2017, 218, 359–369.

2 T. N. Obee and R. T. Brown, Environ. Sci. Technol., 1995, 29,1223–1231.

3 J. Y. Chen, X. Nie, H. X. Shi, G. Y. Li and T. C. An, Chem.Eng. J., 2013, 228, 834–842.

4 M. C. Wen, G. Y. Li, H. L. Liu, J. Y. Chen, T. C. An and H.Yamashita, Environ. Sci.: Nano, DOI: 10.1039/c8en01167b.

5 Q. Guo, C. Y. Zhou, Z. B. Ma, Z. F. Ren, H. J. Fan and X. M.Yang, Chem. Soc. Rev., 2016, 45, 3701–3730.

6 Z. F. Bian, T. Tachikawa, P. Zhang, M. Fujitsuka and T.Majima, J. Am. Chem. Soc., 2014, 136, 458–465.

7 P. F. Xu, X. F. Shen, L. Luo, Z. Shi, Z. X. Liu, Z. G. Chen,M. F. Zhu and L. S. Zhang, Environ. Sci.: Nano, 2018, 5,327–337.

8 W. J. Wang, G. Y. Li, D. H. Xia, T. C. An, H. J. Zhao and P. K.Wong, Environ. Sci.: Nano, 2017, 4, 782–799.

9 H. Wang, T. T. You, W. W. Shi, J. H. Li and L. Guo, J. Phys.Chem. C, 2012, 116, 6490–6494.

10 R. Amrollahi, M. S. Hamdy and G. Mul, J. Catal., 2014, 319,194–199.

11 J. H. Yang, Y. Z. Guo, R. B. Jiang, F. Qin, H. Zhang, W. Z. Lu,J. F. Wang and J. C. Yu, J. Am. Chem. Soc., 2018, 140,8497–8508.

Fig. 10 Possible enhanced photocatalytic mechanism associated withphotogenerated e−/h+ pairs on the Au/TiO2@CNT composite undersolar light irradiation.

Environmental Science: NanoPaper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

Environ. Sci.: Nano, 2019, 6, 948–958 | 957This journal is © The Royal Society of Chemistry 2019

12 R. B. Jiang, B. X. Li, C. H. Fang and J. F. Wang, Adv. Mater.,2014, 26, 5274–5309.

13 X. Q. Deng, J. L. Liu, X. S. Li, B. Zhu, X. B. Zhu and A. M.Zhu, Catal. Today, 2017, 281, 630–635.

14 S. Ghasemi, S. J. Hashemian, A. A. Alamolhoda, I. Gochevaand S. R. Setayesh, Mater. Res. Bull., 2017, 87, 40–47.

15 R. Fiorenza, M. Bellardita, L. D'Urso, G. Compagnini, L.Palmisano and S. Scire, Catalysts, 2016, 6, 121.

16 K. Demeestere, J. Dewulf and H. Van Langenhove, Crit. Rev.Environ. Sci. Technol., 2007, 37, 489–538.

17 J. Y. Chen, G. Y. Li, Y. Huang, H. M. Zhang, H. J. Zhao andT. C. An, Appl. Catal., B, 2012, 123–124, 69–77.

18 T. C. An, J. Y. Chen, X. Nie, G. Y. Li, H. M. Zhang, X. L. Liuand H. J. Zhao, ACS Appl. Mater. Interfaces, 2012, 4,5988–5996.

19 J. Y. Shi, J. Y. Chen, G. Y. Li, T. C. An and H. Yamashita,Catal. Today, 2017, 281, 621–629.

20 C. Wang, J. Kang, P. Liang, H. Y. Zhang, H. Q. Sun, M. O.Tadéa and S. B. Wang, Environ. Sci.: Nano, 2017, 4,170–179.

21 W. Wang, C. H. Lu, Y. R. Ni and Z. Z. Xu, Appl. Catal., B,2013, 129, 606–613.

22 D. Ma, J. Wu, M. C. Gao, Y. J. Xin and C. Chai, Chem. Eng. J.,2017, 316, 461–470.

23 K. C. Nguyen, M. P. Ngoc and M. Van Nguyen, Mater. Lett.,2016, 165, 247–251.

24 S. G. Wang, Z. R. Du, C. X. Kong, P. F. Li, J. L. Xu, C. X.Wang and J. H. Wang, Nano, 2014, 9, 1450003.

25 H. P. Yang, Z. M. Shi, K. J. Dai, Y. P. Duan and J. M. Wu,Acta Chim. Sin., 2011, 69, 536–542.

26 Y. X. Zhang, B. Gao, G. Li Puma, A. K. Ray and H. C. Zeng,Sci. Adv. Mater., 2010, 2, 503–513.

27 J. Li, S. B. Tang, L. Lu and H. C. Zeng, J. Am. Chem. Soc.,2007, 129, 9401–9409.

28 J. X. Sun, G. Chen, J. Z. Wu, H. J. Dong and G. H. Xiong,Appl. Catal., B, 2013, 132–133, 304–314.

29 L. Zhao, F. Q. Tao, Z. Quan, X. L. Zhou, Y. H. Yuan and J. C.Hu, Mater. Lett., 2012, 68, 28–31.

30 X. K. Cheng, Z. L. Huang, J. Q. Li, Y. Liu, C. L. Chen, R. A.Chi and Y. H. Hu, Cryst. Growth Des., 2010, 10, 1180–1188.

31 Y. S. Han, M. Fuji, D. Shchukin, H. Möhalwd and M.Takahashi, Cryst. Growth Des., 2009, 9, 3771–3775.

32 R. Malik, V. K. Tomer, N. Joshi, T. Dankwort, L. W. Linand L. Kienle, ACS Appl. Mater. Interfaces, 2018, 10,34087–34097.

33 X. Feng, C. Matranga, R. Vidic and E. Borguet, J. Phys. Chem.B, 2004, 108, 19949–19954.

34 Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Am. Chem. Soc.,2012, 134, 6575–6578.

35 J. X. Low, S. Q. Qiu, D. F. Xu, C. J. Jiang and B. Cheng, Appl.Surf. Sci., 2018, 434, 423–432.

36 B. Bharti, S. Kumar, H. N. Lee and R. Kumar, Sci. Rep.,2016, 6, 32355.

37 X. D. Jiang, Y. P. Zhang, J. Jiang, Y. S. Rong, Y. C. Wang,Y. C. Wu and C. X. Pan, J. Phys. Chem. C, 2012, 116,22619–22624.

38 C. H. Wu, C. Y. Kuo and S. T. Chen, Environ. Technol.,2013, 34, 2513–2519.

39 Z. Lian, P. Xu, W. Wang, D. Zhang, S. Xiao, X. Li and G. Li,ACS Appl. Mater. Interfaces, 2015, 7, 4533–4540.

40 A. R. Khataee, M. Fathinia, S. Aber and M. Zarei, J. Hazard.Mater., 2010, 181, 886–897.

41 Y. J. You, Z. Q. Jiang, Y. X. Zhang, C. H. Ge and R. R. Li, Res.Chem. Intermed., 2015, 41, 3079–3087.

42 M. Zarei, A. Niaei, D. Salari and A. Khataee, J. Hazard.Mater., 2010, 173, 544–551.

43 D. W. Ding, K. Liu, S. N. He, C. B. Gao and Y. D. Yin, NanoLett., 2014, 14, 6731–6736.

44 O. Carp, C. L. Huisman and A. Reller, Prog. Solid StateChem., 2004, 32, 33–177.

45 C. Lin, Y. Song, L. X. Cao and S. W. Chen, Nanoscale,2013, 5, 4986–4992.

46 R. Fiorenza, M. Bellardita, L. D'Urso, G. Compagnini, L.Palmisano and S. Scirè, Catalysts, 2016, 6, 121.

47 L. M. Guo, K. Liang, K. Marcus, Z. Li, L. Zhou, P. D. Mani,H. Chen, C. Shen, Y. J. Dong, L. Zhai, K. R. Coffey, N.Orlovskaya, Y. H. Sohn and Y. Yang, ACS Appl. Mater.Interfaces, 2016, 8, 34970–34977.

48 Y. H. Xiao, X. H. Song, Z. Liu, R. P. Li, X. R. Zhao and Y. P.Huang, J. Ind. Eng. Chem., 2017, 45, 248–256.

49 X. M. Yu, B. Kim and Y. K. Kim, ACS Catal., 2013, 3,2479–2486.

50 S. M. Wu, X. L. Liu, X. L. Lian, G. Tian, C. Janiak, Y. X.Zhang, Y. Lu, H. Z. Yu, J. Hu, H. Wei, H. Zhao, G. G. Chang,G. V. Tendeloo, L. Y. Wang, X. Y. Yang and B. L. Su, Adv.Mater., 2018, 30, 1802173.

51 A. R. Adhikari, H. Bakhru, P. M. Ajayan, R. Benson and M.Chipara, Nucl. Instrum. Methods Phys. Res., Sect. B,2007, 265, 347–351.

52 B. Gao, G. Z. Chen and G. L. Puma, Appl. Catal., B, 2009, 89,503–509.

53 F. Raziq, L. Q. Sun, Y. Y. Wang, X. L. Zhang, M. Humayun,S. Ali, L. L. Bai, Y. Qu, H. T. Yu and L. Q. Jing, Adv. EnergyMater., 2018, 8, 1701580.

54 Y. Wang, J. G. Yu, W. Xiao and Q. Li, J. Mater. Chem. A,2014, 2, 3847–3855.

55 H. M. Shi, S. Zhang, X. P. Zhu, Y. Liu, T. Wang, T. Jiang,G. H. Zhang and H. G. Duan, ACS Appl. Mater. Interfaces,2017, 9, 36907–36916.

56 H. M. Qin, X. S. Qian, T. Meng, Y. Lin and Z. Ma, Catalysts,2015, 5, 606–633.

57 Y. N. Chen, D. S. Liu, L. J. Yang, M. Meng, J. Zhang, L. R.Zheng, S. Q. Chu and T. D. Hu, Chem. Eng. J., 2013, 234,88–98.

58 L. D. Li, C. Z. Xu, M. Q. Zheng and X. H. Chen, RanliaoHuaxue Xuebao, 2015, 43, 990–997.

59 C. Q. Li, Z. M. Sun, A. K. Song, X. B. Dong, S. L. Zheng andD. D. Dionysiou, Appl. Catal., B, 2018, 236, 76–87.

60 Y. Hou, X. Y. Li, Q. D. Zhao, G. H. Chen and C. L. Raston,Environ. Sci. Technol., 2012, 46, 4042–4050.

61 P. Zhou, J. G. Yu and M. Jaroniec, Adv. Mater., 2014, 26,4920–4935.

Environmental Science: Nano Paper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online

958 | Environ. Sci.: Nano, 2019, 6, 948–958 This journal is © The Royal Society of Chemistry 2019

62 Z. W. Zhao, W. D. Zhang, X. S. Lv, Y. J. Sun, F. Dong andY. X. Zhang, Environ. Sci.: Nano, 2016, 3, 1306–1317.

63 W. P. Zhang, G. Y. Li, W. J. Wang, Y. X. Qin, T. C. An, X. Y.Xiao and W. Y. Choi, Appl. Catal., B, 2018, 232, 11–18.

64 E. Grabowska, M. Marchelek, T. Klimczuk, W. Lisowski andA. Zaleska-Medynska, J. Catal., 2017, 350, 159–173.

65 R. X. Cai, Y. Kubota and A. Fujishima, J. Catal., 2003, 219,214–218.

Environmental Science: NanoPaper

Publ

ishe

d on

28

Janu

ary

2019

. Dow

nloa

ded

by G

uang

dong

Tec

hnol

ogy

Uni

vers

ity L

ibra

ry o

n 3/

18/2

019

6:46

:27

AM

. View Article Online