| 557 |
|
|
| 558 |
|
\begin{figure} |
| 559 |
|
\includegraphics[width=\linewidth]{thermalHist} |
| 560 |
< |
\caption{Speed distribution for thermal conductivity using |
| 561 |
< |
``swapping'' and NIVS-RNEMD methods. Shown is from simulations under |
| 562 |
< |
${\langle T^* \rangle = 0.8}$ with an swapping interval of 200 |
| 563 |
< |
time steps (equivalent ${J_z^*\sim 0.2}$). In circled areas, |
| 564 |
< |
distributions from ``swapping'' RNEMD simulation have deviations |
| 565 |
< |
from ideal Maxwell-Boltzmann distributions.} |
| 560 |
> |
\caption{Velocity and speed distributions that develop under the |
| 561 |
> |
swapping and NIVS-RNEMD methods at high flux. The distributions for |
| 562 |
> |
the hot bins (upper panels) and cold bins (lower panels) were |
| 563 |
> |
obtained from Lennard-Jones simulations with $\langle T^* \rangle = |
| 564 |
> |
4.5$ with a flux of $J_z^* \sim $ (equivalent to a swapping interval |
| 565 |
> |
of 10 time steps). This is a relatively large flux which shows the |
| 566 |
> |
non-thermal distributions that develop under the swapping method. |
| 567 |
> |
NIVS does a better job of producing near-thermal distributions in |
| 568 |
> |
the bins.} |
| 569 |
|
\label{thermalHist} |
| 570 |
|
\end{figure} |
| 571 |
|
|
