Gallenkamp, Charlotte (2022)
Developing molecular models for the prediction of spectroscopic properties of FeNC electrocatalyst active sites using DFT.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00021590
Ph.D. Thesis, Primary publication, Publisher's Version
Text
PhD_Thesis_CGallenkamp_published.pdf Copyright Information: CC BY-SA 4.0 International - Creative Commons, Attribution ShareAlike. Download (111MB) |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Developing molecular models for the prediction of spectroscopic properties of FeNC electrocatalyst active sites using DFT | ||||
Language: | English | ||||
Referees: | Kramm, Prof. Dr. Ulrike I. ; Krewald, Prof. Dr. Vera | ||||
Date: | 2022 | ||||
Place of Publication: | Darmstadt | ||||
Collation: | xx, 203 Seiten | ||||
Date of oral examination: | 7 October 2022 | ||||
DOI: | 10.26083/tuprints-00021590 | ||||
Abstract: | The transition metal iron is a cheap and abundant element. It has a versatile chemistry, which makes iron compounds ideal candidates for substituting critical elements such as platinum or palladium. These noble metals are frequently used as electrocatalysts, for which iron compounds have been investigated as possible alternatives. In particular, iron and nitrogen doped carbon (FeNC) materials have attracted significant interest due to their high electrocatalytic activity towards the oxygen reduction reaction (ORR). The ORR, the cathode reaction in fuel cells, is kinetically hindered in comparison to the oxidation of fuel on the anode side, and therefore requires high amounts of catalysts. By substituting expensive platinum cathode catalysts by FeNC catalysts, hydrogen fuel cell technology for green automotive applications can be made broadly accessible. Even though FeNC catalysts have been optimized over the last decades, they still lack long-term stability which hinders their utilization on an industrial scale. FeNC catalysts are synthesized by a mixture of iron, nitrogen and carbon precursors including at least one pyrolysis step, and are thus rendered as highly amorphous materials with functionalized carbon and several iron phases. The iron phase that dominates ORR activity contains single iron sites, so-called FeN4 centers, where iron is coordinated by nitrogen and embedded within a carbon matrix. However, their exact structure, their ORR mechanism and possible degradation mechanisms are not yet fully understood. It is still under debate whether the FeN4 sites are coordinated by pyrrolic (five-membered rings) or pyridinic nitrogen (six-membered rings), and furthermore whether axial ligands are present, and if so, which type. The exact structure determines the iron oxidation and spin states, crucial information for the catalytic mechanism that remains to be clarified. To unravel the exact nature of the FeNC active site(s), several spectroscopic techniques are used to characterize FeNC catalysts. A very successful example is Mössbauer spectroscopy, which probes the coordination environment of iron and its electronic structure. It is not hindered by the amorphousness of the samples or by the highly functionalized carbon matrix. The Mössbauer spectra of FeNC catalysts are often interpreted by comparison to the spectra of inorganic reference complexes, such as iron porphyrins (pyrrolic type FeN4 sites). Synthetic reference compounds with an extended pi-system, as it is likely present in the carbon matrix of the FeNC materials, are missing. In this thesis, a solution to this gap in knowledge is offered by theory: computational chemistry can develop models to mimic the FeN4 active site environments and to predict their spectroscopic properties which will lead to refined assignments of the active sites. With such FeNC models, all intermediates of the ORR catalytic cycle and their spectroscopic properties can be predicted. This facilitates the interpretation of in situ and operando spectroelectrochemistry experiments on FeNC catalysts to identify the precise nature of the active site(s). Herein, density functional theory calculations were used to develop FeNC active site models. The method for predicting Mössbauer spectra was calibrated using 20 reference complexes with FeN4-6 coordination environments. In this study, computational trust regions for the Mössbauer parameters were defined that were used throughout the thesis. To assess the spectroscopic discernibility of small chemical changes in a series of pyridinic model complexes, seven complementary spectroscopies were predicted and evaluated with respect to experimental resolution and applicability to FeNC catalysts. For model complexes with pyrrolic and pyridinic nitrogens, the predicted spectroscopic changes upon one-electron reduction in presence and absence of axial ligands were studied. Due to the weaker ligand field and the resulting stabilization of high spin states, the pyrrolic models show more distinct spectroscopic changes than the pyridinic models. The distinct changes observed in in situ Mössbauer experiments, namely the formation of a ferrous high spin doublet D3b, were qualitatively explained by small pyrrolic models, whereas the reduction of a side product, hydrogen peroxide, was linked to pyridinic models. Systematically increasing the size of the graphene matrix for three different active site models showed that the Mössbauer parameters converge at around 50 carbon atoms. Furthermore, energetically close-lying electronic structures with unpaired electrons delocalized in the extended pi-system were observed. Using a suitably large model, the spectroscopic changes observed in operando Mössbauer experiments were successfully explained. The active site was identified to be of pyrrolic Fe(II)N4 nature, for which the ORR rate-determining step in acidic electrolyte was determined as either the first proton-coupled electron transfer or the O-O bond breaking step. This thesis forms the basis for further computational studies targeted at optimising the active sites and their catalytic reaction paths, and provides guidance for experimental studies regarding the expected spectroscopic resolution of structural nuances. |
||||
Alternative Abstract: |
|
||||
Status: | Publisher's Version | ||||
URN: | urn:nbn:de:tuda-tuprints-215901 | ||||
Classification DDC: | 500 Science and mathematics > 540 Chemistry | ||||
Divisions: | 07 Department of Chemistry > Eduard Zintl-Institut > Fachgebiet Anorganische Chemie > Catalysts and Electrocatalysts 07 Department of Chemistry > Theoretische Chemie (am 07.02.2024 umbenannt in Quantenchemie) |
||||
Date Deposited: | 31 Oct 2022 13:22 | ||||
Last Modified: | 03 Nov 2022 10:20 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/21590 | ||||
PPN: | 501038388 | ||||
Export: |
View Item |