Ultrathin catalyst layers

Fig. 26 Pt-coated nanowhiskers (Fig. 25) embedded into the ionomeric membrane and backed by a GDL to form an MEA with an ultrathin catalyst layer [53b].

Fig. 27 Hydrogen/air PEFC performance at ambient pressure of both fuel and air, achieved at total Pt loading of only 0.12 mg Pt/cm2 with an MEA based on an ultrathin catalyst layer of the type described in Figs 24 and 25 [53b].

Spherical and planar model geometries of agglomerates will be considered. The planar geometry leads to important conclusions about the performance of ultrathin catalyst layers. 

The nonuniform distribution of protons and potential in water-filled agglomerates and ultrathin catalyst layers is predominantly an electrostatic effect. It is determined by the Debye length. Ad- Resulting reaction rate distributions and effectiveness factors depend on the characteristic sizes of agglomerates (i a) or ultrathin CCLs ( L) and on the transfer coefficient a. 

As a major conclusion, primary pores inside agglomerates and ultrathin catalyst layers should be hydrophilic (maximum wetting). Under such conditions effectiveness of catalyst utilization can approach 100%. Moreover, the microscopic mechanism of the electrochemical reaction, represented by the transfer coefficient a, is essential for the effectiveness of catalyst utilization. 

Effectiveness factors of agglomerates and ultrathin catalyst layers Thickness of the effective layer in the CCL Relative permittivity (of water 78)... 

Here, A max is the voltage loss tolerance due to finite electronic conductivity, b = RgT/ aeffF) is the Tafel parameter with the effective electronic transfer coefficient Ueff of the ORR, and Jq is the operating current density. For instance, at Icl= 10 qm, Jo = 1 A cm , Ueff = I, T = 333 K, and Ai max = 1 mV, the electronic conductivity requirement of the CL is Oei > 0.01 S cm In an ultrathin catalyst layer (UTCL) with thickness L = 100 nm, this bound on Oei is lower, namely, uei > lO- Scm-i. This estimate explains why, in CLs fabricated with the NSTF of the company 3M, a thin film of sputter-deposited Pt provides sufficient electronic conductivity. UTCLs are much less sensitive to the support conductivity. 

FIG U RE 3.17 The self-consistency problem in Pt electrocatalysis. The metal phase potential determines oxidation state and charging properties at the catalyst surface. These properties in turn determine the local reaction conditions at the Hehnholtz or reaction plane. At this point, structural design and transport properties of the catalyst layer come into play (as illustrated for conventional and ultrathin catalyst layers). Newly developed methods in the emerging field of first-principles electrochemistry attempt to find self-consistent solutions for this conpled problem. 

FIGURES.22 An illustration of design and key properties of ionomer-free ultrathin catalyst layers with insulating or electronically conductive support materials. The typical thickness is in the range of 200 nm. 

Chan, K. and Eikerling, M. 2014. Water balance model for polymer electrolyte fuel cells with ultrathin catalyst layers. 16, 2106-2117. 

Moving up the scale to the level of flooded nanoporous electrodes, Michael s group has developed the first theoretical model of ionomer-free ultrathin catalyst layers—a type of layer that promises drastic savings in catalyst loading. Based on the Poisson-Nernst-Planck theory, the model rationalized the impact of interfacial charging effects at pore walls and nanoporosity on electrochemical performance. In the end, this model links fundamental material properties, kinetic parameters, and transport properties with current generation in nanoporous electrodes. 

The length scale i a 50 — 100 nm determines the effectiveness of catalyst utilization for spherical agglomerates. Analogous relations apply for ultrathin planar catalyst layers with similar thickness, L 100 — 200 nm. We consider layers that consist of Pt, water-filled pores and potentially an electronically conducting substrate. With these assumptions, we can put/(dfptc, XfXptc = 1 and g Sr) = 1. The volumetric exchange current density is, thus. 

The alternative is to fabricate CCLs as ultrathin two-phase composites 100 mn -200 mn), in which electroactive Pt could form the electronically conducting phase, or Pt nanoparticles could be supported on a conductive substrate. The remaining volume should be filled with liquid water, as the sole medium for proton and reactant transport. The ultra-thin two-phase catalyst layer was explored by using the Poisson-Nemst-Planck (PNP) equations as employed for water-filled spherical agglomerates [69, 118]. The equations in Section 8.5.2 can be rewritten for the ID planar situation... 

Figure 8.17. Effectiveness factor of ultrathin two-phase catalyst layers as a function of thickness, Lcl. at different current densities. (Reproduced by permission of ECS — The Electrochemical Society, from Wang Q, Eikerling M, Song D, Liu S. Modeling of ultrathin two-phase catalyst layers in PEFCs.)...

As an example of how to use the insights conveyed in this chapter we provide an explicit comparison of overall effectiveness of Pt utilization (by atom number or catalyst weight) for conventional 3-phase composite and ultrathin CCLs, in Table 8.2. For conventional catalyst layers, the main detrimental factors arise at the nanoparticle scale and at the macroscopic scale due to triple-phase boundary requirements. For the nanostructured ultrathin CCLs it is assumed that a sputter-deposited continuous Pt layer is needed to provide electronic conductivity. It was suggested in [153] on the basis of cyclic voltammetry measurements that the irregular surface morphology of such catalysts corresponds to grain sizes of 10 nmwith... 

Wang Q, EikerUng M, Song D, Lin S. Modeling of ultrathin two-phase catalyst layers in PEFCs. J Electrochem Soc 2007 154(6) F95-101. 

An ultra-shallow-bed CAVERN device (Fig. 11) was developed to accommodate occasions where an extremely homogeneous distribution of adsorbates is required. Here the same amount of catalyst is loaded onto a much larger surface. After catalyst activation, reactants are introduced onto an ultrathin layer of catalyst bed (<0.5 mm) to achieve a homogeneous distribution of adsorbates on catalyst. 

---