betaFeedback

WorksPhysics and Astronomy

Nodal Knotted Finite-Energy Spatiotemporally Localized Scalar Wave Solutions

Abstract

Significant amount of information is available in wave optics about monochromatic paraxial beams characterized by nodal knots, the latter being threads of complete darkness (lines of zero intensity) tied into knots. Also, well -known are finite-energy spatiotemporally localized vector-valued wave solutions characterized by knotted to-pologies. Unfortunately, except for solutions to the Schrödinger-like equation describing anomalous dispersion, no scalar-valued spatiotemporally localized waves with analogous knotted topological structures are known to date. It is the goal in this article to fill in this gap by deriving finite-energy spatiotemporally localized scalar-valued wave solutions characterized specifically by Hopf and trefoil nodal knot topologies. This will extend to spatiotemporally localized pulsed fields the diverse important applications of mono-chromatic light beams, having nontrivial topological structures, in trapping and guiding particles, image processing and optical communications; also, in applications to biology and medicine.

Introduction

Luminal, or focus wave modes (FWM), superluminal, or X waves (XWs), and subluminal spatiotemporally localized solutions to various hyperbolic equations governing acoustic, electromagnetic and quantum wave phenomena have been studied intensively during the past thirty years (see [1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15] [8] for pertinent literature). Further details can be found in two edited monographs on the subject [16, 17] and in the recent thorough review article [18]. In general, both linear and nonlinear localized wave (LW) pulses exhibit distinct advantages in comparison to conventional quasi-monochromatic and pulsed signals. Their spatiotemporal confinement and extended field depths render them especially useful in diverse physical applications. Experimental demonstrations have been performed in the acoustic and optical regimes [5, 7, 10, 20, 21, 22, 23]. Work, however, has been carried out at microwave frequencies recently [24, 25, 26].

A class of finite-energy space-time localized electromagnetic waves is known to have interesting topological properties, specifically with respect to linkages of electric and magnetic field lines. Among them are the basic Hopfion, as well as higher-order Hopf-Ranada solutions [27, 28, 29, 30, 31].

An outline of our knowledge of finite-energy spatiotemporally localized knotted vector-valued wave solutions is given in § 2. Unfortunately, although huge amount of information is known in wave optics about monochromatic paraxial beams characterized by nodal knots – the latter being threads of complete darkness (lines of zero intensity) tied into knots [32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44], no scalar-valued spatiotemporally localized waves with analogous knotted topological structures are known to date. It is the goal in this article to fill in this gap by deriving finite-energy spatiotemporally localized scalar-valued wave solutions characterized especially by Hopf and trefoil nodal knot topologies. This work is presented and discussed in § 3-§ 8. Concluding remarks are made in § 9.

Finite-energy spatiotemporally localized knotted vector-valued wave solutions

Bateman Conjugate Functions

Possibly motivated by Whittaker’s two-scalar potential theory [45], Bateman [46] introduced two complex-valued scalar functions α(r⃗,t)\alpha(\vec{r},t) and β(r⃗,t)\beta(\vec{r},t) that can be used to derive electromagnetic fields in vacuum characterized by interesting topological properties. Known as the Bateman conjugate functions, an arbitrary functional ϕ[α(r⃗,t),β(r⃗,t)]\phi[\alpha(\vec{r},t),\beta(\vec{r},t)] of them obeys the nonlinear characteristic (eikonal) equation

(∂ϕ∂x)2+(∂ϕ∂y)2+(∂ϕ∂z)2−1c2(∂ϕ∂t)2=0.(1)\left(\frac{\partial\phi}{\partial x}\right)^2+\left(\frac{\partial\phi}{\partial y}\right)^2+\left(\frac{\partial\phi}{\partial z}\right)^2-\frac{1}{c^2}\left(\frac{\partial\phi}{\partial t}\right)^2=0. \tag*{(1)}

Furthermore, two arbitrary functionals ϕ1,2[α(r⃗,t),β(r⃗,t)]\phi_{1,2}[\alpha(\vec{r},t),\beta(\vec{r},t)] obey the Bateman constraint relation

∇ϕ1×∇ϕ2=i1c(∂ϕ1∂t∇ϕ2−∂ϕ2∂t∇ϕ1).(2)\nabla\phi_1\times\nabla\phi_2=i\frac{1}{c}\left(\frac{\partial\phi_1}{\partial t}\nabla\phi_2-\frac{\partial\phi_2}{\partial t}\nabla\phi_1\right). \tag*{(2)}

Riemann-Silberstein Complex Vector

Real-valued electromagnetic fields E⃗(r⃗,t)\vec{E}(\vec{r},t) and B⃗(r⃗,t)\vec{B}(\vec{r},t) governed by Maxwell equations in vacuum are conveniently expressed in terms of a complex-valued vector

F⃗=ε02(E⃗+icB⃗).(3)\vec{F}=\sqrt{\frac{\varepsilon_0}{2}}\left(\vec{E}+ic\vec{B}\right). \tag*{(3)}

The Riemann-Silberstein vector [47, 48] obeys the equations

∇×F⃗=i1c∂∂tF⃗,∇⋅F⃗=0,(4)\nabla\times\vec{F}=i\frac{1}{c}\frac{\partial}{\partial t}\vec{F},\qquad\nabla\cdot\vec{F}=0, \tag*{(4)}

which are exactly equivalent to the original Maxwell equations for the real fields. Several important physical quantities associated with the real fields can be expressed conveniently in terms of F⃗(r⃗,t)\vec{F}(\vec{r},t). Specifically, the Poynting vector, the electromagnetic field energy density, the electromagnetic momentum density and the electromagnetic angular momentum density can be written as

P⃗=E⃗×H⃗=−icF⃗∗×F⃗,wem=12(ε0E⃗⋅E⃗+μ0H⃗⋅H⃗)=F⃗∗⋅F⃗,M⃗=1c2E⃗×H⃗=−icF⃗∗×F⃗,J⃗=r⃗×M⃗=−icr⃗×(F⃗∗×F⃗).(5)\begin{aligned} \vec{P} &= \vec{E} \times\vec{H} = -ic\vec{F}^{*} \times\vec{F},\\ w_{\mathrm{em}} &= \frac{1}{2}\left(\varepsilon_{0}\vec{E}\cdot\vec{E}+\mu_{0}\vec{H}\cdot\vec{H}\right)=\vec{F}^{*}\cdot\vec{F},\\ \vec{M} &= \frac{1}{c^{2}}\vec{E}\times\vec{H}=-\frac{i}{c}\vec{F}^{*}\times\vec{F},\\ \vec{J} &= \vec{r}\times\vec{M}=-\frac{i}{c}\vec{r}\times\left(\vec{F}^{*}\times\vec{F}\right). \tag*{(5)} \end{aligned}

For the following discussion, the Riemann-Silberstein vector is related to the Bateman conjugate functions in terms of a complex-valued vector potential as follows:

F⃗=ε02c∇×A⃗=12ε0cαp−1βq−1∇α×∇β;A⃗pq=(ε02)1/2c(pq)−1αp∇βq=A⃗e+iA⃗m; p,q=1,2,3,…(6)\begin{aligned} \vec{F} &= \sqrt{\frac{\varepsilon_{0}}{2}}c\nabla\times\vec{A}=\frac{1}{2}\sqrt{\varepsilon_{0}}c\alpha^{p-1}\beta^{q-1}\nabla\alpha\times\nabla\beta;\\ \vec{A}_{pq} &= \left(\frac{\varepsilon_{0}}{2}\right)^{1/2}c\left(pq\right)^{-1}\alpha^{p}\nabla\beta^{q}=\vec{A}_{e}+i\vec{A}_{m};\ p,q=1,2,3,\ldots \tag*{(6)} \end{aligned}

By virtue of its construction, the Riemann-Silberstein vector in Equation (6) obeys Equations (4), and from its definition in Equation (3) it follows that E⃗=c∇×A⃗e, B⃗=∇×A⃗m\vec{E}=c\nabla\times\vec{A}_{e},\ \vec{B}=\nabla\times\vec{A}_{m}.

Another consequence of the specific construction of the Riemann-Silberstein vector according to Equation (6) is the nullity property: 76

F⃗(r⃗,t)⋅F⃗(r⃗,t)=iε0I2+ε02I1=0;I1=∣E⃗(r⃗,t)∣2−c2∣B⃗(r⃗,t)∣2=0,I2=E⃗(r⃗,t)⋅B⃗(r⃗,t)=0.(7)\begin{aligned} \vec{F}(\vec{r},t)\cdot\vec{F}(\vec{r},t) &= i\varepsilon_{0}I_{2}+\frac{\varepsilon_{0}}{2}I_{1}=0;\\ I_{1} &= \left|\vec{E}(\vec{r},t)\right|^{2}-c^{2}\left|\vec{B}(\vec{r},t)\right|^{2}=0,\\ I_{2} &= \vec{E}(\vec{r},t)\cdot\vec{B}(\vec{r},t)=0. \tag*{(7)} \end{aligned}

It requires, specifically, that the two relativistic invariants I1,2I_{1,2} vanish identically. The corresponding real fields E⃗(r⃗,t)\vec{E}(\vec{r},t) and B⃗(r⃗,t)\vec{B}(\vec{r},t) are called null electromagnetic fields. Such fields are transverse with respect to the local flow of energy, which is equipartitioned between the electric and magnetic fields, and the modulus of the local energy transport velocity, defined as the ratio of the Poynting vector and the electromagnetic energy density, equals the speed of light in vacuo, although this velocity can vary both spatially and temporally. The simplest electromagnetic null field is a plane wave. More complex null fields are characterized by “helical” or vortex structures on planes transverse to the direction of propagation, and, in general, are relatively simple so that explicit calculations can be made of the total energy and the total angular momentum they carry.

Specific Bateman Conjugate Functions

Irvine and Bouwmeester [27] explored the physical properties of an exceptional luminal localized solution to the homogeneous Maxwell equations in vacuo that exhibits linked and knotted field line structures. The Irvine-Bouwmeester solution was constructed originally by Ranãda [28] using Hopf maps (fibrations) and belongs to a more general class of solutions to linear and nonlinear equations known as Hopfions. Besieris and Shaarawi [29] have linked the Irvine-Bouwmeester expression to a known null luminal electromagnetic solution.

Consider the luminal Bateman conjugate functions

α=x2+y2+(z+ia)2−c2t2x2+y2+z2−(ct−ia)2,β=2(x+iy)x2+y2+z2−(ct−ia)2(8)\alpha= \frac{x^2+y^2+(z+ia)^2-c^2t^2}{x^2+y^2+z^2-(ct-ia)^2}, \qquad \beta= \frac{2(x+iy)}{x^2+y^2+z^2-(ct-ia)^2} \tag*{(8)}

where aa is a constant parameter. It should be noted that they qualify as Euler potentials because the vector ∇α×∇β\nabla\alpha\times\nabla\beta has zero divergence. Any functionals of these two functions obey the nonlinear characteristic equation and the Bateman constraint. With c=1,a=1c=1,a=1 (See Ref. [30]; also [31]), they have the property αα∗+ββ∗=1\alpha\alpha^*+\beta\beta^*=1. The surface ∣α∣=const.\lvert\alpha\rvert=const. represents a torus for any value of tt. The surface ∣α2−β2∣=const.\lvert\alpha^2-\beta^2\rvert=const. represents two linked tori, ∣α2−β3∣=const.\lvert\alpha^2-\beta^3\rvert=const. represents a trefoil knot, and ∣α2−β5∣=const.\lvert\alpha^2-\beta^5\rvert=const. depicts a cinquefoil, for any value of time (see Figure 1).

Plots of $\lvert\alpha^p-\beta^q\rvert=const.$ for $p=q=2$, $p=2,q=3$, and $p=2,q=5$

Figure 1. Plots of ∣αp−βq∣=const.\lvert\alpha^p-\beta^q\rvert=const. for p=q=2p=q=2 (torus link), p=2,q=3p=2,q=3 (trefoil) and p=2,q=5p=2,q=5 (cinquefoil) at t=0t=0.

When the expressions for the Bateman conjugate functions given in Equation (8) are used in Equation (6), the case p=1p=1 and q=1q=1, corresponding to simplest Hopf-Rañada solution, or the basic Hopfion, yields linked electromagnetic fields. Any two electric field lines are linked, any two magnetic field lines are linked, and any electric field line is linked to any magnetic field line for any instant of time. Electric and magnetic field line linkages for p=1,q=1p=1,q=1 and p=2,q=3p=2,q=3 are shown in Figure 2. The corresponding fields have finite energy.

Plots of field line linkages for p = q = 1 and p = 2, q = 3

Figure 2. Plots of E⃗\vec{E} (red) and B⃗\vec{B} (blue) field line linkages for p=q=1p=q=1 (left) and p=2,q=3p=2,q=3 (right) at t=0t=0.

Hopf nodal knot for a spatiotemporally localized splash mode

Time-independent Bateman conjugate functions

At t=0t=0, the expressions α\alpha and β\beta in Equation (8), with c=1,a=1c=1,a=1, are denoted by uu and vv. The latter were used by Brauer in 1928 [49] to construct explicit realizations of complex-valued scalar knotted fields for each (p,q)(p,q) torus knot or link ∣up−vq∣=const|u^p-v^q|=const. The surface ∣u2−v2∣=const.|u^2-v^2|=const. represents two linked tori and ∣u2−v3∣=const.|u^2-v^3|=const. represents a trefoil linkage.

Hopf nodal knot for the paraxial equation

Consider the expression

u2−v2=N(x,y,z)D(x,y,z)=−4(x+iy)2+(x2+y2+(i+z)2)2(1+x2+y2+z2)2.(9)u^2-v^2=\frac{N(x,y,z)}{D(x,y,z)}=\frac{-4(x+iy)^2+\left(x^2+y^2+(i+z)^2\right)^2}{\left(1+x^2+y^2+z^2\right)^2}. \tag*{(9)}

The numerator contains the information on the Hopf nodal linkage as shown in the figure below.

Next, a paraxial nodal Hopf knot is formed as follows. First the Fourier transform

ψ^(kx,ky)=F{N(x,y,0)exp⁡[−κ(x2+y2)]}(10)\hat{\psi}(k_x,k_y)=\mathcal{F}\left\{N(x,y,0)\exp\left[-\kappa\left(x^2+y^2\right)\right]\right\} \tag*{(10)}

is computed (see [38, 39] for an alternative approach). Then, a specific paraxial Hopf nodal knot is obtained by means of the Fourier inversion

ψ(x,y,z)=F−1{ψ^(kx,ky)exp⁡[−iz2(kx2+ky2)]}=1(1+i2κz)5exp⁡[−κx2+y21+i2κz]×{x4+y4+8ixy(i−2zκ)2+y2(2+8iz(i+κ)−8z2κ(2+κ))+(i−2zκ)2(−1−4iz(−1+κ)+4z2(2+(−2+κ)κ))+2x2(y2+(−i+2zκ)(−3i+z(−4+6κ)))}.(11)\begin{aligned} \psi(x,y,z)&=\mathcal{F}^{-1}\left\{\hat{\psi}(k_x,k_y)\exp\left[-\frac{i z}{2}\left(k_x^2+k_y^2\right)\right]\right\}\\ &=\frac{1}{(1+i2\kappa z)^5}\exp\left[-\kappa\frac{x^2+y^2}{1+i2\kappa z}\right]\\ &\quad\times\left\{\begin{aligned} &x^4+y^4+8ixy(i-2z\kappa)^2+y^2\left(2+8iz(i+\kappa)-8z^2\kappa(2+\kappa)\right)\\ &\quad+(i-2z\kappa)^2\left(-1-4iz(-1+\kappa)+4z^2\left(2+(-2+\kappa)\kappa\right)\right)\\ &\quad+2x^2\left(y^2+(-i+2z\kappa)(-3i+z(-4+6\kappa))\right) \end{aligned}\right\}. \tag*{(11)} \end{aligned}

It obeys the paraxial equation

2i∂∂zψ(x,y,z)+∇t2ψ(x,y,z)=0.(12)2i\frac{\partial}{\partial z}\psi(x,y,z)+\nabla_t^2\psi(x,y,z)=0. \tag*{(12)}

The equality

N(x,y,z)=−4(x+iy)2+(x2+y2+(i+z)2)2(13)N(x,y,z)=-4(x+iy)^2+\left(x^2+y^2+(i+z)^2\right)^2 \tag*{(13)}

in Equation (9) is known as the Hopf Milnor polynomial. It does not satisfy the paraxial equation. However, the procedure followed above allows one to obtain an extended polynomial that satisfies the paraxial equation and, also, inherits the nodal topology of the Hopf Milnor polynomial. If κ\kappa is set equal to zero in Equation (11), one obtains

w(x,y,z)=1+x4−8ixy+y4+2x2(y2−i(−3i−4z))+y2(2+8iz)−4iz−8z2.(14)w(x,y,z)=1+x^{4}-8ixy+y^{4}+2x^{2}\left(y^{2}-i(-3i-4z)\right)+y^{2}(2+8iz)-4iz-8z^{2}. \tag*{(14)}

This expression, known as the Hopf paraxial polynomial, obeys the paraxial equation (12). (see [35], [38] and [39] for alternative derivations of such polynomials). Of course, for z=0z=0, it is equal to N(x,y,0)N(x,y,0). However, N(x,y,z)N(x,y,z) and w(x,y,z)w(x,y,z) differ significantly in their dependence on zz. The former is a polynomial of order four in zz, whereas the latter is of order two. Once the Hopf paraxial polynomial is known, a paraxial Hopf nodal knot -- different from the one in Equation (11) but having the same topology -- can be derived by a different approach based on a conformal symmetry of the Gaussian paraxial solution. Specifically,

ϕ(x,y,z)=1a1+izexp⁡(−12x2+y2a1+iz)w(a1xa1+iz,a1ya1+iz,a1za1+iz),(15)\phi(x,y,z)=\frac{1}{a_{1}+iz}\exp\left(-\frac{1}{2}\frac{x^{2}+y^{2}}{a_{1}+iz}\right)w\left(\frac{a_{1}x}{a_{1}+iz},\frac{a_{1}y}{a_{1}+iz},\frac{a_{1}z}{a_{1}+iz}\right), \tag*{(15)}

where a1a_{1} is a positive free parameter.

Hopf nodal knot for the focus wave solution of the scalar wave equation

In the Hopf Milnor polynomial in Equation (15), zz is replaced by ζ/2b\zeta/2b, where ζ=z−t\zeta=z-t and bb is a free positive parameter. Using the change w(x,y,z)→W(x,y,ζ)w(x,y,z)\to W(x,y,\zeta), the expression

f(x,y,ζ,η)=a1a1+iζexp⁡(ibη)exp⁡(−bx2+y2a1+iζ)×W(a1xa1+iζ,a1ya1+iζ,a1ζa1+iζ),(16)\begin{aligned} f(x,y,\zeta,\eta)={}&\frac{a_{1}}{a_{1}+i\zeta}\exp(ib\eta)\exp\left(-b\frac{x^{2}+y^{2}}{a_{1}+i\zeta}\right)\\ &\times W\left(\frac{a_{1}x}{a_{1}+i\zeta},\frac{a_{1}y}{a_{1}+i\zeta},\frac{a_{1}\zeta}{a_{1}+i\zeta}\right), \tag*{(16)} \end{aligned}

with η=z+t\eta=z+t, is a solution of the scalar wave equation, viz.,

(∇t2+4∂2∂ζ∂η)f(x,y,ζ,η)=0,(17)\left(\nabla_{t}^{2}+4\frac{\partial^{2}}{\partial\zeta\partial\eta}\right)f(x,y,\zeta,\eta)=0, \tag*{(17)}

that retains the Hopf nodal knot topological structure with respect to the variables xx, yy and ζ\zeta.

Hopf nodal knot for the fine-energy splash mode solution of the scalar wave equation

Although the solution in Equation (16) is spatiotemporally localized, it has infinite energy. A finite-energy solution can be derived by an appropriate superposition over the free parameter bb. A particularly simple exact solution can be derived by multiplying the right-hand side of Equation (16) by b2exp⁡(−a2b)b^{2}\exp(-a_{2}b), with a2>0a_{2}>0, and carrying out the integrations

I1=a1a1+iζ∫0∞db b2exp⁡[−b(a2−iη+x2+y2a1+iζ)](−2a12ζ2b2(a1+iζ));I2=a1a1+iζ∫0∞db b2exp⁡[−b(a2−iη+x2+y2a1+iζ)]×(−2ia1ζ+4ia1(a12(x2+y2)(a1+iζ)2)ζb(a1+iζ));I3=a1a1+iζ∫0∞db b2exp⁡[−b(a2−iη+x2+y2a1+iζ)]×(1−2a12(x2+y2)(a1+iζ)2+(a12(x2+y2)(a1+iζ)2)2−4(a1(x+iy)a1+iζ)2).(18)\begin{aligned} I_1 &= \frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^2\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\left(-\frac{2a_1^2\zeta^2}{b^2(a_1+i\zeta)}\right);\\ I_2 &= \frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^2\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\\ &\quad\times\left(\frac{-2ia_1\zeta+4ia_1\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)\zeta}{b(a_1+i\zeta)}\right);\\ I_3 &= \frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^2\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\\ &\quad\times\left(1-2\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}+\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)^2-4\left(\frac{a_1(x+iy)}{a_1+i\zeta}\right)^2\right). \tag*{(18)} \end{aligned}

The three integrals can be carried out explicitly using a table of Laplace transforms, resulting in the Hopf nodal splash mode solution

IH(x,y,ζ,η)=I1+I2+I3.(19)I_H(x,y,\zeta,\eta)=I_1+I_2+I_3. \tag*{(19)}

Setting η=ζ+2t\eta=\zeta+2t, the intersection of the real and imaginary parts of IH(x,y,ζ,t)I_H(x,y,\zeta,t) is shown below relative to the pulse center (ζ=0\zeta=0) for two values of tt.

Hopf nodal linkage at t=0 and t=2

Figure 4. Hopf nodal linkage formed from the intersection of the real and imaginary parts of the scalar splash mode Hopf nodal knot solution for t=0t=0 and t=2t=2, with parameter values a1=1a_1=1 and a2=10a_2=10.

Due to its bidirectionality, the spatiotemporal localization, as well as the Hopf nodal knot topology, of the solution IH(a1,a2,x,y,ζ;t)I_H(a_1,a_2,x,y,\zeta;t) cannot be maintained for relatively large values of tt. However, for a2≫a1a_2 \gg a_1, the pulse is essentially invariant. The energy density of the scalar splash mode Hopf nodal knot solution is given by

UH(x,y,ζ,η)=12(∂∂tIH∂∂tIH∗+∇IH⋅∇IH∗).(20)U_H(x,y,\zeta,\eta)=\frac{1}{2}\left(\frac{\partial}{\partial t}I_H\frac{\partial}{\partial t}I_H^*+\nabla I_H\cdot\nabla I_H^*\right). \tag*{(20)}

Resorting to η=ζ+2t\eta=\zeta+2t, plots of UH(x,y,ζ,t)U_H(x,y,\zeta,t) are shown in the figure below for three values of time.

Energy density of the scalar splash mode Hopf nodal knot solution for $t=-10,0,10$

Figure 5. Energy density of the scalar splash mode Hopf nodal knot solution for t=−10,0,10t=-10,0,10, with parameter values a1=1a_1=1 and a2=10a_2=10.

Hopf nodal knot for a finite-energy splash mode solution of the scalar wave equation based on a superposition of Gauss-Laguerre focus wave modes

The numerator in (9) at z=0z=0 is expressed in polar form, viz.,

N(ρ,ϕ,0)=1−2ρ2−4ρ2e2iϕ+ρ4,(21)N(\rho,\phi,0)=1-2\rho^2-4\rho^2e^{2i\phi}+\rho^4, \tag*{(21)}

and multiplied by a Gaussian function as follows:

FH(ρ,ϕ;w)=N(ρ,ϕ,0)exp⁡[−ρ22w],(22)F_H(\rho,\phi;\mathsf{w})=N(\rho,\phi,0)\exp\left[-\frac{\rho^2}{2\mathsf{w}}\right], \tag*{(22)}

where w\mathsf{w} is a constant. Using a standard procedure (see, e. g., [35, 36, 43]), FH(ρ,ϕ)F_H(\rho,\phi) is used in a superposition of Gauss-Laguerre modes as

fH(ρ,ϕ,z)=∑m,ncmnLGm,n(ρ,ϕ,z;w0);cmn=∬d2r⊥LG−m,n(ρ,ϕ,0;w)LG0,0(ρ,ϕ,0;w)FH(ρ,ϕ;w),(23)\begin{aligned} f_H(\rho,\phi,z)&=\sum_{m,n}c_{mn}LG_{m,n}(\rho,\phi,z;w_0);\\ c_{mn}&=\iint d^2r_\perp LG_{-m,n}(\rho,\phi,0;w)LG_{0,0}(\rho,\phi,0;w)F_H(\rho,\phi;w), \tag*{(23)} \end{aligned}

with the Gauss-Laguerre modes defined by

LGm,n(ρ,ϕ,z;w0)=n!π(∣m∣+n)!ρ∣m∣eimϕw0∣m∣+1(1−iz/zR)n(1+iz/zR)n+∣m∣+1×exp⁡[−ρ22w02(1+iz/zR)]Ln∣m∣(ρ2w02(1+z2/zR2));zR=k0w02.(24)\begin{aligned} LG_{m,n}(\rho,\phi,z;w_0)&=\sqrt{\frac{n!}{\pi(|m|+n)!}}\frac{\rho^{|m|}e^{im\phi}}{w_0^{|m|+1}}\frac{(1-iz/z_R)^n}{(1+iz/z_R)^{n+|m|+1}}\\ &\quad\times\exp\left[-\frac{\rho^2}{2w_0^2(1+iz/z_R)}\right]L_n^{|m|}\left(\frac{\rho^2}{w_0^2(1+z^2/z_R^2)}\right); \quad z_R=k_0w_0^2. \tag*{(24)} \end{aligned}

The coefficients required in (23) are given as follows:

c00=1−2w2+2w4, c01=2(w2−2w4), c02=2w4, c20=−42w2.(25)c_{00}=1-2w^{2}+2w^{4},\ c_{01}=2\left(w^{2}-2w^{4}\right),\ c_{02}=2w^{4},\ c_{20}=-4\sqrt{2}w^{2}. \tag*{(25)}

With the changes k=1k=1, w0=a1/(2b)w_{0}=\sqrt{a_{1}/(2b)}, z→ζ/(2b)z\to\zeta/(2b), where a1a_{1} and bb are free parameters, the expression

gH(ρ,ϕ,ζ,η)=eibηfH(ρ,ϕ,ζ)g_{H}(\rho,\phi,\zeta,\eta)=e^{ib\eta}f_{H}(\rho,\phi,\zeta)
=2π{a1(a1+iζ)3(a12−2ia1(−1+2w2)ζ+(−1+4w2−8w4)ζ2)−4a13/2(a1+iζ)4bw2(a1+2a1e2iϕ+i(1+2e2iϕ−4w2)ζ)ρ2+4a15/2(a1+iζ)5w4ρ4}×exp⁡[ibη−bρ2a1+iζ].(26)\begin{aligned} ={}&\sqrt{\frac{2}{\pi}}\left\{ \frac{\sqrt{a_{1}}}{(a_{1}+i\zeta)^{3}}\left(a_{1}^{2}-2ia_{1}(-1+2w^{2})\zeta+(-1+4w^{2}-8w^{4})\zeta^{2}\right)\right.\\ &-\frac{4a_{1}^{3/2}}{(a_{1}+i\zeta)^{4}}bw^{2}\left(a_{1}+2a_{1}e^{2i\phi}+i(1+2e^{2i\phi}-4w^{2})\zeta\right)\rho^{2}\\ &\left.+\frac{4a_{1}^{5/2}}{(a_{1}+i\zeta)^{5}}w^{4}\rho^{4} \right\}\\ &\times\exp\left[ib\eta-b\frac{\rho^{2}}{a_{1}+i\zeta}\right]. \tag*{(26)} \end{aligned}

is an infinite-energy superposition of focus wave mode solutions to the scalar wave equation [see Equation (17)] characterized by a Hopf nodal link topology. A finite-energy spatiotemporally localized splash mode maintaining the same topology can be found by a superposition over the free parameter bb using the spectrum exp⁡(−a2b)\exp(-a_{2}b), with a2>0a_{2}>0. Specifically,

hH(ρ,ϕ,ζ,η)=∫0∞db e−a2bgH(ρ,ϕ,ζ,η)h_{H}(\rho,\phi,\zeta,\eta)=\int_{0}^{\infty}db\,e^{-a_{2}b}g_{H}(\rho,\phi,\zeta,\eta)
=2πa11(a1+iζ)3(a2−iη+ρ2a1+iζ)(a12−2ia1(1+2w2)ζ+(−1+4w2−8w4)ζ2)+2π4a13/21(a1+iζ)4(a2−iη+ρ2a1+iζ)2×w2(a1+2a1e2iϕ+i(1+2e2iϕ−4w2)ζ)ρ2+2π4a15/2w4ρ4(a1+iζ)5(a2−iη+ρ2a1+iζ)3.(27)\begin{aligned} ={}&\sqrt{\frac{2}{\pi}}\sqrt{a_{1}}\frac{1}{(a_{1}+i\zeta)^{3}\left(a_{2}-i\eta+\frac{\rho^{2}}{a_{1}+i\zeta}\right)} \left(a_{1}^{2}-2ia_{1}(1+2w^{2})\zeta+(-1+4w^{2}-8w^{4})\zeta^{2}\right)\\ &+\sqrt{\frac{2}{\pi}}4a_{1}^{3/2}\frac{1}{(a_{1}+i\zeta)^{4}\left(a_{2}-i\eta+\frac{\rho^{2}}{a_{1}+i\zeta}\right)^{2}}\\ &\quad\times w^{2}\left(a_{1}+2a_{1}e^{2i\phi}+i(1+2e^{2i\phi}-4w^{2})\zeta\right)\rho^{2}\\ &+\sqrt{\frac{2}{\pi}}4a_{1}^{5/2}\frac{w^{4}\rho^{4}}{(a_{1}+i\zeta)^{5}\left(a_{2}-i\eta+\frac{\rho^{2}}{a_{1}+i\zeta}\right)^{3}}. \tag*{(27)} \end{aligned}

Setting η=ζ+2t\eta=\zeta+2t, the intersection of the real and imaginary parts of hH(x,y,ζ,t)h_{H}(x,y,\zeta,t) is shown below for two values of tt.

Hopf nodal linkage formed from the intersection of the real and imaginary parts of the scalar splash mode Hopf nodal knot solution

Figure 6. Hopf nodal linkage formed from the intersection of the real and imaginary parts of the scalar splash mode Hopf nodal knot solution for t=0t=0 and t=1/2t=1/2, with parameter values a1=1a_1=1, a2=2a_2=2 and w=2w=2.

Trefoil nodal knot for a spatiotemporally localized splash mode

Consider the expression

u2−v3=N(x,y,z)D(x,y,z)=−8(x+iy)3+(1+x2+y2+z2)(x2+y2+(i+z)2)2(1+x2+y2+z2)3.(28)u^2-v^3=\frac{N(x,y,z)}{D(x,y,z)} =\frac{-8(x+\mathrm{i}y)^3+\left(1+x^2+y^2+z^2\right)\left(x^2+y^2+(\mathrm{i}+z)^2\right)^2}{\left(1+x^2+y^2+z^2\right)^3}. \tag*{(28)}

The numerator contains the information on the trefoil nodal linkage as shown in the figure below

Trefoil nodal linkage

Figure 7.

Image not extracted

Figure 7 (PDF p. 12). Trefoil nodal linkage formed from the intersection of the real and imaginary parts of the numerator N(x,y,z)N(x,y,z).

The paraxial trefoil polynomial can be derived by a procedure analogous to that followed in the previous section (see, also, [35]). It is given explicitly as follows:

w(x,y,z)=1−8x3+x6+24xy2+8iy3+y6+y4(−1+18iz)+x4(−1+3y2+18iz)−2iz+8z2−48iz3+x2(−1−24iy+3y4+y2(−2+36iz)−8iz−72z2)−y2(1+8iz+72z2).(29)\begin{aligned} w(x,y,z)={}&1-8x^{3}+x^{6}+24xy^{2}+8iy^{3}\\ &+y^{6}+y^{4}(-1+18iz)+x^{4}(-1+3y^{2}+18iz)\\ &-2iz+8z^{2}-48iz^{3}+x^{2}\left(-1-24iy+3y^{4}\right.\\ &\left.\qquad+y^{2}(-2+36iz)-8iz-72z^{2}\right)\\ &-y^{2}(1+8iz+72z^{2}). \tag*{(29)} \end{aligned}

This is converted to W(x,y,ζ)W(x,y,\zeta) and the expression f(x,y,ζ,η)f(x,y,\zeta,\eta) in Equation (16) satisfies the scalar wave equation for a normalized speed c=1c=1 in vacuum.

An exact finite-energy splash mode solution exhibiting a trefoil nodal knot topology can be derived by multiplying the right-hand side of Equation (16) by b3exp⁡(−a2b)b^{3}\exp(-a_{2}b), with a2>0a_{2}>0, and carrying out the integrations

I1=a1a1+iζ∫0∞db b3exp⁡[−b(a2−iη+x2+y2a1+iζ)](−6ia13ζ3b3(a1+iζ)3);I2=a1a1+iζ∫0∞db b3exp⁡[−b(a2−iη+x2+y2a1+iζ)]×(2a12ζ2−18a14(x2+y2(a1+iζ)2)ζ2b2(a1+iζ)2);I3=a1a1+iζ∫0∞db b3exp⁡[−b(a2−iη+x2+y2a1+iζ)]×(−ia1ζ+4ia1(a12(x2+y2)(a1+iζ)2)ζ−9ia1(a12(x2+y2)(a1+iζ)2)2ζb(a1+iζ));I4=a1a1+iζ∫0∞db b3exp⁡[−b(a2−iη+x2+y2a1+iζ)]×(11−(a12(x2+y2)(a1+iζ)2)2+(a12(x2+y2)(a1+iζ)2)3+8(a1(x+iy)a1+iζ)3−a12(x2+y2)(a1+iζ)2).(30)\begin{aligned} I_1&=\frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^3\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\left(-\frac{6ia_1^3\zeta^3}{b^3(a_1+i\zeta)^3}\right);\\ I_2&=\frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^3\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\\ &\quad\times\left(\frac{2a_1^2\zeta^2-18a_1^4\left(\frac{x^2+y^2}{(a_1+i\zeta)^2}\right)\zeta^2}{b^2(a_1+i\zeta)^2}\right);\\ I_3&=\frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^3\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\\ &\quad\times\left(\frac{-ia_1\zeta+4ia_1\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)\zeta-9ia_1\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)^2\zeta}{b(a_1+i\zeta)}\right);\\ I_4&=\frac{a_1}{a_1+i\zeta}\int_0^\infty db\,b^3\exp\left[-b\left(a_2-i\eta+\frac{x^2+y^2}{a_1+i\zeta}\right)\right]\\ &\quad\times\left(11-\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)^2+\left(\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right)^3\right.\\ &\qquad\left.+8\left(\frac{a_1(x+iy)}{a_1+i\zeta}\right)^3-\frac{a_1^2(x^2+y^2)}{(a_1+i\zeta)^2}\right). \tag*{(30)} \end{aligned}

The four integrals can be carried out explicitly using a table of Laplace transforms, resulting in the scalar splash mode trefoil nodal knot solution

It(x,y,ζ,η)=I1+I2+I3+I4.(31)I_t(x,y,\zeta,\eta)=I_1+I_2+I_3+I_4. \tag*{(31)}

Setting η=ζ+2t\eta=\zeta+2t, the intersection of the real and imaginary parts of It(x,y,ζ;t)I_t(x,y,\zeta;t) is shown below for two values of tt.

Trefoil nodal linkage formed from the intersection of the real and imaginary parts of the scalar splash mode Hopf nodal knot solution for $t=0$ and $t=1$

Figure 8. Trefoil nodal linkage formed from the intersection of the real and imaginary parts of the scalar splash mode Hopf nodal knot solution for t=0t=0 and t=1t=1, with parameter values a1=10−1a_1=10^{-1} and a2=1a_2=1.

Resorting to η=ζ+2t\eta=\zeta+2t, plots of the energy density Ut(x,y,ζ,t)U_t(x,y,\zeta,t) are shown in the figure below for three values of time.

Energy density of the scalar splash mode Hopf nodal knot solution for $t=-10$, $0$, and $10$

Figure 9. Energy density of the scalar splash mode Hopf nodal knot solution for t=−10t=-10, 00, 1010, with parameter values a1=1a_1=1 and a2=10a_2=10.

Hopf nodal knotted solution of the Schrödinger equation

Consider the time-dependent Bateman conjugate functions in Equation (8) with a=1a=1 and c=1c=1. Then,

α2−β2=N(x,y,z,t)D(x,y,z,t)=−4(x+iy)2+(−1−t2+x2+y2+z2+2iz)2(−(−i+t)2+x2+y2+z2)2(32)\alpha^2-\beta^2=\frac{N(x,y,z,t)}{D(x,y,z,t)}=\frac{-4(x+iy)^2+(-1-t^2+x^2+y^2+z^2+2iz)^2}{\left(-( -i+t)^2+x^2+y^2+z^2\right)^2} \tag*{(32)}

The numerator contains the information on the Hopf nodal linkage for all values of time. Figure 3 shows the Hopf nodal knot at t=0t=0.

Hopf nodal linkage formed from the intersection of the real and imaginary parts of the numerator

Figure 3. Hopf nodal linkage formed from the intersection of the real and imaginary parts of the numerator N(x,y,z)N(x,y,z); specifically, Re⁡{N(x,y,z)}=0\operatorname{Re}\{N(x,y,z)\}=0 and Im⁡{N(x,y,z)}=0\operatorname{Im}\{N(x,y,z)\}=0.

One can proceed to determine a Hopf nodal knotted solution to the Schrödinger equation

2i∂∂tψ(r,t)+∇2ψ(r,t)=0(33)2i\frac{\partial}{\partial t}\psi(\mathbf{r},t)+\nabla^2\psi(\mathbf{r},t)=0 \tag*{(33)}

by extending the method used in Section 3. Specifically, first the Fourier transform

ψ^(kx,ky,kz)=F{N(x,y,z,0)exp⁡[−κ(x2+y2+z2)]}(34)\hat{\psi}(k_x,k_y,k_z)=\mathcal{F}\left\{N(x,y,z,0)\exp\left[-\kappa\left(x^2+y^2+z^2\right)\right]\right\} \tag*{(34)}

is computed. Then, a specific Hopf nodal knot is obtained by means of the Fourier inversion

ψ(x,y,z,t)=F−1{ψ^(kx,ky,kz)exp⁡[−it2(kx2+ky2+kz2)]}=1(1+it)11/2exp⁡(−12x2+y2+z21+it)×{6t4+4t3(−i+4z)−t2(−9+12x2+8ixy+4y2+28iz+4z2)+10it(x2+y2+(i+z)2)+(x(2i+x)+(2+y)y+(i+z)2)×(x(−2i+x)+y(2+y)+(i+z)2)−4t(4xy+x2(−i+z)+(i+z)2+y2(3i+z))}.(35)\begin{aligned} \psi(x,y,z,t)&=F^{-1}\left\{\hat{\psi}(k_x,k_y,k_z)\exp\left[-i\frac{t}{2}\left(k_x^2+k_y^2+k_z^2\right)\right]\right\}\\ &=\frac{1}{(1+it)^{11/2}}\exp\left(-\frac{1}{2}\frac{x^2+y^2+z^2}{1+it}\right)\\ &\quad\times\left\{6t^4+4t^3(-i+4z)-t^2\left(-9+12x^2+8ixy+4y^2+28iz+4z^2\right)\right.\\ &\qquad+10it\left(x^2+y^2+(i+z)^2\right)+\left(x(2i+x)+(2+y)y+(i+z)^2\right)\\ &\qquad\left.\times\left(x(-2i+x)+y(2+y)+(i+z)^2\right)-4t\left(4xy+x^2(-i+z)+(i+z)^2+y^2(3i+z)\right)\right\}. \tag*{(35)} \end{aligned}

The intersection of the real and imaginary parts of this solution is shown in the figure below for two times.

The intersection of the real and imaginary parts of the solution at two times

Figure 10. The intersection of the real and imaginary parts of the solution in Eq. (29) at t=0t=0 and t=1/4t=1/4.

The Schrödinger Hopf polynomial following from the solution in Equation (35) is given by (see, also, [35])

w(x,y,z,t)=−1+10it+15t2−2x2−10itx2−x4−8ixy+6y2−10ity2−2x2y2−y4+4iz+20tz−4ix2z−4iy2z+6z2−10itz2−2x2z2−2y2z2−4iz3−z4.(36)w(x,y,z,t)=-1+10it+15t^2-2x^2-10itx^2-x^4-8ixy +6y^2-10ity^2-2x^2y^2-y^4+4iz+20tz-4ix^2z -4iy^2z+6z^2-10itz^2-2x^2z^2-2y^2z^2-4iz^3-z^4. \tag*{(36)}

Of course, for t=0t=0, it is equal to N(x,y,z,0)N(x,y,z,0). More complicated solutions, such as a solution characterized by a trefoil nodal knot topology can be derived by extending the procedure followed in § 3.4.

It should be noted that if the following changes are made in the solution in Equation (35)

t→z,z→Z=τ/β‾2,x→β0X,y→β0Y;β0=β(ω0);t \to z,\quad z \to Z=\tau/\sqrt{\overline{\beta}_2},\quad x \to\sqrt{\beta_0}X,\quad y \to\sqrt{\beta_0}Y;\quad\beta_0=\beta(\omega_0);
τ=t−(z/vgr),vgr=1/β1;β1=dβ(ω)dω∣ω=ω0,\tau=t-(z/v_{gr}),\quad v_{gr}=1/\beta_1;\quad\beta_1=\left.\frac{\mathrm{d}\beta(\omega)}{\mathrm{d}\omega}\right|_{\omega=\omega_0},
β‾2=∣β2∣,β2=d2β(ω)dω2∣ω=ω0(37)\overline{\beta}_2=|\beta_2|,\quad\beta_2=\left.\frac{\mathrm{d}^2\beta(\omega)}{\mathrm{d}\omega^2}\right|_{\omega=\omega_0} \tag*{(37)}

one would obtain a solution to the anomalous bidispersive equation

(i∂∂z+12β‾2∂2∂τ2+12β0∇⊥2)φ(x,y,τ,z)=0(38)\left(i\frac{\partial}{\partial z}+\frac{1}{2}\overline{\beta}_2\frac{\partial^2}{\partial\tau^2}+\frac{1}{2\beta_0}\nabla_\perp^2\right)\varphi(x,y,\tau,z)=0 \tag*{(38)}

within the framework of the slowly varying envelope approximation (SVEA). In Equation (37), β(ω)\beta(\omega) is the frequency-dependent wave number, ω0\omega_0 is the carrier angular frequency, vgrv_{gr} denotes the group speed and β2\beta_2 is the index of dispersion. The latter is positive for anomalous dispersion and negative for normal dispersion. The solution in Equation (35) is characterized by a Hopf nodal knot topology in the variables X,Y,ZX,Y,Z for relatively small values of zz (see, e.g., [40, 41, 42]).

Unknot solution of the Klein-Gordon equation

Consider the focus wave mode solution of the scalar wave equation in free space given in Equation (16) of § 3.3. With the modification shown below, viz.,

f(x,y,ζ,η)=a1a1+iζexp⁡(ibη)exp⁡[−p24b(a1+iζ)]exp⁡(−bx2+y2a1+iζ)×W(a1xa1+iζ,a1ya1+iζ,a1ζa1+iζ),(39)f(x,y,\zeta,\eta)=\frac{a_1}{a_1+i\zeta}\exp(ib\eta)\exp\left[-\frac{p^2}{4b}(a_1+i\zeta)\right]\exp\left(-b\frac{x^2+y^2}{a_1+i\zeta}\right) \times W\left(\frac{a_1x}{a_1+i\zeta},\frac{a_1y}{a_1+i\zeta},\frac{a_1\zeta}{a_1+i\zeta}\right), \tag*{(39)}

one has a solution of the Klein-Gordon equation

(∇t2+4∂2∂ζ∂η−p2)f(x,y,ζ,η)=0(40)\left(\nabla_t^2+4\frac{\partial^2}{\partial\zeta\partial\eta}-p^2\right)f(x,y,\zeta,\eta)=0 \tag*{(40)}

governing a cold plasma. It corresponds to a situation where the square of the frequency-dependent wave number (see § 6) is of second order; specifically, β2(ω)=(ω/c)2−(ωp/c)2=(ω/c)2−p2\beta^2(\omega)=(\omega/c)^2-(\omega_p/c)^2=(\omega/c)^2-p^2; p=(ωp/c)p=(\omega_p/c), ωp\omega_p being the plasma frequency.

W(x,y,ζ)W(x,y,\zeta) in Equation (39) is an appropriate polynomial satisfying the paraxial equation

4i∂∂ζW(x,y,ζ)+∇t2W(x,y,ζ)=0.(41)4i\frac{\partial}{\partial\zeta}W(x,y,\zeta)+\nabla_t^2W(x,y,\zeta)=0. \tag*{(41)}

In the following discussion a finite-energy spatiotemporally localized solution to the Klein-Gordon equation will be determined characterized by a simple unknot topology, essentially a closed loop. Toward this goal, consider the expression

u−v=N(x,y,z)D(x,y,z)=x2−2x+2iy+(i+z)1+x2+y2+z2.(42)u-v=\frac{N(x,y,z)}{D(x,y,z)}=\frac{x^2-2x+2iy+(i+z)}{1+x^2+y^2+z^2}. \tag*{(42)}

The numerator contains the information on the unknot property shown in the figure below.

Hopf unknot formed from the intersection of the real and imaginary parts of the numerator

Figure 11. Hopf unknot formed from the intersection of the real and imaginary parts of the numerator N(x,y,z)N(x,y,z).

The polynomial required in Eq. (39) is given by

W(x,y,ζ)=−1+x2−2(x+iy)+y2+iζ/b.(43)W(x,y,\zeta)=-1+x^2-2(x+iy)+y^2+i\zeta/b. \tag*{(43)}

The unknot focus wave mode solution in (39) is given explicitly as

f(x,y,ζ,η)=a1a1+iζexp⁡(ibη)exp⁡[−p24b(a1+iζ)]exp⁡(−bx2+y2a1+iζ)×[−1−2a1(x+iya1+iζ)+a12(x2+y2a1+iζ)+ia1bζa1+iζ].(44)\begin{aligned} f(x,y,\zeta,\eta)={}&\frac{a_1}{a_1+i\zeta}\exp(ib\eta)\exp\left[-\frac{p^2}{4b}(a_1+i\zeta)\right]\exp\left(-b\frac{x^2+y^2}{a_1+i\zeta}\right)\\ &\times\left[-1-2a_1\left(\frac{x+iy}{a_1+i\zeta}\right)+a_1^2\left(\frac{x^2+y^2}{a_1+i\zeta}\right)+i\frac{a_1}{b}\frac{\zeta}{a_1+i\zeta}\right]. \tag*{(44)} \end{aligned}

As mentioned earlier, this solution is nonsingular, but it contains infinite energy. A finite-energy solution is obtained by a superposition over the free parameter bb, with the spectrum shown below

ψ(x,y,ζ,η)=∫0∞db bν−1e−a2bf(x,y,ζ,η;b),a2>0.(45)\psi(x,y,\zeta,\eta)=\int_{0}^{\infty}db\,b^{\nu-1}e^{-a_{2}b}f(x,y,\zeta,\eta;b),\quad a_{2}>0. \tag*{(45)}

The required integrations can be carried out explicitly using the Laplace transform identity

∫0∞db bν−1exp⁡(−bP)exp⁡(−A4b)=2(A4P)ν/2Kν(AP),(46)\int_{0}^{\infty}db\,b^{\nu-1}\exp(-bP)\exp\left(-\frac{A}{4b}\right)=2\left(\frac{A}{4P}\right)^{\nu/2}K_{\nu}\left(\sqrt{AP}\right), \tag*{(46)}

with the restrictions Re⁡{A}>0\operatorname{Re}\{A\}>0 and Re⁡{P}>0\operatorname{Re}\{P\}>0. KνK_{\nu} denotes the ν\nu-order of the modified Bessel function of the second kind. The intersection of the real and imaginary parts of this solution is shown in the figure below for two times.

Hopf unknot formed from the intersection of the real and imaginary parts of the solution at two times

Figure 12. Hopf unknot formed from the intersection of the real and imaginary parts of the solution ψ(x,y,ζ,η=ζ+2ct)\psi(x,y,\zeta,\eta=\zeta+2ct) at ct=0ct=0 and 55. The following parameter values were used: ν=2\nu=2, a1=1a_{1}=1, a2=10a_{2}=10 and p=10−1p=10^{-1}.

Discussion

In the case of knotted vector-valued electromagnetic fields (see § 2), any two electric fields lines are linked, any two magnetic field lines are linked, and any electric field line is linked to any magnetic field line, not only simply, but also in complicated knots (Hopf, trefoil, cinquefoil, etc.) In the case of paraxial monochromatic light beams, on the other hand, one is interested in nodal knots, the latter being threads of complete darkness (lines of zero intensity) tied into diverse topological structures. Several such topological types have been successfully implemented experimentally, e.g., those arising from a superposition of Gauss-Laguerre beams [35, 36, 43].

The aim in this article has been to derive finite-energy spatiotemporally localized scalar-valued wave solutions characterized by nodal knotted topologies. This will extend to spatiotemporally localized pulsed fields the diverse important applications of monochromatic light beams, having nontrivial topological structures, in trapping and guiding particles, image processing and optical communications; also, in applications to biology and medicine.

The work in § 3, § 4 and § 5 has been confined to the basic Hopf and trefoil nodal knots. However, finite-energy spatiotemporally localized splash mode solution to the scalar wave equation with more complicated knotted nodal topologies can be easily derived.

The basic approach used § 3-§ 5 has been used in § 6 to obtain a Hopf nodal solution the {3+1)D Schrödinger equation. It has also been pointed out how to obtain, within the framework of the slowly varying envelope paraxial approximation, the same type of solution in the case of anomalous temporal dispersion by a simple change of variables. Finally, a finite-energy spatiotemporally localized splash mode solution to the Klein-Gordon equation has been derived in § 7. Although for simplicity it is confined to an unknotted nodal topology, more complicated solutions can be easily derived.

Concluding remarks

When dealing with monochromatic light beams, not only bright regions but also those of darkness, arising from exact destructive interference, are of important physical interest, e.g., as traps for particles. Regions of zero light intensity (characterized also by phase singularities) with diverse topological structures, such as nodal knots, have been studied extensively recently. These are scalar monochromatic solutions based on the paraxial approximation. With the exception of solutions to the Schrödinger-like equation describing anomalous dispersion, no scalar-valued spatiotemporally localized waves with analogous knotted topological structures are known to date. In this article we have filled in this gap by deriving finite-energy spatiotemporally localized scalar-valued, knotted wave solutions in free space and in the presence of temporal dispersion.

References

  1. [1]Brittingham, J.N. Focus wave modes in homogeneous Maxwell equations: Transverse electric mode. J. Appl. Phys. 1983, 54, 1179-1189. https://doi.org/10.1063/1.332196
  2. [2]Kiselev, A.P. Modulated Gaussian beams. Radio Phys. Quant. Electron. 1983, 26, 1014-1020. https://doi.org/10.1007/bf01034890
  3. [3]3. Ziolkowski, R. Localized transmission of electromagnetic energy. Phys. Rev. A 1989, 39, 2005-2033. https://doi.org/10.1103/physreva.39.2005
  4. [4]Besieris, I.M.; Shaarawi, A.M.; Ziolkowski, R.W. A bidirectional traveling plane wave representation of exact solutions of the scalar wave equation. J. Math. Phys. 1989, 30, 1254-1269. https://doi.org/10.1063/1.528301
  5. [5]Lu, J.Y.; Greenleaf, J.F. Nondiffracting X-waves exact solutions to the free space scalar wave equation and their finite aperture realization. IEEE Trans. Ultrason. Ferroelectr. Freq. Contr. 1992, 39, 19-31. https://doi.org/10.1109/58.166806
  6. [6]Ziolkowski, R.W.; Besieris, I.M.; Shaarawi, A.M. Aperture realizations of the exact solutions to homogeneous-wave equations. J. Opt. Soc. Am. A 1993, 10, 75-87. https://doi.org/10.1364/josaa.10.000075
  7. [7]Saari, P.; Reivelt, K. Evidence of X-shaped propagation-invariant localized light waves. Phys. Rev. Lett. 1997, 79, 4135-4137. ..... https://doi.org/10.1103/physrevlett.79.4135 ... Besieris, I.M.; Abdel-Rahman, M.; Shaarawi, A.M.; Chatzipetros, A. Two fundamental representations of localized pulse solutions to the scalar wave equation. Progr. Electromagn. Res. (PIER) 1998, 19, 1-48. https://doi.org/10.2528/pier97072900
  8. [9]Salo, J.; Fagerholm, J.; Friberg, A.T.; Saloma, M.M. Unified description of X and Y waves. Phys. Rev. E 2000, 62, 4261. https://doi.org/10.1103/physreve.62.4261
  9. [10]Grunwald, R.; Kebbel, V.; Neumann, U.; Kummrow, A.; Rini, M.; Nibbering, R.T.; Piche, M.; Rousseau, G.; Fortin, M. Generation and characterization of spatially and temporally localized few-cycle optical wave packets. Phys. Rev. A 3820, 67, 06. https://doi.org/10.1103/physreva.67.063820
  10. [11]Saari, P.; Reivelt, K. Generation and classification of localized waves by Lorentz transformations in Fourier space. Phys. Rev. E 6612, 68, 03. https://doi.org/10.1103/physreve.69.036612
  11. [12]Longhi, S. Spatial-temporal Gauss-Laguerre waves in dispersive media. Phys. Rev. E 6612, 68, 06. https://doi.org/10.1103/physreve.68.066612
  12. [13]Conti, C.; Trillo, S.; di Trapani, P.; Valiulis, G.; Piskarskas, A.; Jedrkiewicz, O.; Trull, J. Nonlinear electromagnetic X waves. Phys. Rev. Lett. 0406, 90, 17. https://doi.org/10.1103/physrevlett.90.170406
  13. [14]Kiselev, A.P. Localized light waves: Paraxial and exact solutions of the wave equation (review). Opt. Spectrosc. 2007, 102, 603-622. https://doi.org/10.1134/s0030400x07040200
  14. [15]Yessemov, M.; Bhaduri, B.; Kondakci, H.E.; Abouraddy, A.F. Classification of propagation-invariant space-time wavepackets in free space: Theory and experiments. Phys. Rev. A 2019, 99, 023856. https://doi.org/10.1103/physreva.99.023856
  15. [16]Hernandez-Figueroa, H.E.; Zamboni-Rached, M.; Recami, E.; Eds. Localized Waves; Wiley-VCH, New York, 2008.
  16. [17]Hernandez-Figueroa, H.E.; Recami, E.; Zamboni-Rached, M.; Eds. Non-Diffracting Waves; Wiley-VCH, New York, 2014. https://doi.org/10.1002/9783527671519
  17. [18]Yessemov, M.; Hall, L.A.; Schepler, K.L.; Abouraddy, A.F. Space-time wavepackets. Adv. Opt. Photon. 2022, 14, 455-470. https://doi.org/10.1364/aop.450016
  18. [20]Bowlan, P.; Valtna-Lukner, H.; Lõhmus, M.; Piksarv, P.; Saari, P.; Trebino, R. Measurement of the spatio-temporal field of ultrashort Bessel-X pulses. Opt. Lett. 2009, 34, 2276. https://doi.org/10.1364/fio.2009.ftuo5
  19. [21]Saari, P., “X-Type Waves in Ultrafast Optics,” in Non-Diffracting Waves, H. E. Hernandez-Figueroa, E. Recami, and M. Zamboni-Rached, Eds.; Wiley, New York, 2013, pp. 109-134. https://doi.org/10.1002/9783527671519
  20. [22]Kondakci, H.F.; Abouraddy, A.F. Diffraction-free space-time light sheets. Nat. Photon. 2017, 11, 733-740. https://doi.org/10.1038/s41566-017-0028-9
  21. [23]Bhaduri, B.; Yessemov, M.; Abouraddy, A.F. Space-time wave packets that travel in optical materials at the speed of light in vacuum. Optica 2019, 6, 139-145. https://doi.org/10.1364/optica.6.000139
  22. [24]Papasimakis, N.; Raybould, T.; Fedotov, V.A.; Tsai, D.P.; Youngs, I.; Zheludev, N.I. Pulse generation scheme for flying electromagnetic doughnuts. Phys. Rev. B 2018, 97, 201409-1-6. https://doi.org/10.1103/physrevb.97.201409
  23. [25]Comite, D.; Fuscaldo, W.; Podilchak, S.K.; Gómez-Guillamón Buendia, V. Microwave generation of X-waves by means of planar leaky-wave antenna. Appl. Phys. Lett. 2018, 113, 144102-1-5. https://doi.org/10.1063/1.5047397
  24. [26]Fuscaldo, W.; Comite, D.; Boesso, A.; Baccarelli, P.; Bughinogli, P.; Galli, A. Focusing leaky waves: a class of electromagnetic localized waves with complex spectra. Phys. Rev. A 2018, 9, 054005-1-15. https://doi.org/10.1103/physrevapplied.9.054005
  25. [27]Irvine, W.T.M.; Bouwmeester, D. Linked and knotted beams of light. Nat. Phys. 2008, 4, 716-720. https://doi.org/10.1038/nphys1056
  26. [28]Rañada, A.F. Knotted solutions of Maxwell equations in vacuum. J. Phys. A: Math. Gen. 1990, 23, L815-L820. https://doi.org/10.1088/0305-4470/23/16/007
  27. [29]Besieris, I.M.; Shaarawi, A.M. Hopf-Rañada linked and knotted light beam solution viewed as a null electromagnetic field. Opt. Lett. 2009, 34, 3887-3889. https://doi.org/10.1364/ol.34.003887
  28. [30]Bialynicki-Birula, I. Electromagnetic vortex lines riding atop null solutions of the Maxwell equations. J. Pure Appl. Opt. 2004, 6, S181-S183. https://doi.org/10.1088/1464-4258/6/5/007
  29. [31]Kedia, H.; Bialynicki-Birula, I.; Peralta-Salas, D.; Irvine, W.T.M. Tying knots in light fields. Phys. Rev. Lett. 2017, 111, 150404. https://doi.org/10.1103/physrevlett.111.150404
  30. [32]Berry, M.V.; Dennis, M.R. Knotted and linked phase singularities in monochromatic waves. Proc. R. Soc. Lond. A 2001, 457, 2251-2263. https://doi.org/10.1098/rspa.2001.0826
  31. [33]Dennis, M.R. Braided nodal lines in wave superpositions. New J. Phys. 2003, 5, 134. https://doi.org/10.1088/1367-2630/5/1/134
  32. [34]Dennis, M.R.; King, R.P.; Jack, B.; O’Halleran, K.; Padgett, M.J. Isolated optical knots. Nat. Phys. 2010, 6, 118-121. https://doi.org/10.1038/nphys1504
  33. [35]King, R.P. Knotting of optical vortices. Ph. D. thesis, University of Southampton, England, 2010.
  34. [36]Bode, B.; Dennis, M.R.; Foster, D.; King, R.P. Knotted fields and explicit fibrations for lemniscate knots. Proc. R. Soc. Lond. A 2018, 473, 2016.0829. https://doi.org/10.1098/rspa.2016.0829
  35. [37]Radożycki, T. Knotted trajectories of neutral and charged particles in Gaussian light beams. arXiv:2008.13045v1[physics. Optics] 2020. https://arxiv.org/abs/2008.13045
  36. [38]T. Radożycki, T. Knotted nodal lines in superpositions of Bessel-Gaussian light beams. Phys. Rev. A 2021, 103, 013509. https://doi.org/10.1103/physreva.103.013509
  37. [39]Wan, C.; Shen, Y.; Chong, A.; Zhan, Q. Scalar optical Hopfions. eLight 2022, 2. https://doi.org/10.1186/s43593-022-00030-2
  38. [40]Wan, C.; Cao, Q.; Chen, J.; Chong; Zhan, Q. Toroidal vortices of light. Nat. Photon. 2022, 16, 519-522. https://doi.org/10.1038/s41566-022-01013-y
  39. [41]Li, C.; Wang, S.; X; Li. Spatiotemporal pulse weaving scalar optical Hopfions. Light: Sci. Appl. 2023, 12, 54. https://doi.org/10.1038/s41377-023-01101-w
  40. [42]Lyu, Z.; Fang, Y.; Liu, Y. Formation and controlling of optical Hopfions in high harmonic generation. Phys. Rev. Lett 2024, 133, 133801. https://doi.org/10.1103/physrevlett.133.133801
  41. [43]Tsvetkov, D.; Pipes, D.G.; Sedeh, H.B.; Litchinitser, N.M. Sculpting isolated optical vortex knots on demand. Photon. Res. 2025, 13, 527-540. https://doi.org/10.1364/prj.533264
  42. [44]Zhong, J.; Zhang, Q. Structured optical toroidal vortices with rotational symmetry. Photonics 2025, 12, 288. https://doi.org/10.3390/photonics12030288
  43. [45]Whittaker, E.T. On the expressions of the electromagnetic field due to electrons by means of two scalar potential functions. Proc. Lond. Math. Soc. 1904, 1, 367-372. https://doi.org/10.1112/plms/s2-1.1.367
  44. [46]Bateman, H. The Mathematical Analysis of Electrical and Optical Wave-Motion on the Basis of Maxwell’s Equations, Dover, New York, 1955.
  45. [47]Weber; H. Die partiellen Differential-Gleichungen der mathematischen Physik nach Riemann’s Vorlesungen. Friedrich Vieweg und Sohn, Brunschweig, 1901.
  46. [48]Silberstein, L. Electromagnetische Grundgleichungen in bivectorieller Behandlung. Ann. D. Phys. 1907, 22, 579-586.
  47. [49]Brauner, K. Zur Geometrie der Functionen zweir komplexen Veränderlichen I, III, IV. Abh. Math. Sem. Hamburg 1928, 6, 8-54.

Paper details

vlnc:W_8nxk3sjj
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 10

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents