Appendix. Supplementary materials for Investigation of the properties influencing the deactivation of iron electrodes in iron-air batteries [Dataset]
S1.- Electrode manufacturing scheme: Figure S1. Scheme of the process of electrode manufacturing. S2.- Mössbauer spectroscopy: Figure S2. Mössbauer effect spectra of the studied iron oxides with fitted components and distribution of doublets and sextets as explained in the manuscript. Table S1. Fitt...
| Autores: | , , , , , , |
|---|---|
| Tipo de recurso: | conjunto de datos |
| Fecha de publicación: | 2023 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/334397 |
| Acceso en línea: | http://hdl.handle.net/10261/334397 |
| Access Level: | acceso abierto |
| Palabra clave: | Deactivation Electrochemical impedance spectroscopy Iron-air batteries Porous electrodes http://metadata.un.org/sdg/7 Ensure access to affordable, reliable, sustainable and modern energy for all |
| Sumario: | S1.- Electrode manufacturing scheme: Figure S1. Scheme of the process of electrode manufacturing. S2.- Mössbauer spectroscopy: Figure S2. Mössbauer effect spectra of the studied iron oxides with fitted components and distribution of doublets and sextets as explained in the manuscript. Table S1. Fitted parameters of the Mössbauer spectra using WinNormos software. Isomer shifts (δ) given with respect to α-Fe. Errors in brackets. For distributions, the average value of δ is given. Max1 and max2 indicate the maxima in the distribution values of quadrupole splitting ∆ (doublets) or hyperfine field Bhf (sextets). S3.- Determination of crystallite size: The size of the crystallites of each phase, hematite and maghemite, was determined as follows. The powder-diffractograms were fitted by the Rietveld method using FullProf software. In the fits we used the Thompson-Cox-Hastings (TCH) pseudo-Voigt shape for reflections. Instead of the full width of the reflection and pseudo-Voigt mixing parameter, the parametrization of the TCH pseudo-Voigt function allows to calculate the Gaussian and Lorentzian widths, which can be readily identified with crystallite-size effects, microstrains and instrumental resolution broadening. The integral breadth corresponding to size effects finally gives the crystallite size through the Scherrer formula, thus excluding contributions from the instrumental resolution. No strain contribution to the reflections broadening was considered in the present case. Table S2. Relative molar fractions of iron in hematite and maghemite in %, assuming identical recoilless fractions for both oxides. The values for Fe2O3-TAR-N2 were obtained by fitting the distribution of hyperfine values above 45 T with two Gaussians. The area of the Gaussian centered at ≈51 T yields the contribution of hematite; the rest of the distribution is then assigned to maghemite. S4.- Nitrogen physisorption: Figure S3. Nitrogen physisorption isotherms at 77 K over a) Fe2O3-TAR-air and S-Fe2O3-TAR-air, b) Fe2O3-TAR-N2 and S-Fe2O3-TAR-N2, and c) Fe2O3-SHX-air. S5.- XPS spectra: Figure S4. XPS spectra of samples: a) Fe2O3-TAR-air, b) S-Fe2O3-TAR-air, c) Fe2O3-TAR-N2, d) S-Fe2O3-TAR-N2, and e) Fe2O3-SHX-air. Carbon percentages are overestimated because of adventitious carbon. Figure S5. XPS spectra of orbitals Fe2p (left) and S2p (right) of samples: a) S-Fe2O3-TAR-air, and b) S-Fe2O3-TAR-N2. Figure S6. XPS spectra of orbital O1s of samples: a) Fe2O3-TAR-air, b) S-Fe2O3-TAR-air, c) Fe2O3-TAR-N2, d) S-Fe2O3-TAR-N2, and e) Fe2O3-SHX-air. S6.- FESEM and TEM/STEM images of the iron oxides: Figure S7. FESEM micrographs at 50k magnification of iron oxides a) Fe2O3-TAR-air, b) S-Fe2O3-TAR-air, c) Fe2O3-TAR-N2, d) S-Fe2O3-TAR-N2, and e) Fe2O3-SHX-air. Figure S8. TEM micrographs of iron oxides a) Fe2O3-TAR-air, b) S-Fe2O3-TAR-air. STEM images for c) Fe2O3-TAR-air, d) S-Fe2O3-TAR-air. S7.- Electrochemical reactions on the electrodes and electrodes deactivation During charge (dotted lines), all of the composites show a first plateau at around -0.95 V vs Hg|HgO, associated to the reduction of iron (III) to iron (II), and a second longer plateau where both the reduction of iron (II) to metallic iron and the HER take place at -1.18 V vs Hg|HgO. When discharging (solid lines), the plateau corresponding to the oxidation of Fe to Fe(OH)2 is visible at around -0.92 V vs Hg|HgO and the plateau of the formation of iron (III) appears between -0.75 and -0.70 V vs Hg|HgO. Two interesting facts related to all of the electrodes must be noted: first, that both discharge plateaus are roughly the same length. While the stoichiometry of the reactions indicates that two electrons are transferred in the first step and only one in the second and so, the first discharge plateau should be twice the length of the second one. This means that not all the iron (II) hydroxide molecules are reducing to metallic iron and oxidizing to Fe(OH)2 again. The second remarkable fact is that the second discharge plateau appears to be divided into two: a first shorter plateau and a second longer one. This suggests that the oxidation of Fe(OH)2 to FeOOH or Fe2O3 occurs through an intermediary, probably Fe3O4. Figure S9. Detail of the processes of charge and discharge of an iron oxide electrode. Figure S10. Model applied to the discharge capacity of electrodes a) Fe2O3-TAR-air, b) S-Fe2O3-TAR-air, c) Fe2O3-TAR-N2, and d) Fe2O3-SHX-air. Figure S11. Post-mortem XPS analyses of orbitals a) Fe2p and b) O1s of electrode Fe2O3-TAR-N2. S8.- Influence of the physical-chemical properties on rate capability The rate capability of the electrodes S-Fe2O3-TAR-air and S-Fe2O3-TAR-N2 was tested performing charge-discharge cycles at C-rates of 0.8 - 0.4 C and 1.6 C - 0.8 C. The electrode S-Fe2O3-TAR-N2, with lower porosity and surface area, is more affected by higher C-rates, as its f factor decreases from 0.962 to 0.891 when increasing the charge-discharge rates from 0.4 - 0.2 C to 1.6 - 0.8 C. This effect is not as strong in electrode S-Fe2O3-TAR-air (f factor barely decreases, from 0.976 to 0.970). Figure S12. Model applied to the discharge capacity of electrodes a) S-Fe2O3-TAR-air and, b) S-Fe2O3-TAR-N2 at higher C-rates. S9.- Additional EIS data: The equivalent circuit fitted to the Nyquist diagrams of samples Fe2O3-TAR-air and Fe2O3-TAR-N2 is the same as in Figure 7 in the main text. The Nyquist diagram and fitted parameters of Fe2O3-TAR-air show little variability, with the charge transfer resistance increasing ca. 10% after 15 cycles. Fe2O3-TAR-N2, by contrast, shows an increment of 120 mV, more than threefold. Figure S13. Nyquist diagrams of EIS tests after 1 and after 15 cycles of electrodes: a) Fe2O3-TAR-air and b) Fe2O3-TAR-N2 Table S3. Optimized parameters of equivalent circuit (Figure 7 in the main text) for the electrodes in Figure S13. S9.- Post-mortem textural characterization: Figure S14 shows the N2 physisorption obtained for two electrodes, considering a fresh electrode (without cycling) and an electrode cycled 20 times. Both of the tested electrodes (S-Fe2O3-TAR-air and S-Fe2O3-TAR-N2) showed a similar decrease in their surface area and pore volume, as the iron oxides expanded, occupying the pores of the carbon matrix. This phenomenon was investigated by Yang et al. Coincidently with what they found, the electrode S-Fe2O3-TAR-air showed a greater decline in pore size (see Table S4), from 14.1 nm to 6.8 nm, while S-Fe2O3-TAR-N2 average pore diameter increased from 13.8 nm to 16.6 nm. This indicates a better utilization and better contact between the carbon and iron phases in S-Fe2O3-TAR-air electrode, and thus, a greater stability. Figure S14. Adsorption isotherms of electrodes: a) S-Fe2O3-TAR-air, and b) S-Fe2O3-TAR-N2; before cycling and after 20 cycles. Table S4. Textural properties of the fresh and cycled electrodes.-- Under a Creative Commons license BY-NC-ND 4.0. |
|---|