Bra-Ket Notation and Dual Spaces

In our last lesson, we established the mathematical setting for quantum mechanics: the Hilbert space. We saw that the state of a quantum system is represented by a vector in this complete inner product space. Now that we have defined the stage, we need to introduce the language we'll use to describe the actors and their interactions. This language, a powerful and elegant notation developed by Paul Dirac, is the focus of today's lesson.

Your goal for this lesson is to master the use of Dirac's bra-ket notation. We will define how it represents state vectors, inner products, and the actions of operators. Crucially, we will delve into the underlying structure connecting these elements by relating the "bra" to its corresponding vector in the dual space. This formalism is not just a notational convenience; it encapsulates deep mathematical relationships, making complex calculations in quantum mechanics remarkably intuitive.

Dual Spaces and Linear Functionals

Before introducing the notation itself, we must first understand the concept of a dual space. In the previous lesson, we identified quantum states with vectors in a Hilbert space $H$ . The dual space, denoted $H^{*}$ , is the set of all continuous linear functionals on $H$ .

A linear functional is simply a linear map from the vector space to its field of scalars. In our case, it's a map $f : H \to C$ such that for any vectors $∣ ψ ⟩, ∣ ϕ ⟩ \in H$ and any scalars $a, b \in C$ :

f (a ∣ ψ ⟩ + b ∣ ϕ ⟩) = a f (∣ ψ ⟩) + b f (∣ ϕ ⟩)

Essentially, a linear functional "eats" a vector and produces a number, and it does so linearly.

The collection of all such functionals on $H$ itself forms a vector space—the dual space $H^{*}$ . The following video provides an excellent conceptual introduction, using the familiar example of $R^{2}$ to build intuition before abstracting to quantum state spaces.

Ch 6: What are bras and bra-ket notation? | Maths of Quantum Mechanics

Watch the initial segments of this video from Quantum Sense to understand the concepts of linear functionals and the dual space.

Focus on these key ideas: Linear Functionals: The video defines a linear functional as a map from a vector to a scalar and shows how, in a finite-dimensional space, it can be represented by a row matrix acting on a column vector. The Dual Space: This part explains that the set of all linear functionals (represented by row matrices) forms its own vector space, called the dual space.

Kets, Bras, and the Inner Product

Dirac's great insight was to create a notation that seamlessly integrates the Hilbert space $H$ and its dual $H^{*}$ .

Kets: A vector $ψ$ in the Hilbert space $H$ is denoted by a ket, written as $∣ ψ ⟩$ . This is purely a notational change. All the properties of vectors we discussed (addition, scalar multiplication) apply to kets.
Bras: A linear functional $f \in H^{*}$ is denoted by a bra, written as $⟨ f ∣$ . A bra acts on a ket to produce a complex number. The action of the bra $⟨ f ∣$ on the ket $∣ ψ ⟩$ is written as $⟨ f ∣ ψ ⟩$ .

This structure might seem to split our world into two distinct types of objects, kets and bras. However, the power of the formalism lies in the profound connection between them, which is guaranteed by the structure of a Hilbert space.

The Riesz Representation Theorem

For every continuous linear functional on a Hilbert space, there is a unique vector in that space associated with it. More formally, the Riesz Representation Theorem states that for every bra $⟨ ϕ ∣ \in H^{*}$ , there exists a unique ket $∣ ϕ ⟩ \in H$ such that the action of the bra on any other ket $∣ ψ ⟩$ is given by the inner product:

⟨ ϕ ∣ ψ ⟩ = (ϕ, ψ)

where $(ϕ, ψ)$ is the inner product on $H$ that we defined in the previous lesson.

This theorem is the cornerstone of bra-ket notation. It provides a one-to-one correspondence between the Hilbert space $H$ and its dual $H^{*}$ , justifying why we label a bra $⟨ ϕ ∣$ with the same symbol as its corresponding ket $∣ ϕ ⟩$ . The bra is effectively the "inner product-taker" associated with its ket.

The following reading formalizes this connection.

2 Hilbert Space

This text provides a rigorous definition of dual spaces and the introduction of Dirac notation.

Please read the sections titled "Dual Spaces" and "Dirac Notation and Continuum States". Focus on the following points: In "Dual Spaces", read about how any element of the dual space H^* can be constructed using the inner product with a fixed vector from H. The text mentions that this is due to the Riesz Representation Theorem. In "Dirac Notation...", pay close attention to how a ket and bra are defined as elements of H and H^* respectively. Note how the inner product is naturally written as a "bra-ket" or bracket, \langle\phi|\psi\rangle, implicitly using the Riesz theorem.

This correspondence also clarifies the anti-linear property of the inner product. Recall from the last lesson that the inner product in physics is anti-linear in its first argument. The relationship between a ket and its corresponding bra is therefore anti-linear:

(a ∣ ϕ ⟩ + b ∣ ψ ⟩) ⟷ a^{*} ⟨ ϕ ∣ + b^{*} ⟨ ψ ∣

This ensures that the inner product behaves correctly:

⟨ a ϕ + b ψ ∣ χ ⟩ = a^{*} ⟨ ϕ ∣ χ ⟩ + b^{*} ⟨ ψ ∣ χ ⟩

Operators in Dirac Notation

With this notation for states and inner products, we can now express the action of linear operators. A linear operator $A$ is a map from $H$ to $H$ that transforms one ket into another: $A ∣ ψ ⟩ = ∣ ϕ ⟩$ .

The most common quantity we compute is the inner product of some state $⟨ χ ∣$ with the state that results from $A$ acting on $∣ ψ ⟩$ . This is written as $⟨ χ ∣ (A ∣ ψ ⟩)$ , which is conventionally shortened to the "sandwich" notation $⟨ χ ∣ A ∣ ψ ⟩$ . This quantity is often called a matrix element of the operator $A$ .

The Adjoint Operator

For every linear operator $A$ , we can define its adjoint, denoted $A^{†}$ . The adjoint is defined by the following relation for all kets $∣ ϕ ⟩$ and $∣ ψ ⟩$ :

⟨ ϕ ∣ A^{†} ∣ ψ ⟩ = (⟨ ψ ∣ A ∣ ϕ ⟩)^{*}

A useful mnemonic is to see that taking the adjoint of a full matrix element is equivalent to swapping the bra and ket and applying the adjoint to the operator: $(⟨ ϕ ∣ A ∣ ψ ⟩)^{†} = ⟨ ψ ∣ A^{†} ∣ ϕ ⟩$ .

This also means we can define the action of an operator "to the left" on a bra. The bra that results from $A$ acting on $⟨ ψ ∣$ is denoted $⟨ ψ ∣ A$ , and it is defined as the bra corresponding to the ket $A^{†} ∣ ψ ⟩$ .
In other words, the dagger operation connects the action on kets and bras:

(A ∣ ψ ⟩)^{†} = ⟨ ψ ∣ A^{†}

This is a powerful feature of the notation that greatly simplifies manipulations.

Linear Algebra In Dirac Notation

This text gives a clear definition of operators, their matrix elements, and the adjoint in Dirac notation.

In Section 3.3, "Operators, Dyads", please read from the beginning to understand how operators are defined. Focus on the introduction of the sandwich notation \langle\phi|A|\psi\rangle. Then, skip to the definition of the adjoint operator and study the rule for its matrix elements.

Representations in a Basis

The abstract notation becomes concrete when we introduce a basis. Let ${∣ n ⟩}$ be an orthonormal basis for our Hilbert space $H$ , meaning $⟨ n ∣ m ⟩ = δ_{nm}$ .

Any ket $∣ ψ ⟩$ can be expanded in this basis:

∣ ψ ⟩ = n \sum c_{n} ∣ n ⟩

The coefficients $c_{n}$ are found by taking the inner product with the basis bra $⟨ m ∣$ :

⟨ m ∣ ψ ⟩ = n \sum c_{n} ⟨ m ∣ n ⟩ = n \sum c_{n} δ_{mn} = c_{m}

So, the coefficients are just $c_{n} = ⟨ n ∣ ψ ⟩$ . Substituting this back gives:

∣ ψ ⟩ = n \sum ⟨ n ∣ ψ ⟩ ∣ n ⟩ = n \sum ∣ n ⟩ ⟨ n ∣ ψ ⟩

Since this must hold for any $∣ ψ ⟩$ , we arrive at the immensely useful completeness relation or resolution of the identity:

I = n \sum ∣ n ⟩ ⟨ n ∣

This shows that the identity operator can be decomposed into a sum of outer products of the basis vectors. Each term $∣ n ⟩ ⟨ n ∣$ is itself an operator—specifically, it's the projection operator onto the state $∣ n ⟩$ .

The following video provides a concise derivation of this identity, highlighting the power and elegance of the notation.

Ch 6: What are bras and bra-ket notation? | Maths of Quantum Mechanics

Watch this final segment from the Quantum Sense video to see a practical application of the notation's power.

The segment from Deriving the Identity demonstrates how seamlessly the notation allows you to manipulate expressions to prove the completeness relation.

With a basis chosen, we can represent these abstract objects with familiar arrays of numbers:

A ket $∣ ψ ⟩$ is represented by a column vector of its components $⟨ n ∣ ψ ⟩$ .
A bra $⟨ ϕ ∣$ is represented by a row vector of its components $⟨ ϕ ∣ n ⟩ = ⟨ n ∣ ϕ ⟩^{*}$ .
An operator $A$ is represented by a matrix with elements $A_{mn} = ⟨ m ∣ A ∣ n ⟩$ .

Finally, this connects the abstract formalism back to the wave functions you are familiar with from your previous studies. For a particle in one dimension, the position eigenstates $∣ x ⟩$ form a continuous basis. The expansion coefficient of a state $∣ ψ ⟩$ in this basis is precisely the position-space wave function:

ψ (x) = ⟨ x ∣ ψ ⟩

The inner product of two states $∣ ϕ ⟩$ and $∣ ψ ⟩$ becomes the familiar overlap integral:

⟨ ϕ ∣ ψ ⟩ = \int ⟨ ϕ ∣ x ⟩ ⟨ x ∣ ψ ⟩ d x = \int ϕ^{*} (x) ψ (x) d x

Here, we have inserted the completeness relation for the continuous position basis, $I = \int ∣ x ⟩ ⟨ x ∣ d x$ . The abstract vector $∣ ψ ⟩$ in Hilbert space is the fundamental object; its wave function $ψ (x)$ is merely its representation in a particular basis.

Conclusion

In this lesson, we have translated the abstract properties of Hilbert spaces into the practical and powerful language of Dirac notation. Let's review the key ideas:

Quantum states are kets $∣ ψ ⟩$ , vectors in a Hilbert space $H$ .
Linear functionals on $H$ are bras $⟨ ϕ ∣$ , vectors in the dual space $H^{*}$ .
The Riesz Representation Theorem provides the crucial link: for every ket, there is a unique corresponding bra, and their combined action $⟨ ϕ ∣ ψ ⟩$ is the inner product.
This notation elegantly handles operator actions $⟨ ϕ ∣ A ∣ ψ ⟩$ and adjoints $(A ∣ ψ ⟩)^{†} = ⟨ ψ ∣ A^{†}$ .
The completeness relation $I = \sum_{n} ∣ n ⟩ ⟨ n ∣$ is a powerful tool for changing bases and performing calculations.
Wave functions, like $ψ (x) = ⟨ x ∣ ψ ⟩$ , are simply the components of an abstract state vector in a specific basis.

Having established the notation for states and operators, our next step is to examine the properties of the operators themselves. In the next lesson, we will classify key types of linear operators—Hermitian, unitary, and projection operators—and derive the properties that make them central to the postulates of quantum mechanics.

Recommended resources