Table 5.
RMSDs, MUEs and Average Errors of all benchmarked DFT methods (in kcal/mol). B3LYP-MM represents results obtained with the correction scheme developed in this work.
| Basis Set | Counter- poise corrected |
DFT-Method | Whole Interaction Energy DB (1715 datapointsa) |
Conformational Energies DB (700 data points) |
||||
|---|---|---|---|---|---|---|---|---|
| MUEb | RMSDc | Avgd | MUEb | RMSDc | Avgd | |||
| LACVP* | no | B3LYP-D3 | 2.11 | 2.87 | −2.10 | 0.82 | 1.10 | 0.02 |
| B3LYP-MM | 0.41 | 0.68 | −0.04 | 0.55 | 0.71 | 0.05 | ||
| M06-2X | 1.20 | 1.97 | −0.94 | 0.78 | 1.04 | 0.10 | ||
| B3LYP | 2.75 | 3.66 | 1.35 | 1.54 | 2.03 | 0.32 | ||
| yes | B3LYP-D3 | 1.21 | 1.91 | −0.64 | nae | |||
| B3LYP-MM | 0.41 | 0.65 | 0.18 | 0.67f | 0.87f | −0.03f | ||
| M06-2X | 1.20 | 1.97 | −0.94 | nae | ||||
| B3LYP | 3.53 | 4.87 | 2.81 | |||||
| aug-cc-pVDZ | no | B3LYP-D3 | 0.95 | 1.48 | −0.72 | 0.39 | 0.49 | −0.05 |
| B3LYP-MM | 0.37 | 0.53 | 0.09 | 0.37 | 0.47 | −0.07 | ||
| M06-2X | 0.73 | 1.17 | −0.30 | 0.61 | 0.79 | −0.02 | ||
| B3LYP | 3.11 | 4.36 | 2.73 | 1.57 | 2.15 | 0.25 | ||
| yes | B3LYP-D3 | 0.87 | 1.41 | −0.56 | nae | |||
| B3LYP-MM | 0.32 | 0.48 | 0.11 | 0.38f | 0.49f | −0.09f | ||
| M06-2X | 0.67 | 1.08 | −0.10 | nae | ||||
| B3LYP | 3.25 | 4.57 | 2.89 | |||||
For the following levels of theory fewer number of cases were included due to SCF convergence problems: B3LYP/aug-cc-pVDZ without counterpoise correction (2 cases missing); M06-2X/LACVP* with explicit counterpoise correction (11 cases missing); M06-2X/aug-cc-pVDZ with explicit counterpoise correction (3 cases missing); M06-2X/aug-cc-pVDZ without counterpoise correction (4 cases missing)
Mean Unsigned Error.
Root Mean Square Deviation.
Average Error.
No consistent way to apply counterpoise corrections.
B3LYP energies were obtained without counterpoise corrections, however, the MM correction part was parametrized with counterpoise corrected interaction energies.