Sabatier Principle¶
The foundation of activity scoring in ASCICat.
Historical Background¶
Paul Sabatier (Nobel Prize 1912) proposed:
"An ideal catalyst binds reaction intermediates with intermediate strength."
Too weak binding → reactants don't stick Too strong binding → products don't desorb
The Volcano Plot¶
Plotting catalytic activity vs. binding energy produces a "volcano":
Activity
↑
| ★ Optimum
| /|\
| / | \
| / | \
| Pt / | \ Au
| / | \
| Ni / | \ Ag
|____/______|______\____→ Binding Energy
strong optimal weak
Theoretical Basis¶
Brønsted-Evans-Polanyi (BEP) Relations¶
Activation energy correlates with reaction energy:
\[E_a = E_a^0 + \alpha \Delta E\]
- Stronger binding → Lower adsorption barrier
- Stronger binding → Higher desorption barrier
Scaling Relations¶
Different intermediates scale together:
\[\Delta E_{*OH} = a \cdot \Delta E_{*O} + b\]
This enables single-descriptor screening.
Application to HER¶
For Hydrogen Evolution Reaction:
\[\Delta E_{opt} = -0.27 \text{ eV}\]
Derivation:
- Thermoneutral point: \(\Delta G_{H*} \approx 0\)
- ZPE and entropy corrections: \(\Delta E_{H*} \approx -0.27\) eV
- Validated by Pt(111) experiments
Application to CO2RR¶
| Pathway | Intermediate | \(\Delta E_{opt}\) |
|---|---|---|
| CO | *CO | -0.67 eV |
| CHO | *CHO | -0.48 eV |
| COCOH | *COOH | -0.32 eV |
Linear Scoring Implementation¶
\[S_a = \max\left(0, 1 - \frac{|\Delta E - \Delta E_{opt}|}{\sigma_a}\right)\]
Parameters:
- \(\Delta E_{opt}\) = Volcano peak position
- \(\sigma_a\) = Width parameter (typically 0.15 eV)
Why 0.15 eV?¶
The width parameter accounts for:
- DFT uncertainty (~0.1 eV)
- Functional variation (~0.05-0.1 eV)
- Surface coverage effects
References¶
- Sabatier, P. Ber. Dtsch. Chem. Ges. 44, 1984 (1911)
- Nørskov, J. K. et al. J. Catal. 209, 275 (2002)
- Greeley, J. et al. Nat. Mater. 5, 909 (2006)