3.1 Characterization
3.1.1 XRD analysis
Fig.1A displayed XRD patterns of prepared samples, including original ZnAlLa-LDH, ZnAlLa-CLDH, CLDH-AM and CM-CLDH (CLDH-AM was ZnAlLa-CLDH after adsorption in amaranth solution). As depicted in pattern, spiculate and symmetric peaks at 11.5°, 23.1°, 34.4° which were assigned to (003), (006) and (009) planes indicated prepared materials’ lamellar structure26. These aforementioned peaks, together with (105), (108), (110), (113) planes’ peaks located at 38.9°, 44.6°, 59.9° and 61.2°, illustrated typical crystalline structure of LDH27. After calcination, these typical peaks disappeared, manifesting collapse of the original lamellar crystalline structure. With the increase of calcination temperature, peaks of obtained products became sharper, which may attribute to the formation of metal oxide28. After adsorption in amaranth solution, CLDH represented characteristic peaks of LDH again, which meant rehydration lamellar hydrotalcite-like structure reconstruction had been accomplished. Referred to former research, the main reason was LDH’s intrinsic “memory effect” 29.
3.1.2 FTIR
As shown in Fig. 1B, peaks at around 3450 cm-1 were belonged to O-H stretching vibration of water on surface and intercalated into layered double hydroxide30. Intensity was weaker in calcined LDH, caused by water loss via calcination. The absorption band stood at 1367 cm-1 was connected with asymmetric stretching of interlayer carbonate31. Bands located at 500-1000 cm-1 region were related to metal-oxygen and metal-hydroxyl vibrations32. Asymmetric stretching vibration of S-O was observed at around 1200 cm-133, and vibrations at region of 1100-1300 cm-1 of CLDH-AM were corresponding to SO32- vibrations31, manifesting amaranth was adsorbed by CLDH.
3.1.3 XPS
Fig. 1C and D expounded sectional high solution XPS spectra of CM-CLDH and used CM-CLDH (catalyst after photocatalysis). Full XPS spectra and high solution XPS spectra of residual elements were delivered in supporting information (Fig. S1). Signals of Zn 2p, Al 2p, La 3d, C 1s and O 1s could be detected in both samples. Zn 2p had two peaks at 1021.4 and 1044.6 eV, indicating Zn 2p1/2 and Zn 2p3/2 of Zn2+, which could be assigned ZnO and ZnAl2O428. Al 2p peak at 74.1 eV proved the existence of Al3+34.
Original La3+ peak was at 834.4 eV35. O 1s peak at 530.9 eV was assigned to metal-oxide chemical bonds18. C 1s peaks at 285.0 and 289.0 eV were attibuted to C-C and C=O species18, 36, indicating that adsorbed amaranth has been transformed into the carbon based material, which could also be proved by elemental mapping images. After photocatalytic process, binding energy peaks had positive shifts, the probable cause was interaction between Ibuprofen anions and CM-CLDH during reaction, increase of electronegativity promoted attraction between extranuclear electrons and nuclear, which led to augment of binding energy35.
3.1.4 Morphology
Fig. 2a and b demonstrated exterior structure of synthesized materials. ZnAlLa-CLDH got a sphere-like morphology, stacked by a large number of platelet-like units, which was in accordance with relevant research results37. After reaction in amaranth solution, the size of single nanoflakes became larger. Furthermore, a part of nanoflakes exhibited rougher edges, together with some little particles appeared on their surface, these phenomena were caused by adsorption of amaranth.
As displayed in Fig. 2c and d, TEM patterns of CM-CLDH certified this material had a multi-layered structure. Moreover, interplanar spacing was measured by Digital Micrograph software, three obvious interplanar distances were severally 0.251, 0.193 and 0.243 nm, corresponding to (101), (102) planes of ZnO, and that of (311) planes attributing to ZnAl2O4. The selected area electron diffraction (SAED) pattern demonstrated through calcination, polycrystalline phase had been constructed in CM-CLDH38, which could also be proved by XRD patterns.
Elemental mapping analysis had been conducted simultaneously, the results of Fig. 3 elucidated that zinc, aluminum, lanthanum, oxide and carbon were the main components of CM-CLDH. These five elements uniformly dispersed throughout sample, only a part of zinc and lanthanum accompanied with oxygen constructed sphere-like structure, which might relate to the formation of ZnO and La2O3.
3.1.5 Thermal analysis
Fig. S2 exhibited thermogravimetry curves of synthesized materials and amaranth. As for LDH, weight loss from 50 ℃ to 168 ℃ and 168 ℃ to 242 ℃ were related to evaporation of water absorbed on the crystalline surface and intercalated into layers successively4. The weight of water evaporated accounted for 17.6 % of the total weight of LDH. During 242 ℃ to 336 ℃, dehydroxylation and decomposition of some interlaminar carbonate led to 3.7 % weight loss. With continuous temperature rise from 336 ℃ to 800 ℃, residual carbonate decomposed and the lamellar structure collapsed39.
Amaranth kept stable during 50 ℃ to 300 ℃, then a rapid weight loss could be spotted with temperature risen from 300 ℃ to 800 ℃. A great number of aromatic groups might provide amaranth with a prominent thermal stability. When temperature rose to a relevant high range, the aromatic structure triggered carbonation process, which could explain sharp decrease of weight loss40. The carbonaceous products remained stable, so there was still 51.9 % residue of amaranth left at 800 ℃.
Compared with pristine LDH, CLDH-AM suffered from less weight loss in the first step, indicating weaker interaction between LDH and water. In the end, total mass loss of the composite CLDH-AM material was 21.3 %, much less than LDH (33.05 %), demonstrating the intercalation of amaranth could improve thermal stability and hydrophobicity of LDH14.
3.2 Amaranth adsorption experiments
As shown in Fig. 4a, CLDHs calcined at 400, 500, 600 ℃ showed obviously higher adsorption capacity of amaranth than LDH, driven by larger specific surface area and “memory effect” as mentioned previously.
No significant difference had been found on CLDHs’ adsorption capacity and rate, calcination temperature had a negligible effect on amaranth adsorption in this experiment. All three adsorbents could reach adsorption equilibrium at 180 min, and ZnAlLa-400CLDH got a relevant higher reacion rate, so this material was chosen for follow-up experiments and characterizaions.
Adsorption kinetics and adsorption isotherms of amaranth on ZnAlLa-400CLDH was fitted by pseudo-first-order model and Freundlich isothem model, respectively. Pseudo-first-order model was proposed by Lagergren, in this model, adsorbate-diffusion via a boundary occurred in advance of adsorption41. Adsorbent owing heterogeneous surface is the basic assumption of Freundlich isotherm. In this empirical model, adosption was a heterogenetic process, interactions between adsorbates were taken into account42. Besides, the increase of uptake capacity would lead to exponential reduction in binding energy of surficial multilayers from adsorbed ions43. Hence, a large amount of amaranth molecules might be adsorbed on the surface of ZnAlLa-400CLDH stacks by stacks, which was in consonance with SEM images.
3.3 Photocatalysis experiments
After getting adsorption equilibrium of 50 mg/L amaranth, the composite material CLDH-AM were sent to tubular furnace for calcination, temperature was set at 400, 500 and 600℃, respectively. The resulting materials were noted as 400CM-CLDH, 500CM-CLDH and 600CM-CLDH, then utilized as photocatalysts in Ibuprofen degradation experiments, detailed procedures were recorded in 2.5. Fig. 5 recorded data of photodegradation process, X axis represented time (min) and Y axis represented the ratio of sampled solution’s concentration (C) and initial concentration (C0). The consequence (Fig. 5a) inlucidated that when amount of carbon source was equal, composites calcined at 500 ℃ had a higher photodegradation efficiency.
To evaluate the effect of carbon source amount on photocatalytic efficiency, in the following step, ZnAlLa-400CLDH was stirred in amaranth solution (concentration controlled at 10, 20, 50, 100 respectively) overnight for amaranth thoroughly adsorbed by ZnAlLa-400CLDH. Then resulting composites were calcined at 500 ℃ under nitrogen flow. Samples were correspondingly noted as CM-CLDH10, CM-CLDH20, CM-CLDH50, CM-CLDH100. As displayed in Fig. 5b, with the decrease of adsorbed amaranth, the resulting catalysts exhibited higher working efficiency. As pointed out in the previous publication, excessive carbon material might trigger light harvesting competition between carbon material and semiconductor. Also, instead of improving electron transportation, excessive carbon material could act as a center for photo-induced carriers’ recombination44.
When compared with pristine CLDHs calcined at 400, 500 and 600 ℃, CM-CLDH10 still presented obvious advantages, curves of C/C0 dropped rapidly especially in the first 60 minutes (Fig. 5c). Pseudo-first-order equation derived from Langmuir-Hinshelwood model was used to evaluate kinetics of photocatalytic process45, the formula could be expressed as follows:
See formula 5 in the supplementary files.
where kapp represented the apparent pseudo-first-order rate constant, C0 and C were pollutant concentration at the start and the end of reaction. Higher kapp value normally indicated faster photodegradation process of a catalyst, kapp of CM-CLDH10, pure ZnAlLa-CLDHs and their pseudo-first-order kinetic plots were shown in Fig. 5d. Obviously, CM-CLDH10 displayed the highest photodegradation rate. Additionally, in CM-CLDH10 systhem, it removed more than 90 % Ibuprofen with the least time comsumption, which could be regarded as the most effective catalysts in this work. Moreover, CM-CLDH10 was collected via suction filtration after photocatalysis, recycled sample was used in IBF photodegradation experiments for another four times to evaluate its recyclability and stability. As depicted in Fig. S3, after five runs, photodegradation efficiency of CM-CLDH10 descreased from 94.9 % to 89.7 %, indicating promising recyclability and stability of synthezised catalyst.
3.4 Photodegradation mechanisms
3.4.1 Effect of solution pH
Original pH of 5 mg/L Ibuprofen solution in this experiment was 5.90. Nitric acid and sodium hydroxide (0.5 mol/L) were injected by syringe to regulate pH of Ibuprofen at 3, 5, 7 and 9 (3.06, 4.96, 6.96, 9.05 in effect). Photodegradation efficiencies over CM-CLDH10 under these four pH conditions were provided by Fig. 6a. With the decrease of pH values, CM-CLDH10 presented higher photodegradation efficiency, at pH = 3.06 more than 95 % of target comtaminant could be removed in 60 minutes, faster than that of other conditions. Interstingly, photodegradation rate gradually declined with the increase of pH value. On the basis of former researches, acidic solutions could promote photodegradation through acceleration of ·OH manufacture45. Moreover, H+ would react with ·O2- to generate HO·2, both ·O2- and HO·2 had clear advantage in ring clevage of aromatic groups46-47. According to scavenger experiment, effect of ·OH could be neglected, so attack of ·O2- and HO·2 on Ibuprofen’s aromatic ring might be indispensible during photodegradation. On the other hand, with the increase of pH, holes’ oxidation ability would diminish due to cathodic displacement happened on valence band position45. And too many OH- could also form hydrogen oxidation film during photocatalytic process, which hampered efficiency of catalysts17.
3.4.2 Radical species detection
Radical species (RS) (mainly hydroxyl radical (·OH) and superoxide radical (·O2-)) and singlet oxygen (1O2) play a vital role in photodegradion on organic contaminants. RS scavengers like tertiery buthanol (BuOH), ethylenediaminetetraacetic acid disodium (EDTA-2Na), benzoquine (BQ), K2Cr2O7 and L-histidine were added into reaction system respectively to probe ·OH, holes (h+), ·O2-, electrons (e-) and 1O2. As depicted in Fig. 6b, involvement of L-histidine, K2Cr2O7 and BQ distinctly hampered efficiency and rate of photodegradation, which reflected singlet oxygen, electron and superoxide radical were dominant radicals during photocatalytic process48-50. On the contrary, addition of EDTA-2Na and BuOH showed limited inhibiting effect, revealing photo-induced holes and hydroxyl radicals’ attack on Ibuprofen was relatively weak50-51.
Electronic spin resonance (ESR) test was also performed to investigate main radical species during photocatalytic process. As for hydroxyl radical and singlet oxygen, 5 mL water and 5 mg catalyst were added into quartz reacion dish, then severally injected 5 μL 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) and 2,2,6,6-tetramethylpiperidine (TEMP) as spin-trapping agent. The resulting mixture was irradiated under a ultraviolet lamp for 5 minutes before being sampled in capillary tube for detection. Method for superoxide radical was similar, the only difference was 5 mL water being substituted by 5 mL DMSO (dimethyl sulfoxide).
As shown in Fig. 6c and d, both ZnAlLa-500CLDH and CM-CLDH10 exhibited characteristic peaks of DMPO-·O2- and TEMP-1O2 with intensity ratio of 1:1:1:1:1:1 and 1:1:1. And relative intensity of DMPO-·O2- was higher in CM-CLDH10, suggesting involvement of carbonaceous material might lead to improvement of photodegradation. Moveover, after ultraviolet irradiation for 5 minutes, only CM-CLDH10 generated sufficient ·OH to be captured by DMPO, this evidence also reflected that combination with carbonaceous material could make CLDH present higher radical productivity, which might promote its photocatalytic performance.
3.5 Photoelectronic property
Photoluminescence (PL) singal would be generated at the time of photo-induced electrons and holes’ recombination52, the higher intensity indicated lower separation efficiency of photo-generated carriers. With excitation wavelength set at 350 nm, PL signals of ZnAlLa-500CLDH and CM-CLDH10 were detected in the wavelength range of 365–600 nm (Fig. 7a). The consequence reflected that PL signal intensity of CM-CLDH10 was lower, manifesting better separation ability of photo-induced electrons and holes, which might be an important factor for better photocatalytic performance over CM-CLDH10. Photocurrent transient curves were obtained, the transient photocurrent responses of CM-CLDH10 was much higher than that of ZnAlLa-500CLDH, revealling carbonaceous material had a positive effect on separation of electrons and holes53. This conclusion was further proved by transient photocurrent response. As reflected in Fig. 7b, stronger intensity has been spotted on CM-CLDH10, which was triggered by higher separation efficiency of photo-induced h+ and e-54.
Electrochemical impedance spectroscopy (EIS) curves were applied to assess electrochemical performances of synthesized materials. In this test, 1 mol·L-1 sodium hydroxide carbon rod and saturated calomel electrode were utilized as electrolyte, counter electrode and reference electrode.
Fig. 7c exhibited EIS plots of both catalysts, applied potential was 1.5 V (versus RHE) from 1000 kHz to 0.01 Hz in 1.0 mol·L-1 sodium hydroxide electrolyte. Much larger semicircle’s radius in Nyquist plot was connected with worse charge transfer resistance55. On the basis of PL, photocurrent and EIS test, photo-induced electrons and holes were generated mainly by CLDH’s MO6 octahedra which shared Metal-Oxygen bonds56, introduction of carbonaceous material could reduce recombination of photo-induced carriers. Nevertheless, its conductive activity was lower than pure metal or metal oxide, which might explain why photodegradation efficiency was falling down with the increase amount of carbonaceous material on ZnAlLa-CLDH.
According to UV-vis/DRS spectra, with increase of adsorbed amaranth amount, UV absorbance spectrum of resulting calcination product shifted to longer wavelength region, which suggested combination of carbonaceous material improved light adsorption57. Kubelka-Munk method driven transformed reflectance spectras were attained to measure energy band (Eg) of ZnAlLa-500CLDH and CM-CLDH1058, Eg values were 3.37 eV and 3.21 eV, involvement of carbonaceous material narrowed band gap, which could improve photocatalytic activity. Mott-Schottky plots via implementing Mott-Schottky measurement was applied to estimate conduction band potential (ECB) of prepared materials (saturated calomel electrode as reference electrode, 0.2 M Na2SO4 as electrolyte, pH = 7). As shown in Fig. 8c and d, ECB values of n-type ZnAlLa-500CLDH and CM-CLDH10 were -0.71 V and -1.27 V (V vs NHE), after transformation from E (SCE) to E (NHE)59. As is known to all, Eg = EVB – ECB, calculated value of valence band potential (EVB) were 2.66 V and 1.9 V.
According to aforementioned analysis, potential mechanism of photodegradation could be listed as follows:
See formulas 6 - 12 in the supplementary files.