<aside> 📌 Previously, we studied in detail how density operator works for definite states. Now, we look for a way to describe an ensemble (A system which is a statistical mixture of states) via the density operator.
</aside>
The following steps are followed in order to construct a density matrix describing an ensemble:
Consider an ensemble with $n$ possible states $\ket{\psi_i}$ each with probability $p_i$ (where $i\in n$). Further, as studied earlier, density operator for each state $\ket{\psi_i}$ is given by:
$\rho_i = \ket{\psi_i}\bra{\psi_i}$
Then, density operator for the ensemble if given by:
$\rho = \sum_{i=1}^{n}p_i\ket{\psi_i}\bra{\psi_i}$
An operator $\rho$ is a density operator iff it satisfies the following three requirements:
$\rho = \rho^\dagger$ -(hermitian)
$Tr(\rho) = 1$
$\rho$ is a positive operator : $\braket{u|\rho|u} \ge 0$ for any state vector $\ket{u}$.
(Recall that an operator is positive if and only if it is Hermitian and has non-negative eigenvalues.)
Additional Points to be noted:
$Tr(\rho^2) = 1$ ———(For Pure State)
$Tr(\rho^2) < 1$ ———(For Mixed State)
Density operator for a Mixed State is not a projection operator (unlike pure state).
The expectation value of a statistical mixture (ensemble) is the same as that of pure state, i.e.:
$\braket{A} = Tr(\rho A)$
Consider the matrix representation of the density operator (representing an ensemble)
$\rho = \left(\begin{matrix} \braket{0|\rho|0} & \braket{0|\rho|1} \\ \braket{1|\rho|0} & \braket{1|\rho|1} \end{matrix}\right)$
Then, the probability that the system is found in state:
i.e., in the same fashion as in the case of a pure state.