Estimating the Time-Dependent Performance of Nanocatalysts in Fuel Cells Based on a Cost-Normalization Approach

Abstract

Researchers have developed new catalysts for fuel cells (FC), whose performances are compared after applying different normalization procedures. However, there is not a standard procedure. The current produced from CO electrooxidation was compared for Pt 4 Ru 5 Sn 1 /C and homemade Pt/C nanoparticles (NPs) normalized by different methods and the use of different methods renders different interpretations. Since the whole field aims to maximize the cost-effectiveness, a complementary method to normalize currents and power in terms of the total cost of the nanocatalyst, the Catalyst-based Cost method (CbC), was proposed. CbC considers the cost of all metals employed to build the catalyst, not only those ones with available surfaces. By applying a simple smoothing method on the prices in a time series, we were able to forecast the prices and consequently the power density of a FC. CbC provides tools for industrials forecast the designing of nanomaterials with improved efficiency and low cost.Keywords: fuel cell, catalyst, normalization of current and power, cost of metal, forecasting activity IntroductionThe continuous growth of the energy demand opened a wide field of research for energy converters, among which fuel cells have been exhaustively studied. Fuel cells can produce energy with low environmental impact since they generate power by the oxidation of a fuel at the anode and reduction of O 2 at the cathode. Fuel cells can be fed by H 2 or small alcohols, as methanol and ethanol. The catalysts, both anodes and cathodes, are mainly based on Pt, which is historically an expensive material.1 This is the main reason why many efforts have been made to replace Pt (at least partially) in catalysts used in fuel cells.When a new nanoparticle catalyst is produced, the normalization of the electrochemical current generated by its use is imperative in electrocatalysis, since this procedure allows distinct surfaces to be directly compared for a given reaction in terms of their intrinsic electroactivities. In this context, different normalization methods generate multiple interpretations about the activity of a catalyst. This lack of consensus hinders the comparison of different materials in terms of their electrochemical performances and generates impasses that prevent a faster development of the research area.A method commonly used consists in normalizing the currents by the electrochemically active surface area (ECSA), which is highly useful but not easily accessed for many materials. For Pt-based catalysts (as PtRu/C Zanata et al. 1981 Vol. 27, No. 11, 2016 and PtRuSn/C) the ECSA is sometimes estimated by the charge involved in the H desorption region.2-5 For Pd-based catalysts, the ECSA is usually calculated by the charge involved in the reduction of a PdO monolayer. [6][7][8] The main concern about using the ECSA for multi-metallic catalysts calculated by these methods lies on the fact that the metal lattice parameter suffers intense modifications when an additional element is inserted into its structure. 9,10 In t...

Highly active Ag/C nanoparticles containing ultra-low quantities of sub-surface Pt for the electrooxidation of glycerol in alkaline media

“…In any case, the clear enhancement in mass activity points out the superior activity of the modified Ag material. Normalizing the data by the Pt mass is widely used as it is motivated for an economic point of view [27] because Pt is by far a much more expensive material than Ag. To understand if the activity improvement is due to a better Pt utilization or if in fact the activity per Pt atom in Pt(1%)Ag/C is higher, another option to normalize the activity is considering only the Pt atoms at the surface of the catalyst.…”

Gold and gold-palladium branched nanocrystals for applications in plasmonic catalysis and electrocatalysis

“…79,80 In the next step, the electrocatalysts were evaluated for methanol oxidation reaction (MOR) and CO oxidation in absence and presence of green light irradiation (532 nm -200 mW of maximum total source intensity), as showed in Figure 26A-E. The obtained oxidation currents were normalized by Pd and Au loading (for the Au/C) 81 . For the MOR, it is very common to give voltammograms with two oxidation peaks, the forward peak related to the fresh methanol molecule oxidation and the backward peak that can be associated to incompletely oxidized methanol molecule or intermediates of the oxidation reaction.…”