Physlib

Physlib.QuantumMechanics.Operators.Multiplication

Multiplication operators on `SpaceDHilbertSpace`

i. Overview

In this module we introduce unbounded operators defined by multiplication by a function `f : Space d → ℂ`. The domain is defined to be as large as possible, namely a vector `ψ ∈ SpaceDHilbertSpace d μ` is in the domain iff `f • ψ ∈ SpaceDHilbertSpace d μ`.

ii. Key results

- `mulOperator f` : Given a function `f : Space d → ℂ`, the operator defined by `ψ ↦ f • ψ` (with maximal domain) with notation `𝓜 f`. - `mulOperator_adjoint_eq_conj` : For a.e. strongly measurable `f`, `(𝓜 f)† = 𝓜 (conj ∘ f)` - `mulOperator_isUnbounded` : For a.e. strongly measurable `f`, `𝓜 f` is an unbounded operator. - `mulOperator_compRestricted_le` : The composition `𝓜 f ∘ᵣ 𝓜 g` is contained in `𝓜 (f • g)`. - `mulOperator_compRestricted_eq` : The composition `𝓜 f ∘ᵣ 𝓜 g` is equal to `𝓜 (f • g)` when `(𝓜 g).domain = ⊤`.

iii. Table of contents

- A. Definition - B. Domain - C. Adjoint - C.1. Self-adjoint - D. Closable & unbounded - E. Composition

iv. References

See examples 1.3 and 3.8 in - [Konrad Schmüdgen, *Unbounded Self-Adjoint Operators on Hilbert Space*][Schmudgen2012]

A. Definition

B. Domain

C. Adjoint

C.1. Self-adjoint

D. Closable & unbounded

E. Composition

F. Spectrum

17 declarations

theorem

Essential boundedness of ff implies dom(Mf)=\text{dom}(\mathcal{M}_f) = \top

Let μ\mu be a measure on Space d\text{Space } d. For any μ\mu-a.e. strongly measurable function f:Space dCf : \text{Space } d \to \mathbb{C}, if there exists a constant cRc \in \mathbb{R} such that f(x)c\|f(x)\| \le c for μ\mu-almost every xSpace dx \in \text{Space } d, then the domain of the multiplication operator Mf\mathcal{M}_f on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}) is the entire space (denoted by \top).

theorem

gf    dom(Mf)dom(Mg)\|g\| \leq \|f\| \implies \text{dom}(\mathcal{M}_f) \subseteq \text{dom}(\mathcal{M}_g)

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d. Let f,g:Space dCf, g : \text{Space } d \to \mathbb{C} be complex-valued functions, and suppose gg is μ\mu-almost everywhere strongly measurable. If g(x)f(x)\|g(x)\| \leq \|f(x)\| for μ\mu-almost every xx, then the domain of the multiplication operator Mf\mathcal{M}_f is a subset of the domain of the multiplication operator Mg\mathcal{M}_g, that is, dom(Mf)dom(Mg)\text{dom}(\mathcal{M}_f) \subseteq \text{dom}(\mathcal{M}_g).

theorem

dom(Mf)=dom(Mg)\text{dom}(\mathcal{M}_f) = \text{dom}(\mathcal{M}_g) for functions with equal norms almost everywhere

Let μ\mu be a measure on Space d\text{Space } d. For any two μ\mu-almost everywhere strongly measurable functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C} such that their norms are equal μ\mu-almost everywhere (i.e., f(x)=g(x)\|f(x)\| = \|g(x)\| for μ\mu-almost every xx), the domain of the multiplication operator Mf\mathcal{M}_f is equal to the domain of the multiplication operator Mg\mathcal{M}_g on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}).

theorem

The multiplication operator Mf\mathcal{M}_f is closed

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d that is finite on compact sets. If f:Space dCf : \text{Space } d \to \mathbb{C} is a function that is μ\mu-almost everywhere strongly measurable, then the multiplication operator Mf\mathcal{M}_f acting on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}) is a closed operator.

theorem

The multiplication operator of the zero function is the zero operator (M0=0\mathcal{M}_0 = 0)

For any measure μ\mu on Space d\text{Space } d, the multiplication operator M0\mathcal{M}_0 associated with the constant zero function f(x)=0f(x) = 0 on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu) is equal to the zero operator. The domain of this operator is the entire Hilbert space.

theorem

f=gf = g a.e.     Mf=Mg\implies \mathcal{M}_f = \mathcal{M}_g

Let μ\mu be a measure on Space d\text{Space } d. For any two functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C}, if f=gf = g almost everywhere with respect to μ\mu, then the corresponding multiplication operators Mf\mathcal{M}_f and Mg\mathcal{M}_g on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}) are equal.

theorem

cMfMcfc \cdot \mathcal{M}_f \le \mathcal{M}_{cf}

Let μ\mu be a measure on Space d\text{Space } d, f:Space dCf : \text{Space } d \to \mathbb{C} be a complex-valued function, and cCc \in \mathbb{C} be a scalar. Let Mf\mathcal{M}_f denote the multiplication operator on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu). The operator cMfc \cdot \mathcal{M}_f is a restriction of the multiplication operator Mcf\mathcal{M}_{cf}, i.e., cMfMcfc \cdot \mathcal{M}_f \le \mathcal{M}_{cf}. Equality holds when c0c \neq 0, but for c=0c = 0, the domain of 0Mf0 \cdot \mathcal{M}_f (which is the domain of Mf\mathcal{M}_f) may be smaller than the domain of M0\mathcal{M}_0 (which is the entire space).

theorem

Mcf=cMf\mathcal{M}_{cf} = c\mathcal{M}_f for c0c \neq 0

Let μ\mu be a measure on Space d\text{Space } d, f:Space dCf : \text{Space } d \to \mathbb{C} be a function, and cCc \in \mathbb{C} be a non-zero complex scalar (c0c \neq 0). Then the multiplication operator Mcf\mathcal{M}_{cf} (defined by the function xcf(x)x \mapsto c \cdot f(x)) is equal to the scalar multiplication of the operator Mf\mathcal{M}_f by cc, satisfying Mcf=cMf\mathcal{M}_{cf} = c \mathcal{M}_f.

theorem

Mf=Mf\mathcal{M}_{-f} = -\mathcal{M}_f

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d. For any complex-valued function f:Space dCf : \text{Space } d \to \mathbb{C}, let Mf\mathcal{M}_f denote the multiplication operator on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}). Then, negation commutes with the multiplication operator, such that the multiplication operator by f-f is equal to the negation of the multiplication operator by ff: Mf=Mf\mathcal{M}_{-f} = -\mathcal{M}_f

theorem

Mf+g\mathcal{M}_{f+g} extends Mf+Mg\mathcal{M}_f + \mathcal{M}_g

For a measure μ\mu on Space d\text{Space } d and functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C}, the multiplication operator associated with the sum f+gf + g, denoted Mf+g\mathcal{M}_{f+g}, is an extension of the sum of the individual multiplication operators Mf+Mg\mathcal{M}_f + \mathcal{M}_g on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu). This means that the domain of the sum dom(Mf+Mg)\text{dom}(\mathcal{M}_f + \mathcal{M}_g), which consists of vectors ψ\psi such that both fψf \cdot \psi and gψg \cdot \psi are square-integrable, is contained within the domain dom(Mf+g)\text{dom}(\mathcal{M}_{f+g}), which only requires the weaker condition that (f+g)ψ(f+g) \cdot \psi be square-integrable. On the shared domain, the operators agree: (Mf+Mg)ψ=Mf+gψ(\mathcal{M}_f + \mathcal{M}_g)\psi = \mathcal{M}_{f+g}\psi.

theorem

Mf+g=Mf+Mg\mathcal{M}_{f+g} = \mathcal{M}_f + \mathcal{M}_g if dom(Mg)=\text{dom}(\mathcal{M}_g) = \top

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d. Let f,g:Space dCf, g : \text{Space } d \to \mathbb{C} be complex-valued functions, and let Mf\mathcal{M}_f and Mg\mathcal{M}_g be the corresponding multiplication operators on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu). If the domain of Mg\mathcal{M}_g is the entire Hilbert space (i.e., dom(Mg)=\text{dom}(\mathcal{M}_g) = \top), then the multiplication operator of the sum of the functions is equal to the sum of the individual multiplication operators: Mf+g=Mf+Mg\mathcal{M}_{f+g} = \mathcal{M}_f + \mathcal{M}_g

theorem

Mf+g=Mf+Mg\mathcal{M}_{f+g} = \mathcal{M}_f + \mathcal{M}_g for functions pointwise bounded by their sum

Consider the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}) with measure μ\mu. Let f,g:Space dCf, g: \text{Space } d \to \mathbb{C} be μ\mu-almost everywhere strongly measurable functions. Suppose there exist constants c,cRc, c' \in \mathbb{R} such that the inequalities f(x)cf(x)+g(x)\|f(x)\| \le c \|f(x) + g(x)\| and g(x)cf(x)+g(x)\|g(x)\| \le c' \|f(x) + g(x)\| hold for μ\mu-almost every xSpace dx \in \text{Space } d. Then the multiplication operator associated with the sum of the functions is equal to the sum of the individual multiplication operators, i.e., Mf+g=Mf+Mg\mathcal{M}_{f+g} = \mathcal{M}_f + \mathcal{M}_g.

theorem

Mfg\mathcal{M}_{f-g} extends MfMg\mathcal{M}_f - \mathcal{M}_g

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d. For any complex-valued functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C}, the multiplication operator Mfg\mathcal{M}_{f-g} is an extension of the difference of operators MfMg\mathcal{M}_f - \mathcal{M}_g. That is, dom(MfMg)dom(Mfg)\text{dom}(\mathcal{M}_f - \mathcal{M}_g) \subseteq \text{dom}(\mathcal{M}_{f-g}) and for all ψdom(MfMg)\psi \in \text{dom}(\mathcal{M}_f - \mathcal{M}_g), (MfMg)ψ=Mfgψ(\mathcal{M}_f - \mathcal{M}_g)\psi = \mathcal{M}_{f-g}\psi. The domain of the difference MfMg\mathcal{M}_f - \mathcal{M}_g is the intersection of the individual domains, dom(Mf)dom(Mg)\text{dom}(\mathcal{M}_f) \cap \text{dom}(\mathcal{M}_g), requiring both fψf \cdot \psi and gψg \cdot \psi to be in L2(Space d,μ)L^2(\text{Space } d, \mu). In contrast, the domain of Mfg\mathcal{M}_{f-g} only requires the difference (fg)ψ(f - g) \cdot \psi to be in L2(Space d,μ)L^2(\text{Space } d, \mu), which is a weaker condition.

theorem

Mfg=MfMg\mathcal{M}_{f-g} = \mathcal{M}_f - \mathcal{M}_g if dom(Mg)=\text{dom}(\mathcal{M}_g) = \top

Let μ\mu be a measure on Space d\text{Space } d and let L2(Space d,C)L^2(\text{Space } d, \mathbb{C}) be the Hilbert space of square-integrable complex-valued functions. For any two functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C}, if the domain of the multiplication operator Mg\mathcal{M}_g is the entire Hilbert space (i.e., dom(Mg)=L2(Space d,C)\text{dom}(\mathcal{M}_g) = L^2(\text{Space } d, \mathbb{C})), then the multiplication operator associated with the difference fgf - g is equal to the difference of the individual multiplication operators: Mfg=MfMg\mathcal{M}_{f-g} = \mathcal{M}_f - \mathcal{M}_g

theorem

Mfg=MfMg\mathcal{M}_{f-g} = \mathcal{M}_f - \mathcal{M}_g when ff and gg are pointwise bounded by their difference

Let μ\mu be a measure on the space Space d\text{Space } d. Let f,g:Space dCf, g: \text{Space } d \to \mathbb{C} be almost everywhere strongly measurable functions. Suppose there exist constants c,cRc, c' \in \mathbb{R} such that for μ\mu-almost every xx, the following conditions hold: f(x)cf(x)g(x)andg(x)cf(x)g(x)\|f(x)\| \leq c \|f(x) - g(x)\| \quad \text{and} \quad \|g(x)\| \leq c' \|f(x) - g(x)\| Then the multiplication operator Mfg\mathcal{M}_{f-g} is equal to the difference of the multiplication operators Mf\mathcal{M}_f and Mg\mathcal{M}_g on the Hilbert space L2(Space d,C)L^2(\text{Space } d, \mathbb{C}). This equality implies that the domain of Mfg\mathcal{M}_{f-g} coincides with the intersection of the domains of Mf\mathcal{M}_f and Mg\mathcal{M}_g.

theorem

Mfg\mathcal{M}_{f \cdot g} Extends the Composition MfrMg\mathcal{M}_f \circ_r \mathcal{M}_g

Let dd be a natural number and μ\mu be a measure on Space d\text{Space } d. For any complex-valued functions f,g:Space dCf, g : \text{Space } d \to \mathbb{C}, let Mf\mathcal{M}_f and Mg\mathcal{M}_g be the corresponding multiplication operators on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu). Then the multiplication operator associated with the pointwise product fgf \cdot g, denoted Mfg\mathcal{M}_{f \cdot g}, is an extension of the restricted composition of the individual operators MfrMg\mathcal{M}_f \circ_r \mathcal{M}_g. Specifically, the domain of the composition satisfies dom(MfrMg)dom(Mfg)\text{dom}(\mathcal{M}_f \circ_r \mathcal{M}_g) \subseteq \text{dom}(\mathcal{M}_{f \cdot g}), and for any ψ\psi in the domain of the composition, (MfrMg)ψ=Mfgψ(\mathcal{M}_f \circ_r \mathcal{M}_g)\psi = \mathcal{M}_{f \cdot g}\psi.

theorem

MfrMg=Mfg\mathcal{M}_f \circ_r \mathcal{M}_g = \mathcal{M}_{f \cdot g} when dom(Mg)=\text{dom}(\mathcal{M}_g) = \top

Let μ\mu be a measure on Space d\text{Space } d, and let f,g:Space dCf, g: \text{Space } d \to \mathbb{C} be functions. Let Mf\mathcal{M}_f and Mg\mathcal{M}_g be the multiplication operators on the Hilbert space L2(Space d,μ)L^2(\text{Space } d, \mu) associated with ff and gg respectively. If the domain of Mg\mathcal{M}_g is the entire Hilbert space (i.e., dom(Mg)=\text{dom}(\mathcal{M}_g) = \top), then the restricted composition MfrMg\mathcal{M}_f \circ_r \mathcal{M}_g is equal to the multiplication operator Mfg\mathcal{M}_{f \cdot g} of the pointwise product fgf \cdot g: MfrMg=Mfg\mathcal{M}_f \circ_r \mathcal{M}_g = \mathcal{M}_{f \cdot g} This theorem provides a sufficient condition (the totality of the domain of the second operator) to turn the operator inclusion MfrMgMfg\mathcal{M}_f \circ_r \mathcal{M}_g \subseteq \mathcal{M}_{f \cdot g} into an equality.