Fabrication of AgCl/Ag3PO4/graphitic carbon nitride heterojunctions for enhanced visible light photocatalytic decomposition of methylene blue, methylparaben and E. coli

Herein, a novel ternary nanocomposite AgCl/Ag3PO4/g-C3N4 was successfully synthesized via sedimentation precipitation and ion exchange method. The photocatalytic performance of the as-prepared AgCl/Ag3PO4/g-C3N4 nanocomposite was investigated via photocatalytic degradation of methylene blue (MB), methylparaben (MPB) and inactivation of E. coli under visible light irradiation. The AgCl/Ag3PO4/g-C3N4 composite presented the optimal photocatalytic performance, degrading almost 100% MB and 100% MPB, respectively. The excellent stability of AgCl/Ag3PO4/g-C3N4 was also verified in the cycle operations; the degradation efficiency of MPB could still be maintained at 85.3% after five cycles of experiments. Moreover, the AgCl/Ag3PO4/g-C3N4 composite displayed more superior photocatalytic inactivation efficiency with 100% removal of E. coli (7-log) in 20 min under visible light irradiation. The efficient photo-generated charge separation originated from a strong interaction in the intimate contact interface, which was confirmed by the results of photocurrent and EIS measurements. In addition, radical trapping experiments revealed that hole (h+) was the predominant active species in the photocatalytic system. Based on the experimental results, a photocatalytic mechanism for the degradation of parabens over AgCl/Ag3PO4/g-C3N4 was also proposed. We believe that this work provides new insights into the multifunctional composite materials for the applications in solar photocatalytic degradation of harmful organic compounds and common pathogenic bacteria in wastewater.


Introduction
Water pollution has become a serious hazard to public health and ecosystems in the world. [1][2][3] The composition of wastewater contains many kinds of toxic chemicals, including organic dyes, endocrine disrupting compounds (EDCs), and common pathogenic bacteria, and various strategies have been explored for solving the issue. 1,4 The increasing emergence of organic dyes from textile and food industries has become one of the most important types of water contaminants. Among the EDCs, methylparaben (MPB) with estrogenic or androgenic activity at very low concentrations may cause potential and real detrimental effects on the endocrine systems of humans and wildlife. 5,6 Moreover, infectious diseases caused by the harmful waterborne pathogens threaten the public health. According to the latest report of the World Health Organization, about 884 million people in the world use untreated water, which make waterborne diseases the leading cause of death. 7,8 Therefore, it is a great challenge to eliminate toxic chemicals and diseasecausing waterborne pathogens using a single material, simultaneously. One way of achieving this objective is to explore new bifunctional nanocomposites capable of efficiently degrading organic pollutants and eradicating pathogenic bacteria via an eco-friendly technique. 9,10 Over the past decades, photocatalytic technology has been widely regarded as the most promising technology to solve environmental pollution and energy shortage issues. 11,12 Semiconducting photocatalysts, especially those with high catalytic efficiency and good stability under visible light irradiation, have been widely applied in the degradation of organic contaminants owing to their ability to directly harvest solar energy and excellent high-visible-light-driven photocatalytic activities. 13 Recently, Ag-based photocatalysts have been widely reported for the treatment of persistent pollutants in wastewater, such as AgX (X ¼ Cl, Br, and I), Ag 3 CO 3 , Ag 3 PO 4 and CdS, which show a much faster photodegradation rate than the conventional TiO 2 . 14-17 Among them, silver orthophosphate (Ag 3 PO 4 ), an efficient n-type photocatalytic material, has high quantum efficiency (up to 90%) and indirect band gap of 2.36 eV, making it a prospective visible-light induced photocatalyst. 18,19 However,

Synthesis of samples
All materials (analytical purity) were obtained from Macklin and used without further purication.
Synthesis of g-C 3 N 4 . 80 mmol of melamine was placed in a vacuum tube furnace, heated to 550 C in a muffle furnace and maintained for 4 h, and then cooled to room temperature naturally. The product was washed several times with deionized water and ethanol to remove the soluble reactants and impurities. Aer drying in vacuum at 60 C, a yellow agglomerate was obtained and ground into powder for further use.
Synthesis of Ag 3 PO 4 . 5 mmol NaH 2 PO 4 was dissolved in 50 mL deionized water at room temperature, and then 0.015 mmol AgNO 3 was added dropwise to the NaH 2 PO 4 solution. The color of the solution changed to yellow and aer stirring for 5 h, the obtained yellow Ag 3 PO 4 precipitate was washed, and then dried at 60 C to obtain Ag 3 PO 4 nanoparticles.
Synthesis of AgCl/Ag 3 PO 4 . 1 g Ag 3 PO 4 was dispersed in 50 mL deionized water by ultrasonication, then 0.1 M NaCl solution was dropped into the Ag 3 PO 4 dispersion, and aer vigorously stirring for 2 h the precipitate was collected, washed and dried. The nal photocatalyst AgCl/Ag 3 PO 4 was obtained.
Synthesis of AgCl/Ag 3 PO 4 /g-C 3 N 4 . 0.2 M g-C 3 N 4 solution was added to the Ag 3 PO 4 /AgCl solution under mechanical stirring, followed by continuous stirring for 5 h. Subsequently, the precipitate was collected, washed with water and ethanol to neutrality in a low-speed centrifuge and dried in an electro thermal blast drying oven at 75 C. Finally, the product was collected, ground and labeled as AgCl/Ag 3 PO 4 /g-C 3 N 4 . The schematic diagram of the preparation process is shown in Scheme 1.

Characterization of the photocatalyst
The crystallographic properties of AgCl/Ag 3 PO 4 /g-C 3 N 4 composites were characterized on a DX-2700 X-ray diffractometer (XRD) at a scanning step size of 0.03 in the 2q range of 20 to 80 . The surface morphology of the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite was characterized by scanning electron microscopy (JSM-7800F) and transmission electron microscopy (JEOL JEM-2010). Ultraviolet-vis (UV-vis) diffuse reectance spectra were measured on a UV-vis Cary 50 Bio with a scanning step in the wavelength range of 200-800 nm. The photoluminescence (PL) spectra were investigated on a F-4600 spectrophotometer with an excitation wavelength of 390 nm. The XPS analyses were carried out on a K-Alpha Photoelectron Spectrometer with an Xray source of Al Ka (hv ¼ 1253.6 eV). All the binding energies were calibrated internally by C 1s at 284.8 eV. The photocurrent and resistance measurements were performed on an electrochemical workstation (CHI660E, CHI Instruments, Inc., China).

Photocatalytic degradation experiments
The photocatalytic activities of resultant nanocomposites were estimated by the degradation of MB, MPB and E. coli under visible light irradiation.
In a typical procedure, 50 mg photocatalyst was dispersed in 100 mL 20 mg L À1 MB and MPB solution. The light source was a 300 W metal-halide lamp (PLS-SXE300, Shanghai Bilang Co., Ltd., China) with a UV cut-off lter (l < 400 nm) and the illumination intensity was kept ca. 10 mW cm À2 . In all experiments, the temperature of the reaction was maintained at 25 AE 1 C by the water continuously circulated in the jacket surrounding the reactor. Before irradiation, all the reaction samples were stirred for 30 min in dark to obtain an adsorption-desorption equilibrium. Aerwards, aliquots (2 mL) of dispersion was collected and ltered at regular intervals of the irradiation time, and the MB and MPB concentrations were detected by using a UV-vis spectrophotometer at the wavelengths of 235 and 369 nm, respectively. For the assessment of photocatalytic activities of the resultant photocatalyst, the degradation efficiency was calculated by C/C 0 Â 100%, where C and C 0 are the concentrations of MB and MPB at a real-time t and the initial concentration, respectively. For comparison, light (without photocatalyst) and dark controls (without light) were also performed.
The photocatalytic disinfection was carried out using a 300 W xenon lamp with a cut-off lter of 420 nm and the illumination intensity at around 10 mW cm À2 . All glass apparatuses were autoclaved at 120 C for 20 min for the disinfection experiments to ensure sterility. Aer incubation in 10% nutrient broth solution at 30 C for 18 h with shaking, the bacterial cell was washed with sterilized saline. The cell density was adjusted to 1.5 Â 10 7 colony forming units per milliliter (cfu mL À1 ). Aer the photocatalytic treatment, an aliquot of the reaction solution (5 mL) was taken out at different time and immediately diluted with sterilized saline solution (0.9% NaCl). The appropriate dilution of the sample was spread on the nutrient agar and incubated for 24 h at 37 C. All the experiments were performed in triplicates.
Fluorescence spectroscopy. Each aliquot of bacteria collected before and aer the photocatalytic treatment at different time was stained with typical cell-labeling dye mixtures of SYTO 9 (a green-uorescent nucleic acid dye) according to the recommended procedure in the bacterial viability kit to detect living and dead bacterial cells, respectively. Aer the incubation in the dark at 25 C for 15 min, the samples were transferred to cover slips. Fluorescence spectroscopy of the samples was performed with a uorescence microscope (Nikon ECCLIPSE 80i, Japan), which was equipped with a Spot-K slide CCD camera (Diagnostic Instruments Inc., USA) and a lter block N UV-2A consisting of excitation lter Ex 400-680 (Nikon, Japan). A FIT with the intensity of 100 mW m À2 was selected as the visible light source.  Fig. 1. It could be found clearly ( Fig. 1a) that the structure of AgCl/Ag 3 PO 4 particles was regular ellipse, and g-C 3 N 4 act as a support for both AgCl and Ag 3 PO 4 particles, when it was introduced into the AgCl/Ag 3 PO 4 composites; the composite displayed an agglomerated bulk with a rough surface, which could enhance the surface area and led to a better photocatalytic performance. 36 The elemental mapping images (Fig. 1c-i) displayed the uniform distribution of elements C, Ag, Cl, P, O and N in the Ag 3 PO 4 /AgCl/g-C 3 N 4 composite.

Results and discussion
XRD. The crystalline structure and phase composition information of the as-prepared samples were further characterized by XRD. As shown in  38 The diffraction peak at 27.6 can be attributed to the (110) plane of g-C 3 N 4 . 18 It was clear that the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite consisted of AgCl, Ag 3 PO 4 and g-C 3 N 4 phases, indicating successful synthesis of a high-purity composite.
BET. N 2 adsorption-desorption test was performed to study the textural properties of the materials. Fig. 2 shows the nitrogen adsorption/desorption isotherms of AgCl/Ag 3 PO 4 /g-C 3 N 4 , and Table 1 shows the BET specic surface areas obtained based on the results shown in Fig. 2. The isotherms of AgCl/Ag 3 PO 4 /g-C 3 N 4 are of type-II classication and it has the typical hysteresis loops of mesoporous materials. At the same time, it can be concluded from the pore size distribution diagram that the pore size distribution is 80-110 nm, which shows that we have prepared a ternary mesoporous material to greatly improve its structural stability.
XPS. To further investigate the surface elemental composition and the valence states of Ag 3 PO 4 /AgCl/g-C 3 N 4 composite, XPS was carried out. It can be seen from Fig. 3a that the sample is composed of P, Cl, C, N, O and Ag elements, which is consistent with the results of EDS mapping. In the Ag 3d spectrum (Fig. 3b), two peaks were observed at 367.77 and 373.77 eV in the Ag 3d region, which were assigned to Ag 3d 5/2 and Ag 3d 3/2 orbitals typical of Ag(I), respectively. 39 The binding energy peaks of Cl 2p (Fig. 3c) at 198.08 and 199.68 eV were associated with Cl 2p 3/2 and Cl 2p 1/2 of Cl À in AgCl. 40,41 In addition, N 1s (Fig. 3d) showed a peak at 398.8 eV, and two photoelectron peaks were tted at 398.7 eV and 399.7 eV, which  were derived from the sp 2 hybridized N in C]N-C and the sp 3 hybridized N in N-[C] 3 . 42 Additionally, as shown in Fig. 3e, the peak at 531.24 eV could be assigned to O 1s of the lattice oxygen of Ag 3 PO 4 . 43 As shown in Fig. 3f, two individual bands at 132.8 and 133.4 eV could be attributed to the electron orbitals of P 2p 3/2 and P 2p 1/2 , respectively.

Optical and electrical properties of the material
UV-vis and band gap analyses. The optical properties of the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite were recorded using UV-vis spectroscopy and the results are displayed in Fig. 4a. The absorption band edge of pure Ag 3 PO 4 was found at around 530 nm, and the absorption capability was diminished at wavelengths >530 nm. 11,44,45 The light absorption capability of g-C 3 N 4 is weak in both UV and visible regions. When g-C 3 N 4 was introduced into the Ag 3 PO 4 /AgCl composite, the band edge of the absorption spectra also showed an obvious red-shi, and  the visible light absorption capacity in the range of 400-500 nm was further enhanced, resulting in the promotion of photocatalytic activity. The absorption of Ag 3 PO 4 /AgCl/g-C 3 N 4 in the visible light region is apparently enhanced, which can be attributed to the interaction between the valence band and conduction band of g-C 3 N 4 and Ag 3 PO 4 . Compared to the pure Ag 3 PO 4 , AgCl and g-C 3 N 4 , the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite photocatalyst has a stronger absorption capacity for visible light, improving the conversion efficiency of visible light and further improving the photocatalytic activity of the composite material. Moreover, the band gap energy can be estimated from a plot of (ahv) 1/2 versus photo energy (hv). The x-intercept of the tangent line gives an approximation of the band gap energy of the samples. As shown in Fig. 4b, the band gap energy of the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite is about 2.3 eV, which is smaller than that of AgCl and Ag 3 PO 4 , indicating that the AgCl/Ag 3 PO 4 / g-C 3 N 4 composite can respond better in the visible region of 540 nm.

Photoelectrochemical (PEC) measurements.
Photoelectrochemical properties were measured to investigate the excitation and transfer of photo-generated carriers and the interface reaction ability of the charges in photocatalytic materials. Generally, the smaller arc size reects smaller charge transfer resistance at the interface. 46 In Fig. 4c, the AgCl/Ag 3 PO 4 / g-C 3 N 4 composite shows the smallest radius of arc, suggesting that the heterostructure has a lower resistance and the fastest interface charge transfer as well as the best photo-generated carrier separation efficiency. Additionally, Fig. 4d displays the current-time (I-t) curves of AgCl/Ag 3 PO 4 /g-C 3 N 4 , AgCl and g-C 3 N 4 with several 30 s light on/off cycles. Compared to g-C 3 N 4 and AgCl/Ag 3 PO 4 composite, the ternary composite AgCl/ Ag 3 PO 4 /g-C 3 N 4 showed an increased photocurrent response. The formation of photocurrent is mainly by the separation and diffusion of photo-generated electron-hole pairs from the internal structure of the photocatalyst to its surface and free charge acceptors in the electrolyte. 47 Therefore, the increased photocurrent of AgCl/Ag 3 PO 4 /g-C 3 N 4 indicates more efficient separation and less recombination of the photo-generated electron-hole pairs, which is conducive to enhance photocatalytic performance.

Photocatalytic degradation and disinfection performance
The photocatalytic performance of AgCl/Ag 3 PO 4 /g-C 3 N 4 composite was evaluated by the photocatalytic degradation of MB and MPB under visible-light irradiation. Fig. 5a demonstrate that 91% of MB could be degraded by AgCl within 30 min, while Ag 3 PO 4 and AgCl/Ag 3 PO 4 could completely degrade MB within 30 min; however, aer combining with g-C 3 N 4 , the AgCl/ Ag 3 PO 4 /g-C 3 N 4 composite showed enhanced photocatalytic activity, leading to 100% degradation of MB within 20 min. To quantitatively explore the corresponding MB degradation kinetic curves of the samples, the data were matched with a rst-order model. As shown in Fig. 5b, the pseudo-rst-order rate constants (k) of AgCl, Ag 3 PO 4 , AgCl/Ag 3 PO 4 and AgCl/ Ag 3 PO 4 /g-C 3 N 4 composites were calculated to be 0.07, 0.08, 0.16 and 0.24 min À1 , respectively. The k value of the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite was about 3.42 and 3.4 times higher than that of pure AgCl and Ag 3 PO 4 , respectively. Moreover, the timedependent UV-vis absorption spectra of the photocatalytic degradation of MB by the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite was studied, as shown in Fig. 5c, with the absorption peak at about 660 nm corresponding to the characteristic absorption of MB. Aer 10 min of visible light irradiation, the intensity of the absorption peak at 660 nm was signicantly decreased, and with further increase of the illumination duration, the absorption peak at 660 nm fades away. The results indicate that the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite can completely oxidize parabens under visible light.
The samples was also estimated by degradation of MPB to evaluate the photocatalytic capacity of the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite, the samples were also estimated by the degradation of MPB. As shown in Fig. 5d, only very little amount of MPB was degraded in the dark and in the absence of catalyst, indicating that MPB was considerably stable and self-photolysis can be neglected. Among Ag 3 PO 4 , AgCl, AgCl/Ag 3 PO 4 and AgCl/Ag 3 PO 4 / g-C 3 N 4 composite, the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite exhibited the highest photocatalytic performance, and about 100% of MPB was degraded within 30 min; in contrast, 94.8%, 92.3% and 60% of MPB was degraded within 40 min by AgCl/Ag 3 PO 4 , Ag 3 PO 4 and AgCl, respectively. The excellent photocatalytic performance of AgCl/Ag 3 PO 4 /g-C 3 N 4 could be ascribed to the high migration efficiency of the photo-induced electron-hole pairs. As shown in Fig. 5e, the rate constant for the photocatalytic degradation of MPB with AgCl/Ag 3 PO 4 /g-C 3 N 4 was 0.1 min À1 , which was 1.43, 1.67, and 10 times higher than those with AgCl/Ag 3 PO 4 (0.07 min À1 ), Ag 3 PO 4 (0.06 min À1 ) and AgCl (0.01 min À1 ), respectively. We can conrm that the AgCl/ Ag 3 PO 4 /g-C 3 N 4 composite is more effective than the other samples. Additionally, the total organic carbon (TOC) experiment was further carried out to track the degradation of MPB during the photocatalytic reaction process. Fig. 5f shows the results of TOC mineralization efficiency on the photocatalytic degradation of MPB with different samples. The mineralization yield of Ag 3 PO 4 /AgCl/g-C 3 N 4 can reach up to 80.27% within 40 min, which is higher than that of Ag 3 PO 4 /AgCl (72.3%), Ag 3 PO 4 (68%) and AgCl (36.8%). The superiority of AgCl/ Ag 3 PO 4 /g-C 3 N 4 is understandable since it has stronger light absorption and faster charge transfer rate. This result also indicated that MPB was indeed photocatalytically degraded into inorganic substances (such as H 2 O and CO 2 ). Therefore, AgCl/ Ag 3 PO 4 /g-C 3 N 4 is an excellent photocatalytic composite material that could degrade organic parabens into inorganic substances.
It is well-known that the stability of a photocatalyst is essential for practical applications. The recycling experiments were performed for ve times for the degradation of MPB over AgCl/Ag 3 PO 4 /g-C 3 N 4 to evaluate the photocatalytic stability. As shown in Fig. 6a, the degradation of MPB was 100%, 94.8%, 92.6%, 91.9% and 85.3%, respectively. In contrast, the degradation over AgCl/Ag 3 PO 4 decreased from 92% to 76% aer ve recycling runs. It can be found that the stability of the ternary composite Ag 3 PO 4 /AgCl/g-C 3 N 4 material with g-C 3 N 4 as the matrix material is further improved compared to that of the Ag 3 PO 4 /AgCl. In addition, no additional characteristic peaks were observed in the XRD patterns (Fig. 6b) of Ag 3 PO 4 /AgCl/g-C 3 N 4 aer cycling, which indicates that the crystal structure did not change signicantly aer the photocatalytic reaction. This further proved that the sample has adequate stability without remarkable reduction of photocatalytic activity under visible light irradiation. Fig. 7 shows the disinfection efficiency of E. coli by AgCl/ Ag 3 PO 4 /g-C 3 N 4 for different irradiation time. A mass of E. coli was alive without light irradiation. However, more than half of the E. coli was killed aer being irradiated for 10 min, and all the E. coli was killed aer 20 min irradiation. This result indicates that the AgCl/Ag 3 PO 4 /g-C 3 N 4 composite has excellent photocatalytic anti-bacterial properties.

Photocatalytic mechanism analysis
In order to get insights into the photocatalytic mechanism of the Ag 3 PO 4 /AgCl/g-C 3 N 4 composite, it is necessary to identify and assess which reactive species played the most prominent role in the photodecomposition of MPB. The trapping experiments of radicals were conducted to measure the effect of the active substances on the nal degradation results. In the present work, sodium oxalate, catalase, ascorbic acid (VC), isopropanol and potassium dichromate were used as scavengers for photo-generated holes (h + ), hydrogen peroxide (H 2 O 2 ), superoxide radicals (cO 2À ), hydroxyl radicals (cOH) and photogenerated electrons (e À ), respectively. [48][49][50] As shown in Fig. 8a, the degradation of MPB was 34.4%, 65.6%, 88.6%, 82.1% and 83.5%, respectively, indicating that h + was the main reactive species in the MPB photodecomposition process. These results demonstrate that photo-generated h + are the dominant species for the decomposition of MPB, while cOH, cO 2À , H 2 O 2 and e À played relatively minor roles in the MPB photocatalytic decomposition process.
Based on above analyses, the possible photocatalytic mechanism of the AgCl/Ag 3 PO 4 /g-C 3 N 4 nanocomposite is shown in Fig. 8b. AgCl/Ag 3 PO 4 /g-C 3 N 4 was excited to photogenerate electrons and holes under visible light irradiation. The staggered band gaps promote electron transfer, so g-C 3 N 4 with a more negative CB (À1.15 eV) could easily transfer the photogenerated electrons to the CB of Ag 3 PO 4 (+0.45 eV) and AgCl (À0.22 eV). At the same time, Ag 3 PO 4 with more positive VB energy would transfer the photogenerated holes to the VB of g-C 3 N 4 . Because AgCl has a wide band gap and cannot be activated by visible light, it is used as an important electron acceptor to capture and shuttle electrons that would further promote the separation of electron-hole pairs of Ag 3 PO 4 and g- C 3 N 4 . 51,52 Therefore, the AgCl/Ag 3 PO 4 /g-C 3 N 4 photocatalyst shows enhanced photocatalytic activity compared with pure AgCl or Ag 3 PO 4 . Under visible light irradiation, the excited photo-generated holes could directly oxidize the pollutants; it also could react with O 2 and nally generate reactive cOH, which will induce the degradation of the organic pollutant. Then, the accumulated electrons on the CB could reduce the O 2 to form superoxide radicals (cO 2 ), and then participate in photooxidation. In this way, the accumulation of electrons in the CB of Ag 3 PO 4 can be transfer effectively, which prevents the decomposition of photo-induced corrosion and improves the stability of Ag 3 PO 4 .

Conclusion
In summary, a novel ternary AgCl/Ag 3 PO 4 /g-C 3 N 4 composite was prepared by loading Ag 3 PO 4 /AgCl onto g-C 3 N 4 with a large specic surface area. The AgCl/Ag 3 PO 4 /g-C 3 N 4 composite showed excellent photocatalytic efficiency for the removal of MB, MPB and E. coli. The degradation rates of MB and MPB over Ag 3 PO 4 /AgCl/g-C 3 N 4 composite can both reach 100% within 20 min under visible light irradiation, and the degradation ratio of MPB remained 85.3% even aer ve cycles. Besides, it also showed good performance in the inactivation of E. coli within 20 min. The enhanced photocatalytic performance and stability could be ascribed to the combination of AgCl/Ag 3 PO 4 and g-C 3 N 4 , which effectively promotes the transfer efficiency of the photogenerated carriers and inhibits the recombination of the photo-generated charge carriers during the photocatalytic reaction. Due to the larger specic surface area of g-C 3 N 4 , efficient separation of the photo-generated electron-hole can be achieved, which would improve the visible light conversion efficiency. This work not only demonstrates that the composite of Ag 3 PO 4 and new carbon materials can enhance the photocatalytic properties and stability, but also provides an insight for the preparation of new high-performance photocatalysts.

Authorship contribution statement
Haishuai Li conceived the experiments and wrote the manuscript. Linlin Cai wrote the manuscript. Xin Wang conducted the EIS and transient photocurrent response measurement and analysis. Huixian Shi supervised the project and reviewed the manuscript. All the author contributed to the data analyses.

Conflicts of interest
The authors declared that they have no conicts of interest to this work. We declare that we do not have any commercial or associative interest that represents a conict of interest in connection with the work submitted.