61st Annual Report on Research 2016

Significance: Polymer electrolyte membrane fuel cells (PEMFCs) and direct formic acid fuel cells (DFAFCs) are a highly promising and well-developed technologies that will likely support a future hydrogen economy. A major strategy in this regard is the development and application of sub-5 nm diameter bimetallic catalysts instead of bulk platinum. Background: Clusters of nanoparticles (NPs) offer important advantages over isolated NPs. An interparticle gap with decreasing nanoscale dimensions will permit increased readsorbtion of chemical intermediates in a fuel cell half reaction and hence a more efficient power output. Furthermore, the potential drop in an interparticle gap can approach 300 mV suggesting that fuel cell reactions may be catalyzed with less overpotential and less interference from other ions. Controlling the composition in NPs in a cluster is particularly challenging and hence the role of Pt-based clusters in fuel cell catalysis is largely unknown. Further, as the interparticle distance in neighboring Pt NPs in clusters approaches 1 nm, the specific activity of Pt NPs for the oxygen reduction reaction can be increased 6 fold compared to that of a commercial Pt/C catalyst. The challenge in this scientific pursuit however lies in the controlled synthesis of uniformly dispersed Pt-based NP clusters with a controlled NPs population. Achievements: This research project has established a highly versatile method for the preparation of a self-assembled polystyrene-block-poly(4-vinylpyridine) block copolymer (BCP) template capable of metal ion-loading for NP synthesis (Scheme 2). We have shown that selection of the BCP template determines the length scale of the array of NP clusters while the distribution of metal anions co-dissolved in a common loading bath specifies for the composition of the NPs (Scheme 2, step iii). Using this approach, our group has studied three different bimetallic NP systems in great detail, and importantly, determined important structure-property relationship between the arrays of NP clusters and their electrocatalytic activity. In the case of PtAu (Figure 1), the BCP template synthesis yields PtAu with a nanostructure comprised of a Pt-rich shell is located on a PtAu alloy core. The PtAu core-shell bimetallic NPs have a very high density of electrochemically active Pt surface sites. Accordingly, the activity of Pt-rich core-shell PtAu nanocatalysts for the electrocatalytic oxidation methanol was approximately 2-4 fold that of a monometallic Pt benchmark catalyst (ETEK) and only 28% less than that of the PtRu bimetallic benchmark catalyst. The Au that is present in the PtAu NPs was beneficial to Pt-rich electrocatalysts as the carbonaceous poisoning If/Ib metric was greatly enhanced relative to both benchmark materials (~ 2-3 fold increase; Figure 3B). In the case of the PtIr system (Figure 2), the BCP template synthesis yields an array of alloyed PtIr NP. Many of the bimetallic PtIr NPs isolated from the BCP template approach are much higher in activity than typical industry/research standards. The most active array of Pt88Ir12 NPs had mass activities for formic acid oxidation that were four-fold higher that a PtRu industrial research standard (Figure 3C). Overall, we anticipate that BCP template method of preparing highly active clusters of low-diameter, Pt-contianing NPs will have a positive impact on several emerging areas in nanoscience such as energy devices, solar fuels, and heterogeneous catalysis. Current research is focused on the assessment of the electrocatalytical activity of these catalysts for the oxygen reduction reaction. We have recently extended the BCP template approach to the synthesis of bimetallic catalysts with non-precious metals. The morphology of the arrays of the PtCu bimetallic NPs are largely similar to that of the PtAu and PtIr NPs as observed by atomic force microscopy (AFM, see Figure 4A). The periodicity and the size of the NP clusters are specified by the BCP template. X-ray photoelectron spectroscopy (XPS) was used to quantify the composition, the oxidation state and the mixing of the metals (see the high resolution XPS data for the Pt 4f region in Figure 4B). The binding energy (BE) of the Pt 4f 5/2 and 7/2 peaks indicate Pt(0) (similar observations for the Cu 2p 1/2 and 3/2 peaks were made). As the copper content in the bimetallic NPs increases there is a decrease in the Pt signal intensities. The XPS-determined composition is reported as subscripts in the labels in Figure 4A and 4B. Importantly, as the amount of copper increased in the bimetallic NPs, a subtle shift to lower BE was found for the Pt4f 5/2 peak. This, in combination with the nanostructure observed by AFM, is strong evidence for a high level of mixing of the two metals in the NPs. The Pt:Cu ratio in the bimetallic NPs was tuned according to the ratio of the metals in the loading bath. A complex trend in the Pt-content in the NPs with the Pt-content in the loading bath was found (Figure 4C) and appears to follow a near-linear trend in the lower Pt-content NPs (< 50% Pt /mol). A non-linear trend was found for high Pt-content NPs (> 80% Pt/mol). The complexity in the metal-Pt and Cu loading trends are the subject of current research. The PtCu NPs in Figure 4A are electrocatalytically active. We are currently characterizing the series of NPs using electrochemical features such as the charge associated with the adsorption of hydrogen, the desorption of hydrogen and the charge for the oxidation of adsorbed monolayers of CO (Figure 5A). Additional features such as the current associated with the oxidation of methanol in the forward and backward traces (If and Ib; see Figure 5B) of CV curves are also determined to describe the electrocatalytic activity of these NPs. Further structure-property relationships for the PtCu system are a current research focus.