Oh my... so many things....
A) The major reason Earth rotation speed is getting smaller every year "due to the tides" is not friction. It's gravitational energy exchange with the moon. The tidal bulge gives the moon a slight tangential (with respect to its orbit around Earth) component to gravity, which kicks it into a slightly higher orbit. It's on the order of single-digit centimeters per year. The reaction force to this slows earth's rotation.
B) Conservation of momentum always applies, even if kinetic energy is lost. The bulk Earth does exchange some angular momentum with its atmosphere, but this is, in general, not something that causes meaningful slowing. It is a measurable effect, but sometimes it is Earth-slower, wind-faster and sometimes it is Earth-faster, wind-slower. I think somewhere computed that global warming, with higher average wind speeds, will be something like a few milliseconds slower rotation per day per year kind of thing.
C) The Earth can "lose" angular momentum to space, if atmospheric molecules with high angular momentum also have escape velocity, because those particles will never be able to crash back to earth and then give the momentum back. But this is a small effect.
D) If Venus is slowing by 24
seconds a year, it's due to gravitational effects or perhaps differences in the core and crust rotation, not because of atmospheric "drag". Atmospheric wind speeds are not increasing that much per year, and Venus is probably not losing that much atmospheric mass per year (see C above).
E) A system of objects has two "components" to angular momentum: the angular momentum of the particles spinning about their own axes
and rx
p - the cross product of the displacement vector of the object from the system center of mass and the linear momentum of each object. This is why a collision of two non-spinning objects in space generally ends up causing those objects to start spinning.
F) Asymmetric thermal radiation is an effect, but is really small. We saw this on ... whatever probe that was a couple years ago. But for a planet....yes it's a real effect, but decimal dust. And in general unless it's an outer planet, the effect is going to be dwarfed by radiation pressure from the star anyway.
G) "To be specific"... at least in the academia from which I came, there is no such thing as "heat energy". There is heat
transfer (transfer of energy between two objects due to the difference in their temperature", but there is only kinetic and potential energy. The energy that gives rise to temperature is kinetic energy.
Or maybe I'm misunderstanding what was meant by trying to distinguish between "heat generated due to collisions between particles within the system" and "heat radiated from the sun." Internal collisions between particles in an object do not "generate" heat - they are actually conservative (unless you are talking about collisions that result in chemical or nuclear reactions, which convert potential energy into kinetic?). The key is that energy transferred from an unbalanced external source (like the sun)
might change the momentum of the center of mass of the object while the energy exchanges internal to the object
cannot change the momentum of the center of mass of the object.
A neat - and not intuitive - exercise is to consider a system of two masses connected by an ideal spring. Start initial conditions where both masses have zero velocity. Apply an impulse to the system exactly in line with the spring. Say both masses are mass 1kg (for a total mass of 2kg) and the impulse applied is 2 N-s. This will give the velocity of the center of mass of the system 1 m/s. BUT - what is the kinetic energy of the system? Do you think it's only 1 joule? You can show that it is at least one joule, but it might be much more! It actually depends on how that impulse is applied - the only way to get 1J of kinetic energy is to apply the impulse equally and simultaneously to both masses. ANY other application will require more than 1J input energy, and will result in oscillation between the two
masses as well as translation of the center of mass.
Also note: the velocity of the center of mass will be constant in all cases you pick an inertial reference frame - it won't oscillate!*
*I'm pretty sure on this one - I haven't checked. I think if you pick one of the masses, you might get an apparent oscillating CG velocity, but that's because the masses are an accelerating frame.