| 439 |
|
\subsubsection*{Thermal Conductivity} |
| 440 |
|
|
| 441 |
|
Our thermal conductivity calculations show that the NIVS method agrees |
| 442 |
< |
well with the swapping method. Four different swap intervals were |
| 442 |
> |
well with the swapping method. Five different swap intervals were |
| 443 |
|
tested (Table \ref{LJ}). With a fixed scaling interval of 10 time steps, |
| 444 |
|
the target exchange kinetic energy produced equivalent kinetic energy |
| 445 |
|
flux as in the swapping method. Similar thermal gradients were |
| 446 |
|
observed with similar thermal flux under the two different methods |
| 447 |
< |
(Figure \ref{thermalGrad}). |
| 447 |
> |
(Figure \ref{thermalGrad}). Furthermore, with appropriate choice of |
| 448 |
> |
scaling variables, temperature along $x$, $y$ and $z$ axis has no |
| 449 |
> |
observable difference(Figure TO BE ADDED). The system is able |
| 450 |
> |
to maintain temperature homogeneity even under high flux. |
| 451 |
|
|
| 452 |
|
\begin{table*} |
| 453 |
|
\begin{minipage}{\linewidth} |
