Chapter 11 Key Takeaways: Tensor Products and Composite Systems
The Big Ideas
-
Quantum systems combine via the tensor product. The Hilbert space of a composite system $AB$ is $\mathcal{H}_A \otimes \mathcal{H}_B$, with dimension $d_A \cdot d_B$ — multiplicative, not additive. This exponential growth of state space with the number of subsystems is the mathematical source of both quantum computing's power and the difficulty of classical simulation.
-
Entanglement is the rule, not the exception. A state that cannot be written as $|\psi\rangle_A \otimes |\phi\rangle_B$ is entangled. In a composite Hilbert space, product states form a set of measure zero — almost every state is entangled. Entanglement means the whole has a definite state while its parts do not.
-
The coefficient matrix encodes entanglement. For a bipartite state $|\Psi\rangle = \sum_{ij}\alpha_{ij}|ij\rangle$, the matrix $C_{ij} = \alpha_{ij}$ has rank 1 if and only if the state is separable. For two qubits, $\det(C) \neq 0$ is equivalent to entanglement.
-
The Schmidt decomposition is the canonical form. Any bipartite pure state can be written as $|\Psi\rangle = \sum_k \lambda_k |a_k\rangle|b_k\rangle$ — a single sum with orthonormal bases and positive coefficients. The Schmidt coefficients are the singular values of $C$, and the Schmidt rank equals the number of nonzero coefficients.
-
The partial trace gives the correct subsystem description. The reduced density matrix $\hat{\rho}_A = \text{Tr}_B(\hat{\rho}_{AB})$ is the unique operator that reproduces all local measurement statistics. For entangled pure states, the reduced state is mixed — reflecting quantum correlations, not classical ignorance.
-
Bell states are the maximally entangled basis. The four Bell states $|\Phi^{\pm}\rangle$, $|\Psi^{\pm}\rangle$ span the two-qubit space. Each has Schmidt coefficients $\frac{1}{\sqrt{2}}, \frac{1}{\sqrt{2}}$, entanglement entropy $S = 1$ ebit, and reduced density matrix $\frac{1}{2}\hat{I}$.
-
Entanglement is a resource. Bell pairs power quantum teleportation (1 ebit + 2 cbits = 1 qubit), superdense coding (1 ebit + 1 qubit = 2 cbits), and quantum key distribution (security from monogamy of entanglement).
Essential Formulas
| Formula | Meaning |
|---|---|
| $\dim(\mathcal{H}_A \otimes \mathcal{H}_B) = d_A \cdot d_B$ | Dimension of composite space |
| $(\hat{O}_A \otimes \hat{O}_B)(\|\psi\rangle \otimes \|\phi\rangle) = (\hat{O}_A\|\psi\rangle) \otimes (\hat{O}_B\|\phi\rangle)$ | Action of product operators |
| $\text{Tr}_B(\|a_1\rangle\langle a_2\| \otimes \|b_1\rangle\langle b_2\|) = \|a_1\rangle\langle a_2\| \cdot \langle b_2\|b_1\rangle$ | Partial trace definition |
| $\hat{\rho}_A = \sum_k \lambda_k^2 \|a_k\rangle\langle a_k\|$ | Reduced state from Schmidt decomposition |
| $S(\hat{\rho}_A) = -\sum_k \lambda_k^2 \log_2 \lambda_k^2$ | Entanglement entropy |
Common Mistakes to Avoid
- Confusing dimension addition with multiplication. The composite space has $d_A \times d_B$ dimensions, not $d_A + d_B$.
- Assuming every two-term state is entangled. The state $\frac{1}{\sqrt{2}}(|00\rangle + |01\rangle) = |0\rangle|+\rangle$ has two terms but is separable. Always check the coefficient matrix rank.
- Thinking a mixed reduced state implies a mixed composite state. A pure entangled state of $AB$ gives mixed reduced states. The mixedness is from entanglement, not classical ignorance.
- Confusing the Schmidt decomposition with a change of basis. The Schmidt decomposition uses different bases for $A$ and $B$, chosen to diagonalize the correlation structure. It is not just rewriting in a new product basis.
- Forgetting that the partial trace depends on the chosen subsystem. $\text{Tr}_A$ and $\text{Tr}_B$ are different operations yielding operators on different spaces.
Connections to Previous Chapters
| Chapter | Connection |
|---|---|
| Ch 8 (Dirac Notation) | All tensor product manipulations use Dirac bra-ket formalism |
| Ch 9 (Spectral Decomposition) | Schmidt decomposition generalizes eigendecomposition via SVD |
| Ch 10 (Density Matrices) | Reduced density matrices from partial trace; pure vs. mixed distinction |
What Comes Next
The tensor product and entanglement formalism from this chapter is the foundation for:
- Angular momentum addition (Clebsch-Gordan coefficients generalize the singlet-triplet decomposition)
- Bell's theorem (quantitative Bell inequalities use expectation values of product operators)
- Quantum gates and circuits (multi-qubit operations are operators on tensor product spaces)
- Decoherence (entanglement of a system with its environment explains the quantum-to-classical transition)
- Quantum error correction (encoding logical qubits in entangled states of many physical qubits)