• Issue 1  |
  • August 2009

Synthesis Science

Exploring the Role of Charge in Formation of Metal Clusters on Nanostructured Carbons

In the search for an energy carrier for renewable energy, hydrogen has great appeal. It is both abundant and environmentally friendly. There is a problem, however: currently, there is no way to store the hydrogen with high gravimetric and volumetric densities. Without this density, the hydrogen will take up too much room to be practical. In addition, the hydrogen recycling needs to be performable and reversible near ambient conditions. Researchers at ORNL have been using ab initio (spin) density functional theory (DFT) to study the way charge affects the formation of metal clusters on nanostructured carbons, which might be able to effectively store hydrogen.

Hydrogen interaction with Ca32C60

FIG. 1 Hydrogen interaction with Ca32C60.
(a) The isosurface of charge density
differences caused upon adsorption of the
32 Ca atoms onto the carbon cage. (b) The
optimized hydrogenorganometallic complex
of 92H2-Ca32C60 with a hydrogen uptake
of 8.4 wt%.


Researchers at ORNL demonstrated the use of DFT predicting the qualities of C60 fullerenes coated with light alkaline-earth metals.1 While Be and Mg can only bind weakly onto the C60, Ca and Sr adsorb strongly – and highly prefer a monolayer coating, important in hydrogen storages efficiency. The strong binding leads to a charge redistribution that causes electric fields to form around the fullerenes, which attracts the hydrogen. With the Ca, this gives a hydrogen uptake of over 8.4 wt %. This makes Ca one of the most promising coating elements.

In a following study, ORNL scientists systematically studied the electronic structure of metallofullerenes, which consist of valence electrons from metal atom(s) transferred to the enclosing carbon fullerene cage (this is caused by the higher electron affinity of the cage)2. For lanthanum inside carbon, the first type they studied, this is written as La @ Cn.


FIG. 2. Structures of (a) La@C28,
(b) La@C82, (c) Ca@C82, and (d)
Li@C82. The C, La, Ca, and Li atoms
are denoted by yellow, purple,
green, and gray balls, respectively.

The traditional view has been that this carbon cage is homogeneously negatively charged. In contrast, it was found that for large metallofullerenes (n >32) the transferred charges are not spread homogeneously, but instead are highly localized inside the fullerene, near the metal atom. They attributed this to the strong metal-fullerene interaction. With this new understanding, the physical properties of fullerene-based nanomaterials can be evaluated more reliably.

Transition metal (TM) coated nanostructures are considered a very promising material for storing hydrogen to be used as an energy carrier. However, the TM tends to cluster into multi-layer packets instead of staying in one layer on the fullerene. In a follow on experiment using both neutral and charged C60 surfaces, the researchers studied the energetics of Ti clusters as a function of the number of Ti atoms, with N ≤ 12. They found a critical cluster size, NC = 5, below which the Ti should tend to stay in one layer.3 Surprisingly, moderately doping it with boron doesn’t change NC for neutral C60, or only increases it by 1. It does, however, provide a higher kinetic barrier hindering Ti atoms from climbing on top of each other and stabilizing single layer structures with N ≥ NC at low temperatures. While this may not help enough at higher temperatures, it suggests that for other metals with strong interactions with the surface of C60 but with lower cohesive energies such as scandium, it may be enough – leading to an efficient way to store hydrogen for energy.


FIG. 3. (a) Two possible mechanisms of 2D layer to 3D cluster transformation for Ti5C60 complex, i.e., direct hopping (left) and position exchange (right). Both processes have similar kinetic barrier heights. Energy barriers of direct hopping for Ti5C60 (b) and for Ti5C48B12 (c).

Research supported by the Division of Materials Sciences and Engineering, Office of Basic Energy Sciences, DOE, and by the U.S. NSF and the NSF of China. The calculations were performed at DOE’s NERSC. Research for Yoon et al. was also funded by the DOE Hydrogen Sorption Center of Excellence.

1Yoon et al., “Calcium as the Superior Coating Metal in Functionalization of Carbon Fullerenes for High-Capacity Hydrogen Storage” Phys. Rev. Lett. 100, 206806 (2008)

2Yang et al., “Electron transfer and localization in endohedral metallofullerenes: Ab initio density functional theory calculations” Phys. Rev. B 78, 115435 (2008)

3Yang et al., “Energetics and kinetics of Ti clustering on neutral and charged C60 surfaces” J. Chem. Phys., 129, 134707 (2008)