Making Fuel Cells More Efficient and Affordable

Fuel cells sound like an ideal power source—they are efficient and quiet, have few moving parts, run on hydrogen, and emit only water vapor. It’s no wonder that President Bush proposed new research into hydrogen-powered automobiles in his 2003 State of the Union address. But for this dream to become reality, scientists and engineers need to make fuel cells less expensive.

While there are several types of fuel cells, proton exchange membrane (PEM) cells are the leading candidates for powering light-duty vehicles and buildings and for replacing batteries in portable electronic devices. The PEM is a polymer electrolyte that allows hydrogen ions (protons) to pass through it. As hydrogen (H₂) is fed to the anode side of the fuel cell, a catalyst (usually platinum or a platinum-ruthenium alloy) encourages the hydrogen atoms to separate into electrons and protons. The electrons travel from anode to cathode along an external circuit, creating a usable electric current. Meanwhile, on the cathode side, oxygen molecules (O₂) are being separated by the catalyst into separate atoms in a process called oxygen reduction. The negatively charged oxygen atoms (adsorbed on the catalyst) attract the positively charged hydrogen ions through the PEM, and together they combine with the returning electrons to form water molecules (H₂O).

While splitting the hydrogen molecules is relatively easy for platinum-alloy catalysts, splitting the oxygen molecules is more difficult. Although platinum catalysts are the best known for this task, they are still not efficient enough for widespread commercial use in fuel cells, and their high price is one of the reasons why fuel cells are not yet economically competitive.

Finding more efficient and affordable catalysts is one of the goals of a research group led by Perla Balbuena, Associate Professor of Chemical Engineering at the University of South Carolina. Using computational chemistry and physics techniques, they are trying to predict the properties of electrodes consisting of bimetallic and trimetallic nanoparticles embedded in a polymer system and dispersed on a carbon substrate. They are working to identify the atomic distribution, vibrational modes, and diffusion coefficients of a variety of alloy nanoparticles.

“The rationale behind the use of alloy nanoparticles,” Balbuena says, “is that each of the elements, or particular combinations of them, may catalyze different aspects of the reaction. The large surface-to-volume area of nanoparticles may also contribute to their effectiveness.” Experimental data indicates that alloys can be as effective or better than platinum catalysts, but nobody fully understands why this is so. Little is known about the physical, chemical, and electrical properties of alloy nanoparticles, and in particular about the atomic distribution of the various elements on the catalyst surface, which is difficult to determine experimentally. Computational studies play an essential role in answering these questions.

For example, Figure 1 shows a test simulation of a system involving only platinum nanoparticles on a carbon substrate and surrounded by a hydrated polymer. The goal of this simulation was to test the stability of the nano-size catalyst in such an environment—that is, how the polymer groups, water, and substrate influence the shape and mobility of the nanoparticles, and how they influence the exposed surfaces where the oxygen molecules will be adsorbed and dissociated into oxygen atoms.

Figure 1 Platinum nanoparticles (gold-colored atoms) are deposited on a graphite support. The long chains are Nafion fragments composed of fluorine (green), carbon (blue), oxygen (red), and sulfur (yellow) atoms; some of these chains become adsorbed over the metallic clusters. The background contains water molecules (red for oxygen and white for hydrogen). Nafion, widely used for proton exchange membranes, has both hydrophilic and hydrophobic chemical groups, so some degree of water segregation is observed. Image: Eduardo J. Lamas.

Balbuena and her colleagues recently developed a new computational procedure to investigate the behavior of active catalytic sites, including the effect of the bulk environment on the reaction. They demonstrated this procedure in simulations of oxygen dissociation on platinum-based clusters alloyed with cobalt, nickel, or chromium, and embedded in a platinum matrix. To determine the optimal alloy, they studied a variety of compositions and geometries for alloy ensembles embedded into the bulk metal (Figure 2).

Figure 2 Oxygen molecules (red) adsorb on a bimetallic surface of platinum (purple) and cobalt (green). The oxygen molecules most frequently adsorb on “bridge” sites composed of two cobalt atoms or one cobalt and one platinum atom. The oxygen molecules will subsequently dissociate into individual atoms. From P. B. Balbuena, D. Altomare, L. Agapito, and J. M. Seminario, “Theoretical analysis of oxygen adsorption on Pt-based clusters alloyed with Co, Ni, or Cr embedded in a Pt matrix,” J. Phys. Chem. B (in press).

The researchers concluded that cobalt, nickel, and chromium make platinum-based complexes more reactive because they enable a higher number of unpaired electrons. Their computations yielded extensive new information that would be very hard to obtain experimentally, such as showing for the first time how alloying results in changes in the density of electronic states, and how those changes influence the oxygen reduction process. One of the questions they addressed was whether nickel, cobalt, and chromium might become oxidized and act as “sacrificial sites” where other species can adsorb, leaving the platinum sites available for oxygen dissociation. Their study shows that this may be the case for nickel, but that alternative mechanisms are possible, especially for cobalt and chromium, which instead of being just sacrificial sites, act as active sites where oxygen dissociates, in some cases more efficiently than on platinum sites. They found that the most effective active sites that promote oxygen dissociation are O₂CoPt, O₂Co₂Pt, O₂CrPt, and O₂Cr₂Pt, while ensembles involving nickel atoms are as catalytically effective as pure platinum.

The Balbuena group’s research is now focusing on the time evolution of the complete reaction on the best catalytic ensembles, including the combination of oxygen atoms with protons leading to water formation. Future work will address the possibilities of reducing further the catalyst size, nowadays in the order of tens of nanometers, as well as developing a full understanding of the role of the rest of the environment (polymer, water, and substrate) on the outcome of the reaction. The ultimate goal of these studies is to enable the design of new catalysts from first principles rather than trial-and-error experiments, reducing the cost of catalysts for fuel cells through the use of optimum alloy combinations requiring only minimal amounts of expensive materials.

Research funding: BES

Top