Phantom energy is something that appears pathological in some ways. The Chandra data has w=-0.98+-0.07, and yet From Reiss et al data the value for the important parameter is $w = −1.007 ± 0.081$ . The data is not precise enough to rule out phantom energy from straight dark energy with $w~=~-1$. So one might then give this some serious consideration, for it could turn out to be the case.
The scale factor for the evolution of the spacetime, which I work out in a Newtonian framework here How did the universe shift from "dark matter dominated" to "dark energy dominated"? within a Newtonian context, is
$$
H^2~=~\left(\frac{\dot a}{a}\right)^2~=~H_0\left[\frac{\Omega_m}{a^3}~+~(1~-~\Omega_m)a^{-3(1+w)}\right]
$$
Here $\Omega_m$ holds for matter, which is $.26$ for dark matter and $.04$ for luminous matter so $\Omega_m~=~.3$. Obviously $1~-~\Omega_m$ pertains to dark energy, or phantom energy. Below is a graph of what happens to the scale factor for different values of $w$
The case $w~=~-1/3$ is the coasting cosmology and for smaller $w$ there is exponential expansion. For $w~=~-1$ this is the dark energy condition, and for smaller we have phantom energy.
This has the effect of increasing the vacuum energy or dark energy so that within a finite time the vacuum energy approaches Planck energy density and the universe effectively becomes a singularity. This big rip occurs at a time $t~\simeq~t_0~-~(2/3)H_0^{-1}(1~-~\Omega_m)^{-1/2}$ the Hubble paramter is about $70km/sec-Mpc$. If we assume that the Riess value is real $w~=~-1.007$ then the big rip will occur around $1.6\times 10^{12}$ years. This would be around the time the last stars wink out.
We may think of this as well according to an increase in the cosmological constant. By some mechanism the vacuum energy increases and the cosmological constant increases. The cosmological constant increases by $\Lambda~\simeq~\Lambda_0(1~-~w)t$. For the cosmological constant $H^2~=~\Lambda/3c^2$ $=~8\pi G\rho/3$, it is clear that the vacuum energy increases and the cosmological horizon distance constricts as $r_h~=~\sqrt{3/\Lambda}$. The temperature of the horizon is then $T~=~(2\pi\hbar/k)\sqrt{\Lambda/3}$ or proportional to $1/r_h$ as it constricts.
The question of equilibrium is similar to the problem of quantum black holes. Given a black hole of mass $M$ it has temperature $T~=~1/8\pi M$. Assume this sits in a world with the same background temperature. Now if the black hole absorbs a photon of mass-energy $\delta m$ then by the increase in black hole mass it puts the black hole at a colder temperature and away from this equilibrated (not equilibrium) condition. The converse of course means if the black hole emits a photon it is not hotter and will not preferentially radiate photons into the background. Does something similar happen to a big rip cosmology?
Let us look at a black hole in the de Sitter spacetime. The stationary metric element $g_{tt}~=~1~-~2m/r~-~\lambda r^3/3$ is plotted below for various values of the cosmological constant.
The funny thing is that the horizons of black hole and the cosmology merge and beneath the horizontal black line the metric is effectively spacelike. Before then the horizon of the black hole appears to move out, increasing the horizon area and by extension decreasing its temperature. For $w~=~-1$ there is a nice case of where the energy density $\rho$ and the “work” due to pressure $pc$ are equal in magnitude and opposite in sign. So there is no net energy from dark energy or the vacuum that pumps up black holes to larger mass. However, for $w~<~-1$ we can think of the black hole as gaining mass right from the vacuum until it reaches a type of equilibrium condition with the cosmological horizon. However, if the cosmological horizon is constricting in this is only a temporary condition.
The constriction of the cosmological horizon means it temperature increases, and by extension matter is heated by the frame dragging of extended objects and from the background. The constriction however is not abated by matter reaching some temperature limit, unless that is the string Hagedorn or Planck temperature.
