Supplementary information for Towards sustainable TiO2 photoelectrodes based on cellulose nanocrystals as a processing adjuvant [Dataset]
20 figures, 2 tables.-- TEM, size distribution and elemental analysis of CNC: TEM images of the type II CNC are presented in Figure S1 and the distribution of diameters and heights of the CNC are in Figure S2. It can be observed the characteristic shape and size of this type of CNC, with a mean diam...
| Autores: | , , , , , , , |
|---|---|
| Tipo de recurso: | conjunto de datos |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/360680 |
| Acceso en línea: | http://hdl.handle.net/10261/360680 |
| Access Level: | acceso abierto |
| Palabra clave: | Cellulose nanocrystals TiO2 Photoelectrochemistry UV-vis Spectroscopy http://metadata.un.org/sdg/7 http://metadata.un.org/sdg/13 Ensure access to affordable, reliable, sustainable and modern energy for all Ensure sustainable consumption and production patterns Take urgent action to combat climate change and its impacts nanocellulose |
| Sumario: | 20 figures, 2 tables.-- TEM, size distribution and elemental analysis of CNC: TEM images of the type II CNC are presented in Figure S1 and the distribution of diameters and heights of the CNC are in Figure S2. It can be observed the characteristic shape and size of this type of CNC, with a mean diameter 28 ± 13 nm and a length of 58 ± 15 nm. Type II CNCs were analyzed to determine their composition by elemental analysis and the results are presented in Table S1. The most characteristic result is the relative high mass percentage of sulfur of 3.45%, which is clearly higher than the value in type I CNCs prepared by the same approach. The elemental composition (C, H, N and S) of the CNCs was determined using a LECO 628 elemental analyzer (Velp Scientifica). The elemental analysis was performed in triplicate to ensure reproducibility, and the average values were reported. Commercial TiO2 paste (TiO2-P): Film fabrication: The commercial paste (TiO2-P) was dried in an oven at 120 ºC overnight to remove organic solvents and get a material ready for subsequent solid stated characterization such as XRD and TGA measurements. For the preparation of the photoelectrodes using the commercial paste, an optimized screen-printing procedure was employed following the instructions provided by the supplier. The paste was applied to cover a 1 cm2 surface area of the FTO substrates. Subsequently, the electrode was thermally sintered in an oven with the following temperature profile: 5 minutes at 325 ºC, 5 minutes at 375 ºC, 5 minutes at 450 ºC, and 15 minutes at 500 ºC, under an air atmosphere, according to the instructions from the provider. Prior to the photoelectrochemical (PEC) evaluation, the TiO2 film was activated by treating it at 500 ºC in air for 30 minutes. Preparation of TiO2(NH4OH) and solid material: TiO2(NH4OH) dispersions were prepared by mixing 50 mg of anatase powder, 300 µL of commercial aqueous ammonia (30%) and 19.7 mL of ultrapure water. Then, the resulting mixture was homogenized in an ultrasonic bath for 1 h. Dispersions were freeze-dried to obtain powder materials for further characterization. UV-Vis of the employed materials: Figure S5a shows the transmittance curves of various aqueous dispersions, including freshly prepared TiO2-NPs, TiO2(NH4OH), CNC materials, and the TiO2(NH4OH) material after 24 hours of dispersion preparation. At the selected wavelength of 360 nm, the TiO2- NPs dispersion exhibits the highest transmittance, indicating an unstable system with most of the material settling down. Conversely, the TiO2(NH4OH) material shows lower transmittance values, which moderately increases after 24 hours, evidencing successful dispersion of the TiO2-NPs when ammonia was used. Figure S5b shows the variation of the transmittance for the TiO2(NH4OH), TiO2-CNC and TiO2-CNC(NH4OH) materials with time. The results demonstrate the effective role of ammonia in facilitating the dispersion of TiO2-NPs, although CNC exhibited superior stabilization efficiency. Notably, the combination of ammonia and CNC yields the most stable aqueous dispersions, as evidenced by consistent transmittance values even after 24 hours. TEM of the TiO2(NH4OH) material: TEM images in Figure S6 reveal the presence of TiO2 aggregates, typically smaller than 100 nm. It seems that the addition of NH4OH during the preparation of the TiO2-NPs dispersion effectively disrupt the formation of large aggregates. Thermogravimetric analysis of CNC: All thermogravimetric analysis (TGA) of powder materials were carried out under air atmosphere in a Libra F1 (Netzsch) thermobalance using a ramp of 10 ºC/min. Figure S7 confirms that CNC have been completely eliminated, as there is no residual mass after the experiment. Thermogravimetric analysis of the TiO2-CNC material: As the CNC residue is 0% at 800 ºC and approximately of 5% at 450 ºC, it can be deduced that CNC have been completely removed from the TiO2 matrix during the sintering step (450 ºC, 2 hours). The TiO2-CNC material reveals a 50% of residual mass, which confirms that all CNC were removed at a 1:1 TiO2 : CNC ratio (Figure S8). Differences in surface morphology between films prepared from commercial TiO2- P and TiO2-CNC(NH4OH): Figure S9 shows the results from the profilometry measurements of TiO2-P and TiO2- CNC(NH4OH) films. Both films displayed remarkable differences in surface characteristics. While TiO2-P film shows a smooth surface, the film obtained from the TiO2-CNC(NH4OH) dispersion exhibits a significant higher roughness. This disparity is clearly a consequence of the respective fabrication methodologies. In the case of the TiO2-P, screen printing was employed, allowing the particles to accommodate under the gentle pressure of the printing blade, resulting in a roughness of 110 nm (Figure S9a). By contrast, the spray coating process immobilized the TiO2-NPs upon contact with the hot substrate, leading to rapid droplet evaporation and the formation of a film with a pronounced roughness of 1300 nm (Figure S9b). Effect of CNC in the surface morphology. Comparison between films prepared from commercial TiO2-P and TiO2(NH4OH): To gain more insight into the effect of film processing, namely screen printing and spray coating, a comparison between TiO2 photoanodes of the same thickness (~3.5 µm) prepared by both techniques is herein shown. The screen-printed TiO2 photoanode was prepared with the GreatCell® paste, whereas the spray-coated TiO2 photoanode was fabricated from the TiO2(NH4OH) dispersions in order to discard the CNC effect upon sintering, as described in the main article. In terms of surface morphology, it is of great interest the study of such property according to the followed fabrication procedure. As commented before, the screen-printed films display low Rq values (Figure S10a) whereas the TiO2(NH4OH) ones show a higher roughness (Figure S10b). These differences directly arise from the film fabrication method followed. Furthermore, the addition of CNC clearly influences the morphology of the film, with the bare TiO2(NH4OH) film showing lower roughness (900 nm, Figure S10b) compared to the TiO2-CNC(NH4OH) one (1300 nm, Figure S9b). Gas physisorption of the TiO2 materials (N2 isotherms): To further explore the effect of the CNC on the macroporous structure of the TiO2-CNC(NH4OH) material used as photoanode, physisorption measurements were conducted on both bare TiO2-NPs and TiO2-CNC(NH4OH) powder materials after sintering. N2 adsorption−desorption at -196 °C (Quantachrome Autosorb-6B Instrument) was measured after sample degassing (250 °C, 4 h) to characterize the porous texture and the equivalent Brunauer−Emmett−Teller (BET) specific surface area (SBET). Figure S11 shows the N2 isotherms of the employed materials. Both materials exhibit type II isotherms according to the IUPAC classification, typical of non-porous solids.The isotherms (Figure S11) reveal an initial increase at low relative pressure values. When it comes to the intermediate region of the isotherms, it is important to note that the TiO2 CNC hybrid has a slightly increased adsorption due to the removal of the biopolymer during the thermal treatment. This probably refers to an enlarged separation between the solid TiO2 particles. A narrow hysteresis loop appears at very high relative pressures, around p/p0 = 0.9, which is commonly ascribed to the capillary condensation taking place within the interstitial pores between TiO2 particles. The BET specific surface area of the TiO2-CNC(NH4OH) powder material is marginally higher (42 m2/g) to the observed for the bare TiO2 nanoparticles (35 m2 /g), evidencingnot too much influence of the CNC in the final internal porosity. Pore size distribution (DFT and BJH methods): Pore size distribution has been calculated for both samples from their N2 adsorption isotherms using the density functional theory (DFT) (Figure S12) and the Barrett-JoynerHalenda (BJH) method (Figure S13), which uses the Kelvin model of pore filling. X-Ray diffraction: X-ray diffraction results of the employed materials, namely CNC (type-II), TiO2-NPs (anatase), TiO2-CNC, and the commercial TiO2 paste (GreatCell®) are shown in Figure S14. The diffractogram of CNC (type-II) is in agreement with literature. The synthesized TiO2-NPs exhibits the characteristic profile of anatase NPs, evidencing a comparable crystal phase composition and crystallite size to the commercial TiO2 paste. The average crystallite size of the employed TiO2-NPs is 25 nm,3 whereas the commercial TiO2 paste is composed of particles distributed in two sizes: 20 and 300 nm. Notably, the XRD analysis of the prepared TiO2-CNC(NH4OH) hybrid shows a combined pattern from TiO2 and CNC, not showing any additional peaks. Thermogravimetric analysis of the commercial TiO2 paste: The concentration of TiO2 in the commercial paste is somewhat higher (62%, Figure S15) to that of the TiO2-CNC. This mass loss is ascribed to the removal of alkylated celluloses from the paste. Scanning electron microscopy of photoanodes from commercial TiO2-P based: Figure S16 shows SEM images of the film obtained from TiO2 commercial paste. The TiO2-P based photoelectrode film displays a smooth surface, both before and after air sintering, despite having a similar content of cellulose derivatives (62 wt.% of TiO2) to our TiO2-CNC(NH4OH) hybrid (50 wt.%) (Figure S8). Photoelectrochemical characterization of the films prepared from TiO2(NH4OH) and commercial TiO2-P: Figure S17 displays the CV profiles of the TiO2(NH4OH) and TiO2-P films, both under dark and illumination. In the absence of light, the photoelectrodes exhibit the characteristic reversible redox behavior of TiO2 electrodes. A cathodic current is observed at more negative potentials, and a nearly symmetric positive current during the backward scan. The voltammograms exhibited an accumulation region at approximately -0.8 V, and a depletion region at higher potentials, -0.4 V. Under illumination conditions, both materials show similar photocurrent values (~ 46 µA·cm2). This observation is further supported by transient photocurrent measurements (Figure S18), which reveal slight differences between the two electrodes (from 32 µA·cm-2 to 43 µA·cm-2), highlighting the significant improvement of the TiO2 electrodes when using CNC and ammonia for their fabrication. 5-hour photocurrent measurements: Aiming to study the stability of the TiO2-CNC(NH4OH) and TiO2-P photoanodes, 5-hour experiments at a constant potential (0 V vs. Ag/AgCl) were performed (Figure S19). A photocurrent decay is observed in both cases, mainly caused by the blocking of the TiO2 active sites due to parasitic redox processes (see references 33, 34 and 35). Nevertheless, the TiO2-CNC(NH4OH) retained a higher PEC performance, even after 5 hours under operation conditions due to its specific morphology that arises from the CNC processing. Electrochemical impedance spectroscopy (EIS): Equivalent circuit: EIS spectra were analyzed according to the model circuit shown in Figure S20. Resistances (RS, Rct and Rsc) and constant phase elements, i.e. non-ideal capacitors with phase angle <90, (CPEdl and CPEsc) are calculated as model fitting parameters to the experimental data.-- Under a Creative Commons license BY-NC 3.0. |
|---|