ρ

The lowercase Greek letter $\rho$, pronounced “rho,” denotes a density operator by default in quantum-state contexts. The same glyph also commonly denotes a probability density, number or charge density, density of states, radial coordinate, or resistivity. Its arguments, indices, units, and mathematical type are decisive.

Default Meaning

For a quantum state, $\rho$ is a positive trace-one operator:

$$\rho\ge0, \qquad \operatorname{Tr}\rho=1.$$

In finite dimensions,

$$\rho^\dagger=\rho, \qquad \rho=\sum_j\lambda_j \lvert j\rangle\langle j\rvert, \qquad \lambda_j\ge0, \qquad \sum_j\lambda_j=1.$$

The abstract density operator is basis independent. Its matrix elements in a basis $\{\lvert n\rangle\}$ are

$$\rho_{mn} =\langle m\vert\rho\vert n\rangle.$$

The phrase density matrix often refers informally to the operator itself, but a matrix is one representation of that operator.

Mathematical Type and Units

Use	Mathematical type	Typical normalization or units
$\rho$	Positive trace-class state operator	Dimensionless; $\operatorname{Tr}\rho=1$
$\rho_{mn}$	Matrix element in a discrete orthonormal basis	Dimensionless
$\rho(x,x')$	Coordinate kernel $\langle x\vert\rho\vert x'\rangle$	$L^{-1}$ in one dimension
$\rho(x)$	Position probability density	$L^{-1}$ in one dimension
$n(\mathbf x)$ or $\rho_N(\mathbf x)$	Number density	$L^{-3}$
$\rho_q(\mathbf x)$	Charge density	charge per volume
$\rho_m(\mathbf x)$	Mass density	mass per volume
$\rho(E)$	Density of states	states per energy, possibly also per volume
$\rho$ in cylindrical coordinates	Radial distance from an axis	length
$\rho_{\mathrm{el}}$	Electrical resistivity	$\Omega\,\mathrm m$ in SI

The coordinate kernel’s units follow from

$$\operatorname{Tr}\rho =\int\rho(x,x)\,dx=1.$$

Thus a dimensionless abstract operator can have dimensionful continuous-basis matrix elements because generalized basis kets and integration measures carry dimensions.

Common State Forms

Form	Meaning
$\rho(t)$	Time-dependent density operator
$\rho_A$	State reduced to subsystem $A$
$\rho_{AB}$	Joint state of subsystems $A$ and $B$
$\rho_{\mathrm{ss}}$	Steady-state density operator
$\rho_\beta$	Thermal state at inverse temperature $\beta$
$\rho_{\mathrm{in}}$, $\rho_{\mathrm{out}}$	Input and output states
$\rho^{(k)}$	Member of an ensemble or indexed family
$\rho_{mn}$	Basis matrix element, not usually a subsystem label

Subsystem subscripts and matrix indices play different roles. $\rho_A$ is itself an operator on $\mathcal H_A$, while $\rho_{mn}$ is a scalar component after choosing a basis.

Prediction Formulas

For an observable $A$,

$$\langle A\rangle =\operatorname{Tr}(\rho A).$$

For a measurement effect $E_k$,

$$p(k)=\operatorname{Tr}(\rho E_k).$$

For a bipartite state,

$$\rho_A=\operatorname{Tr}_B(\rho_{AB}).$$

Closed-system evolution is

$$i\hbar\frac{d\rho}{dt} =[H,\rho],$$

or

$$\rho(t) =U(t,t_0)\rho(t_0)U^\dagger(t,t_0).$$

These equations identify $\rho$ as an operator. A scalar spatial density does not enter a commutator in this way.

Pure, Mixed, and Thermal Forms

A pure state has

$$\rho =\lvert\psi\rangle\langle\psi\rvert, \qquad \rho^2=\rho, \qquad \operatorname{Tr}(\rho^2)=1.$$

A mixed state has $\operatorname{Tr}(\rho^2)<1$. A thermal Gibbs state is often written

$$\rho_\beta =\frac{e^{-\beta H}}{Z}, \qquad Z=\operatorname{Tr}(e^{-\beta H}),$$

where

$$\beta=\frac{1}{k_{\mathrm B}T}$$

unless thermal units $k_{\mathrm B}=1$ have been declared.

Coordinate Kernel

The density operator in position representation has kernel

$$\rho(x,x') =\langle x\vert\rho\vert x'\rangle.$$

For a pure state,

$$\rho(x,x') =\psi(x)\psi^*(x').$$

The diagonal

$$\rho(x,x)=\lvert\psi(x)\rvert^2$$

is the position probability density for a scalar pure wavefunction. The off-diagonal kernel contains spatial coherence information.

For a mixed state,

$$\rho(x,x') =\sum_jp_j \psi_j(x)\psi_j^*(x'),$$

for any chosen ensemble representation. The operator is independent of which ensemble decomposition is used.

Probability and Number Densities

In wave mechanics, authors sometimes write

$$\rho_{\mathrm{prob}}(x,t) =\lvert\psi(x,t)\rvert^2,$$

with

$$\int\rho_{\mathrm{prob}}(x,t)\,dx=1.$$

For a spinor, the density includes a component sum:

$$\rho_{\mathrm{prob}}(\mathbf x,t) =\psi^\dagger(\mathbf x,t)\psi(\mathbf x,t).$$

A particle-number density instead often satisfies

$$\int n(\mathbf x)\,d^3x=N.$$

The notation $\rho(\mathbf x)$ does not reveal whether the integral is one, total charge, total mass, or particle number. Units and the defining integral do.

Density of States

The notation

$$\rho(E)$$

often denotes a density of energy levels. In a discrete formal form,

$$\rho(E) =\sum_n\delta(E-E_n).$$

It has units of inverse energy. A density of states per volume has units of inverse energy per volume. This $\rho(E)$ is a scalar distribution, not a quantum state operator.

Radial Coordinate

In cylindrical coordinates,

$$\rho=\sqrt{x^2+y^2}$$

is commonly the distance from the symmetry axis. The volume element is

$$d^3x=\rho\,d\rho\,d\phi\,dz.$$

Here $\rho$ has units of length and appears without an operator trace. Some texts instead use $r_\perp$ to avoid collision with density notation.

Canonical Home

Density Operators owns the default quantum-state meaning. Density-Matrix Conventions defines subscripts, matrix elements, ensembles, and Bloch-vector notation.

Reduced Density Matrices owns $\rho_A$ and $\rho_{AB}$. The Density Matrix Glossary Entry gives the concept-level summary, and the Density-Matrix Expectation Formula Card provides the compact trace rule.

Other Notations

• $\varrho$ is sometimes used for a density operator when $\rho$ is needed for a spatial density.
• $\hat\rho$ makes operator status explicit; this documentation usually omits the hat when the context is unambiguous.
• $D$ or $W$ can denote density matrices in some older or specialized sources.
• $\rho^{(1)}$ and $\rho^{(2)}$ can denote reduced one- and two-body density matrices rather than powers.

Convention Warnings

• Do not confuse a density operator $\rho$ with a spatial density $\rho(x)$.
• Matrix entries of $\rho$ depend on the chosen basis.
• The notation “density matrix” is common even when the abstract object is a density operator.
• A density operator is dimensionless, but its continuous-basis kernel can carry units.
• $\rho(x,x)$ is a diagonal kernel; $\rho(x)$ may instead be an independently defined scalar density.
• A subsystem label such as $A$ is not a matrix index.
• Superscripts in $\rho^{(k)}$ are not automatically powers.
• A diagonal density matrix is diagonal only relative to the stated basis.
• $\rho(E)$ as a density of states is not positive trace one as an operator.
• The cylindrical radial coordinate $\rho$ is distinguished by its geometric definition and length units.
• A number, charge, or mass density integrates to the corresponding total, not necessarily to one.

References

• L. E. Ballentine, Quantum Mechanics: A Modern Development, 2nd ed., World Scientific, 2014, chs. 2 and 3.
• M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information, 10th anniversary ed., Cambridge University Press, 2010, sec. 2.4.
• R. Shankar, Principles of Quantum Mechanics, 2nd ed., Springer, 1994, ch. 12.
• J. J. Sakurai and J. Napolitano, Modern Quantum Mechanics, 3rd ed., Cambridge University Press, 2020.