46 |
|
calculated densities which were were significantly lower than |
47 |
|
experimental densities. Analysis of self-diffusion constants shows |
48 |
|
that the original SSD model captures the transport properties of |
49 |
< |
experimental water very well in both the normal and super-cooled |
49 |
> |
experimental water very well in both the normal and supercooled |
50 |
|
liquid regimes. We also present our reparameterized versions of SSD |
51 |
|
for use both with the reaction field or without any long-range |
52 |
|
electrostatic corrections. These are called the SSD/RF and SSD/E |
739 |
|
\end{center} |
740 |
|
\end{figure} |
741 |
|
|
742 |
< |
Fig. \ref{ssdedense} shows the density profile for the SSD/E |
742 |
> |
Figure \ref{ssdedense} shows the density profile for the SSD/E |
743 |
|
model in comparison to SSD1 without a reaction field, other |
744 |
|
common water models, and experimental results. The calculated |
745 |
|
densities for both SSD/E and SSD1 have increased |
752 |
|
better than the SSD value of 0.967$\pm$0.003 g/cm$^3$. The |
753 |
|
changes to the dipole moment and sticky switching functions have |
754 |
|
improved the structuring of the liquid (as seen in figure |
755 |
< |
\ref{grcompare}, but they have shifted the density maximum to much |
755 |
> |
\ref{grcompare}), but they have shifted the density maximum to much |
756 |
|
lower temperatures. This comes about via an increase in the liquid |
757 |
|
disorder through the weakening of the sticky potential and |
758 |
|
strengthening of the dipolar character. However, this increasing |