Abstract
Semiclassical dispersion corrections developed by Grimme and co-workers have become indispensable in applications of Kohn-Sham density functional theory. A deeper understanding of the underlying parametrization might be crucial for well-founded further improvements of this successful approach. To this end, we present an in-depth assessment of the fit parameters present in semiclassical (D3-type) dispersion corrections by means of a statistically rigorous analysis. We find that the choice of the cost function generally has a small effect on the empirical parameters of D3-type dispersion corrections with respect to the reference set under consideration. Only in a few cases, the choice of cost function has a surprisingly large effect on the total dispersion energies. In particular, the weighting scheme in the cost function can significantly affect the reliability of predictions. In order to obtain unbiased (data-independent) uncertainty estimates for both the empirical fit parameters and the corresponding predictions, we carried out a nonparametric bootstrap analysis. This analysis reveals that the standard deviation of the mean of the empirical D3 parameters is small. Moreover, the mean prediction uncertainty obtained by bootstrapping is not much larger than previously reported error measures. On the basis of a jackknife analysis, we find that the original reference set is slightly skewed, but our results also suggest that this feature hardly affects the prediction of dispersion energies. Furthermore, we find that the introduction of small uncertainties to the reference data does not change the conclusions drawn in this work. However, a rigorous analysis of error accumulation arising from different parametrizations reveals that error cancellation does not necessarily occur, leading to a monotonically increasing deviation in the dispersion energy with increasing molecule size. We discuss this issue in detail at the prominent example of the C<sub>60</sub> "buckycatcher". We find deviations between individual parametrizations of several tens of kilocalories per mole in some cases. Hence, in combination with any calculation of dispersion energies, we recommend to always determine the associated uncertainties for which we will provide a software tool.
Introduction
Standard Kohn–Sham Density Functional Theory (DFT) often fails to account for London dispersion—the weak, attractive van der Waals forces essential for molecular stability. To address this, the "D3" semiclassical correction by Stefan Grimme has become the industry standard. However, D3 relies on empirical fit parameters calibrated against small-molecule reference data. This paper provides a rigorous statistical "stress test" for these corrections, revealing that while they are robust for small molecules, small uncertainties can accumulate into massive errors in large, complex systems.
Methodology and Parameter Landscape
The researchers employed two powerful statistical techniques to evaluate the D3-BJ and D3-zero variants: Nonparametric Bootstrapping, generating 10,000 synthetic datasets to create a distribution of parameter values, and Jackknife Analysis, systematically removing data points to identify influential outliers.
As shown in Figure 3, the parameter space for D3-BJ is well-defined for the reference sets used (such as GMTKN30), but the confidence intervals highlight the underlying statistical sensitivity. The study also introduced the effective system size ($\Omega_{\text{eff}}$) to quantify how dispersion interactions scale with molecular complexity.
Key Results: The Risk of Error Accumulation
While the standard deviation of D3 parameters is small for small molecules, the study proves that errors often accumulate monotonically rather than cancelling out. This is most evident in large systems like the C$_{60}$ "buckycatcher."
Table 1 illustrates the optimized parameters across various functionals. A critical insight is the "Weighting Trap": the original D3 calibration weighted rare gas dimers 20 times more than other molecules. Removing this non-uniform weighting (parenthetical values in Table 1) resulted in a staggering 63.47 kcal/mol difference in the buckycatcher's dispersion energy. This deviation far exceeds "chemical accuracy" (~1 kcal/mol) and could lead to qualitatively incorrect conclusions in research.
Conclusion
Researchers should not treat dispersion corrections as "black boxes." The accumulation of parameter uncertainty can skyrocket as molecules grow. To mitigate this, the authors have released BootD3, a software tool that allows users to calculate "virtual error bars" for their own molecular systems, ensuring that reported energies are statistically meaningful. Future work will extend this analysis to the newer D4 model.
Summary updated