Okay. I'm taking an astrophysics class and upon hearing that the centers of large stars can reach temperatures in excess of 100 million K, I did some rough calculations. I found that the average velocity of a hydrogen nucleus at a temperature of hundreds of millions of degrees, not accounting for relativity, would be over the speed of light. They would of course actually be moving near the speed of light with an increased mass from the mass of their kinetic energy (E=MC^2).

So, at the centers of the biggest stars you have temperatures so high that the particles involved gain relativistic mass.

I got to thinking - in the largest stars, could the thermal energy of the particles making up the star be high enough that its mass is a significant fraction of the star's total mass?

I calculated the total thermal energy present inside a 60 solar mass, 1,000,000 solar luminosity star. Assuming an average path length for a photon of 100,000 lightyears (like for our sun) of scattering before it is emitted from the surface of the star, you get a figure of 1.23*10^45 Joules of heat energy and radiation present inside the star. Thats 1/146 of a SOLAR MASS when you convert energy to mass - meaning that more than 1/9,000 of the mass of this large star comes only from its thermal energy!!!

If you assume that because the star is more massive the photons will scatter even more on their way out, you get an even more incredible result. I took the cube root of 60 for my multiplier for the path length, thinking that the total amount of matter from the core to the surface along any one line would be the deciding factor, I could be wrong. Anyways, if you multiply the path length by the cube root of 60 (3.91) you get that a full 1/37.5 solar mass is present in the form of thermal energy and radiation. This would mean that a full 1/2241 of the star's mass comes from its thermal energy alone.

Am I totally off my rocker here?

In addition this got me thinking - say you have a large gas cloud that collapses. It will heat up, as the gravitational potential energy is converted into heat. Since conservation of energy holds and this heat energy can be considered mass (due to the relativistic effects on the atoms if nothing else) then does this mean that this gravitational potential energy has mass as well? And therefore, could large diffuse objects have larger masses than small compact objects with the same matter content and temperature?