Case Study 2: Bose-Einstein Condensation — When Bosons Gang Up

Overview

In 1924, Satyendra Nath Bose sent a short paper to Albert Einstein, proposing a new derivation of the Planck blackbody radiation formula that treated photons as genuinely indistinguishable particles. Einstein immediately recognized the significance and extended the idea to massive particles, predicting a new phase of matter: a gas of bosons that, at sufficiently low temperatures, would macroscopically occupy a single quantum state. This prediction — Bose-Einstein condensation (BEC) — was confirmed experimentally 71 years later, in 1995, earning Eric Cornell, Carl Wieman, and Wolfgang Ketterle the 2001 Nobel Prize in Physics.

This case study explores BEC as the most dramatic consequence of bosonic exchange symmetry: the tendency of identical bosons to "gang up" in the same quantum state.

Part 1: The Theoretical Prediction

Bose's Insight

In June 1924, Bose derived the Planck radiation law using a new counting method. The key idea: when distributing identical photons among energy states, configurations that differ only by which photon is in which state should not be counted separately. Photons are not distinguishable balls in distinguishable boxes — they are indistinguishable quanta in distinguishable modes.

This "Bose counting" naturally produces the Planck spectrum. The older derivation by Einstein (1917) had required the ad hoc assumption of stimulated emission. Bose showed that stimulated emission is a consequence of indistinguishability.

Einstein extended Bose's counting to massive particles (atoms) and derived the Bose-Einstein distribution:

$$\bar{n}(\epsilon) = \frac{1}{e^{(\epsilon - \mu)/k_BT} - 1}$$

where $\epsilon$ is the single-particle energy, $\mu$ is the chemical potential (always $\mu \leq 0$ for non-relativistic bosons), $k_B$ is the Boltzmann constant, and $T$ is the temperature.

The Condensation Prediction

Einstein realized that at sufficiently low temperatures, a remarkable thing happens. As the temperature decreases, the chemical potential $\mu$ must increase (become less negative) to maintain the correct total particle number. But $\mu$ cannot exceed zero (if it did, the ground-state occupation would be negative, which is unphysical). When $\mu$ reaches zero, the ground-state occupation:

$$N_0 = \frac{1}{e^{-\mu/k_BT} - 1} \approx \frac{k_BT}{|\mu|}$$

diverges. This signals a phase transition: a macroscopic fraction of the particles "condenses" into the single-particle ground state.

Critical Temperature

For a gas of $N$ non-interacting bosons of mass $m$ in a box of volume $V$, the critical temperature for BEC is:

$$T_c = \frac{2\pi\hbar^2}{m k_B}\left(\frac{N/V}{2.612}\right)^{2/3} = \frac{2\pi\hbar^2}{m k_B}\left(\frac{n}{\zeta(3/2)}\right)^{2/3}$$

where $n = N/V$ is the number density and $\zeta(3/2) \approx 2.612$ is the Riemann zeta function.

The condition for BEC is equivalent to requiring that the thermal de Broglie wavelength $\lambda_{\text{th}} = h/\sqrt{2\pi m k_B T}$ become comparable to the inter-particle spacing $n^{-1/3}$:

$$n \lambda_{\text{th}}^3 \gtrsim 2.612$$

When the wave packets of individual bosons overlap, the particles "realize" they are identical, and quantum statistics takes over.

Below $T_c$, the fraction of particles in the ground state is:

$$\frac{N_0}{N} = 1 - \left(\frac{T}{T_c}\right)^{3/2}$$

At $T = 0$, all particles are in the ground state: $N_0 = N$.

📊 By the Numbers: For a gas of $^{87}$Rb atoms at the density used in the original BEC experiments ($n \sim 10^{13}$ cm$^{-3}$), the critical temperature is $T_c \sim 170$ nK. For liquid $^4$He ($n \sim 2 \times 10^{22}$ cm$^{-3}$, $m$ about 50 times larger), the predicted $T_c \sim 3.1$ K, remarkably close to the observed superfluid transition at 2.17 K (the lambda point). The discrepancy arises because liquid helium is a strongly interacting system, not an ideal gas.

Part 2: Why Fermions Cannot Condense (And Why Cooper Pairs Can)

The Fermi Sea

Contrast the bosonic ground state with the fermionic one. At $T = 0$:

• Bosons: All $N$ particles occupy the lowest single-particle state. The $N$-particle ground state is $|0\rangle^{\otimes N}$ (symmetrized).
• Fermions: Each particle must occupy a distinct single-particle state (Pauli exclusion). The $N$ particles fill the lowest $N$ states, up to the Fermi energy $\epsilon_F$. This is the Fermi sea.

The Fermi sea is the exact opposite of a condensate: instead of all particles in one state, each particle is in a different state. The Fermi energy for a typical metal is $\epsilon_F \sim 5$ eV, corresponding to a Fermi temperature $T_F = \epsilon_F/k_B \sim 60{,}000$ K. Even at room temperature, the electron gas in a metal is deeply degenerate — quantum statistics completely dominates.

The Cooper Pair Trick

In 1957, Bardeen, Cooper, and Schrieffer (BCS) showed that in certain metals at low temperatures, electrons can form bound pairs — Cooper pairs — through an attractive interaction mediated by phonons (lattice vibrations). Each Cooper pair consists of two electrons with opposite spin and opposite momentum:

$$(k\!\uparrow, \, -k\!\downarrow)$$

Since each Cooper pair contains an even number of fermions (two), it is a composite boson. A gas of Cooper pairs can undergo Bose-Einstein condensation, and this is the mechanism of superconductivity. The BCS ground state is a coherent superposition of Cooper pairs, all condensed into the same pair state.

This explains why superconductors exhibit zero electrical resistance and the Meissner effect (expulsion of magnetic fields): the macroscopic quantum coherence of the condensate prevents the scattering processes that cause resistance.

💡 Key Insight: BEC requires bosons, and the Pauli exclusion principle prevents fermions from condensing. But fermions can form bosons (Cooper pairs), and those composite bosons can condense. Nature finds a way: even in a system of fermions, bosonic behavior can emerge through pairing. This is one of the deepest connections between the physics of identical particles and macroscopic quantum phenomena.

Part 3: The Experimental Achievement (1995)

The Challenge

Einstein predicted BEC in 1925, but experimental realization took 70 years. The challenge was formidable:

1. Low temperature: $T_c \sim 100$ nK requires cooling atoms to within a billionth of a degree above absolute zero.
2. Dilute gas: At such low temperatures, most materials are solid. A BEC must be a dilute gas to avoid solidification.
3. No container: A material container would heat the gas. The atoms must be trapped using magnetic or optical fields.

Laser Cooling and Evaporative Cooling

The breakthrough came through a combination of two cooling techniques:

Laser cooling (1985–1995): Atoms are slowed by the radiation pressure of laser beams tuned slightly below an atomic resonance. An atom moving toward the laser sees the light Doppler-shifted into resonance, absorbs a photon, and receives a momentum kick opposing its motion. This "optical molasses" can cool atoms to $\sim 100$ $\mu$K — still far above $T_c$.

Steven Chu, Claude Cohen-Tannoudji, and William Phillips received the 1997 Nobel Prize for developing laser cooling.

Evaporative cooling: The laser-cooled atoms are transferred to a magnetic trap, and the trap depth is gradually reduced, allowing the most energetic atoms to escape. The remaining atoms re-thermalize at a lower temperature. This is the same principle by which a cup of coffee cools — the fastest molecules evaporate, leaving the rest cooler.

The combination of laser cooling and evaporative cooling achieved temperatures below 100 nK.

The First BEC

On June 5, 1995, Eric Cornell and Carl Wieman at JILA (University of Colorado, Boulder) produced the first gaseous BEC using approximately 2000 $^{87}$Rb atoms at a temperature of about 170 nK. Four months later, Wolfgang Ketterle at MIT produced a much larger BEC of $^{23}$Na atoms, enabling more detailed studies.

The signature of BEC is unmistakable: when the trap is turned off and the gas expands, the velocity distribution is measured by absorption imaging. Above $T_c$, the distribution is a broad thermal cloud. Below $T_c$, a sharp, narrow peak appears at the center — the condensate — sitting atop the thermal background. As the temperature decreases further, the condensate peak grows while the thermal cloud shrinks, until at the lowest temperatures essentially all atoms are in the condensate.

🧪 Experiment: The "smoking gun" of BEC is the anisotropic expansion of the condensate. A thermal gas in an elongated trap expands isotropically when released (the velocities are random). But a condensate, described by a single macroscopic wavefunction, expands faster in the direction that was most tightly confined (due to the uncertainty principle: $\Delta x$ was smallest, so $\Delta p$ is largest). This "inversion of aspect ratio" was observed in both the JILA and MIT experiments and is a direct signature of macroscopic quantum coherence.

Part 4: What Does a BEC Look Like?

A Macroscopic Wavefunction

In a BEC, all $N$ bosons occupy the same single-particle state $\phi_0(\mathbf{r})$. The $N$-particle wavefunction is:

$$\Psi(\mathbf{r}_1, \mathbf{r}_2, \ldots, \mathbf{r}_N) = \prod_{i=1}^{N} \phi_0(\mathbf{r}_i)$$

This is already symmetric (as required for bosons) — no symmetrization is needed when all particles are in the same state.

The system is described by a single macroscopic wavefunction $\Phi(\mathbf{r}) = \sqrt{N}\phi_0(\mathbf{r})$, satisfying $\int |\Phi|^2 d^3r = N$. This is the order parameter of the condensate.

For a gas in a harmonic trap $V(\mathbf{r}) = \frac{1}{2}m\omega^2 r^2$, the ground state of the non-interacting system is a Gaussian:

$$\phi_0(\mathbf{r}) = \left(\frac{m\omega}{\pi\hbar}\right)^{3/4} \exp\left(-\frac{m\omega r^2}{2\hbar}\right)$$

With interactions (which are repulsive for most alkali gases), the condensate spreads out. In the Thomas-Fermi limit (large $N$, strong interactions), the density profile becomes an inverted parabola:

$$n(\mathbf{r}) = |\Phi(\mathbf{r})|^2 = \frac{\mu - V(\mathbf{r})}{g} \quad \text{for } V(\mathbf{r}) < \mu$$

where $g = 4\pi\hbar^2 a_s/m$ is the interaction strength ($a_s$ is the $s$-wave scattering length) and $\mu$ is the chemical potential.

Coherence

A BEC exhibits long-range phase coherence: the macroscopic wavefunction has a well-defined phase $\theta(\mathbf{r})$ across the entire condensate. This has been demonstrated by:

• Interference experiments: Two BECs released from adjacent traps produce interference fringes, just like the double-slit experiment with light. The fringe spacing is set by the de Broglie wavelength.

• Vortices: When a BEC is rotated, it cannot rotate rigidly (that would require a continuous range of angular momenta). Instead, it develops quantized vortices — lines of zero density around which the phase winds by $2\pi$. The circulation is quantized in units of $h/m$:

$$\oint \mathbf{v} \cdot d\mathbf{l} = \frac{nh}{m}, \qquad n = 0, \pm 1, \pm 2, \ldots$$

This is a direct consequence of the single-valuedness of the macroscopic wavefunction.

Part 5: BEC and Superfluid Helium

The Lambda Transition

Liquid $^4$He undergoes a phase transition at $T_\lambda = 2.17$ K (the "lambda point," named for the shape of the specific heat curve). Below $T_\lambda$, helium becomes a superfluid: it flows without viscosity, creeps up container walls, and exhibits quantized circulation.

London (1938) proposed that the lambda transition is a form of Bose-Einstein condensation. However, liquid helium is a strongly interacting system (unlike the dilute gases in modern BEC experiments), so the ideal-gas BEC theory is only qualitatively applicable:

The fact that only about 8% of the helium atoms are in the $k=0$ state at $T=0$ (due to interactions depleting the condensate into higher momentum states) shows that BEC in liquid helium is a more complex phenomenon than in the ideal gas. Nevertheless, the superfluid properties — zero viscosity, quantized vortices, second sound — are all consequences of macroscopic quantum coherence rooted in bosonic exchange symmetry.

$^3$He vs. $^4$He

The contrast between the two helium isotopes is one of the most striking demonstrations of the boson-fermion distinction:

• $^4$He (2 protons + 2 neutrons + 2 electrons = 6 fermions → composite boson): Superfluid below 2.17 K via Bose-Einstein condensation.
• $^3$He (2 protons + 1 neutron + 2 electrons = 5 fermions → composite fermion): Does NOT become superfluid until 0.0025 K, and only through the formation of Cooper-like pairs ($^3$He-$^3$He pairs analogous to Cooper pairs in superconductors).

Same element. Nearly identical mass (3 vs. 4 amu). Completely different quantum behavior at low temperatures — because one is a boson and the other is a fermion.

📊 By the Numbers: The superfluid transition temperatures of $^4$He and $^3$He differ by a factor of nearly 1000: $T_c(^4\text{He}) / T_c(^3\text{He}) \approx 870$. This dramatic ratio reflects the fundamental difference between direct BEC (bosons) and BCS-like pairing (fermions). Pair formation is an exponentially weak effect, suppressed by $\sim e^{-1/|V|g(\epsilon_F)}$, where $V$ is the pairing interaction strength.

Part 6: Modern Frontiers

BEC as a Quantum Laboratory

Since 1995, BEC experiments have become routine in atomic physics laboratories worldwide. BECs are now used as:

• Quantum simulators: Ultracold atoms in optical lattices (periodic potentials created by laser standing waves) simulate condensed matter systems like the Hubbard model, enabling the study of quantum phase transitions, superfluidity, and magnetism in controlled settings.

• Precision measurements: BEC interferometers are used for ultra-precise measurements of gravity, rotations, and fundamental constants. Atom interferometers based on BECs can measure gravitational acceleration to $10^{-9}g$ precision.

• Tests of fundamental physics: BECs have been used to create "artificial" magnetic fields for neutral atoms, observe the Efimov effect (universal three-body bound states), and study quantum turbulence (the analog of classical turbulence in a superfluid).

Fermionic Condensates

In 2003, Deborah Jin's group at JILA achieved the first fermionic condensate — a degenerate Fermi gas of $^{40}$K atoms in which atom pairs form a BEC-like state. By tuning the interaction strength using a Feshbach resonance, they could smoothly cross over from a BEC of tightly bound molecules to a BCS-like state of loosely bound Cooper pairs. This BEC-BCS crossover had been predicted theoretically but never observed.

BEC in Exotic Systems

Bose-Einstein condensation has been observed in systems beyond cold atoms:

• Exciton-polaritons in semiconductor microcavities (half-light, half-matter quasi-particles)
• Magnons (quantized spin waves) in ferromagnetic films
• Photons in a dye-filled optical microcavity (Klaers et al., 2010)

Each of these demonstrates the universality of bosonic statistics: wherever identical bosons exist, they tend to pile up in the same state.

Discussion Questions

1. Einstein predicted BEC in 1925, but it was not experimentally realized until 1995. What were the key technical barriers, and what breakthroughs enabled the eventual realization? What does this 70-year gap tell us about the relationship between theoretical prediction and experimental capability?

2. Liquid $^4$He becomes a superfluid, but only $\sim 8\%$ of the atoms are actually in the $k=0$ condensate at $T = 0$. How can a system be a superfluid if most of the particles are NOT in the condensate? What is the distinction between condensation and superfluidity?

3. The BEC-BCS crossover shows that there is a smooth connection between Bose-Einstein condensation (bosons) and Cooper pairing (fermions). Does this blur the sharp distinction between bosons and fermions, or does it reinforce it? Explain.

4. A photon BEC was created in 2010 by Klaers et al. But photons already "condense" in a laser — all the photons are in the same mode. What is the difference between a laser and a photon BEC? (Hint: think about thermodynamic equilibrium.)

5. Bose-Einstein condensation requires that the thermal de Broglie wavelength $\lambda_{\text{th}}$ be comparable to the inter-particle spacing. For electrons in a metal, $\lambda_{\text{th}} \sim 1$ nm and the inter-particle spacing is $\sim 0.2$ nm — so the wavefunctions strongly overlap. Why don't electrons in metals undergo BEC?