density of states
/ DEN-si-tee uhv stayts /
Climb a staircase and the steps near the bottom may be far apart, but higher up they crowd closer and closer until the staircase feels almost like a ramp. The density of states tells you how tightly packed the available energy levels are at each height — how many states lie within each thin slice of the energy range.
More precisely, the density of states (often written g(E)) is the number of microscopic states per unit energy interval at a given energy E. For systems with countless, finely spaced levels — the molecules in a litre of gas, electrons in a metal — it is far more useful to ask 'how many states per small energy band' than to list each level by hand. Multiply the density of states by a Boltzmann factor and integrate, and you recover the partition function.
Why it matters: the density of states is the practical engine behind heat capacities, the colours of solids, the behaviour of semiconductors and the thermodynamics of large systems. It turns an impossible sum over individual states into a smooth integral. The honest caveat is that it is an approximation that works only when levels are so dense that treating them as a continuum introduces negligible error — at very low temperatures, the discreteness can bite back.
For a free particle in a box, the higher the energy, the more ways its motion can be divided among the three directions, so the density of states grows with energy — proportional to the square root of E in three dimensions. That rising count is why translational motion dominates a gas's thermal energy.
In three dimensions the density of states grows as the square root of the energy.
Density of states is the continuum cousin of degeneracy: where degeneracy counts states sitting exactly at one energy, density of states counts states per unit energy across a range. Use the first for sparse levels, the second for dense ones.