Journal of Materials Science & Technology  2019 , 35 (10): 2288-2296 https://doi.org/10.1016/j.jmst.2019.05.057

Orginal Article

Chlorine doped graphitic carbon nitride nanorings as an efficient photoresponsive catalyst for water oxidation and organic decomposition

Er-Xun Hana, Yuan-Yuan Lia, Qi-Hao Wanga, Wei-Qing Huanga*, Leng Luoa, Wangyu Hub, Gui-Fang Huanga*

a Key Laboratory for Micro/Nano Optoelectronic Devices of Ministry of Education & Hunan Provincial Key Laboratory of Low-Dimensional Structural Physics and Devices, School of Physics and Electronics, Hunan University, Changsha 410082, China
bSchool of Materials Science and Engineering, Hunan University, Changsha 410082, China

Corresponding authors:   *Corresponding authors.E-mail addresses: wqhuang@hnu.edu.cn (W.-Q. Huang), gfhuang@hnu.edu.cn(G.-F. Huang).*Corresponding authors.E-mail addresses: wqhuang@hnu.edu.cn (W.-Q. Huang), gfhuang@hnu.edu.cn(G.-F. Huang).

Received: 2019-04-15

Revised:  2019-05-15

Accepted:  2019-05-23

Online:  2019-10-05

Copyright:  2019 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

Rationally engineering the microstructure and electronic structure of catalysts to induce high activity for versatile applications remains a challenge. Herein, chlorine doped graphitic carbon nitride (Cl-doped g-C3N4) nanorings have been designed as a superior photocatalyst for pollutant degradation and oxygen evolution reaction (OER). Remarkably, Cl-doped g-C3N4 nanorings display enhanced OER performance with a small overpotential of approximately 290 mV at current density of 10 mA cm-2 and Tafel slope of 83 mV dec-1, possessing comparable OER activity to precious metal oxides RuO2 and IrO2/C. The excellent catalytic performance of Cl-doped g-C3N4 nanorings originates from the strong oxidation capability, abundant active sites exposed and efficient charge transfer. More importantly, visible light irradiation gives rise to a prominent improvement of the OER performance, reducing the OER overpotential and Tafel slope by 140 mV and 28 mV dec-1, respectively, demonstrating the striking photo-responsive OER activity of Cl-doped g-C3N4 nanorings. The great photo-induced improvement in OER activity would be related to the efficient charge transfer and the •OH radicals arising spontaneously on CN-Cl100 catalyst upon light irradiation. This work establishes Cl-doped g-C3N4 nanorings as a highly competitive metal-free candidate for photoelectrochemical energy conversion and environmental cleaning application.

Keywords: Cl-doped g-C3N4 ; Nanoring ; Electrocatalysts ; Oxygen evolution reaction ; Pollutant degradation

0

PDF (3252KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Er-Xun Han, Yuan-Yuan Li, Qi-Hao Wang, Wei-Qing Huang, Leng Luo, Wangyu Hu, Gui-Fang Huang. Chlorine doped graphitic carbon nitride nanorings as an efficient photoresponsive catalyst for water oxidation and organic decomposition[J]. Journal of Materials Science & Technology, 2019, 35(10): 2288-2296 https://doi.org/10.1016/j.jmst.2019.05.057

1. Introduction

Energy and environment concerns are among the most serious challenges of current age [[1], [2], [3]]. Hydrogen generation via water splitting using renewable solar energy offers an appealing and promising possibility towards clean energy. Nevertheless, the high activation barriers associated with the rate-determining step and sluggish kinetics of oxygen evolution reaction (OER) in the opposite half reaction due to the four-electron reaction process restrains largely the overall water splitting performance [3,4]. Noble metal-based catalysts (RuO2 and IrO2) are the desired OER catalysts with high activity. Unfortunately, their scarcity and high cost prevent them from large-scale applications. Therefore, it is highly desirable to develop low-cost alternatives for the ever-increasing energy demand and environmental sustainability.

Graphitic carbon nitride (g-C3N4) has inspired great research enthusiasm since it emerged as a charming metal-free catalyst in water splitting, pollutant removal and CO2 reduction, owing to its stability, non-toxicity and suitable band gap [[5], [6], [7]]. While the low charge migration rate, weak oxidization ability and low conductivity of g-C3N4 are the major dilemmas in its catalytic applications. As a typical 2D polymeric material, g-C3N4 offers diversified choice of chemical protocols to design the texture, molecular and electronic structure to ameliorate the performance [8,9]. Considerable efforts, such as nanostructure engineering [10,11], heterostructure construction [12] and doping [13,14], have been made to modifying g-C3N4. Among them, doping is especially promising choice with the advantages of simple operation, effective modification on texture and structure of g-C3N4 for desired catalytic efficiency. Various elements, such as phosphor [15], sulfur [16,17], and boron [18,19], have been doped into g-C3N4 to manipulate the catalytic properties. For example, boron-doped g-C3N4 prepared by heating the mixture of dicyanodiamide and BH3NH3 display enhanced photocurrent response for CO2 reduction [18]. Phosphor-doped g-C3N4 shows improved photocatalytic activity in hydrogen evolution [13], pollutant degradation [15], and oxidation of aromatic alcohols [20]. Specially, halogen elements play effective role in manipulating the optical absorption, charge-carrier transfer rate, as well as catalytic properties of halogen-doped g-C3N4, which are synthesized by saturated NH4Br or NH4Cl hydrothermal post-treatment [21], co-condensation approach [22]. However, the traditional doping method may cause disorganized agglomerated morphology and low separation efficiency of electrons-holes.

Herein, we develop a facile approach, which provides the opportunity to construct g-C3N4 nanorings and introduce chlorine heteroatoms simultaneously through the preorganization of melamine and cyanuric chloride and subsequent thermal polycondensation. The introduction of chlorine in g-C3N4 leads to enhanced electronic conductivity and strong oxidation capability. The integration of texture engineering by coupling with chloride doping generates the functional Cl-doped g-C3N4 nanorings with highly efficient photooxidation ability for OER and pollutant degradation. The impressive photoinduced OER improvement of CN-Cl100 nanorings correlates with the generation of •OH radicals upon visible light irradiation.

2. Experimental

2.1. Preparation of Cl-doped g-C3N4 nanorings

Cl-doped g-C3N4 nanorings are synthesized through the preorganization of melamine and cyanuric chloride under oil bath condition and subsequent thermal polycondensation. Typically, 2.7675 g cyanuric chloride and 0.9460 g melamine are separately dissolved in appropriate amount of ethyl alcohol which is marked as solution A and solution B, respectively. Then, solution B is heated to 100 ℃ in an oil bath under stirring and solution A is stirred until both of them are dissolved completely. After that, solution A is added dropwise into solution B under stirring, and the mixture is left to be dried to obtain solid supramolecular aggregates, which are denoted as CM100. Finally, the white supramolecular aggregates are heated at 500 ℃ for 2 h with a ramping rate of 5 ℃ min-1 and the product is labelled as CN-Cl100. In addition, Cl-doped g-C3N4 is also synthesized through heating the supramolecular aggregates obtained at 80 ℃ or 120 ℃, which are denoted as CN-Cl80, and CN-Cl120, respectively. For comparision, bulk g-C3N4 is prepared by directly heating melamine under the identical heating condition, which is marked as bulk CN.

2.2. Characterization

Scanning electron microscopy (SEM) and X-Ray diffraction (XRD) are used to analyze the morphology and crystal structure of the samples, respectively. Fourier transform infrared (FTIR) spectra are also recorded to further characterize the sample structure. The energy-dispersive X-ray spectroscopy (EDS) measurement is performed to detect the content and distribution of elements. The UV-vis absorption and photoluminescence (PL) sepctra are recorded on UV-vis and fluorescence spectrophotometer, respectively.

2.3. Electrochemical measurement

Electrochemical measurement is performed on CHI 660E electrochemical work station. The catalyst ink is prepared by mixing 5 mg catalysts, 500 μL ethyl alcohol and 15 u L nafion under sonication for 1 h. Then, 150 μL catalyst ink is extracted and coated on the support with the loading area of 1 cm2 and dried at room temperature. Linear sweep voltammetry (LSV) measurements are carried out with nickel foam as the support of the catalysts in 1.0 mol/L KOH solution using typical three-electrode system at a scan rate of 2 mV s-1. The stability tests for OER are evaluated using a chronoamperometry technique with an overpotential of 560 mV in dark or under visible light irradiation. In addition, the photocurrent response, Mott-Schottky curves and electrochemical impedance spectroscopy (EIS) are performed with conductive glass as the support of catalysts in 0.5 mol/L Na2SO4 solution. The saturated Ag/AgCl and graphite rod is applied as the reference and counter electrodes, respectively. The potentials are showed versus reversible hydrogen electrode (RHE) through the Nernst equation: ERHE= EAg/AgCl + 0.059 pH + 0.197 V.

2.4. Photocatalytic activity test

Rhodamine B (RhB) is used as organic pollutant model for photocatalysis under the irradiation of 300 W halogen tungsten lamp, 10 mg of catalysts is used to degrade 40 mL of RhB solution (10 mg/L). To make sure the adsorption-desorption equilibrium of RhB on the surface of catalysts, the suspension is ultrasonicated for 20 min before irradiation. 3 mL of suspension is taken out per hour during irradiation process and centrifuged to separate catalysts from RhB solution, and analyzed by UV-vis spectrophotometer.

In order to determine the generation and role of •OH under light irradiation, terephthalic acid photoluminescence (TA-PL) probing is used to detect the production of •OH under 50 W fluorescent lamp illumination via analyzing the PL intensity of 2-hydroxyterephthalic acid (TAOH) transformed from TA combining with •OH.

3. Results and discussion

3.1. Morphology and structure characterization

The formation of supramolecular aggregates is the key to construct g-C3N4 nanorings and introduce chlorine heteroatoms simultaneously since the hydrogen bond between melamine and cyanuric chloride allow them to self-assemble into ordered supramolecular assembly. The evidence for the formation of supramolecular aggregates structure is achieved by the FTIR spectra as shown in Fig. 1(a). It can be observed that the C-Cl stretching vibration of cyanuric chloride at 849 cm-1 disappear while that at 667 cm-1 remain in the supramolecular aggregates(CM80,CM100). In addition, the triazine ring vibration of melamine shifts from 814 cm-1 to 768 cm-1. The obvious shift of the vibration reflects the strong interaction between cyanuric chloride and melamine and the formation of supramolecular aggregates, which is then heated at 500 ℃ to produce g-C3N4 catalysts. Fig. 1(b) displays the FTIR spectra of bulk CN, CN-Cl80 and CN-Cl100. Several bands observed in the region 1300-1700 cm-1 are related to the typical stretching vibration modes of C-N heterocycles [22,23]. Additionally, the characteristic peak of triazine unit at around 816 cm-1 is also found [24,25]. It is clear that the FTIR spectra of CN-Cl80 and CN-Cl100 retain the main characteristic peaks of g-C3N4. Interestingly, vibration band at about 667 cm-1 is observed only in CN-Cl80 and CN-Cl100, which is likely related to the characteristic stretching vibration of C-Cl, reflecting the existence of chlorine in the CN-Clx.

Fig. 1.   (a) FTIR spectrum of precursors (a: melamine, b: cyanuric chloride, c: CM-80 and d: CM-100), (b) products (Ⅰ: bulk CN, Ⅱ: CN-Cl80 and Ⅲ: CN-Cl100) and (c) XRD graphs for CN-Clx and bulk CN.

The formation of g-C3N4 is further confirmed via XRD analysis. Fig. 1(c) shows that all the products display similar XRD patterns with two characteristic peaks at about 27.1° and 13.0° corresponding to (002) and (100) crystal planes for g-C3N4, respectively [[26], [27], [28]]. The strong peak located at 27.1° could be assigned to the typical interplanar stacking peak of conjugated aromatic systems. While the weak peak at 13.0° is associated with the in-planar structural packing motif. Both XRD and FTIR analysis indicate that the Cl-doped products retain the chemical skeleton and crystalline phase of g-C3N4. Comparing with bulk CN, Cl-doped g-C3N4 exhibits wider and weaker diffraction peaks, which might be attributed to the size decrease. This can be expected some special properties and performance.

The morphologies of bulk CN, CN-Cl80, CN-Cl100, and CN-Cl120 are detected by SEM as shown in Fig. 2. Bulk CN displays aggregated structures with irregular morphology (Fig. 2(a)). Obviously, a short fiber-like structure can be observed in CN-Cl80 (Fig. 2(b)). CN-Cl100 is composed of nanofibers with diameter of approximately 50 nm which curl into nanorings (Fig. 2(c)). Whereas, the morphology of CN-Cl120 appears to be some aggregative. This demonstrates that the temperature applied to produce supramolecular aggregates greatly influences the morphologies of g-C3N4 products. The nanorings of CN-Cl100 may provide adequate active sites and high charge-carrier mobility, leading to improved catalytic activity. Moreover, the elemental composition details of CN-Cl100 is detected and the atomic percentages of C, N and Cl in CN-Cl100 are determined to be 37.01%, 62.89% and 0.10%, respectively. Furthermore, the elemental mapping (Fig. 2(e)-(g)) exhibits the uniform distribution of C, N and Cl in CN-Cl100, providing the preliminary evidence for the successful doping of chlorine in CN-Clx samples.

Fig. 2.   SEM image of (a) bulk CN, (b) CN-Cl80, (c) CN-Cl100, (d) CN-Cl120 and (e, f, g) elemental mappings of CN-Cl100.

Fig. 3 schematically presents the formation process of Cl-doped g-C3N4 nanorings. To introduce Cl into g-C3N4 nanostructure, cyanuric chloride solution is dropwise added into hot melamine solution under stirring. The combination of melamine and cyanuric chloride results in the formation flake-like supramolecular aggregates through hydrogen bonds between melamine and cyanuric chloride. To prepare active g-C3N4, the hydrogen-bonded melamine-cyanuric chloride aggregates are further heated. During the thermal polycondensation process, various gases including CO2, HCl and NH3 will release and escape from the surface of CN flakes, which trend to curl up driven by the force of the gas flow and produce the Cl-doped g-C3N4 nanorings.

Fig. 3.   Schematic illustration for formation of Cl-doped g-C3N4 nanorings.

3.2. Optical analysis

The optical properties of bulk CN and CN-Clx are investigated through UV-vis diffuse reflectance spectra (DRS) and PL spectra. Fig. 4(a) shows that all samples display typical semiconductor absorption of g-C3N4, further indicating the reserved intrinsic backbone structures of g-C3N4 after Cl modification. It is significant that the absorption edge of CN-Clx shows an obvious blue shift in comparison to that of bulk CN. This blue shift can be presumably attributed to the quantum confinement effect owing to the CN-Clx nanomaterials. The band gap estimated from the Tauc plot is 2.61 eV, 2.72 eV, 2.69 and 2.72 eV for bulk CN, CN-Cl80, CN-Cl100 and CN-Cl120, respectively, as shown in the inset of Fig. 4(a).

Fig. 4.   (a) UV-vis diffuse reflectance spectra (inset: Tauc plots for estimating the band gap (Eg) values) and (b) the room-temperature PL spectra of bulk CN and CN-Clx (x = 80, 100 and 120).

PL spectra of bulk CN and CN-Clx are recorded as illustrated in Fig. 4(b). The emission peak of bulk CN locates at about 460 nm under photoexcitation at 320 nm, which could be attributed to the band-band PL phenomenon with the energy of light approximately equal to the energy gap of g-C3N4 [29]. While the emission peak is slight blue-shift in the case of CN-Clx. The blue shift of emission peak is in well agreement with DRS measurement. Additionally, it can be found that the peak intensity of CN-Clx are lower than that of bulk CN, reflecting a lower recombination rate of the electron-hole pairs [30,31], which is desirable in photocatalytic process.

3.3. Photoelectrocatalytic performance

Water oxidation reaction is considered as a more challenge half reaction in water-splitting process owing to its sluggish 4e transfer kinetics, which greatly impedes the large scale application. OER analysis of the products is performed in a typical three-electrode setup. Fig. 5(a) displays the LSV curves of bulk CN and CN-Clx drop-casted on nickel foams in 1 mol/L KOH with a scan rate of 2 mV s-1. For comparision, bare nickel foam is also tested. As can be observed from Fig. 5(a), all the products show higher OER activity than bare nickel foam. Note that chlorine doping CN not only increases the current density owing to its improved electroconductivity, but also decreases the onset potential of OER, verifying that chlorine doping CN could lead to an improved OER activity. It is noticeable that CN-Cl100 demonstrates the highest catalytic activity with the minimum onset potential of about 1.47 V and a small overpotential (290 mV) at the current density of 10 mA cm-2, which are superior to bulk CN, g-C3N4 catalysts [32,33], and the related samples previously reported as compared in Table 1, as well as commercial RuO2 and IrO2/C [34]. In addition, the OER catalytic kinetics are investigated via the corresponding Tafel plots (Fig. 5(b)). The smaller Tafel slope observed for CN-Clx than that of bulk CN indicates that chlorine doping faciliates the electron transport and demonstrates favorable reaction kinetics. As expected, CN-Cl100 nanorings exhibits the lowest Tafel slope (83 mV dec-1), which is lower than that reported IrO2-CNT (90 mV dec-1) [32].

Fig. 5.   (a) LSVs of CN-Clx and bulk CN in 1 M KOH with a scan rate of 2 mV s-1 and (b) Tafel plots of CN-Clx and bulk CN.

Table 1   Comparison on catalytic activities of as-obtained CN-Cl100 and related samples in the literature.

CatalystElectrolyteOnset Potential (V vs. RHE)Overpotential (mV)Tafel slope (mV dec-1)Reference
CN0.1 M KOH1.70720384.1[38]
CN@C0.1 M KOH1.40_254.2[38]
SH-g-C3N40.5 M KOH1.47340128[39]
63.6(in light)
Cl-doped g-C3N41.0 M KOH1.4729083This work
55(in light)
CoO@Co-NC/KB1.0 M KOH_26672[40]
Co@NCNT HMS1.0 M KOH_31779[41]
GCNTs1.0 M KOH1.5036055[42]
BCN1.0 M KOH1.6241670[42]
Ni@g-C3N4 CNT1.0 M KOH1.5032667[43]
O-N-CNs1.0 M KOH_381442[44]
Ni-CN-2001.0 M KOH1.54_60[45]

New window

Remarkably, the OER catalytic properties of CN-Cl100 nanorings can be greatly improved upon irradiation with 300 W halogen lamp as visible light source. It can be found from Fig. 5(a) that the onset potential of CN-Cl100 nanorings shifts from 1.47 V to 1.35 V vs. RHE upon light irradiation. Meanwhile, the current density of CN-Cl100 nanorings increases sharply with the bias potential. It needs extremely low overpotentials of about 155 mV to deliver the current density of 10 mA cm-2. Moreover, Tafel slope is significantly decreased to 55 mV dec-1 (Fig. 5(b)), indicating the fast electrocatalytic reaction kinetics. These parameters reflect the massive improvement in OER catalytic performance upon light irradiation, suggesting that CN-Cl100 nanorings could be served as a superior photo-responsive OER electrocatalyst. The photo-responsive phenomenon is similar to that observed in Ni12P5@NCNT hybrid [35]. The origin of the impressive photoinduced OER improvement could be related with the •OH radicals arising on CN-Cl100 nanorings under light irradiation and will be discussed in detail below.

To rule out the effect of pre-condensation of melamine itself on the ultimate OER activity of CN-Cl100, the activities of g-C3N4 formed from the condensation of melamine in 100 ℃ ethyl alcohol is also characterized. Fig. 6(a) demonstrates that the OER activity of g-C3N4 formed from the condensation of melamine in 100 °C ethyl alcohol is similar to that of bulk CN. Whereas, Cl-doped CN-Cl100 displays much enhanced OER performance. These results suggest that the superior OER catalytic activity of CN-Cl100 may be related to the doping of Cl, which leads to the formation of 1D nanorings, avoids the agglomeration of nanomaterials with more active sites exposed and promotes the charge separation.

Fig. 6.   (a) OER comparision of CN (in ethyl alcohol), bulk CN and CN-Cl100 and (b) current-time chronoamperometric responses of CN-Cl100 in dark or under visible light irradiation.

Besides the catalytic activity, stability and reusability of catalyst are also important factors that determine its practical application. To evaluate the stability of Cl-doped g-C3N4 catalysts, controlled chronoamperometry technique with an overpotential of 290 mV in dark or under 50 W fluorescent lamp irradiation is performed over CN-Cl100 catalyst. Fig. 6(b) clearly shows that when the CN-Cl100 catalyst is irradiated, the current density is twice higher than that in dark, further demonstrating the dramatically photoinduced OER improvement. In addition, the relatively stable current density is observed both in dark and under light irradiation during the durability test, reflecting its excellent long-term operating stability. The minor fluctuation of current density appeared during the durability test can be attributed to the formation and releasing of oxygen bubbles on the surface of catalyst. This result reveals the high OER durability of Cl-doped g-C3N4 catalysts.

3.4. Photocatalytic activity

Photocatalytic activity of bulk CN and CN-Cl100 is evaluated towards RhB photodegradation under visible light irradiation. To ensure that the equilibrium of adsorption and desorption of RhB on the surface of the catalysts, solutions with suspended photocatalysts are sonicated in the dark for 20 min before irradiation. Fig. 7(a) exhibits the variation of RhB concentration as a function of irradiation time. As Fig. 7(a) demonstrated, as adsorption equilibrium is reached after sonication in dark for 20 min, the amount of RhB adsorption is about 17% on CN-Cl100 nanorings, which is higher than that on bulk CN (7%) and could be associated with the larger exposed active area of Cl-doped products. Furthermore, CN-Cl100 nanorings show higher photocatalytic activity toward RhB degradation with the rate constant k of 0.150 h-1, which is about 4 times higher than that of bulk CN as is shown in Fig. 7(b), suggesting the enhanced photocatalytic activity of the CN-Clx nanorings.

Fig. 7.   (a) Photocatalytic degradation rate of RhB under visible light irradiation in the presence of photocatalysts and (b) first-order plots for the photogradation of RhB over CN-Cl100 and bulk CN photocatalysts.

4. Mechanism of improved performance in OER and photocatalytic activity

The catalytic performance is highly related to the interfacial charge transfer and the electrotric structure of catalysts. To illustrate the enhanced charge transfer in CN-Clx nanorings, photoelectrochemical measurements are conducted. Fig. 8(a) shows a comparison of photocurrent-time curves measured with 30 mV bias voltage for products upon three on-off cycles of visible light irradiation. It can be found that the photoresponsive current density is reversible and all CN-Clx products show enhanced photocurrent response in compare with bulk CN. As expected, CN-Cl100 nanorings exhibit the largest photocurrent benefiting from the synergistic effect of chlorine doping and 1D nanostructure formation, further demonstrating that chloride modification promotes the charge separation. Additionally, EIS is recorded to further confirm the improved charge transfer of CN-Clx catalysts. It is clearly observed from Fig. 8(b) that the arc radius of the EIS Nyquist plots for CN-Clx catalysts is smaller relative to that of bulk CN. The small arc radius reflects the decreased resistance, highlighting the promoted electrical conductivity and charge transport of CN-Clx catalysts [34]. The EIS results are consistent well with the PL spectra and photocurrent response measurement.

Fig. 8.   (a) Transient photocurrent response, (b) EIS changes of bulk CN and CN-Clx, (c) Mott-Schottky plots of bulk CN and CN-Clx and (d) band structure of bulk CN and CN-Clx.

To determine the band position of CN-Clx, the flat potential of the products is estimated by the traditional Mott-Schottky approaches. The Mott-Schottky plots and the corresponding linear fits are illustrated in Fig. 8(c). The positive slopes of all Mott-Schottky plots reflect that bulk CN and CN-Clx are n-type semiconductors. The flat potential determined from the intersection points by extrapolating their corresponding straight line to potential axis is -2.05 V, -1.74 V, -1.26 V and -1.60 V vs. Ag/AgCl (pH 6.5) for bulk CN, CN-Cl80, CN-Cl100, and CN-Cl120, respectively. As is well known, the conduction band (CB) potential is very close to the flat band potential in n-type semiconductors, so the CB potential can be estimated to be -1.40 eV, -1.09 eV, -0.61 eV and -0.95 eV vs. RHE for bulk CN, CN-Cl80, CN-Cl100, and CN-Cl120, respectively. Moreover, the Mott-Schottky plots of CN-Clx display smaller slope than that of bulk CN, further confirming that CN-Clx catalysts have higher carrier density and much faster carrier transfer. Combining with the CB potential and band gap derived from UV-vis spectra, the valence bands (VB) potentials can be calculated and the band structure of the catalysts is illustrated in Fig. 8(d). The obvious positive shift of the VB potential in CN-Clx catalysts could enhance the oxidation capability of photogenerated holes, thus is favor of the OER and pollutant degradation. Based on the above results and analysis, the more active sites exposed, enhanced charge transfer and strong oxidation capability of Cl-doped g-C3N4 nanorings directly contribute to the improved OER and polutant degradation performance.

As is generally known, four steps take place in the alkaline OER process and three intermediates will arise sequentially [36]. The first step is the adsorbed OH- captures electrons to produce •OH radicals, followed by the rearrangement of •OH radical to generate adsorbed *O in the second step, which is captured by another adsorbed OH- to form *OOH and finally further release oxygen. The high activation energy of electron transfer in the first step make it the rate-limiting step in OER process. In the absence of light irradiation, the Tafel slope for bulk CN and CN-Clx catalysts is in the range of 83-182 mV dec-1 (Fig. 5(b)), suggesting that the rate- limiting step is associated with the formation of •OH radicals in the first electron transfer step in OER [33,37]. Under visible light irradiation, photoinduced electrons and holes could be generated in CB and VB of Cl-doped g-C3N4 nanorings. As observed in Fig. 8(d), the VB potential of Cl-doped g-C3N4 nanorings shiftes downward compared with that of bulk CN, suggesting that the driving force of the photogenerated holes for oxidation process increases. The photogenerated holes in the VB of Cl-doped g-C3N4 nanorings with strong oxidative power would directly react with the surface adsorbed H2O/OH- to produce •OH radicals spontaneously. These generated •OH radicals will rearrange and directly take part in the OER process, no additional high activation energy associated with the first electron transfer step is needed. As can be observed in Fig. 5(b), the Tafel slope greatly decreases to 55 mV dec-1 upon light irradiation, indicating that the rearrangement of •OH radical instead of the first electron transfer step becomes the rate-determining step [33]. As expected, light irradiation is favor of promoting OER, and the impressive photoinduced OER improvement of CN-Cl100 nanorings correlates with the generation of •OH radicals spontaneously upon light irradiation.

To confirm the generation of •OH radicals and the promoting effect on OER, the •OH radicals produced upon light irradiation are detected by the fluorescence spectrometer with TA as a probe molecule, which reacts withOH and generates highly fluorescent TAOH with the PL peak intensity in proportion to the amount of •OH radicals. Fig. 9 illustrates the TAOH PL intensities over bulk CN and CN-Cl100. The weak signal of TAOH corresponding •OH radicals from bulk CN can be attributed to the two-electron oxidation pathway. It is worth noting that CN-Cl100 displays much higher PL intensity of TAOH than bulk CN, suggesting that the CN-Cl100 can greatly promote the spontaneous generation of •OH radicals upon light irradiation, which involve in the OER process.

Fig. 9.   Plots of TAOH PL intensities over bulk CN and CN-Cl100 samples under illumination for 4 h.

Therefore, the synergistic effect is triggered to promote the catalytic activity of CN-Cl100 nanorings by exposing more active sites, accelerating the charge transfer and altering the electronic structure via nanoring formation and chlorine doping strategy. More importantly, light irradiation could further boost their OER activities, which is associated with the •OH radicals arising spontaneously. This work opens an opportunity for using inexhaustible solar energy to get higher efficient OER electrocatalysts and photocatalysts in resolving the growing energy crisis and serious environmental issues.

5. Conclusion

Cl-doped g-C3N4 nanorings were synthesized via the preorganization of melamine and cyanuric chloride and subsequent thermal polycondensation. The integration of texture engineering by coupling with chloride doping generates the functional Cl-doped g-C3N4 nanorings with highly efficient photooxidation ability, as demonstrated by the superior OER performance with a small overpotential of approximately 290 mV at current density of 10 mA cm-2 and Tafel slope of 83 mV dec-1, as well as efficient RhB photodegradation. More importantly, the resultant Cl-doped g-C3N4 nanorings could serve as effective photo-responsive catalysts with the OER overpotential and Tafel slope futher decreasing to 150 mV and 55 mV dec-1, respectively, upon visible light irradiation. The improved catalytic performance of Cl-doped g-C3N4 nanorings can be attributed to the abundant active sites exposed, strong oxidation capability and efficient electron transfer owing to chlorine doping and nanoring structure. The further photo-induced enhancement of OER activity originates from the •OH radicals arising spontaneously on CN-Cl100 catalyst under light irradiation. This work offers a cost-effective strategy to develop photo-responsive catalysts for efficient environmental purification and photoelectrochemical energy conversion via using inexhaustible solar energy.

Acknowledgements

This work was supported financially by the National Natural Science Foundation of China (Nos. 51772085, 51471068 and U1530151), and Large instrument fund of Hunan University.


/