A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
A microstate describes the complete quantum or classical state of every particle in a system, while a macrostate describes the system using only measurable properties like temperature, pressure, and volume. Statistical mechanics connects these levels: the multiplicity of microstates corresponding to a single macrostate determines its entropy and thermodynamic properties.
Start with simple systems like a gas with a few distinguishable particles, count the microstates for different energy configurations, and observe how the number grows exponentially with system size.
You already know from thermodynamics that entropy increases in isolated systems and that thermodynamic processes are described by state variables like T, P, V, and U. But thermodynamics doesn't explain *why* entropy increases — it simply asserts it. Statistical mechanics provides the microscopic foundation, and it starts with the distinction between microstates and macrostates.
A macrostate is a description of the system using only the variables you can directly observe or control: total energy, volume, number of particles, pressure, temperature. A microstate is a complete specification of the mechanical state of every particle — positions and momenta for a classical gas, or the occupation of each quantum energy level for a quantum system. The same macrostate (say, 1 liter of nitrogen at 300 K and 1 atm) is consistent with an astronomically large number of different microstates. The key quantity is Ω(E, V, N), the multiplicity — the count of microstates consistent with a given macrostate.
The fundamental postulate of statistical mechanics is that in equilibrium, every accessible microstate is equally probable. From this single assumption, everything follows. The macrostate with the most microstates — the largest Ω — is overwhelmingly the most likely to be observed, because it is compatible with the most microscopic arrangements. Boltzmann's entropy is S = k_B ln Ω: entropy is simply the logarithm of multiplicity. This makes entropy increase trivially understandable — a system evolving from a low-multiplicity state (e.g., all molecules in one half of a box) to a high-multiplicity state (molecules spread throughout the box) is simply moving from an improbable configuration to a vastly more probable one. The second law is not a fundamental constraint of nature so much as a statement about overwhelming statistical likelihood.
To make this concrete, consider 4 distinguishable particles with 4 units of total energy. The configuration where one particle has all the energy (4,0,0,0) can be arranged in 4 ways (any one of the 4 particles holds all the energy). The configuration (1,1,1,1) — energy equally distributed — is just 1 arrangement as written, but accounting for all orderings there are many more microstates compatible with "roughly equal" energy distributions than with "one particle has all the energy." As N grows to Avogadro's number, the ratio of multiplicities between an "ordered" and "disordered" macrostate becomes so enormous (Ω_disordered / Ω_ordered ~ eN) that the ordered state is never observed spontaneously. The arrow of time emerges from counting.