No history yet

Mathematical Tools for Entanglement

Beyond State Vectors

So far, we've described quantum states using vectors like "ψ"|"\psi"\rangle. This works perfectly for isolated, pure quantum states. But in the real world, systems are rarely perfect. We might not know the exact state of a system, or it might be interacting with its environment. For these situations, we need a more powerful tool: the density matrix, represented by the Greek letter "ρ""\rho" (rho).

A density matrix can describe both pure states and mixed states. A mixed state is a statistical mixture, or ensemble, of several possible quantum states. For example, imagine a qubit that has a 50% chance of being in state "ψ"1|"\psi"_1\rangle and a 50% chance of being in state "ψ"2|"\psi"_2\rangle. This is a classic mixed state.

For any pure state "ψ"|"\psi"\rangle, its density matrix is formed by taking the outer product of the state vector with its conjugate transpose, ""ψ"""\langle"\psi"|". This is written as "ρ""\rho" = "ψ"""ψ"|"\psi"\rangle"\langle"\psi"|.

ρpure=ψψ\rho_{pure} = |\psi\rangle\langle\psi|

For a mixed state, the density matrix is a weighted sum of the pure state density matrices that make up the mixture. If a system is in state "ψ"i|"\psi"_i\rangle with probability pip_i, its density matrix is:

ρmixed=ipiψiψi\rho_{mixed} = \sum_{i} p_i |\psi_i\rangle\langle\psi_i|

A key feature of any density matrix is that its trace (the sum of its diagonal elements) is always 1. This is written as Tr("ρ""\rho") = 1. This property ensures that the total probability of the system being in some state is 100%.

We can distinguish between pure and mixed states by calculating the trace of the square of the density matrix, Tr("ρ"2"\rho"^2). For a pure state, Tr("ρ"2"\rho"^2) = 1. For any mixed state, Tr("ρ"2"\rho"^2) < 1. This value, known as the purity, tells us how 'mixed' our knowledge of the system is. A purity of 1 means we know its state exactly.

Isolating a Subsystem

Now, let's consider a system made of two parts, like a pair of qubits shared between Alice and Bob. We call this a bipartite system. If their combined state "ρ"AB"\rho"_{AB} is entangled, we can't simply write down a separate, independent state vector for Alice's qubit. The whole point of entanglement is that their fates are intertwined. So how can we describe just Alice's part of the system?

We use an operation called the to mathematically 'ignore' or 'average over' Bob's part of the system. By taking the partial trace over Bob's subsystem (denoted TrB_B), we are left with a reduced density matrix, "ρ"A"\rho"_A, which describes everything we can possibly know about Alice's qubit alone.

ρA=TrB(ρAB)\rho_A = \text{Tr}_B(\rho_{AB})

Here's the crucial insight: if the original state "ρ"AB"\rho"_{AB} was a pure entangled state, the reduced density matrices for the subsystems, "ρ"A"\rho"_A and "ρ"B"\rho"_B, will be mixed states. Let's see this with an example. Consider the Bell state "Φ"+=1"2"(00+11)|"\Phi"^+\rangle = \frac{1}{"\sqrt{2}"}(|00\rangle + |11\rangle). The joint density matrix is:

ρAB=Φ+Φ+=12(0000+0011+1100+1111)\rho_{AB} = |\Phi^+\rangle\langle\Phi^+| = \frac{1}{2}(|00\rangle\langle00| + |00\rangle\langle11| + |11\rangle\langle00| + |11\rangle\langle11|)

When we perform the partial trace over Bob's system, we are left with Alice's reduced density matrix:

ρA=TrB(ρAB)=12(00+11)=12I\rho_A = \text{Tr}_B(\rho_{AB}) = \frac{1}{2}(|0\rangle\langle0| + |1\rangle\langle1|) = \frac{1}{2}I

This is a profound result. Even though the combined system of Alice and Bob is in a perfectly known pure state, the state of Alice's qubit, when considered alone, is completely random. All the information is stored in the correlations between them, not in the individual parts.

Quantifying Entanglement

We've seen that entanglement in a pure state leads to mixedness in its subsystems. We can use this fact to quantify entanglement. Two key tools for this are the Schmidt decomposition and the Von Neumann entropy.

The is a theorem stating that any pure state "ψ"|"\psi"\rangle of a bipartite system can be written in a special, diagonal form:

ψAB=i=1kλiuiAviB|\psi\rangle_{AB} = \sum_{i=1}^{k} \lambda_i |u_i\rangle_A \otimes |v_i\rangle_B

The number of non-zero Schmidt coefficients, kk, is called the Schmidt rank. This number is a direct indicator of entanglement. If the Schmidt rank is 1, the state is a simple product state with no entanglement. If the Schmidt rank is greater than 1, the state is entangled.

StateSchmidt DecompositionSchmidt RankEntangled?
$0\rangle1\rangle$$1 \cdot
$\frac{1}{\sqrt{2}}(00\rangle +11\rangle)$$\frac{1}{\sqrt{2}}

While Schmidt rank tells us if a state is entangled, we often want to know how much. For this, we use the of a reduced density matrix. It's the quantum analog of Shannon entropy from classical information theory.

S(ρ)=Tr(ρlog2ρ)S(\rho) = -\text{Tr}(\rho \log_2 \rho)

The entanglement of a pure bipartite state "ψ"AB|"\psi"\rangle_{AB} is defined as the Von Neumann entropy of either of its reduced density matrices. Conveniently, they are always equal: S("ρ"A)=S("ρ"B)S("\rho"_A) = S("\rho"_B). This value can be calculated directly from the Schmidt coefficients:

E("ψ"AB)=S("ρ"A)=i=1kλi2log2(λi2)E(|"\psi"\rangle_{AB}) = S("\rho"_A) = -\sum_{i=1}^{k} \lambda_i^2 \log_2 (\lambda_i^2)

For a product state, there is only one "λ"i=1"\lambda"_i = 1, so the entropy is 0. For a maximally entangled state like a Bell state, the coefficients are all equal, maximizing the entropy and thus the entanglement.

Quiz Questions 1/5

What is the primary purpose of a density matrix (ρ\rho) in quantum mechanics?

Quiz Questions 2/5

If a pure bipartite state ψAB|\psi\rangle_{AB} is entangled, taking the partial trace over subsystem B to get Alice's reduced density matrix, ρA=TrB(ψABψAB)\rho_A = \text{Tr}_B(|\psi\rangle_{AB}\langle\psi|_{AB}), will result in a __________ state for Alice.

These mathematical tools, from the density matrix to the Schmidt decomposition, provide the rigorous language needed to describe, analyze, and quantify the strange and powerful connections within entangled quantum systems.