| density matrix | |
|---|---|
| Name | Density matrix |
| Field | Quantum mechanics |
| Introduced | 1927 |
| Introduced by | John von Neumann |
density matrix
A density matrix, or density operator, is an operator used in Quantum mechanics to describe the statistical state of a quantum system, encompassing both pure and mixed states. It provides a complete description of observable expectation values and is fundamental to the study of quantum statistical mechanics, quantum information theory, and open quantum systems.
The density matrix ρ is a positive semidefinite, trace-one operator on a Hilbert space that encodes probabilities and coherences of a quantum system. First formalized by John von Neumann in the context of statistical mechanics, the density operator generalizes the state vector (or wave function) description to statistical ensembles and subsystems. Physically, diagonal elements of ρ in a basis give the probability distribution for outcomes of a corresponding projective measurement, while off-diagonal elements quantify quantum coherences and interference. The formalism links to expectation values via Tr(ρA) for any observable A represented by a self-adjoint operator.
Mathematically, a density matrix ρ on a finite-dimensional Hilbert space H satisfies ρ = ρ†, ρ ≥ 0, and Tr(ρ) = 1. It can be represented by matrices in a chosen basis, and admits a spectral decomposition ρ = Σ_i p_i |ψ_i⟩⟨ψ_i| with probabilities p_i ≥ 0 and orthonormal eigenstates |ψ_i⟩. Important properties include convexity (the set of density matrices is convex), unitary covariance (ρ → UρU† under unitary operator U), and monotonicity under completely positive trace-preserving maps (quantum channels). The purity Tr(ρ^2) ranges from 1/d for the maximally mixed state to 1 for pure states. The density matrix formalism connects to entropy measures such as the von Neumann entropy S(ρ) = −Tr(ρ log ρ), which parallels the Gibbs entropy in classical statistical mechanics.
A pure state corresponds to a rank-one projector ρ = |ψ⟩⟨ψ| and is equivalent to a state vector up to a global phase; pure states have purity Tr(ρ^2)=1 and zero von Neumann entropy. Mixed states represent classical ensembles of pure states, e.g., ρ = Σ_j w_j |ψ_j⟩⟨ψ_j| with weights w_j summing to unity. Different ensembles can yield the same density matrix (ensemble equivalence), a phenomenon tied to the convex structure of the state space and exemplified by the Hughston–Jozsa–Wootters theorem. Mixed states arise naturally from ignorance about preparation, thermalization (the canonical ensemble), or when considering subsystems of entangled composites.
Closed-system unitary evolution of a density matrix follows the von Neumann equation iħ dρ/dt = [H, ρ], where H is the system Hamiltonian and [·,·] denotes the commutator. This equation is equivalent to Schrödinger evolution for pure states and preserves trace and positivity. For open systems interacting with an environment (bath), reduced dynamics are often nonunitary and modeled by quantum master equations such as the Lindblad equation (Gorini–Kossakowski–Sudarshan–Lindblad form), which generate trace-preserving completely positive semigroups. Perturbative derivations use techniques from the Born–Markov approximation and Nakajima–Zwanzig projection operator methods.
For a composite system AB described by ρ_AB on H_A ⊗ H_B, the reduced density matrix for subsystem A is ρ_A = Tr_B(ρ_AB) using the partial trace over subsystem B. Reduced density matrices capture local statistics and often become mixed due to entanglement between subsystems even when ρ_AB is pure. Entanglement measures—such as the entanglement entropy (von Neumann entropy of ρ_A), concurrence, and negativity—are frequently computed from reduced density matrices. The Schmidt decomposition provides a canonical form for pure bipartite states linking nonzero Schmidt coefficients to the spectrum of reduced density operators.
Measurement outcomes and post-measurement states are described using density matrices and Positive Operator-Valued Measures (POVMs). A projective measurement with projectors {P_k} yields outcome probabilities p_k = Tr(ρP_k) and post-measurement states P_k ρ P_k / p_k. Interaction with an environment induces decoherence, suppressing off-diagonal elements in certain bases and driving classical behavior; decoherence is modeled quantitatively with reduced density matrices and master equations. Open quantum systems theory, developed at institutions such as Bell Labs, Los Alamos National Laboratory, and research groups in quantum optics and condensed matter physics, uses completely positive maps, Kraus operator-sum representations, and correlation functions to describe dissipation and noise.
Density matrices are central in quantum information theory for describing mixed-state protocols (quantum communication, quantum cryptography, and quantum error correction). They quantify resources like entanglement and coherence, underpin fidelity measures (Uhlmann fidelity), and appear in performance bounds such as the Holevo bound. In quantum statistical mechanics, density operators describe thermal states (Gibbs states) ρ = e^{−βH}/Z and are used to derive thermodynamic quantities, fluctuations, and phase transitions. Practical applications include state tomography procedures (quantum state tomography), simulations with tensor network methods (e.g., DMRG, matrix product density operators), and experimental implementations in platforms such as IBM Quantum, Google Quantum AI, ion trap quantum computers, and superconducting qubits.
Category:Quantum mechanics Category:Quantum information theory