How do individual, microscopic atoms—which are constantly jiggling around due to thermal energy—manage to align perfectly with one another to form a permanent magnet? The competition between the tendency of neighbouring atoms to align and the disruptive effect of temperature is at the core of statistical mechanics.

The simplest and most profound way to understand this is the Ising Model. Originally proposed to understand ferromagnetism, it has become one of the most studied models in physics because it beautifully demonstrates how a system transitions between order and disorder.

Spins on a Grid

Imagine a grid, like a chessboard. At each square, there is an atom with a "spin". In the simplest version of the Ising model, this spin can only point in one of two directions: "up" (+1) or "down" (−1).

These spins interact with their immediate neighbors. If the interaction is ferromagnetic, neighbouring spins want to align with each other. Pointing in the same direction lowers the total energy of the system. We can also apply an external magnetic field, which acts like a subtle nudge encouraging all spins to align with the field.

The total energy, or Hamiltonian, for the whole grid is calculated by summing up the interactions between all pairs of neighboring spins and the effect of the external field:

$$ H = -J \sum_{\langle ij \rangle} s_i s_j - h \sum_i s_i $$

Here, $J > 0$ means neighbors prefer to align (ferromagnetism), and $h$ is the external field.

The Heat is On: Metropolis-Hastings

If left alone, the spins would eventually all point the same way to reach the lowest possible energy state. However, temperature introduces random fluctuations. As temperature increases, the spins gain thermal energy, allowing them to occasionally flip against the preferred alignment.

We simulate this using the Metropolis-Hastings algorithm. It randomly selects a spin and proposes flipping it. If the flip lowers the total energy, it happens. But even if it increases the energy, the flip might still occur with a probability that depends on the temperature. This allows the system to jiggle out of local energy minimums and represents the thermal noise in a real material.

The Critical Point

The magic happens at a very specific temperature known as the critical temperature ($T_c$).

• Below $T_c$: The aligning force overpowers the thermal jiggling. Huge "domains" of aligned spins form. The system is ordered, acting like a magnet.
• Above $T_c$: Thermal energy dominates. The spins flip rapidly and randomly. The overall magnetization is zero. The system is disordered.
• Exactly at $T_c$: The system is balanced on a knife edge. It exhibits beautiful, scale-free fractal patterns. You will see clusters of aligned spins of every possible size, from single atoms to massive blobs.

Lars Onsager's exact solution of the 2D Ising model in 1944 mathematically proved this phase transition, marking a monumental achievement in 20th-century theoretical physics.

Try It Yourself

You can explore this fundamental model of statistical mechanics in our interactive Ising Model simulation. Start from a cold ordered state and slowly heat it up past the critical temperature to watch the domains fragment, or try applying an external magnetic field to watch a minority phase vanish!