harrylin said:
But it has little to do with the topic I fear...
Why not? The original question was why c² and not any other proportionality factor between energy and mass. Without other information this may refer to rest energy and rest mass or to total energy and relativistic mass. This question can be answered for both cases at once:
Assuming we know that energy is linear correlated with mass (as used in Newton's definition of momentum) but we do not know the proportionality factor k:
[itex]E = k \cdot m[/itex]
Then the change of mechanic energy is
[itex]dE = F \cdot ds = k \cdot dm[/itex]
According to Newton's second law the force is
[itex]F = \frac{{dp}}{{dt}} = \frac{{d\left( {m \cdot v} \right)}}{{dt}} = m \cdot \frac{{dv}}{{dt}} + v \cdot \frac{{dm}}{{dt}}[/itex]
Integration of the resulting differential equation
[itex]\frac{{dm}}{m} = \frac{{v \cdot dv}}{{k - v^2 }}[/itex]
results in
[itex]m = \frac{{m_0 }}{{\sqrt {1 - \frac{{v^2 }}{k}} }}[/itex]
The constant of integration m
0 is the mass of the body at rest and as this equation gives rational results for k<v
2 only the unknown proportionality factor k must be the square of a maximum velocity that no body can reach or exceed. In relativity there is such a velocity: the speed of light in vacuum. Therefore there is only one possibility for a linear correlation of energy and mass:
[itex]E = m \cdot c^2 = \frac{{m_0 \cdot c^2 }}{{\sqrt {1 - \frac{{v^2 }}{c^2}} }}[/itex]