| 46 | | 3a. Currently, I am using electron excitation as high temperature cooling source [http://adsabs.harvard.edu/abs/1998ApJ...497..303M Mastrodemons & Morris (1998)] and H2 and water as low temperature cooling source [http://adsabs.harvard.edu/abs/1993ApJ...418..263N Neufeld & Kaufman (1993)]. |
| | 46 | 3a. Currently, I am using electron excitation as high temperature cooling source [http://adsabs.harvard.edu/abs/1998ApJ...497..303M Mastrodemons & Morris (1998)] and H2 and water as low temperature cooling source [http://adsabs.harvard.edu/abs/1993ApJ...418..263N Neufeld & Kaufman (1993)]. The defect of these method is: it admits the chemical interaction but did not fully take them into account. For example, HII can interact with electron to form HI. This chemical reaction will absorb heat thus making the gas's specific heat change ($\Gamma\sim1$). In [http://adsabs.harvard.edu/abs/2013ApJ...763....6T Tomida et al. (2013)] and [http://adsabs.harvard.edu/abs/2015ApJ...801..117T Tomida et al. (2015)] use statistical mechanics to derive $\Gamma$ table as a function of $\rho$ and $T$. I will use their model in future simulation. Besides, there is new data for high temperature cooling (>10000 K) [http://adsabs.harvard.edu/abs/2007ApJS..168..213G Gnat & Sternberg (2007)]. |
| | 47 | |
| | 48 | 3b. Opacity is another key aspect of this problem. MESA has some subroutines that can derive opacity tables. |
| | 49 | |
| | 50 | 3c. Implicit cooling scheme is useful when $t_{cool}<t_{dynamical}$. [http://adsabs.harvard.edu/abs/2009ApJS..181..391T Townsend R.H.D. (2009)] and the reference therein are useful. |
