Qingchi Xua* , Jiaxin Zenga , Haiqing Wanga , Xingyun Lia , Jun Xua , Jianyang Wua , Guangcan Xiaob, Fang-Xing Xiaob*, Xiangyang Liuca

a Department of Physics, Research Institute for Biomimetics and Soft Matter, Fujian Provincial Key  Laboratory for Soft Functional Materials, Xiamen University, Xiamen, 361005, P. R. China.

b Instrumental Measurement and Analysis Center, Fuzhou University, Fuzhou, People's Republic of  China.

c Department of Physics, National University of Singapore, 2 Science Drive 3, Singapore, 117542,  Singapore. 

Email: xuqingchi@xmu.edu.cn fangxing2010@gmail.com

Abstract

A facile and efficient ligand-triggered electrostatic self-assembly strategy has been developed to  fabricate a series of Au/CdS nanosheets (Ns) (Au-CdS Ns) nanocomposites with varied weight  addition ratios of Au nanoparticles (NPs) by judiciously utilizing the intrinsic surface charge  properties of assembly units, through which uniform dispersion and controllable deposition of Au NPs  on the CdS Ns was achieved. Versatile probe reactions including photocatalytic oxidation of organic  dye pollutant, selective photocatalytic reduction of aromatic nitro compounds and photocatalytic  hydrogen production reactions under visible light irradiation at ambient conditions were used to  systematically evaluate the photoredox performances of as-assembled well-defined Au-CdS Ns  nanocomposites. It was unveiled that photoactivities of Au-CdS Ns nanocomposites is strongly  depend on the weight addition ratio of Au NPs and the addition of excess amount of Au NPs is  detrimental to the separation of photogenerated charge carriers from CdS Ns. Under the optimum  addition amount of Au NPs (1 wt %), it was found that spontaneous assembly of Au NPs on the CdS  Ns remarkably prolonged the lifetime of photogenerated charge carriers from CdS Ns under visible  light irradiation, thus resulting in significantly enhanced photocatalytic redox activities of Au-CdS Ns nanocomposites compared with CdS Ns. The crucial role of Au NPs in the photoredox reactions was  determined to act as efficient electron traps rather than plasmonic sensitizer. Moreover, predominant  active species responsible for the photocatalytic process were unambiguously determined and possible  photocatalytic mechanism was elucidated. It is anticipated that our work could open up a new avenue to rationally prepare various 2D semiconductors-metal nanocomposites by utilizing such a simple and  efficient self-assembly strategy for extensive photocatalytic applications in a myriad of fields.

1. Introduction

Semiconductor-based photocatalysis has been attracting a great deal of interest owing to its universal  applications especially in utilizations of solar energy, environmental remediation and selective organic  transformation to fine chemicals.1-4 Among a large variety of semiconductor materials, CdS, as a  well-known II-VI semiconductor, has attracted particular attention due to its suitable bandgap (2.4 eV) which corresponds well with the solar spectrum, size-dependent electronic and optical properties, and  thus recent years have witnessed their wide-ranging applications in a myriad of fields such as  hydrogen production, environment remediation, and solar cells.5-7 Nonetheless, utilization of CdS  nanomaterials is limited by several issues, e.g., agglomeration is frequently observed in conventional  CdS nanomaterials giving rise to reduced specific surface area. Moreover, high recombination rate of  photogenerated electron-hole pairs and relatively poor photostability of CdS nanomaterials further  restricts their wide-spread applications. In this regard, various strategies have been developed to  surmount these obstacles, such as deposition of noble metal nanoparticles (NPs), integrating with the  second semiconductor components or carbon materials, and forming p-n heterojunctions.8-13

It has been well-established that charge transport in CdS nanomaterials is substantially affected by  the presence of surface defect states which can serve as charge traps leading to a fast recombination of  photogenerated charge carriers.14, 15 To solve this problem, coupling CdS with metal nanocrystals (NCs) (e.g., Au, Ag, Pd, Pt, etc.) have been considered as an effective approach to suppress the  recombination of photogenerated charge carriers by either utilizing surface plasmon resonance (SPR)  effect of metal NCs to enhance light-harvesting efficiency in the visible region or constructing the  Schottky barriers to facilitate the transfer of photogenerated charge carriers.8, 16-23 More importantly,  deposition of metal NCs on CdS could not only provide a heterojunction interface but also passivate CdS surface defects upon contact. Among plentiful metal NCs, Au NCs has been deemed as one of the  most efficient metal NCs with fascinating catalytic activities. 24-26 Thus far, a new class of Au-CdS nanocomposites has been prepared by combining the substantial light-harvesting property of CdS with  the catalytically active Au NCs for photocatalytic applications.27-30 It has been ascertained that  monodispersion of Au NCs on the CdS surface in conjunction with their intimate interfacial  interaction are two key factors affecting the photocatalytic performances of Au-CdS  nanocomposites.15 To this end, various synthetic strategies have been developed to prepare Au-CdS  nanocomposites but most of which are generally confined to conventional methods such as  photo-reduction, deposition-precipitation, spray pyrolysis, sol-gel and electrochemical approaches.8,  29-36 For example, Khon et al.36 reported the synthesis of Au/CdS nanocomposites with tunable  morphology through temperature-controlled reduction of gold-oleate complexes. Kumar et al.23 reported an in-situ synthesis of Au-CdS hybrid nanoparticles using a simple continuous spray  pyrolysis method.33 Most recently, Majeed et al. reported a novel iodide reduction method to prepare  Au/CdS nanocomposites with initial mean Au particle size between 2 nm to 5 nm.8 Despite the  endeavors, rational construction of Au-CdS nanocomposites with controllable deposition amount and  monodispersivity of Au via a facile and green approach still constitutes a long-standing challenge by  virtue of large lattice mismatch between Au and CdS.37

On the other hand, in recent years, 2D transition-metal compounds have received tremendous  attention due to their unique electronic, geometric, and physical-co-chemical properties in comparison  with corresponding bulk materials.38-40 In particular, 2D semiconductors have been widely synthesized  and utilized as photocatalysts for solar energy conversion and environmental remediation.41-48 Inspired  by this and enlightened by the intrinsic structural merits of ultra-thin 2D CdS nanosheets (CdS Ns), it is thus highly desirable to judiciously integrate CdS Ns with Au NCs in a close integration manner which would take full advantage of the synergistic interaction between them giving rise to promising photocatalytic performances

Herein, a facile and efficient electrostatic self-assembly method has been developed to rationally  fabricate well-defined Au-CdS Ns nanocomposites by directly harnessing the intrinsic surface charge  properties of assembly units (i.e., CdS Ns@L-cysteine and Au@4-dimethylaminopyridine), based on  which positively charged Au@4-dimethylaminopyridine (Au@DMAP) NPs were spontaneously and  uniformly deposited on the negatively charged CdS Ns@L-cysteine under pronounced electrostatic  attractive interaction. It was revealed that the thus-assembled Au-CdS Ns nanocomposites demonstrated versatile and significantly enhanced photoredox performances including photocatalytic  oxidation of organic dye pollutant, photocatalytic reduction of aromatic nitro compounds and  photocatalytic hydrogen production under visible light irradiation in comparison with CdS Ns, owing  predominantly to the synergistic interaction between Au and CdS Ns. The Au component in the  nanocomposites played a crucial role as efficient electron traps rather than plasmonic photosensitizer in the photocatalytic reaction, thereby boosting the lifetime of photogenerated electron-hole pairs from  CdS Ns. Moreover, predominant active species responsible for the photocatalytic process were  determined and specific photocatalytic mechanism was elucidated. It is hoped that our work could  open up a new and general synthetic concept to prepare various metal-semiconductor nanomaterials  by such a facial and efficient ligand-triggered self-assembly approach for extensive photocatalytic  applications.

2. Experimental section

2.1 Materials

Cadmium Chloride (CdCl2·2.5H2O), L-cysteine, sodium hydroxide, toluene, sodium sulfate anhydrous,  sodium sulfide nonahydrate, benzoquinone (BQ), tert-butyl alcohol (TBA) and sulfuric acid were  purchased from Xilong Chemical Co., Ltd. Sulfur, N,N-Dimethylformamide, potassium bromide,  4-nitroaniline (4-NA), 2-nitroaniline (2-NA), 4-nitrotoluene (4-NT), 4-nitrophenol (4-NP) and sodium  borohydride were purchased from Sinopharm group Co., Ltd. Diethylenetriamine,  4-dimethylaminopyridine (DMAP), 3-nitroaniline (3-NA), 2-nitrophenol (2-NP), 3-nitrophenol (3-NP) and chlorauric acid were purchased from Aladdin. Tetraoctyl ammonium bromide (TOAB, Xiya  Reagent), 5, 5-diemthyl-1-pyrroline N-oxide (DMPO, Sigma), Deionized water (DI H2O, Millipore,  18.2 MΩ·cm). All reagents were AR grade and used without further purification. .

2.2 Preparation of CdS Ns

In a typical synthetic process,6 0.32 mmol CdCl2∙2.5H2O, 2.0 mmol S powder and 12 mL of  diethylenetriamine (DETA) were mixed and vigorously stirred for 30 min to form a homogenous  suspension. The mixture was transferred to a Teflon-lined autoclave (20 mL) and heated at 80 oC for  48 h. After being cooled down to room temperature naturally, light yellow precipitate was collected by  centrifugation and washed with ethanol/H2O (V : V = 1 : 1) for three times. The obtained light yellow  powder (CdS-DETA) was dried in an oven at 60 oC for 24 h. After that, 20 mg CdS-DETA, 10 mg  L-cysteine, 0.1 mL of DETA and 40 mL of DI H2O were added into a beaker (100 mL) and  ultra-sonicated continuously for 2 h at ambient conditions. The resultant yellow suspension was  centrifuged at 800 rpm for 10 min to remove the large aggregates and the supernatant (top four-fifth of  the centrifuged suspension) was collected to obtain the ultrathin CdS Ns suspension. Subsequently, the  ultrathin CdS Ns suspension was centrifuged at 10000 rpm for 10 min and washed with DI H2O for 3  times to remove free L-cysteine and DETA. Finally, the sample was re-dispersed in DI H2O and the  concentration of the thus-obtained CdS Ns was ca. 1.8 mg/mL.

2.3 Preparation of Au NPs

30 mL of 30 mM HAuCl4 aqueous solution was added to 80 mL of 25 mM TOAB toluene solution.  Afterwards, 25 mL of 0.4 M freshly prepared NaBH4 aqueous solution was rapidly added to the  mixture under vigorously stirring. After 30 min the toluene phase was washed with 0.1 M H2SO4, 0.1  M NaOH and DI H2O for three times, respectively, and dried with anhydrous Na2SO4. 1 mL of 0.1 M  4-dimethylaminopyridine (DMAP) aqueous solution was gradually added into 1 mL of the Au NPs mixtures, by which direct phase transfer across the organic/aqueous boundary was completed within 1  h without stirring or agitation to obtain Au NPs aqueous suspension.49

2.4 Self-assembly of Au-CdS Ns nanocomposites

10 mL of CdS Ns aqueous suspension and different weight addition ratios of Au@DMAP NPs (Au/CdS Ns = 0.5, 1.0 and 2.0 wt.%) were mixed and stirred for 3 h to trigger the self-assembly  process at ambient conditions. Subsequently, the mixture was filtered and washed with DI H2O for 3  times to remove the free organic molecules and the yellow precipitate obtained was dried in an oven at  80 oC for 12 h. Finally, CdS Ns and Au-CdS Ns nanocomposites were sintered in a furnace at a  heating rate of 5 oC/min and kept at 300 oC for 30 min. Au-CdS Ns nanocomposites with different  Au/CdS Ns weight addition ratios of 0, 0.5, 1.0 and 2.0 wt. % were labeled as CdS Ns, 0.5Au-CdS Ns,  1Au-CdS Ns and 2Au-CdS Ns, respectively

2.5 Characterization

X-ray diffraction (XRD) analysis was carried out by a Philips Panalytical X’pert PRO with Cu-Kα  radiation at λ = 0.1542 nm. XRD patterns were recorded with a scan step of 1º min-1 (2θ) in the range  from 20º to 80º. Fourier Transform Infrared Spectroscopy (FTIR) measurements were performed on a

Nicolet 6700 equipped with a MCT/A detector. X-ray photoelectron spectroscopy (XPS)  measurements were collected on an ESCALAB 250 photoelectron spectrometer (Thermo Fisher  Scientific) at 2.4 × 10−10 mbar using a monochromatic Al Kα X-ray beam (1486.60 eV). The binding  energy was calibrated with respect to C (1s) at 284.6 eV. UV-Vis diffuse reflectance spectra (DRS)  were obtained using a UV-Visible spectrophotometer (Macylab UV-1800A). Photoluminescence (PL)  measurements were carried out on a F97 (Lengguang Tech.) fluorescence photometer with an exciting  wavelength of 375 nm. Zeta potential was determined by electrophoretic light scattering (ELS) with  Zetaplus (Brookhaven Instruments Corporation, Holtsville, NY, USA). Transmission electron  microscopy (TEM) images were taken by JEOL JEM-1400 and Higher-resolution TEM (HRTEM)  images were taken by JEM 2100 and F30. The morphology and elemental mapping of the samples were measured by scanning electron microscopy (SEM, Sigma HD). Specific surface areas were determined by a Brunauer-Emmett-Teller (BET) method using nitrogen adsorption and desorption  isotherms on a Micromeritics Instrument Corporation sorption analyzer (TriStar II3020). Photoelectrochemical (PEC) measurements were performed on an electrochemical workstation (CHI  660D). The electrochemical setup was composed of conventional three-electrode, a quartz cell  containing 20 mL of Na2SO4 (0.2 M) aqueous solution, and a potentiostat. A platinum plate (8 ×18 mm) was used as counter electrode and Ag/AgCl/KCl as reference electrode. The sample films with  active area (0.2 cm 2) were vertically dipped into the electrolyte and irradiated with a 300 W Xenon arc  lamp with a UV cutoff filter (λ ≥ 420 nm)

2.6 Photocatalytic activity measurements

Photocatalytic activities of different samples were evaluated by photocatalytic degradation of  methylene blue (MB) under visible light irradiation (λ ≥ 420 nm) at ambient conditions. A 300 W Xe  lamp (CEL-HXF300, 15A) equipped with a 420 nm-cut-off filter was used as light source and held at  25.0 cm away from the glass reactor. 5 mg photocatalyst was dispersed in 50 mL of 10 ppm MB  aqueous solution and the concentration of photocatalyst was 0.1 mg mL-1. Before light irradiation, the  suspension was stirred and kept in the dark for 6 h to achieve the adsorption-desorption equilibrium  between the photocatalyst and MB. After that, the suspension was irradiated with visible light (λ ≥ 420  nm) and 2.5 mL of the sample solution was collected and centrifuged at every 0.5 h for analysis. The  concentration of MB was determined using a UV-Vis spectrophotometer by monitoring the variation  of peak absorbance at 664 nm. To obtain the appearance quantum efficiency (AQE), photocatalytic  activities of different samples were also measured under the same experimental conditions except for  using the monochromatic light with different wavelengths (i.e., 380, 420, 450, 500, 550, 600 and 700  nm) as the light source.

2.7 Determination of the active species in the photocatalytic process

2.7.1 Detection of active species by quenching experiments50, 51

Quenching experiments were conducted under the same experimental conditions to that of the  photodegradation reactions except adding 1.0 mM scavengers in the reaction system. Different  scavengers such as benzoquinone, tert-butyl alcohol, ammonium oxalate and K2S2O8 were utilized as  the scavengers for quenching superoxide radical (O2•−), hydroxyl radical (OH•), hole (h+ ) and electron  (e- ), respectively.

2.7.2 Detection of hydroxyl radicals (OH•)52

In a typical process, 5 mg different samples was dispersed in 50 mL of mixed solution of terephthalic  acid (0.8 mM) and NaOH (0.4 mM) and then the suspension was irradiated with visible light (λ ≥ 420  nm). At different time intervals (0, 30, 60, 90, 120, 150 and 180 min), 3 mL of the sample suspension  was centrifuged (8000 rpm, 10 min) and finally 2.5 mL of the transparent solution was collected for  PL measurement. The concentration of hydroxyl terephthalate anion was determined by a F97  (Lengguang Tech.) fluorescence photometer with an excitation wavelength of 320 nm.

2.7.3 Detection of superoxide radical (O2 •−) using electron paramagnetic resonance (EPR) 53

Electron paramagnetic resonance spectra of spin-trapped radicals with 5, 5-diemthyl-1-pyrroline  N-oxide (DMPO) were recorded by a BrukerEMX-10/12 spectrometer. Briefly, 5 mg sample was  dispersed in 0.5 mL of methanol for superoxide radicals, into which 20 µL DMPO-methanol mixed  solution (V : V= 1 : 10) was added. A 300 W Xe lamp (15 A, Microsolar 300, Perfect light, Beijing)  equipped with a 420 nm-cut-off optical filter (λ ≥ 420 nm) was used as the light source. The settings  for EPR spectrometer were as follows: center field = 3370 G, microwave frequency = 9.450343 GHz  and power = 20.21 mW.

2.8 Photocatalytic reduction of aromatic nitro compounds

For anaerobic photocatalytic reduction of 4-nitroaniline (4-NA) in an inert atmosphere (N2), a 300 W  Xe lamp (CEL-HXF300, 15A) equipped with a 420 nm-cut-off filter (λ ≥ 420 nm) was used as the  light source and kept at 25.0 cm away from the glass reactor. 10 mg photocatalyst and 40 mg  NH4HCO2 were mixed with 30 mL of 4-NA aqueous solution (20 mg L-1 ) in a glass reactor under N2 bubbling. Before light irradiation, the suspension was stirred and kept in the dark for 1 h to achieve  the adsorption-desorption equilibrium between the photocatalyst and 4-NA. After that, the suspension  was irradiated with visible light (λ ≥ 420 nm). At different time intervals (0, 5, 10, 15, 20, 25, 30 and  35 min), 3 mL of the sample solution was collected and centrifuged (12000 rpm) and analyzed on a  UV-Vis spectrophotometer (Macylab UV-1800A). Alternatively, photoreduction of other aromatic  nitro compounds, including 3-NA, 2-NA, 4-NP, 3-NP, 2-NP and 4-NT were also carried out and photoreduction activities were measured under the same experimental conditions

2.9 Photocatalytic hydrogen production

Photocatalytic H2 production reactions were carried out in an online photocatalytic hydrogen  production system (Lab Solar-IIIAG, Perfect Light, Beijing). 50 mg sample was dispersed by  ultrasonication in 100 mL of aqueous solution containing 0.25 M Na2S and 0.25 M Na2SO3 as  sacrificial reagents in a reaction cell. The suspension solution was degassed for 30 min to remove air  completely before light irradiation. Then, the solution was irradiated by a 300W Xe lamp (15 A, Micro solar 300, Perfect light, Beijing) equipped with a 420 nm-cut-off optical filter (λ ≥ 420 nm). The H2 evolution amount was determined using an online gas chromatograph (GC-2018, TCD).

3. Results and Discussion

Scheme 1 illustrates the spontaneous ligand-triggered self-assembly process for fabricating Au-CdS  Ns nanocomposites by mediating intrinsic surface charge properties of CdS Ns and Au NPs assembly  units. It should be emphasized that the CdS Ns are capped by a large amount of L-cysteine molecules  which affords plentiful carboxyl functional groups on the surface and thus the deprotonation of these  carboxyl groups endows CdS Ns with a pronounced negatively charged surface. This can be  corroborated by the zeta potential (-15.4 mV, Fig. S1) and Fourier Transform Infrared Spectroscopy  (FTIR) results (Fig. S2) of CdS Ns.6 On the other hand, with regard to Au NPs which are protected by  DMAP moleculars, partial protonation of exocyclic nitrogen atom in DMAP linker give rise to  positively charged surface of Au@DMAP, 49 which can also be mirrored by the zeta potential (+14.2  mV, Fig. S1) and FTIR results (Fig. S2). Hence, when these two assembly units are mixed together,  positively charged Au@DMAP NPs are able to be spontaneously and rapidly attracted on the  negatively charged CdS Ns@L-cysteine surface under substantial electrostatic interaction resulting in Au-CdS Ns nanocomposites. With a view to reinforcing the interfacial contact between Au and CdS  Ns and removing the organic moieties capped on their surfaces, Au@DMAP/CdS Ns nanocomposites were finally calcined at low temperature yielding the well-defined Au-CdS Ns nanocomposites