Chapter 34: Second Quantization — From Particles to Fields

"The concept of identical particles does not exist classically... In quantum mechanics the situation is quite different and has no classical analogue whatsoever." — Paul Dirac

In Chapter 4, you met the quantum harmonic oscillator and its ladder operators $\hat{a}$ and $\hat{a}^\dagger$. Those operators raised and lowered energy levels, and you may have noticed a peculiar linguistic choice: we called $\hat{a}^\dagger$ the creation operator and $\hat{a}$ the annihilation operator. At the time, this was just convenient language. A creation operator "creates" a quantum of energy $\hbar\omega$, and an annihilation operator "destroys" one.

In this chapter, we discover that the language was not metaphorical. It was literal.

Second quantization is the mathematical framework in which particles themselves — electrons, photons, phonons — are created and destroyed by operators acting on a generalized state space called Fock space. This is not merely a change of notation. It is a conceptual revolution: the shift from asking "where are the particles?" to asking "how is the field excited?" It is the doorway from quantum mechanics to quantum field theory.

The machinery is beautiful and, once mastered, remarkably efficient. Problems that would require pages of Slater determinants (Ch 15) collapse to a few lines of operator algebra. And the framework does not merely simplify calculations — it makes certain physics visible that was invisible before: particle creation in relativistic collisions, quasiparticle excitations in solids, and the vacuum as a seething sea of quantum fluctuations.

Learning paths: - 🏃 Streamlined path: Focus on Sections 34.1–34.3 (creation/annihilation operators and Fock space), then skip to 34.6 (phonons) for a concrete application. Return to 34.4–34.5 when you need many-body calculations. - 🔬 Deep dive path: Work through everything sequentially. Sections 34.4 and 34.5 are essential if you plan to study condensed matter physics or quantum field theory (Ch 37).

34.1 Creation and Annihilation Operators for Bosons

The QHO as Prototype

Let us begin with what we already know. In Chapter 4, we defined the ladder operators for the harmonic oscillator:

$$\hat{a} = \sqrt{\frac{m\omega}{2\hbar}}\left(\hat{x} + \frac{i\hat{p}}{m\omega}\right), \qquad \hat{a}^\dagger = \sqrt{\frac{m\omega}{2\hbar}}\left(\hat{x} - \frac{i\hat{p}}{m\omega}\right)$$

with the fundamental commutation relation:

$$[\hat{a}, \hat{a}^\dagger] = 1$$

The number operator $\hat{n} = \hat{a}^\dagger\hat{a}$ has eigenstates $|n\rangle$ with $\hat{n}|n\rangle = n|n\rangle$, and:

$$\hat{a}^\dagger|n\rangle = \sqrt{n+1}\,|n+1\rangle, \qquad \hat{a}|n\rangle = \sqrt{n}\,|n-1\rangle$$

Now consider a radical reinterpretation. Instead of thinking of $|n\rangle$ as the $n$-th excited state of a single oscillator, think of it as a state containing $n$ identical bosonic particles (photons, phonons, or any boson) in a particular single-particle state. The operator $\hat{a}^\dagger$ does not merely raise the energy level — it creates a particle. The operator $\hat{a}$ destroys one.

This reinterpretation is not ad hoc. It is forced on us by the physics of identical particles and the structure of quantum field theory. Let us see how.

🔗 Connection (Ch 4): We are reinterpreting the ladder operators from Section 4.4. The mathematics is identical; the physical interpretation has expanded enormously. The $\sqrt{n+1}$ and $\sqrt{n}$ factors that seemed like mathematical curiosities now have a physical meaning: they encode the bosonic statistics of identical particles.

Many Modes: The General Bosonic Algebra

In a realistic system, particles can occupy many different single-particle states. Label these states by an index $k$ (which might represent momentum, position, spin, or any complete set of quantum numbers). For each single-particle state $k$, we introduce a pair of operators $\hat{a}_k$ and $\hat{a}_k^\dagger$ satisfying the bosonic commutation relations:

$$[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$$

$$[\hat{a}_k, \hat{a}_{k'}] = 0, \qquad [\hat{a}_k^\dagger, \hat{a}_{k'}^\dagger] = 0$$

These three equations are the complete algebraic specification of bosonic particles. Everything else follows.

The first relation says that creating and then annihilating a particle in the same mode differs from the reverse order by exactly one unit — this is the quantum signature of particle creation. The second and third relations say that creation operators for different modes commute, as do annihilation operators — you can create particles in any order without affecting the result.

💡 Key Insight: The commutation relations $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$ are the defining property of bosons. They encode Bose-Einstein statistics algebraically. You do not need to symmetrize wavefunctions by hand (as in Ch 15) — the algebra does it for you automatically.

Building Multi-Particle States

Starting from the vacuum state $|0\rangle$ (defined by $\hat{a}_k|0\rangle = 0$ for all $k$ — no particles in any mode), we construct multi-particle states by applying creation operators:

One particle in mode $k$: $$|1_k\rangle = \hat{a}_k^\dagger|0\rangle$$

Two particles in mode $k$: $$|2_k\rangle = \frac{(\hat{a}_k^\dagger)^2}{\sqrt{2!}}|0\rangle$$

General state with $n_k$ particles in mode $k$: $$|n_k\rangle = \frac{(\hat{a}_k^\dagger)^{n_k}}{\sqrt{n_k!}}|0\rangle$$

The $\sqrt{n_k!}$ in the denominator ensures proper normalization. This is precisely the structure we found for QHO energy eigenstates in Chapter 4, equation (4.42). The connection is exact — and not a coincidence.

For particles distributed across multiple modes: $$|n_1, n_2, n_3, \ldots\rangle = \prod_k \frac{(\hat{a}_k^\dagger)^{n_k}}{\sqrt{n_k!}}|0\rangle$$

Because all creation operators commute with each other, the order in which we apply them does not matter. This is the bosonic symmetry: swapping any two identical bosons leaves the state unchanged.

The Number Operator

For each mode $k$, the number operator is:

$$\hat{n}_k = \hat{a}_k^\dagger \hat{a}_k$$

with eigenvalue $n_k$ (the number of particles in mode $k$). The total number operator is:

$$\hat{N} = \sum_k \hat{n}_k$$

Unlike in ordinary quantum mechanics where particle number is fixed, in second quantization $\hat{N}$ is an operator whose value can change. This is one of the deepest conceptual shifts: particle number becomes a dynamical quantity.

⚠️ Common Misconception: Students sometimes think the vacuum $|0\rangle$ is "nothing." It is not — it is the state of the quantum field with no excitations. The vacuum has definite physical properties (zero-point energy, vacuum fluctuations, the Casimir effect). It is the "ground state of the field," not the absence of physics.

Useful Commutator Identities

Several commutator identities will be used throughout this chapter. All follow from the fundamental relation $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$ and the general identity $[\hat{A}\hat{B}, \hat{C}] = \hat{A}[\hat{B}, \hat{C}] + [\hat{A}, \hat{C}]\hat{B}$:

$$[\hat{n}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}\hat{a}_{k'}^\dagger \qquad \text{(creation raises particle number by 1)}$$

$$[\hat{n}_k, \hat{a}_{k'}] = -\delta_{kk'}\hat{a}_{k'} \qquad \text{(annihilation lowers particle number by 1)}$$

Proof of the first identity. Using the product rule for commutators:

$$[\hat{n}_k, \hat{a}_{k'}^\dagger] = [\hat{a}_k^\dagger\hat{a}_k, \hat{a}_{k'}^\dagger] = \hat{a}_k^\dagger[\hat{a}_k, \hat{a}_{k'}^\dagger] + [\hat{a}_k^\dagger, \hat{a}_{k'}^\dagger]\hat{a}_k = \hat{a}_k^\dagger \delta_{kk'} + 0 = \delta_{kk'}\hat{a}_k^\dagger$$

For $k = k'$, this gives $[\hat{n}, \hat{a}^\dagger] = \hat{a}^\dagger$, which means: if $|n\rangle$ is an eigenstate of $\hat{n}$ with eigenvalue $n$, then $\hat{a}^\dagger|n\rangle$ is an eigenstate with eigenvalue $n + 1$. This is the rigorous justification for calling $\hat{a}^\dagger$ a "creation" operator.

These commutator identities also immediately give us:

$$[\hat{N}, \hat{a}_k^\dagger] = \hat{a}_k^\dagger, \qquad [\hat{N}, \hat{a}_k] = -\hat{a}_k$$

which means that $\hat{a}_k^\dagger$ increases the total particle number by 1, and $\hat{a}_k$ decreases it by 1. Any operator that commutes with $\hat{N}$ conserves particle number.

Worked Example: Two-Mode Bosonic System

To make the abstract formalism concrete, consider a system with exactly two single-particle modes ($k = 1, 2$). This could represent two energy levels of a quantum dot, two polarization states of a photon, or two sites in a double-well potential.

The Fock space basis states are $|n_1, n_2\rangle$ where $n_1, n_2 = 0, 1, 2, \ldots$

Problem: Compute $\hat{a}_1^\dagger \hat{a}_2 |0, 3\rangle$.

Solution: This operator destroys a particle in mode 2 and creates one in mode 1 (a "hopping" process):

$$\hat{a}_1^\dagger \hat{a}_2 |0, 3\rangle = \hat{a}_1^\dagger \sqrt{3}|0, 2\rangle = \sqrt{3}\sqrt{1}|1, 2\rangle = \sqrt{3}|1, 2\rangle$$

The total particle number is unchanged ($0 + 3 = 3 = 1 + 2$), but one particle has "hopped" from mode 2 to mode 1. This type of operator — $\hat{a}_k^\dagger\hat{a}_{k'}$ with $k \neq k'$ — appears in every interacting many-body Hamiltonian as the mechanism for transitions between single-particle states.

🧪 Experiment: The bosonic commutation relations have been verified directly in quantum optics experiments. In cavity quantum electrodynamics (cavity QED), individual photons are trapped in a superconducting microwave cavity, and the creation and annihilation operators correspond to adding or removing a single photon. The group of Serge Haroche (Nobel Prize 2012) demonstrated the $\sqrt{n}$ and $\sqrt{n+1}$ factors in the matrix elements by measuring the photon number-dependent coupling between the cavity and a Rydberg atom. These experiments provided direct, quantitative confirmation of the bosonic algebra.

34.2 Creation and Annihilation Operators for Fermions

The Pauli Problem

Everything in Section 34.1 assumed that multiple particles could occupy the same single-particle state — which is fine for bosons but fatally wrong for fermions. The Pauli exclusion principle (Ch 15) states that no two identical fermions can occupy the same quantum state. How do we encode this algebraically?

The answer is elegant: replace commutators with anticommutators.

The Fermionic Anticommutation Relations

Define the anticommutator:

$$\{A, B\} \equiv AB + BA$$

For fermionic creation and annihilation operators $\hat{c}_k^\dagger$ and $\hat{c}_k$, the defining relations are:

$$\{\hat{c}_k, \hat{c}_{k'}^\dagger\} = \delta_{kk'}$$

$$\{\hat{c}_k, \hat{c}_{k'}\} = 0, \qquad \{\hat{c}_k^\dagger, \hat{c}_{k'}^\dagger\} = 0$$

These look almost identical to the bosonic relations, but the change from commutator $[\cdot, \cdot]$ to anticommutator $\{\cdot, \cdot\}$ has profound consequences.

The Pauli Exclusion Principle — Automatic

Consider what happens when you try to put two fermions in the same state $k$. From $\{\hat{c}_k^\dagger, \hat{c}_k^\dagger\} = 0$, we get:

$$\hat{c}_k^\dagger \hat{c}_k^\dagger + \hat{c}_k^\dagger \hat{c}_k^\dagger = 0$$

$$2(\hat{c}_k^\dagger)^2 = 0$$

$$(\hat{c}_k^\dagger)^2 = 0$$

Applying this to any state: $(\hat{c}_k^\dagger)^2|\text{anything}\rangle = 0$. You cannot create two fermions in the same state — the algebra forbids it. The Pauli exclusion principle is not imposed as an external rule; it is a theorem of the anticommutation relations.

💡 Key Insight: The choice between commutators (bosons) and anticommutators (fermions) encodes the spin-statistics theorem algebraically. This one-character difference — $[\cdot,\cdot]$ versus $\{\cdot,\cdot\}$ — is responsible for the stability of matter, the structure of the periodic table, and the distinction between metals and insulators.

Fermionic Number States

For fermions, the occupation number $n_k$ can only be 0 or 1 (occupied or unoccupied). The fermionic number operator is:

$$\hat{n}_k = \hat{c}_k^\dagger \hat{c}_k$$

and we can verify that $\hat{n}_k^2 = \hat{n}_k$ (the eigenvalues satisfy $n_k^2 = n_k$, so $n_k = 0$ or $1$). Proof:

$$\hat{n}_k^2 = \hat{c}_k^\dagger \hat{c}_k \hat{c}_k^\dagger \hat{c}_k = \hat{c}_k^\dagger (1 - \hat{c}_k^\dagger \hat{c}_k) \hat{c}_k = \hat{c}_k^\dagger \hat{c}_k - \hat{c}_k^\dagger \hat{c}_k^\dagger \hat{c}_k \hat{c}_k = \hat{n}_k - 0 = \hat{n}_k$$

where we used $\{\hat{c}_k, \hat{c}_k^\dagger\} = 1$ in the second step and $(\hat{c}_k^\dagger)^2 = (\hat{c}_k)^2 = 0$ in the last step.

Multi-Fermion States and the Sign Convention

A general fermionic state is:

$$|n_1, n_2, n_3, \ldots\rangle = (\hat{c}_1^\dagger)^{n_1}(\hat{c}_2^\dagger)^{n_2}(\hat{c}_3^\dagger)^{n_3}\cdots|0\rangle$$

where each $n_k \in \{0, 1\}$. But now order matters! Since $\hat{c}_k^\dagger \hat{c}_{k'}^\dagger = -\hat{c}_{k'}^\dagger \hat{c}_k^\dagger$, swapping the order of two creation operators introduces a minus sign. For example:

$$|1_1, 1_2, 0_3, \ldots\rangle = \hat{c}_1^\dagger \hat{c}_2^\dagger |0\rangle = -\hat{c}_2^\dagger \hat{c}_1^\dagger |0\rangle$$

This sign is the algebraic encoding of the antisymmetry of fermion wavefunctions. In Chapter 15, we built antisymmetric states using Slater determinants. Second quantization replaces all of that bookkeeping with the simple rule: fermionic creation operators anticommute.

🔴 Warning: The sign convention requires that we fix a standard ordering of modes (e.g., increasing $k$). Creating operators in a different order produces an overall sign that must be tracked carefully. This is the main source of errors in fermionic calculations.

Comparison Table: Bosons vs. Fermions

🔗 Connection (Ch 15): In Section 15.4, you constructed antisymmetric wavefunctions using Slater determinants. The second-quantized approach replaces the determinant with anticommuting operators. The two formulations are exactly equivalent — but the operator approach scales far more gracefully to many-particle systems.

Worked Example: Recovering the Slater Determinant

Let us verify that the second-quantized formalism reproduces the antisymmetric two-particle wavefunction. Consider two fermionic modes with single-particle wavefunctions $\phi_1(\mathbf{r})$ and $\phi_2(\mathbf{r})$. In first quantization, the antisymmetric two-particle wavefunction is the Slater determinant:

$$\Psi(\mathbf{r}_1, \mathbf{r}_2) = \frac{1}{\sqrt{2}}\begin{vmatrix}\phi_1(\mathbf{r}_1) & \phi_2(\mathbf{r}_1) \\ \phi_1(\mathbf{r}_2) & \phi_2(\mathbf{r}_2)\end{vmatrix} = \frac{1}{\sqrt{2}}[\phi_1(\mathbf{r}_1)\phi_2(\mathbf{r}_2) - \phi_2(\mathbf{r}_1)\phi_1(\mathbf{r}_2)]$$

In second quantization, the same state is:

$$|1, 1\rangle = \hat{c}_1^\dagger\hat{c}_2^\dagger|0\rangle$$

To extract the wavefunction, we use the field operator $\hat{\psi}^\dagger(\mathbf{r}) = \sum_k \phi_k^*(\mathbf{r})\hat{c}_k^\dagger$. The position-space representation of $|1, 1\rangle$ is:

$$\langle\mathbf{r}_1, \mathbf{r}_2|1, 1\rangle = \langle 0|\hat{\psi}(\mathbf{r}_2)\hat{\psi}(\mathbf{r}_1)\hat{c}_1^\dagger\hat{c}_2^\dagger|0\rangle$$

Expanding $\hat{\psi}(\mathbf{r}) = \sum_k \phi_k(\mathbf{r})\hat{c}_k$ and using the anticommutation relations to evaluate all the operator products, one obtains exactly the Slater determinant above. The minus sign in the determinant comes from the single anticommutation needed to bring $\hat{c}_2\hat{c}_1^\dagger$ into normal order.

This worked example demonstrates the deep equivalence between the Slater determinant (first quantization) and the anticommuting operator (second quantization) approaches. For two particles, the equivalence is easily verified. For 100 particles, the Slater determinant is a $100 \times 100$ determinant, while the second-quantized state is still just a string of creation operators applied to the vacuum — the efficiency advantage is overwhelming.

The Jordan-Wigner Transformation

There is a remarkable mathematical equivalence between fermionic operators and spin-1/2 operators, known as the Jordan-Wigner transformation (1928). For a chain of $N$ sites, define:

$$\hat{c}_j = \left(\prod_{i=1}^{j-1}\hat{\sigma}_i^z\right)\hat{\sigma}_j^-$$

where $\hat{\sigma}^- = (\hat{\sigma}^x - i\hat{\sigma}^y)/2$ is the spin lowering operator. The string of $\hat{\sigma}^z$ operators (called the "Jordan-Wigner string") is essential: it ensures that operators on different sites anticommute rather than commute.

This transformation is not merely a mathematical curiosity. It is the standard tool for mapping spin chain problems (like the Heisenberg model of magnetism) onto fermionic problems, and vice versa. It is also central to the simulation of fermionic systems on quantum computers, where the native operations are on qubits (spin-1/2 systems).

🔵 Historical Note: The anticommutation relations for fermions were introduced by Pascual Jordan and Eugene Wigner in 1928, shortly after Dirac introduced the bosonic creation/annihilation operators for the electromagnetic field. Jordan and Wigner showed that the Pauli exclusion principle could be encoded algebraically through anticommutation, just as Bose-Einstein statistics was encoded through commutation. This completed the second-quantized formalism for all particle types known at the time.

✅ Checkpoint: Before proceeding, verify that you can: 1. Show that $[\hat{a}_k, \hat{a}_k^\dagger] = 1$ implies $\hat{a}^\dagger|n\rangle = \sqrt{n+1}|n+1\rangle$. 2. Show that $\{\hat{c}_k, \hat{c}_k^\dagger\} = 1$ implies $n_k = 0$ or $1$. 3. Explain why bosonic creation operators commute but fermionic creation operators anticommute.

34.3 Fock Space

The Arena for Variable Particle Number

In standard quantum mechanics, you work in a Hilbert space with a fixed number of particles. A two-electron system lives in a different Hilbert space than a three-electron system. But in nature, particle number can change: photons are emitted and absorbed, electron-positron pairs are created, phonons are excited in solids.

Fock space is the mathematical arena that accommodates all of this. It is the direct sum of Hilbert spaces with 0, 1, 2, ... particles:

$$\mathcal{F} = \mathcal{H}_0 \oplus \mathcal{H}_1 \oplus \mathcal{H}_2 \oplus \mathcal{H}_3 \oplus \cdots$$

where $\mathcal{H}_0 = \mathbb{C}$ (the vacuum), $\mathcal{H}_1$ is the one-particle Hilbert space, $\mathcal{H}_2$ is the (appropriately symmetrized or antisymmetrized) two-particle Hilbert space, and so on.

A general state in Fock space is a superposition of states with different particle numbers:

$$|\Psi\rangle = c_0|0\rangle + \sum_k c_k^{(1)} \hat{a}_k^\dagger|0\rangle + \sum_{k,k'} c_{kk'}^{(2)} \hat{a}_k^\dagger \hat{a}_{k'}^\dagger|0\rangle + \cdots$$

The creation and annihilation operators act naturally on Fock space, moving between sectors of different particle number:

$$\hat{a}_k^\dagger: \mathcal{H}_n \to \mathcal{H}_{n+1}, \qquad \hat{a}_k: \mathcal{H}_n \to \mathcal{H}_{n-1}$$

The Occupation Number Representation

The most natural basis for Fock space is the occupation number basis:

$$|n_1, n_2, n_3, \ldots\rangle$$

where $n_k$ is the number of particles in single-particle state $k$. For bosons, each $n_k \in \{0, 1, 2, \ldots\}$. For fermions, each $n_k \in \{0, 1\}$.

This basis is orthonormal:

$$\langle n_1', n_2', \ldots | n_1, n_2, \ldots \rangle = \prod_k \delta_{n_k' n_k}$$

and complete:

$$\sum_{\{n_k\}} |n_1, n_2, \ldots\rangle \langle n_1, n_2, \ldots| = \hat{1}$$

Every operator — the Hamiltonian, observables, interaction terms — can be expressed in terms of creation and annihilation operators acting on this basis. This is what makes second quantization so powerful.

💡 Key Insight: In first quantization, you ask "which particle is in which state?" In second quantization, you ask "how many particles are in each state?" The second question is more natural for identical particles, because identical particles have no individual identity. Second quantization builds this indistinguishability into the formalism from the ground up.

Why Fock Space Is Necessary

One might ask: why not just work in a fixed-$N$ Hilbert space and use Slater determinants or permanents? There are several reasons why Fock space is essential:

1. Particle creation and destruction. In any system coupled to the electromagnetic field, photons are constantly being created and absorbed. In relativistic physics, particle-antiparticle pairs are produced. A fixed-$N$ formalism cannot describe these processes.

2. Grand canonical ensemble. In statistical mechanics at finite temperature and chemical potential, the system samples states with different particle numbers. The grand partition function $\mathcal{Z} = \text{Tr}_{\mathcal{F}}[e^{-\beta(\hat{H} - \mu\hat{N})}]$ naturally involves a trace over all of Fock space.

3. Quasiparticle descriptions. Many-body ground states are often described in terms of quasiparticles that do not conserve the number of bare particles. For example, the BCS ground state of a superconductor is a superposition of states with different electron numbers (Cooper pairs condense), and the Bogoliubov transformation that diagonalizes the BCS Hamiltonian mixes creation and annihilation operators — something that is meaningless outside of Fock space.

4. Computational efficiency. For many algorithms in quantum chemistry and condensed matter physics (configuration interaction, coupled cluster, DMRG), the Fock space formulation in terms of creation and annihilation operators is far more efficient than the equivalent wavefunction-based approach.

Bosonic vs. Fermionic Fock Space

For bosons, the $n$-particle sector $\mathcal{H}_n$ consists of symmetric states. The symmetrization is automatic because bosonic creation operators commute.

For fermions, the $n$-particle sector consists of antisymmetric states. The antisymmetrization is automatic because fermionic creation operators anticommute. The maximum number of particles is limited by the number of available single-particle states (each can hold at most one fermion).

📊 By the Numbers: Consider a system with $M$ available single-particle states. The bosonic Fock space dimension for fixed particle number $N$ is $\binom{N + M - 1}{N}$. For fermions, it is $\binom{M}{N}$. The total Fock space dimension (all particle numbers) is infinite for bosons and $2^M$ for fermions. For 10 fermionic modes, the Fock space has $2^{10} = 1024$ basis states — already a formidable space for exact diagonalization.

The Vacuum and Its Properties

The vacuum state $|0\rangle$ deserves special attention. It is defined by:

$$\hat{a}_k|0\rangle = 0 \quad \text{(bosons)}, \qquad \hat{c}_k|0\rangle = 0 \quad \text{(fermions)}, \qquad \text{for all } k$$

The vacuum is normalized: $\langle 0|0\rangle = 1$.

The vacuum is the unique state annihilated by all annihilation operators. It is not "nothing" — it is the ground state of the quantum field, with measurable properties:

• Zero-point energy: Each bosonic mode contributes $\frac{1}{2}\hbar\omega_k$ (from Ch 4). The total vacuum energy is $E_{\text{vac}} = \sum_k \frac{1}{2}\hbar\omega_k$, which is formally infinite and must be handled by renormalization.
• Vacuum fluctuations: $\langle 0|\hat{a}_k^\dagger \hat{a}_k|0\rangle = 0$ (no particles), but $\langle 0|(\hat{a}_k + \hat{a}_k^\dagger)^2|0\rangle \neq 0$ (field fluctuations are nonzero).
• Physical consequences: The Casimir effect, spontaneous emission, and the Lamb shift are all manifestations of vacuum structure.

🔵 Historical Note: The concept of Fock space is named after Vladimir Fock, who introduced the occupation number representation in 1932. Fock's formalism, combined with Dirac's creation/annihilation operator approach, became the standard language of quantum field theory by the late 1930s. The key insight — that particle number should be a dynamical variable rather than a fixed parameter — was one of the great conceptual advances of 20th-century physics.

34.4 Many-Particle Operators in Second-Quantized Form

Why Rewrite Operators?

The real payoff of second quantization is the ability to express any operator — one-body, two-body, or $n$-body — in a compact form using creation and annihilation operators. This is not merely compact notation; it automatically handles all the symmetrization (for bosons) or antisymmetrization (for fermions) that would otherwise require laborious bookkeeping.

One-Body Operators

A one-body operator is a sum of operators that act on one particle at a time. In first quantization, for $N$ particles:

$$\hat{O}_1 = \sum_{i=1}^{N} \hat{o}(i)$$

where $\hat{o}(i)$ acts on the $i$-th particle. Examples: kinetic energy $\hat{T} = \sum_i \hat{p}_i^2/(2m)$, external potential $\hat{V}_{\text{ext}} = \sum_i V(\hat{x}_i)$.

In second quantization, this becomes:

$$\hat{O}_1 = \sum_{k, k'} \langle k|\hat{o}|k'\rangle\, \hat{a}_k^\dagger \hat{a}_{k'}$$

The matrix elements $\langle k|\hat{o}|k'\rangle$ are computed in the single-particle basis $\{|k\rangle\}$. This formula works identically for bosons and fermions (just replace $\hat{a}$ with $\hat{c}$ for fermions).

Derivation sketch. Consider a one-body operator acting on an $N$-particle state. Insert the resolution of identity $\hat{1} = \sum_k |k\rangle\langle k|$ on either side of $\hat{o}$, and use the fact that creating a particle in state $k$ while removing one from state $k'$ is precisely $\hat{a}_k^\dagger \hat{a}_{k'}$. The sum over particles is automatically handled by the sum over modes.

Example: Kinetic energy. In a plane-wave basis $|k\rangle$ with $\hat{p}|k\rangle = \hbar k|k\rangle$:

$$\hat{T} = \sum_k \frac{\hbar^2 k^2}{2m}\, \hat{a}_k^\dagger \hat{a}_k = \sum_k \epsilon_k\, \hat{n}_k$$

where $\epsilon_k = \hbar^2 k^2 / (2m)$ is the single-particle energy. The kinetic energy is diagonal in the momentum basis — each occupied mode contributes its single-particle energy.

Two-Body Operators

A two-body operator describes pairwise interactions. In first quantization:

$$\hat{O}_2 = \frac{1}{2}\sum_{i \neq j} \hat{v}(i, j)$$

In second quantization:

$$\hat{O}_2 = \frac{1}{2}\sum_{k, k', q, q'} \langle k, k'|\hat{v}|q, q'\rangle\, \hat{a}_k^\dagger \hat{a}_{k'}^\dagger \hat{a}_{q'} \hat{a}_q$$

Notice the operator ordering: two creation operators followed by two annihilation operators, with the annihilation operators in reversed order relative to the bra. This ordering is crucial — it ensures that self-interaction terms ($i = j$) are automatically excluded.

Example: Coulomb interaction. For electrons interacting via the Coulomb potential $v(r_{ij}) = e^2/(4\pi\epsilon_0 |\mathbf{r}_i - \mathbf{r}_j|)$, in a plane-wave basis with momentum transfer $\mathbf{q}$:

$$\hat{V}_{\text{Coulomb}} = \frac{1}{2V}\sum_{\mathbf{k}, \mathbf{k}', \mathbf{q} \neq 0} \frac{4\pi e^2}{q^2}\, \hat{c}_{\mathbf{k}+\mathbf{q}}^\dagger \hat{c}_{\mathbf{k}'-\mathbf{q}}^\dagger \hat{c}_{\mathbf{k}'} \hat{c}_{\mathbf{k}}$$

The $\mathbf{q} = 0$ term is excluded because it is cancelled by the uniform positive background charge (in a neutral system).

🔴 Warning: The ordering of operators in two-body terms is a frequent source of errors. Always write creation operators to the left and annihilation operators to the right (normal ordering), and be careful with the index ordering. The prescription is: the first index of the bra matches the first creation operator, the second index of the bra matches the second creation operator, and the annihilation operators are in reverse order of the ket indices.

The Full Many-Body Hamiltonian

Combining one-body and two-body terms, the generic many-body Hamiltonian in second quantization is:

$$\hat{H} = \sum_{k, k'} t_{kk'}\, \hat{a}_k^\dagger \hat{a}_{k'} + \frac{1}{2}\sum_{k, k', q, q'} V_{kk'qq'}\, \hat{a}_k^\dagger \hat{a}_{k'}^\dagger \hat{a}_{q'} \hat{a}_q$$

where $t_{kk'} = \langle k|\hat{T} + \hat{V}_{\text{ext}}|k'\rangle$ are the one-body matrix elements and $V_{kk'qq'} = \langle k, k'|\hat{v}|q, q'\rangle$ are the two-body matrix elements.

This is the starting point for essentially all of modern many-body quantum physics: condensed matter theory, nuclear physics, quantum chemistry (under the name "second quantization" or "occupation number formalism"), and quantum field theory.

Detailed Derivation: One-Body Operator

Let us derive the second-quantized one-body operator formula carefully. We want to show that $\hat{O}_1 = \sum_i \hat{o}(i)$ (the first-quantized operator summing over all particles) is equivalent to $\hat{O}_1 = \sum_{k,k'}\langle k|\hat{o}|k'\rangle \hat{a}_k^\dagger\hat{a}_{k'}$.

Consider the action of $\hat{O}_1$ on a one-particle state $|q\rangle = \hat{a}_q^\dagger|0\rangle$. In first quantization, $\hat{o}$ acts on this single particle to give:

$$\hat{o}|q\rangle = \sum_k |k\rangle\langle k|\hat{o}|q\rangle = \sum_k o_{kq}|k\rangle$$

where $o_{kq} = \langle k|\hat{o}|q\rangle$.

Now compute the action of the second-quantized form:

$$\sum_{k,k'} o_{kk'}\hat{a}_k^\dagger\hat{a}_{k'}\hat{a}_q^\dagger|0\rangle = \sum_{k,k'} o_{kk'}\hat{a}_k^\dagger\hat{a}_{k'}|1_q\rangle$$

Since $\hat{a}_{k'}|1_q\rangle = \delta_{k'q}|0\rangle$ (for the one-particle sector), this becomes:

$$\sum_k o_{kq}\hat{a}_k^\dagger|0\rangle = \sum_k o_{kq}|1_k\rangle$$

This is exactly $\sum_k o_{kq}|k\rangle$ — the correct result. By linearity, the equivalence extends to all multi-particle states.

The beauty of the formula is that the sum over particles ($\sum_i$) has been replaced by a sum over modes ($\sum_{k,k'}$). The operator $\hat{a}_k^\dagger\hat{a}_{k'}$ with $k \neq k'$ is a "hopping" or "scattering" operator: it removes a particle from mode $k'$ and places it in mode $k$. When $k = k'$, it is the number operator $\hat{n}_k$, which simply counts particles in mode $k$.

Normal Ordering

A product of creation and annihilation operators is said to be in normal order if all creation operators stand to the left of all annihilation operators. For example:

$$:\hat{a}^\dagger\hat{a}\hat{a}^\dagger\hat{a}: = \hat{a}^\dagger\hat{a}^\dagger\hat{a}\hat{a}$$

Normal ordering is denoted by colons $:\cdots:$. It is essential in quantum field theory because the vacuum expectation value of any normal-ordered product vanishes: $\langle 0|:\hat{O}:|0\rangle = 0$. This is the mathematical basis for the subtraction of vacuum energy — the infinite zero-point energy is swept away by normal ordering, leaving only the physically observable particle energies.

In the two-body operator formula, the ordering $\hat{a}_k^\dagger\hat{a}_{k'}^\dagger\hat{a}_{q'}\hat{a}_q$ is already normal ordered. This is not an accident — it ensures that self-interaction terms are excluded and that the vacuum expectation value of the interaction is zero.

🔴 Warning: The ordering of operators in two-body terms is a frequent source of errors. Always write creation operators to the left and annihilation operators to the right (normal ordering), and be careful with the index ordering. The prescription is: the first index of the bra matches the first creation operator, the second index of the bra matches the second creation operator, and the annihilation operators are in reverse order of the ket indices.

✅ Checkpoint: Verify that the second-quantized kinetic energy $\hat{T} = \sum_k \epsilon_k \hat{n}_k$ gives the correct expectation value for a state $|1_a, 1_b, 0, 0, \ldots\rangle$ (one particle in mode $a$, one in mode $b$): $\langle \hat{T}\rangle = \epsilon_a + \epsilon_b$.

34.5 The Free Electron Gas

Setting Up the Problem

The free electron gas (also called the Fermi gas or the homogeneous electron gas) is the simplest many-fermion system: $N$ non-interacting electrons in a box of volume $V = L^3$ with periodic boundary conditions. It is the starting point for understanding metals and is a cornerstone of condensed matter physics.

The single-particle states are plane waves labeled by wavevector $\mathbf{k}$ and spin projection $\sigma \in \{\uparrow, \downarrow\}$:

$$\phi_{\mathbf{k}\sigma}(\mathbf{r}) = \frac{1}{\sqrt{V}} e^{i\mathbf{k}\cdot\mathbf{r}} |\sigma\rangle$$

with $\mathbf{k} = (2\pi/L)(n_x, n_y, n_z)$ for integers $n_x, n_y, n_z$.

The Hamiltonian

In second-quantized form, the free-electron Hamiltonian is:

$$\hat{H} = \sum_{\mathbf{k}, \sigma} \epsilon_\mathbf{k}\, \hat{c}_{\mathbf{k}\sigma}^\dagger \hat{c}_{\mathbf{k}\sigma}, \qquad \epsilon_\mathbf{k} = \frac{\hbar^2 k^2}{2m}$$

This is already diagonal in the momentum-spin basis. Each $\hat{n}_{\mathbf{k}\sigma} = \hat{c}_{\mathbf{k}\sigma}^\dagger \hat{c}_{\mathbf{k}\sigma}$ is either 0 or 1.

The Ground State: Filling the Fermi Sea

At zero temperature, $N$ electrons fill the lowest-energy single-particle states. Since each state $(\mathbf{k}, \sigma)$ can hold at most one electron, we fill all states with $\epsilon_\mathbf{k} < E_F$, where $E_F$ is the Fermi energy — the energy of the highest occupied state.

In second quantization, the ground state is:

$$|\text{FS}\rangle = \prod_{\substack{\mathbf{k}, \sigma \\ |\mathbf{k}| \leq k_F}} \hat{c}_{\mathbf{k}\sigma}^\dagger |0\rangle$$

where $k_F$ is the Fermi wavevector, defined by $\epsilon_F = \hbar^2 k_F^2/(2m)$.

The product runs over all states within the Fermi sphere — a sphere in $\mathbf{k}$-space of radius $k_F$. The condition that this sphere contains exactly $N$ states (counting spin) determines $k_F$:

$$N = 2 \times \frac{4\pi k_F^3/3}{(2\pi/L)^3} = \frac{V k_F^3}{3\pi^2}$$

Solving: $k_F = (3\pi^2 n)^{1/3}$, where $n = N/V$ is the electron density.

Ground State Energy

The total ground state energy is:

$$E_0 = \sum_{\substack{\mathbf{k}, \sigma \\ |\mathbf{k}| \leq k_F}} \epsilon_\mathbf{k} = 2\sum_{|\mathbf{k}| \leq k_F} \frac{\hbar^2 k^2}{2m}$$

Converting the sum to an integral (in the thermodynamic limit $V \to \infty$):

$$E_0 = 2 \cdot \frac{V}{(2\pi)^3}\int_0^{k_F} \frac{\hbar^2 k^2}{2m}\, 4\pi k^2\, dk = \frac{V\hbar^2}{m\pi^2} \int_0^{k_F} k^4\, dk = \frac{V\hbar^2 k_F^5}{5m\pi^2}$$

The energy per particle is:

$$\frac{E_0}{N} = \frac{3}{5}\frac{\hbar^2 k_F^2}{2m} = \frac{3}{5}E_F$$

📊 By the Numbers: For copper, $n \approx 8.5 \times 10^{28}$ m$^{-3}$. This gives $k_F \approx 1.36 \times 10^{10}$ m$^{-1}$, $E_F \approx 7.0$ eV, and $T_F = E_F/k_B \approx 81{,}000$ K. The Fermi energy of a typical metal is enormous — far above room temperature. This is why conduction electrons behave quantum-mechanically even at "high" temperatures.

Excitations Above the Fermi Sea

The elementary excitations of the Fermi sea are particle-hole pairs. To excite the system, you remove an electron from inside the Fermi sphere (creating a hole with $|\mathbf{k}| < k_F$) and place it outside (creating a particle with $|\mathbf{k}'| > k_F$):

$$|\mathbf{k}'\sigma', \overline{\mathbf{k}\sigma}\rangle = \hat{c}_{\mathbf{k}'\sigma'}^\dagger \hat{c}_{\mathbf{k}\sigma}|\text{FS}\rangle$$

The excitation energy is $\epsilon_{\mathbf{k}'} - \epsilon_\mathbf{k} > 0$. These particle-hole excitations are the low-energy degrees of freedom of the metal, responsible for its heat capacity, electrical conductivity, and magnetic response.

The Density of States

To perform calculations in the thermodynamic limit ($V \to \infty$), we need the density of states — the number of single-particle states per unit energy. Converting the sum over $\mathbf{k}$ to an integral using the prescription $\sum_\mathbf{k} \to \frac{V}{(2\pi)^3}\int d^3k$, and changing variables from $k$ to $\epsilon = \hbar^2k^2/(2m)$:

$$g(\epsilon) = \frac{V}{2\pi^2}\left(\frac{2m}{\hbar^2}\right)^{3/2}\epsilon^{1/2}$$

This is the famous "square-root" density of states for free particles in three dimensions. It vanishes at $\epsilon = 0$ (few states at low energy) and increases as $\sqrt{\epsilon}$ (many states at high energy). The density of states is the key ingredient for computing thermodynamic quantities: internal energy, heat capacity, magnetic susceptibility.

Excited States and the Electron-Hole Picture

The Fermi sea is the ground state, but what about excited states? The language of second quantization provides a powerful and intuitive picture.

Define a hole as the absence of an electron from a state below the Fermi surface. If we remove an electron from state $(\mathbf{k}, \sigma)$ with $|\mathbf{k}| < k_F$, we create a hole. Formally, we can define hole creation and annihilation operators:

$$\hat{h}_{\mathbf{k}\sigma}^\dagger = \hat{c}_{\mathbf{k}\sigma}, \qquad \hat{h}_{\mathbf{k}\sigma} = \hat{c}_{\mathbf{k}\sigma}^\dagger \qquad (|\mathbf{k}| < k_F)$$

A hole in a state of energy $\epsilon_\mathbf{k} < E_F$ has excitation energy $E_F - \epsilon_\mathbf{k} > 0$ — it costs energy to remove an electron from below the Fermi surface. Similarly, adding an electron above the Fermi surface costs energy $\epsilon_{\mathbf{k}'} - E_F > 0$.

The elementary excitations are therefore particle-hole pairs: one electron added above the Fermi surface and one removed from below:

$$|\text{excitation}\rangle = \hat{c}_{\mathbf{k}'\sigma'}^\dagger\hat{c}_{\mathbf{k}\sigma}|\text{FS}\rangle, \qquad |\mathbf{k}'| > k_F, \quad |\mathbf{k}| < k_F$$

with excitation energy $\Delta E = \epsilon_{\mathbf{k}'} - \epsilon_\mathbf{k} > 0$.

This particle-hole picture is the starting point for Landau's Fermi liquid theory, which explains why the low-energy properties of interacting electron systems can be described in terms of weakly interacting quasiparticles that behave much like free electrons near the Fermi surface.

🔗 Connection (Ch 26): The free electron gas is the starting point for the band theory of solids developed in Chapter 26. Adding a periodic lattice potential turns the free-electron parabola into energy bands, and the Fermi sphere distorts into a Fermi surface — but the second-quantized machinery carries over directly.

34.6 Phonons: Quantized Sound

From Classical Lattice to Quantum Field

We now turn from fermions to bosons, and from electrons to sound waves. The quantization of lattice vibrations — producing phonons — is one of the most physically important applications of second quantization, and historically one of the first examples of the particle-field duality that defines modern physics.

The Classical Linear Chain

Consider a one-dimensional chain of $N$ atoms, each of mass $m$, connected by springs of constant $\kappa$, with lattice spacing $a$. The equilibrium position of atom $n$ is $x_n^{(0)} = na$, and we write the displacement from equilibrium as $u_n$.

Newton's second law for atom $n$:

$$m\ddot{u}_n = \kappa(u_{n+1} - u_n) - \kappa(u_n - u_{n-1}) = \kappa(u_{n+1} + u_{n-1} - 2u_n)$$

We look for normal-mode solutions $u_n(t) = A e^{i(kna - \omega t)}$, which gives the dispersion relation:

$$\omega(k) = 2\sqrt{\frac{\kappa}{m}}\left|\sin\left(\frac{ka}{2}\right)\right|$$

The allowed wavevectors are $k = 2\pi p/(Na)$ for integers $p$, with the first Brillouin zone: $-\pi/a < k \leq \pi/a$.

Quantization via Second Quantization

The classical normal modes are independent harmonic oscillators. To quantize them, we promote each mode to a quantum harmonic oscillator — exactly as in Chapter 4. The displacement operator becomes:

$$\hat{u}_n = \sum_k \sqrt{\frac{\hbar}{2Nm\omega_k}} \left(\hat{a}_k e^{ikna} + \hat{a}_k^\dagger e^{-ikna}\right)$$

where $\hat{a}_k$ and $\hat{a}_k^\dagger$ are bosonic creation and annihilation operators satisfying $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$.

The Hamiltonian in second-quantized form is:

$$\hat{H} = \sum_k \hbar\omega_k \left(\hat{a}_k^\dagger \hat{a}_k + \frac{1}{2}\right) = \sum_k \hbar\omega_k \left(\hat{n}_k + \frac{1}{2}\right)$$

This is a collection of independent quantum harmonic oscillators, one for each wavevector $k$. The eigenvalues are:

$$E = \sum_k \hbar\omega_k \left(n_k + \frac{1}{2}\right), \qquad n_k = 0, 1, 2, \ldots$$

The derivation deserves a closer look, because it shows how the creation/annihilation operators emerge naturally from the physics. Starting from the classical normal-mode coordinates $Q_k$ and momenta $P_k$ (which are Fourier transforms of the atomic displacements $u_n$ and momenta $p_n$), the classical Hamiltonian for the linear chain can be written as:

$$H = \sum_k \frac{1}{2}\left(P_k P_{-k} + \omega_k^2 Q_k Q_{-k}\right)$$

This is a sum of harmonic oscillator Hamiltonians, one per mode. We then define the creation and annihilation operators exactly as in Chapter 4:

$$\hat{a}_k = \sqrt{\frac{m\omega_k}{2\hbar}}\left(\hat{Q}_k + \frac{i\hat{P}_{-k}}{m\omega_k}\right), \qquad \hat{a}_k^\dagger = \sqrt{\frac{m\omega_k}{2\hbar}}\left(\hat{Q}_{-k} - \frac{i\hat{P}_k}{m\omega_k}\right)$$

The canonical commutation relations $[\hat{Q}_k, \hat{P}_{k'}] = i\hbar\delta_{k,-k'}$ immediately give $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$, and substituting back yields the quantized Hamiltonian. The procedure is identical to Chapter 4, but applied independently to each normal mode — the power of the harmonic decomposition.

What Is a Phonon?

A phonon is a quantum of vibrational energy in mode $k$. The state with $n_k$ phonons in mode $k$ has energy $\hbar\omega_k(n_k + \frac{1}{2})$ in that mode. $\hat{a}_k^\dagger$ creates a phonon with wavevector $k$, and $\hat{a}_k$ destroys one.

Phonons are not particles in the usual sense — no atom is moving from place to place. They are quasiparticles: collective excitations of the lattice that behave, mathematically, exactly like particles. They carry energy $\hbar\omega_k$ and crystal momentum $\hbar k$. They obey Bose-Einstein statistics (because $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$). They can be scattered, absorbed, and emitted.

💡 Key Insight: Phonons illustrate the central theme of second quantization: particles emerge from the quantization of fields. Sound waves are the classical field; phonons are the quantum particles. Electromagnetic waves are the classical field; photons are the quantum particles. This pattern — classical field → quantized oscillator modes → particles — is the universal template of quantum field theory.

Specific Heat: A Triumph of Phonon Theory

One of the greatest early successes of quantum mechanics was explaining the specific heat of solids. Classically, the Dulong-Petit law predicts that each atom contributes $3k_B$ to the heat capacity — a temperature-independent result that fails catastrophically at low temperatures.

The phonon picture gives the correct answer. At temperature $T$, the mean occupation number of phonon mode $k$ is the Bose-Einstein distribution:

$$\langle n_k \rangle = \frac{1}{e^{\hbar\omega_k / k_B T} - 1}$$

The total internal energy is:

$$U(T) = \sum_k \hbar\omega_k \left(\langle n_k \rangle + \frac{1}{2}\right)$$

In the Debye model (which approximates the dispersion as linear, $\omega_k = v|k|$, up to a cutoff $\omega_D$), this gives:

$$C_V \propto T^3 \quad \text{as } T \to 0$$

This $T^3$ law was one of the first quantitative confirmations of quantum theory in solid-state physics.

🧪 Experiment: The $T^3$ specific heat law was first verified experimentally by Walther Nernst and his students around 1910-1911. Einstein's earlier 1907 model (assuming all oscillators have the same frequency) gave qualitatively correct behavior (heat capacity vanishing at low $T$) but the wrong power law. Debye's 1912 improvement, incorporating the frequency distribution, gave the correct $T^3$. Both models rely on quantized oscillators — i.e., phonons, though the word was not coined until the 1930s.

Three Dimensions and Acoustic/Optical Branches

In three dimensions, with $p$ atoms per unit cell, there are $3p$ phonon branches: - 3 acoustic branches (1 longitudinal + 2 transverse): $\omega \to 0$ as $k \to 0$. These correspond to sound waves. - $(3p - 3)$ optical branches: $\omega \neq 0$ at $k = 0$. These involve relative motion of atoms within the unit cell and couple to infrared light (hence "optical").

Each branch is independently quantized via second quantization. The full phonon Hamiltonian is:

$$\hat{H} = \sum_{\mathbf{k}, s} \hbar\omega_s(\mathbf{k})\left(\hat{a}_{\mathbf{k}s}^\dagger \hat{a}_{\mathbf{k}s} + \frac{1}{2}\right)$$

where $s$ labels the branch.

Phonon-Phonon Interactions

The harmonic approximation (keeping only the quadratic term in the Taylor expansion of the interatomic potential) gives independent, non-interacting phonons. Real crystals include anharmonic terms (cubic, quartic, ...) that produce phonon-phonon interactions.

The cubic anharmonic term, in second-quantized form, has the structure:

$$\hat{H}_3 = \sum_{\mathbf{k}_1, \mathbf{k}_2, \mathbf{k}_3} V_{\mathbf{k}_1 \mathbf{k}_2 \mathbf{k}_3}(\hat{a}_{\mathbf{k}_1} + \hat{a}_{-\mathbf{k}_1}^\dagger)(\hat{a}_{\mathbf{k}_2} + \hat{a}_{-\mathbf{k}_2}^\dagger)(\hat{a}_{\mathbf{k}_3} + \hat{a}_{-\mathbf{k}_3}^\dagger)\,\delta_{\mathbf{k}_1 + \mathbf{k}_2 + \mathbf{k}_3, \mathbf{G}}$$

where $\mathbf{G}$ is a reciprocal lattice vector (the crystal momentum is conserved modulo $\mathbf{G}$). Expanding the products generates several types of three-phonon processes:

• $\hat{a}^\dagger\hat{a}^\dagger\hat{a}$: One phonon decays into two (allowed if energy and crystal momentum are conserved).
• $\hat{a}\hat{a}\hat{a}^\dagger$: Two phonons merge into one.
• $\hat{a}^\dagger\hat{a}^\dagger\hat{a}^\dagger$ and $\hat{a}\hat{a}\hat{a}$: Three-phonon creation or annihilation (typically negligible).

When the total crystal momentum is conserved exactly ($\mathbf{G} = 0$), the process is called normal (N-process). When $\mathbf{G} \neq 0$, it is called an Umklapp (U-process). Umklapp processes are crucial for thermal resistance in crystals — without them, a perfect harmonic crystal would have infinite thermal conductivity.

The phonon-phonon interactions described above are also responsible for thermal expansion (the equilibrium lattice constant depends on temperature) and for the finite lifetime of phonon modes (each phonon eventually decays into other phonons, leading to a spectral linewidth).

📊 By the Numbers: In silicon at room temperature, the dominant phonon scattering mechanism is three-phonon Umklapp scattering. The phonon mean free path is approximately 40 nm — about 75 lattice constants. This is long enough for the phonon picture to be a good description (the phonon travels many lattice spacings between scattering events) but short enough to limit the thermal conductivity to a finite value ($\kappa \approx 148$ W/(m$\cdot$K) at 300 K).

34.7 Preview of Quantum Field Theory

The Pattern

Let us step back and observe the pattern that has emerged across this chapter:

1. Start with a classical field (electromagnetic field, displacement field in a crystal, electron field).
2. Decompose into normal modes — each mode is a harmonic oscillator.
3. Quantize each mode — introduce creation/annihilation operators with appropriate commutation (bosons) or anticommutation (fermions) relations.
4. Particles emerge as excitations of the quantized field.

This is the procedure of canonical quantization, and it is the foundational method of quantum field theory (QFT).

From Quantum Mechanics to QFT

In quantum mechanics (this textbook up to this chapter), we have: - A fixed number of particles. - Wavefunctions that assign amplitudes to particle positions (or momenta). - The Schr\u00f6dinger equation governing time evolution.

In quantum field theory, we have: - A variable number of particles (creation and annihilation are dynamical processes). - Fields that assign operator values to each point in spacetime. - The field equation (or the Hamiltonian in second-quantized form) governing time evolution.

Second quantization is the bridge. The techniques you have learned in this chapter — creation/annihilation operators, Fock space, normal ordering, the second-quantized Hamiltonian — are not merely useful tools for many-body quantum mechanics. They are the language of quantum field theory.

The Electromagnetic Field

The most famous example of the pattern is the quantization of the electromagnetic field (previewed in Case Study 2 of Chapter 4). Each mode of the EM field in a cavity is a harmonic oscillator with frequency $\omega_k = c|\mathbf{k}|$. Quantizing:

$$\hat{H}_{\text{EM}} = \sum_{\mathbf{k}, \lambda} \hbar\omega_k \left(\hat{a}_{\mathbf{k}\lambda}^\dagger \hat{a}_{\mathbf{k}\lambda} + \frac{1}{2}\right)$$

where $\lambda = 1, 2$ labels polarization. The photon — the quantum of light — is a bosonic excitation of this quantized field. It is born from the same $\hat{a}^\dagger$ operator you met in Chapter 4.

The Electron Field

For electrons, the "classical field" is the Dirac field $\hat{\psi}(\mathbf{r})$, which is actually an operator (since electrons are always quantum). The field operator is:

$$\hat{\psi}(\mathbf{r}) = \sum_k \phi_k(\mathbf{r})\, \hat{c}_k$$

where $\phi_k(\mathbf{r})$ are single-particle wavefunctions and $\hat{c}_k$ are fermionic annihilation operators. The adjoint:

$$\hat{\psi}^\dagger(\mathbf{r}) = \sum_k \phi_k^*(\mathbf{r})\, \hat{c}_k^\dagger$$

These satisfy the equal-time anticommutation relation:

$$\{\hat{\psi}(\mathbf{r}), \hat{\psi}^\dagger(\mathbf{r}')\} = \delta^{(3)}(\mathbf{r} - \mathbf{r}')$$

The field operator $\hat{\psi}^\dagger(\mathbf{r})$ creates an electron at position $\mathbf{r}$, and $\hat{\psi}(\mathbf{r})$ destroys one. Any one-body operator can be written as:

$$\hat{O} = \int d^3r\, \hat{\psi}^\dagger(\mathbf{r})\, o(\mathbf{r})\, \hat{\psi}(\mathbf{r})$$

This is the continuum version of the formula from Section 34.4.

What Lies Beyond

Second quantization opens the door to several profound developments that we can only glimpse here (and will explore more fully in Ch 37):

• Interacting quantum fields: Adding interaction terms like $\hat{\psi}^\dagger \hat{\psi}^\dagger \hat{\psi} \hat{\psi}$ leads to the full machinery of perturbative QFT: Feynman diagrams, renormalization, and the Standard Model.
• Spontaneous symmetry breaking: When the ground state of a quantum field does not share the symmetry of the Hamiltonian, new physics emerges (Goldstone bosons, the Higgs mechanism, superconductivity).
• Topological quantum matter: The operator algebra of second quantization, combined with the topology of band structures, produces topological insulators and other exotic phases (Ch 36).
• Relativistic particles: Combining second quantization with special relativity leads to the creation and annihilation of particle-antiparticle pairs — the hallmark of relativistic quantum field theory.

The Vacuum Energy Problem

The quantized electromagnetic field Hamiltonian contains the zero-point energy:

$$E_{\text{vac}} = \sum_{\mathbf{k}, \lambda} \frac{1}{2}\hbar\omega_k$$

This sum diverges — there are infinitely many modes with arbitrarily high frequency, and each contributes $\frac{1}{2}\hbar\omega_k$. If we naively compute this sum with an ultraviolet cutoff at the Planck scale ($\omega_{\max} \sim 10^{43}$ rad/s), we get an energy density roughly $10^{113}$ J/m$^3$ — larger than the observed cosmological constant by a factor of $10^{120}$. This is the famous cosmological constant problem, one of the deepest unsolved problems in theoretical physics.

In practice, quantum field theorists handle the divergent vacuum energy by normal ordering — redefining the Hamiltonian to subtract the vacuum contribution:

$$:\hat{H}: = \sum_{\mathbf{k}, \lambda}\hbar\omega_k\,\hat{a}_{\mathbf{k}\lambda}^\dagger\hat{a}_{\mathbf{k}\lambda}$$

The normal-ordered Hamiltonian has $\langle 0|:\hat{H}:|0\rangle = 0$, and all physical energies are measured relative to the vacuum. This is perfectly consistent for all laboratory-scale experiments, but leaves the gravitational effect of the vacuum energy as an open question.

Despite the divergence of the total vacuum energy, differences in vacuum energy between different configurations are finite and measurable. The most famous example is the Casimir effect: two parallel conducting plates in vacuum experience an attractive force proportional to $1/d^4$ (where $d$ is the plate separation), because the field modes between the plates have a different (restricted) spectrum than the modes outside. This force, predicted by Hendrik Casimir in 1948 and measured precisely by Steve Lamoreaux in 1997, is a direct macroscopic manifestation of the quantum vacuum.

🧪 Experiment: The Casimir force between two plates separated by $d = 1\,\mu$m is approximately $F/A \approx \pi^2\hbar c/(240 d^4) \approx 1.3 \times 10^{-3}$ N/m$^2$ — small but measurable with modern atomic force microscopes. Lamoreaux's 1997 measurement confirmed the Casimir formula to within 5%. Subsequent experiments using microspheres and flat plates have confirmed it to better than 1%.

⚖️ Interpretation: Second quantization is sometimes described as the "first quantization of fields" (as opposed to "the second quantization of particles"). In this view: - First quantization: Classical particles → quantum particles (Schr\u00f6dinger equation) - Second quantization: Classical fields → quantum fields (field operators, Fock space)

The name "second quantization" is therefore somewhat misleading — it is not that we quantize twice, but rather that we quantize the field itself. A more accurate (but less catchy) name would be "quantum field theory in the canonical formulation."

🔗 Connection (Ch 37): Chapter 37 will develop these ideas further, constructing the simplest quantum field theory (the free scalar field) from the ground up. Everything in that chapter will use the creation/annihilation operator machinery established here.

34.8 Summary and the Road Ahead

Let us collect what we have learned. Second quantization provides a unified mathematical framework for many-particle quantum mechanics:

1. For bosons: Creation operators $\hat{a}_k^\dagger$ and annihilation operators $\hat{a}_k$ satisfy $[\hat{a}_k, \hat{a}_{k'}^\dagger] = \delta_{kk'}$.
2. For fermions: Creation operators $\hat{c}_k^\dagger$ and annihilation operators $\hat{c}_k$ satisfy $\{\hat{c}_k, \hat{c}_{k'}^\dagger\} = \delta_{kk'}$.
3. Fock space is the natural state space, built by acting with creation operators on the vacuum.
4. All operators can be expressed in terms of creation and annihilation operators.
5. Particles emerge as excitations of quantized fields: phonons from lattice vibrations, photons from the electromagnetic field, electrons from the Dirac field.

The conceptual arc of this chapter — from ladder operators in Chapter 4 to the threshold of quantum field theory — is one of the great narrative arcs of physics. You started with a single harmonic oscillator and ended with the language of particle creation and destruction. The mathematics did not change (it is still $[\hat{a}, \hat{a}^\dagger] = 1$). What changed was the physical interpretation, and with it, our understanding of what particles are.

Particles are not fundamental objects that exist independently of the fields they inhabit. Particles are excitations of quantum fields. This is, arguably, the deepest insight of 20th-century physics, and it begins here, with the second-quantized formalism you have just learned.

🔗 Looking Ahead: - Ch 35 (Quantum Error Correction): Uses the operator formalism in the context of qubits — a very different application of the same mathematical language. - Ch 36 (Topological Phases): Combines second quantization with topology to classify exotic phases of matter. - Ch 37 (QM to QFT): Develops the full free scalar field theory using the creation/annihilation operators from this chapter.

🔄 Check Your Understanding: At this point, you should be able to answer the following questions: 1. Starting from the bosonic commutation relation, derive the energy spectrum of a single harmonic oscillator. (This is Ch 4 revisited, but now with the physical interpretation of particle number.) 2. Show that $[\hat{N}, \hat{H}] = 0$ for the free phonon Hamiltonian. What does this imply physically? (Phonon number is conserved in the harmonic approximation.) 3. Explain why the interaction $\hat{c}_{\mathbf{k}+\mathbf{q}}^\dagger\hat{c}_{\mathbf{k}}\hat{a}_\mathbf{q}$ describes an electron absorbing a phonon. Identify the initial and final states and verify that energy and crystal momentum are conserved. 4. Starting from the field operator $\hat{\psi}(\mathbf{r}) = \sum_k \phi_k(\mathbf{r})\hat{c}_k$, construct the number density operator $\hat{n}(\mathbf{r}) = \hat{\psi}^\dagger(\mathbf{r})\hat{\psi}(\mathbf{r})$ and show that $\int d^3r\,\hat{n}(\mathbf{r}) = \hat{N}$.

Chapter 34 Notation Summary