Improved Stability of Platinum Oxygen Reduction Reaction Catalysts by Implementation of Thermally Treated Carbon Supports with High Biomass Content

Abstract

To improve the lifetime of polymer electrolyte membrane fuel cells (PEMFCs), optimization of the fuel cell components is necessary wherein one research focus is the stability of the platinum catalyst support. Due to the harsh conditions in PEMFCs, e.g. high potentials (≥ 0.85 VRHE) and low pH (≤ 2), the catalyst support, which commonly consists of a high surface area carbon, degrades.[1] This is mainly caused by carbon corrosion, which becomes significant at potentials above 1.1 VRHE.[2] Typically, commercially available and chemically derived raw oil based carbons are used as catalyst supports. The use of multi-walled carbon nanotubes (MWCNTs) as a more stable, graphitic catalyst support with high electrical conductivity is well known in literature.[3, 4] The production of MWCNTs is expensive, which makes upscaling difficult. However, a previous study of our group Schonvogel et al. showed low stability of biomass-based supports, which requires further optimization.[5] Therefore, an approach to increase the stability of the biomass-based carbon support is thermal graphitization to generate a stable and ordered carbon structure. This leads to an increased amount of sp2-hybridized carbon with increased crystallinity and increased graphitic degree. At this point, the use of activated biomasses as carbon support is a cheap and sustainable alternative to MWCNTs and commercially available carbon blacks, for example Vulcan XC-72®.[5, 6] In this work, novel composite supports using thermally treated activated sawdust and MWCNTs will be investigated as Pt-nanoparticle support towards their physical properties and electrochemical ORR activity and stability. Sawdust was steam activated at 750 °C for 1 h in nitrogen atmosphere to create a defined pore system. Afterwards, the activated sawdust (aSD) was mechanically mixed with 20 wt.% MWCNTs. The mixture was treated at high temperatures (1500, 1800 and 2100 °C, under Ar atmosphere) in a furnace. For comparison, MWCNTs only were thermally treated under the same conditions. Physical characterization of the supports was performed using low temperature nitrogen sorption for determination of surface properties and crystallinity powder x-ray diffractometry (PXRD) for analysis of the degree of graphitization. Next, these novel carbon materials were used as support for 40 wt.% platinum nanoparticles. The platinum content was determined using inductive coupled plasma with mass spectrometry (ICP-MS) and the platinum nanoparticle diameter was analyzed with transmission electron microcopy (TEM) resulting in mean diameters ≤ 2 nm. In order to evaluate the activity and stability of the synthesized platinum catalysts, electrochemical characterization was performed for most promising catalysts before and after an accelerated stress test (AST, 0.6 – 1.5 VRHE, 500 mV s-1, 5000 cycles). For electrochemical characterization a three-electrode setup with a rotating ring disc electrode (RRDE) was used. The electrochemical active surface area was obtained by hydrogen underpotential deposition (HUPD) and CO stripping method. In Figure 1 the ORR curves, mass activities (MA) determined at 0.9 VRHE and MA loss of Pt/gaSD_gMWCNTs_2100 and Pt/gMWCNTs_2100 before and after AST are shown.Our study reveals that the mixture of 20 wt.% MWCNTs with sustainable aSD and subsequent thermal treatment results in a catalyst with comparable MA and stability to Pt/MWCNTs_2100. Thus, the use of activated biomass as the main component in catalyst supports constitute a more sustainable and cheaper alternative to MWCNTs. Based on these results, thermal treatment at higher temperatures of ≥ 2500 °C and higher biomass content will be further investigated to increase the stability.[1] F. Ettingshausen, J. Kleemann, A. Marcu, G. Toth, H. Fuess, C. Roth, Fuel Cells 2011, 11, 238-245.[2] V. Bandlamudi, P. Bujlo, C. Sita, S. Pasupathi, Materials Today: Proceedings 2018, 5, 10602-10610.[3] V. N. Popov, Materials Science and Engineering: R: Reports 2004, 43, 61-102.[4] D. Schonvogel, J. Hülstede, P. Wagner, I. Kruusenberg, K. Tammeveski, A. Dyck, C. Agert, M. Wark, J. Electrochem. Soc. 2017, 164, F995-F1004.[5] D. Schonvogel, M. Nowotny, T. Woriescheck, H. Multhaupt, P. Wagner, A. Dyck, C. Agert, M. Wark, Energy Technology 2019, 7, 1900344.[6] M. Iwanow, T. Gartner, V. Sieber, B. Konig, Beilstein J Org Chem 2020, 16, 1188-1202. Figure 1