Monolayered Bi₂WO₆ nanosheets mimicking heterojunction interface with open surfaces for photocatalysis

Abstract

Two-dimensional-layered heterojunctions have attracted extensive interest recently due to their exciting behaviours in electronic/optoelectronic devices as well as solar energy conversion systems. However, layered heterojunction materials, especially those made by stacking different monolayers together by strong chemical bonds rather than by weak van der Waal interactions, are still challenging to fabricate. Here the monolayer Bi₂WO₆ with a sandwich substructure of [BiO]⁺–[WO₄]²⁻–[BiO]⁺ is reported. This material may be characterized as a layered heterojunction with different monolayer oxides held together by chemical bonds. Coordinatively unsaturated Bi atoms are present as active sites on the surface. On irradiation, holes are generated directly on the active surface layer and electrons in the middle layer, which leads to the outstanding performances of the monolayer material in solar energy conversion. Our work provides a general bottom-up route for designing and preparing novel monolayer materials with ultrafast charge separation and active surface.

Introduction

Semiconductor-based heterojunctions are able to facilitate fast charge separation, and thus have emerged as one of important strategies for creation of efficient solar energy conversion systems^1,2,3,4. It is well known that a heterojunction in nature is the interface that occurs between two dissimilar crystalline semiconductors. The heterojunction effect on charge separation are thought to come from the built-in electric field forming at the atomically thin interface¹. Compared with conventional powder heterojunction materials, the layered heterojunctions composed of different ultrathin two-dimensional (2D) materials have sparked widespread interests because of the well-defined heterostructures almost as thin as the interface of conventional heterojunctions. As great successes on monolayer materials have been achieved^5,6,7, the stacks of different monolayers held together by van der Waals forces have been realized^{8,9,10,11,12,13,14,15}. However, the layered heterojunctions composed of different monolayers stacked together by strong chemical bonds have been rarely studied. Since chemical bonding between monolayers can greatly enhance charge separation and even induce new energy band structures, the layered heterojunctions are considerably appealing^16,17. Besides charge separation, exposure of active sites (mainly unsaturated atoms) is also key to affect the material performances in solar energy conversion^{Supplementary Figs 11–12).}

Synthesis of other monolayer materials

To verify the universal applicability of this self-assembly route, preparation of monolayer Bi₂O₂CO₃ was studied also in the same conditions. The result shows that the monolayer Bi₂O₂CO₃ with a thickness of 0.55 nm was successfully obtained in the presence of CTAB, as showed in the Supplementary Figs 13–15. Besides Bi₂WO₆ and Bi₂O₂CO₃, the Aurivillius phase layered oxides (general formula Bi₂A_n−1B_nO_3n+3; A=Ca, Sr, Ba, Pb, Bi, Na, K and B=Ti, Nb, Ta, Mo, W, Fe), an important class of functional materials including photocatalysts, superconductors, and ferroelectric materials^40,41,42, have the common feature of containing [Bi₂O₂] layers in crystal structure. Thus, a huge number of monolayer Aurivillius materials can be formed by the similar mechanism of self-assembly.

Band gap of the monolayer Bi₂WO₆

Figure 3a shows ultraviolet–visible diffuse reflectance spectrum of the Bi₂WO₆ samples. The bandgap energies of the samples were estimated from the plots of (αhv)^1/2 versus the energy of absorbed light (inset of Fig. 3a)⁴³. The bandgap energy of the monolayers is ca. 2.7 eV, lower than that of the nanocrystals by 0.2 eV. To understand the differences, band structures of the bulk (multilayers), pure monolayer (unmodified) and Br-modified monolayer of Bi₂WO₆ were investigated theoretically (Supplementary Fig. 16). The corresponding densities of states are presented in Fig. 3b. Although the bandgaps from DFT calculations are usually smaller than that from the experimental determination, the DFT calculations often provide important insights into the physicochemical properties of the materials^43,44. For the unmodified monolayers and the bulk, the top of valence band and the bottom of conduction band are mainly originated from hybrid Bi 6s–O 2p orbital and W 5d orbital, respectively (Fig. 3b). However for the Br-modified monolayers, the bottom of conduction band is still originated from W 5d orbital, but the top of valence band is composed of the hybridized orbital of Bi 6s, O 2p and Br 4p (Fig. 3b). Due to more negative potential of their top of valence, the Br-modified monolayers show a narrower bandgap than the unmodified monolayers and the nanocrystals.

Figure 3: Bandgap states of Bi₂WO₆ samples.

(a) Diffuse reflectance ultraviolet–visible spectra of Bi₂WO₆ samples. (b) Calculated density of states (DOSs) of Bi₂WO₆ bulk, pristine monolayer and Br-modified monolayer. The Fermi level is taken as the energy zero. (c) Band energy diagrams of Bi₂WO₆ nanocrystals and monolayers. W, O, Bi and Br atoms are represented as blue, red, purple and yarrow spheres, respectively.

The sandwich substructure of [BiO]⁺–[WO₄]²⁻–[BiO]⁺ of monolayer Bi₂WO₆ simulates the heterojunction interface with space charge that promotes separation of the photogenerated carriers in the interface, as shown in Fig. 3c. It should be noted that the bottom of conduction band is originated from W 5d orbital located in the middle layer, and the top of valence band is mainly composed of O 2p orbital from the surface [BiO]⁺ layers (Fig. 3b). On irradiation, holes and electrons can be generated, respectively, on the surface layer and in the middle layer and thus are separated directly. In brief, for the monolayer Bi₂WO₆, the sandwich substructure and the Br surface modification lead to decrease of the bandgap and enhancement of the photogenerated charges separation, which are very favourable for photocatalysis.

Open surface of the monolayer Bi₂WO₆

Different from most reported monolayer materials with closed and inert surfaces²¹, the as-synthesized monolayer Bi₂WO₆ possesses oxygen-depleted and highly active surface. Moreover, the Br⁻ ions on surfaces cause the monolayers negatively charged, as proved by the zeta potential in Supplementary Table 1. Therefore, the monolayers can selectively adsorb positively charged species. Figure 4b shows the adsorption performances of various organic pollutants over the Bi₂WO₆ monolayers. Salicylic acid, as a neutral species, is hardly adsorbed by the monolayers. Methyl orange is positively charged in acid medium (Fig. 4a), and therefore its adsorption is enhanced by changing solution pH from 7 to 3 using H₂SO₄ or HCl. Similar to Br⁻ ions, Cl⁻ could also adsorb at the open metal sites on surface and make the monolayers more negatively charged, as shown in Fig. 4c. Hence, the methyl orange adsorption can be promoted more effectively by HCl than by H₂SO₄. Methylene blue molecules containing both Cl⁻ ions and positive ammonium groups are adsorbed almost entirely. Owing to the open surfaces, the monolayers show much better ability for adsorption of Rhodamine B (RhB; 7.3 mg g⁻¹) than the nanocrystals (0.19 mg g⁻¹), though the specific surface area of the monolayers (43 m² g⁻¹) is less than twice that of the nanocrystals (28 m² g⁻¹). All above results confirm that the surface of monolayer Bi₂WO₆ is open and can strongly interact with surrounding species such as water, anions Cl⁻ or Br⁻ and positively charged organics (Fig. 4c).

Figure 4: Open surface of the Bi₂WO₆ monolayers.

(a) Molecular structures of various organic contaminants. (b) Adsorption of various organic pollutants over the Bi₂WO₆ monolayers. (c) Schematic diagram showing the interactions between the open surface of monolayer Bi₂WO₆ and surrounding species.

Photocatalytic performance of the monolayer Bi₂WO₆

Since the holes are photogenerated directly on the open surface [BiO]⁺ layers (Fig. 5a), the monolayers surfaces are highly active for photocatalytic oxidation reactions, such as the oxidation of OH⁻ to ·OH. This feature was verified by the 5, 5-dimethyl-1-pyrroline-N-oxide (DMPO) spin-trap** electron paramagnetic resonance (EPR) technique (Fig. 5b). The ·OH radicals signal is very weak for the nanocrystals under visible light illumination, which is consistent with the previously observation for Bi₂WO₆ samples⁴⁵. However, for the monolayer Bi₂WO₆, the ·OH radicals signal is much stronger, implying the superior activity of the monolayers for the photocatalytic oxidation reactions. On the other hand, the photoelectrons are formed in the middle layer where they transfer to the edge of monolayers for reduction reactions. Figure 5c shows that the reduction activity of the monolayers for O₂ to O₂⁻ is enhanced likewise, compared with that of the nanocrystals, although the enhancement is lower for the reduction than for the oxidation. Figure 5d is comparison of photocurrents between the monolayers and nanocrystals. It can be seen that the photocurrent from the monolayers is about 10-fold higher than that from the nanocrystals. This indicates that the monolayers mimic the heterojunction interface to result in the ultrafast charge transfer and separation. As for the little decrease of photocurrent on the monolayers with irradiation time, this can be because the photogenerated electrons formed in the middle layer need to undergo a long-distance transfer towards the edge, electron accumulation and partial recombination with holes may occur simultaneously with irradiation time.

Figure 5: Photocatalytic performances of Bi₂WO₆ samples.

(a) Schematic illustration of photocatalytic mechanism over the monolayer Bi₂WO₆. (b,c) Electron spin resonance signals of DMPO-·OH⁻ adducts and DMPO-O₂·⁻ adducts produced by Bi₂WO₆ samples under visible light irradiation (λ≥420 nm). (d) Photocurrent of Bi₂WO₆ samples under visible light illumination. (e) Photodegradation of RhB over Bi₂WO₆ samples (λ≥420 nm). (f) Visible-light-driven photocatalytic H₂ evolution over the nanocrystals, monolayers and RhB sensitized monolayers (λ≥420 nm); 0.3 wt% of Pt was loaded.

The photocatalytic performances of Bi₂WO₆ samples for photodegradation of RhB were evaluated under visible light illumination (λ≥420 nm). As shown in Fig. 5e, the monolayers show much higher visible-light-driven photocatalytic activity than the nanocrystals. Ninety-eight percent of RhB are degraded using the monolayers as a photocatalyst within 25 min, while only 13% RhB are degraded using the nanocrystals within 30 min. Moreover, the monolayers can be efficiently recycled and reused without appreciable loss of activity (Supplementary Figs 17–18). Besides, the monolayers show also an excellent photocatalytic activity for H₂ production from water solution under visible light. Figure 5f shows the H₂ evolution over the nanocrystals, monolayers and RhB sensitized monolayers (λ≥420 nm) with 0.3 wt% Pt loading as a function of time. The H₂ evolution activity of Bi₂WO₆ has been reported to be very low (1.6 μmol h⁻¹ g⁻¹) even under a 450 W high pressure mercury lamp⁴⁶. And the Bi₂WO₆ has been considered to have no activity for visible-light-driven H₂ production⁴⁷. As shown in Fig. 5f, the nanocrystals show indeed no H₂ evolution activity, but surprisingly the monolayers exhibit a good H₂ evolution activity under visible light. As adding RhB into the reaction system, ca. fourfold higher activity is observed due to the sensitization effect of RhB. Both the ultrafast charge separation and the highly active surface make the monolayers possess superior performances in solar energy conversion than other nanostructures composed of two or more layers (Supplementary Figs 19–21).

Discussion

In summary, the freestanding monolayer Bi₂WO₆ can be successfully synthesized via a CTAB-assisted self-assembly route. The as-synthesized monolayer Bi₂WO₆ has a sandwich substructure of [BiO]⁺–[WO₄]²⁻–[BiO]⁺. The Bi atoms in the surface [BiO]⁺ layers are coordinatively unsaturated and thus the monolayers can adsorb Br⁻ ions to avoid stacking into multilayer. The monolayer Bi₂WO₆ shows oxygen-depleted surface, and has a narrower bandgap than the nanocrystals. The sandwich structure simulates the heterojunction interface with space charge to facilitate the ultrafast separation of photogenerated carriers. On irradiation, holes are generated directly on the highly active surfaces, resulting in the excellent activities for the photodegradation of RhB. More interestingly, the monolayering allows Bi₂WO₆ to have a visible-light-driven H₂ evolution activity, which can be efficiently enhanced by dye sensitization. This work provides a simple and efficient bottom-up route to prepare novel monolayer materials with ultrafast charge separation and highly active surface.

Methods

Materials synthesis

In the preparation of Bi₂WO₆ monolayers, the start materials 1 mmol Na₂WO₄·2H₂O, and 2 mmol Bi(NO₃)₃·5H₂O and 0.05 g CTAB were added in 80 ml deionized water. After 30 min stirring, the mixed solution was poured into a 100 ml Teflon-lined autoclave. Then the autoclave was sealed into a stainless steel tank and treated at 120 °C for 24 h. Finally, the product was collected and washed several times with deionized water and dried at 60 °C in air for 10 h. In the preparation of monolayer Bi₂WO₆ companied with few layers, 0.1 mmol KBr instead of CTAB was added in the mixture, while kee** other parameters unchanged. The Bi₂WO₆ nanocrystals were prepared without CTAB and KBr assistances.

Bi₂O₂CO₃ monolayers were prepared by the same route. A total of 2 mmol Bi(NO₃)₃ and 0.05 g CTAB were added in 80 ml of 1 M HNO₃ solution. After the mixture became a clear solution, 10 mmol Na₂CO₃ was added and a white precipitate appeared immediately. This suspension was magnetically stirred for 30 min to complete the precipitation reaction. Then the suspension was poured into a 100 ml Teflon-lined autoclave and the autoclave was sealed into a stainless steel tank. The sample was treated at 60 °C for 24 h. Then the reactor was naturally cooled to room temperature. The obtained sample was collected and washed several times with deionized water and dried at 60 °C in air for 10 h.

Materials characterization

The X-ray diffraction patterns were recorded on a Bruker D8 Advance X-ray diffractometer with Ni filtered Cu Kα radiation at 40 kV and 40 mA. Ultraviolet–visible diffuse reflectance (ultraviolet–visible diffuse reflectance spectrum) spectra were obtained with a self-supporting sample disk on a ultraviolet–visible spectrophotometer (Cary 500), where BaSO₄ was used as a reflectance standard. TEM images were obtained using a JEOL model JEM 2010 EX instrument at an accelerating voltage of 200 kV. AFM images were recorded using Agilent 5,500 AFM (Agilent Technologies, USA). All the images were acquired using tap** mode under ambient conditions (ca. 40−50% relative humidity, 25 °C temperature). The used Si cantilevers/tips (Bruker) have a spring constant of 40 N m⁻¹ and a resonance frequency of 300 kHz. During an AFM experiment, sample was dispersed in ethanol using an ultrasonic bath for 20 min and then the dispersion was diluted in ethanol. A drop of the above diluted dispersion was deposited on a new cleaved mica surface and dried in air. The instrument parameters (set point, amplitude, scan size, scan speed and feedback control) were adjusted for the best resolution of images. Electron spin resonance spectra were obtained over Bruker ESP 300 E electron paramagnetic resonance spectrometer.

Activity evaluation

The photodegradation of dyes were performed in a glass vessel under visible-light irradiation by a 300 W Xe lamp with a 420 nm cutoff filter. Twenty milligram of the photocatalyst was added into 80 ml RhB (10 μmol l⁻¹) solution. The dispersion was stirred in the dark for 30 min to reach the adsorption equilibrium. Then the visible light was turn on for the photodegradation tests. Three millilitres of the sample solution were taken at given time intervals and separated by centrifugation. The residual concentration of the organics in solution was analysed using a Varian Cary 50 Scan ultraviolet–visible spectrophotometer. The selective adsorption experiments were performed by the same method. The methyl orange (20 p.p.m.), methylene blue (10 p.p.m.) and salicylic acid (500 μmol l⁻¹) solutions were used. 50 mg of the catalyst was used to test the adsorption performances.

The photocatalytic H₂ production of samples under visible light irradiation was conducted in a Pyrex reactor connected with a gas-closed circulation system and a vacuum system. In a typical experiment, 20 mg catalyst was dispersed in 100 ml deionized water containing 0.4 g ethylenediamine tetraacetic acid (EDTA) as a sacrificial reagent. This system was evacuated for 30 min to remove air prior to irradiation. A 300-W Xe lamp with an optical cutoff filter (λ≥420 nm) was used as a light source. The produced hydrogen was quantified by gas chromatography (Shimadzu GC-8A, TCD, Ar carrier). For the RhB sensitization system, RhB (10 μmol l⁻¹) was contained in solution.

Calculation details

First-principles DFT calculations were carried out utilizing the Vienna ab initio simulation package^48,49. The Perdew–Wang-type (PW91) exchange–correlation functional was employed to study the energies and structures of different Bi₂WO₆ systems. Vanderbilt ultrasoft pseudo-potentials were used to describe the interactions between the ion cores and valence electrons for all atoms, and the kinetic cutoff energy was set to 400 eV. In the calculations, the convergence energy threshold for self-consistent iteration was set to 10⁻⁴ eV per atom, along with the residual atomic forces were <0.03 eV Å⁻¹. For the bulk Bi₂WO₆, the optimized lattice constants are a=5.480 Å, b=5.468 Å and c=16.760 Å, which agrees well with the experimental values of a=5.457 Å, b=5.436 Å and c=16.427 Å. A (2 × 2) periodic slab model was adopted to simulate different Bi₂WO₆ nanosheets, which were constructed from the Bi₂WO₆ (001) surface. During the structural optimization, the positions of all atoms and the lengths of two translation vectors of the nanosheet were allowed to relax. The spacing between the adjacent slabs was set about 10 Å, and a (3 × 3 × 1) Monkhorst–Pack k-point mesh was used for integration in the reciprocal space. To investigate the thermodynamic stability of different Bi₂WO₆ nanosheets, the surface free energy (γ) was calculated, which is defined as following,

where E^slab is the total energy of the slab, is the energy of per Bi₂WO₆ unit in the bulk, N is the number of Bi₂WO₆ unit found in the slab, N_Br is the number of Br atom, E_Br is the ground state energy of Br atom, and A is the area of the slab surface.