Hilbert Space, Inner Products, and Hermitian Operators

The Geometry of Quantum States

Imagine a state of a quantum system not as a wiggly line on a graph, but as an arrow in a high-dimensional space. This space is called Hilbert space. Just as you can add two arrows in 3D space to get a resultant arrow, you can add two quantum states to create a superposition. The "length" of this arrow is not arbitrary; it is related to the probability of finding the system in that state. To define this length, we need a way to measure the "angle" between two arrows. In ordinary 3D space, we use the dot product. In Hilbert space, we use the inner product. This inner product is the quantum equivalent of the dot product, but it must handle complex numbers, which are essential for describing phase and interference. The inner product gives us a way to calculate probabilities: the probability of finding a system in state \(|a\rangle\) when it is prepared in state \(|b\rangle\) is determined by the inner product \(\langle a | b \rangle\). This geometric view unifies the description of particles in space with the description of intrinsic properties like spin, treating them all as vectors in the same kind of space.

Inner Products and Normalization

The inner product between two state vectors \(|\psi\rangle\) and \(|\phi\rangle\) is denoted as \(\langle \psi | \phi \rangle\). It is a complex number. The inner product of a state with itself, \(\langle \psi | \psi \rangle\), is always a non-negative real number, representing the squared "length" or norm of the vector. For a physical state, we require this norm to be 1, a condition known as normalization. This ensures that the total probability of finding the particle somewhere is 100%.

Here, \(\psi^*(x)\) is the complex conjugate of the wavefunction. The inner product is linear in the second argument and anti-linear in the first. This means that if you multiply a state by a complex number \(c\), the inner product scales as \(\langle c\psi | \phi \rangle = c^* \langle \psi | \phi \rangle\) and \(\langle \psi | c\phi \rangle = c \langle \psi | \phi \rangle\). This property is crucial for maintaining the correct probability interpretation when dealing with superpositions.

Hermitian Operators and Observables

Physical quantities such as position, momentum, and energy are represented by operators acting on the state vectors. For a measurement to yield a real number, the operator must be Hermitian (or self-adjoint). An operator \(\hat{A}\) is Hermitian if it satisfies the condition \(\langle \phi | \hat{A} \psi \rangle = \langle \hat{A} \phi | \psi \rangle\) for all states \(|\phi\rangle\) and \(|\psi\rangle\). This condition ensures that the eigenvalues of the operator, which correspond to the possible measurement outcomes, are strictly real numbers.

Here, \(\hat{A}^\dagger\) denotes the adjoint (or conjugate transpose) of the operator. In the position representation, the momentum operator \(\hat{p} = -i\hbar \frac{d}{dx}\) is Hermitian, provided the wavefunctions vanish at infinity. This mathematical property is not a mere formality; it is the reason why we never observe imaginary momenta or energies in the laboratory. The Hermitian nature of operators also guarantees that eigenvectors corresponding to different eigenvalues are orthogonal, which allows us to decompose any state into a basis of these eigenvectors.

Orthogonality and Completeness

A key consequence of the Hermitian property is the orthogonality of eigenstates. If \(\hat{A} |n\rangle = a_n |n\rangle\) and \(\hat{A} |m\rangle = a_m |m\rangle\) with \(a_n \neq a_m\), then \(\langle n | m \rangle = 0\). This means that states corresponding to distinct measurement outcomes are perpendicular in Hilbert space. This orthogonality is the mathematical expression of the fact that a system cannot be in two distinct eigenstates of the same observable simultaneously.

Furthermore, the set of all eigenstates of a Hermitian operator forms a complete basis for the Hilbert space. This means that any arbitrary state \(|\psi\rangle\) can be expanded as a linear combination of these eigenstates:

The coefficients \(c_n\) are given by the inner product \(c_n = \langle n | \psi \rangle\). The probability of measuring the eigenvalue \(a_n\) is \(|c_n|^2 = |\langle n | \psi \rangle|^2\). The completeness relation, \(\sum_n |n\rangle \langle n| = \hat{I}\), allows us to insert the identity operator into expressions, facilitating calculations of expectation values and transition probabilities. This expansion is the quantum analogue of expressing a vector in 3D space in terms of the unit vectors \(\hat{i}, \hat{j}, \hat{k}\).