The Reactive Spray Deposition Technology (RSDT) is a thin-film deposition process that overcomes many of the shortcomings of traditional vapor deposition techniques while yielding equal or better quality coatings at a lower cost [1, 2]. The RSDT process bypasses traditional wet chemical routes by simultaneously nucleating the catalyst on a support and sequential deposition of catalytic layers via gas-phase. The process does not require intensive electrical power and harnesses the energy stored in a liquid fuel to decompose the metal organic precursor into the desired metallic or bi-metallic product. The process can be tailored to favor both alloying and core-shell formation.

This has allowed our team to reduce the cathode Pt loading below 0.1 mg/cm² with enhanced performance. The Pt/Ketjen Black system surpassed the 2017 DOE target with a mass activity of 0.51 A mg^-1at 0.9V and oxygen pressure of 180kPa [2]. 

In the present work mass transport overpotential is investigated with ultra-low Pt loading (0.1 and 0.05 mg cm^-2 on cathode and anode; catalyst with average particle size of 2 nm and 3 nm, respectively) deposited directly by RSDT onto two supports: Ketjen Black (KB) and multi-walled carbon nanotube (MWCNT).

We conduct fundamental studies of the ORR on low Pt loading C supports to address the durability issues.  The focus is to strengthen the support-Pt anchoring and also lower the total Pt loading in the cathode to 0.05 mg/cm², an enhancement in electro-catalytic activity, mass activity @900 mV iR-free of 0.44 A/mg PGM and durability testing at 30,000 voltage cycles. Long-term compatibility with other cell components in real-world environments, including Accelerated Stress Test (AST) cycling will be demonstrated in MEAs (≥25cm²).

We performed polarization curves in pure Oxygen RH 100%, 80°C, stoich. 9.5/2, for KB and MWCNT with the same catalyst loading but different particle sizes. We noticed that there were no significant differences in the absolute performances due to the variation of particle size for Ketjen-Black samples while MWCNT samples show improvement in performances by increasing the diameter of Pt particles from 2 to 3 nm.

We performed polarization curves in Air with constant flows: 1 L/min Air and 0.5 L/min Hydrogen. Figure 1, shows initial performance and after 30,000 of cycles for both supports and Pt particles size of 3 nm, see figure 1a. The Ketjen-Black sample shows better performance in constant-flow polarizations than the MWCNT sample, as we presume it suffers to a lesser extent from cathode charge transfer limitations.

Differences between the two samples seem not to vary during AST. A microscopy study on post-durability electrodes showed an agglomeration in the cathode layer for KB and MWCNT particles size from 2-15 nm and 2-12, respectively, after 30K cycles, Figure 1b and c.

The evolution during the AST of specific activity and platinum oxide coverage is also investigated showing results in accordance with particle growth mechanisms.

References:

[1] J.M. Roller, M.J. Arellano-Jimenez, H. Yu, R. Jain, C.B. Carter, R. Maric, Electrochim. Acta 107 (2013) 632-655

[2] H. Yu, J.M. Roller, W.E. Mustain, R. Maric J Power Sources 283 (2015) 84-94.