Title: Predicted Structures of the Active Sites Responsible for the Improved Reduction of Carbon Dioxide by Gold Nanoparticles
Authors: Tao Cheng, Yufeng Huang, Hai Xiao, and William A. Goddard, III.
By: Jared Mondschein
Plants have developed an amazing process to store sunlight in the form of chemical bonds. Called photosynthesis, an intricate network of enzyme cascades use energy from the sun to rearrange the bonds in water and carbon dioxide to make sugars. Plants then use these sugars to support growth.
Scientists want to replicate this process, but without all of the complex biology. Dubbed artificial photosynthesis, this development would enable (1) the storage of solar energy and (2) the transformation of carbon dioxide into useful chemical feedstocks. This would help lower atmospheric concentrations of a potent greenhouse gas while also turning a byproduct of many industrial processes into value-added chemicals.
Key to this work is the discovery of materials called catalysts that are capable of performing the chemistry. The ideal catalyst requires minimal energy inputs, is stable over long time periods, and is highly selective for product formation. Many materials have been tested for this application, but the overwhelming majority requires either too much energy or do not perform the desired chemistry. Gold is a promising candidate as many studies have shown that gold nanoparticles can electrochemically reduce carbon dioxide into carbon monoxide, a feedstock chemical for many industrial processes, while avoiding the production of other chemicals. However, the active sites on gold catalysts, where carbon dioxide is reduced, is not well understood. Such knowledge may help researchers further improve gold-based catalysts.
In this study, researchers utilized computational methods to determine the active sites. First, the authors simulated the production of gold nanoparticles and subsequent post-synthesis treatments via reactive molecular dynamics simulations and ReaxFF reactive force fields (Figure 1). The authors then modeled the transformation of carbon dioxide on >260 candidate active sites, calculating the change in energy that occurs when various intermediates bind. This change in energy was then related to how strong the intermediates were bound to the site.
Figure 1. (a) The atomic structure of a computationally synthesized Au nanoparticle and (b) a predicted transmission electron microscopy image. The yellow markers indicate grain boundaries.
These calculations revealed that only a select few sites are capable of serving as the active site. Sites located on the corner of a gold nanoparticle would promote unfavorable chemical reactions due to binding intermediates too loosely, while other sites located on a facet site would require the input of too much energy due to binding intermediates too tightly.
Edge sites, however, were the best sites for promoting the transformation of carbon dioxide to carbon monoxide. Located at grain boundaries, these sites bound the intermediates neither too tightly nor too loosely. As a result, minimal energy inputs would be required while only allowing for the desired reaction to occur.
While further studies are required to confirm that edge sites are the active sites for production of carbon monoxide from carbon dioxide, this study suggests that researchers should aim to synthesize gold nanoparticles with large numbers of grain boundaries. This should enhance the number of edge sites, and therefore active sites, and allow gold nanoparticles to excel as catalysts for the selective electrochemical reduction of carbon dioxide to carbon monoxide.