Correlations between the heat of adsorption and the position of the center of the d-band: Differences between computation and experiment

Previous DFT computations have shown that there should be a correlation between the calculated heat of adsorption of gases such as hydrogen, carbon monoxide, and ethylene and the energy of the center of the d-band. This paper considers whether the trends expected from the calculations agree with the available data. It is found that experimental heats of adsorption of CO on platinum single crystals increase linearly with increases in the energy of the center of the d-band as expected theoretically, but the slope is about one-third of that expected theoretically. Experimental heats of adsorption of hydrogen on platinum show a small decrease as the energy of the center of the d-band becomes less negative. By comparison, recent DFF calculations suggest a substantial increase. Experimental heats of adsorption of ethylene on platinum also show a small decrease as the energy of the center of the d-band becomes less negative. By comparison, recent DFT calculations suggest a substantial increase. Interestingly, the heat of adsorption of oxygen, and the heat of dissociative adsorption of ethylene, do follow the trends expected from the model, although no quantitative comparison is possible. Overall, calculated and experimental heats of adsorption differ by as much as 90 kJ/mol. Ours is the fifth recent paper that shows significant differences between careful experiments and careful DFT calculations. We suggest, therefore, that the functionals used to model adsorption need improvement,