Laguerre-Gauss profile

The LaguerreGaussLaser type contains the parameters required for describing a Laguerre-Gauss laser pulse. The spatial profile of the electric field is given by

\[\newcommand{\ii}{\mathrm{i}} % imaginary unit \begin{aligned} E_x(r,z,φ) &= ξ_x E_g N_{pm} \left(\frac{\sqrt{2}r}{w(z)}\right)^{|m|} ₁F₁\left(-p, |m|+1; 2\left(\frac{r}{w(z)}\right)^2\right) \exp\left[\ii\left((2p+|m|)\arctg{\frac{z}{z_R}}-mφ - \phi_0\right)\right] \\ E_y(r,z,φ) &= \frac{ξ_y}{ξ_x} E_x(r,z,φ) \\ E_z(r,z,φ) &= -\frac{\ii}{k} \left\{\left[-2\frac{1+\ii\frac{z}{z_0}}{w(z)^2} + \frac{4p}{(|m|+1)w(z)^2}\ ₁F₁^{-1}\left(-p, |m|+1; 2\left(\frac{r}{w(z)}\right)^2\right)\right](x E_x + y E_y) - \frac{|m|}{x+\ii y} (E_x \mp \ii E_y)\right\} \end{aligned}\]

where

  • $E_g$ is the amplitude of the electric field form the Gaussian profile, $E_x(r,z)$
  • $N_{pm}$ is a normalization factor given by $\sqrt{(p+1)_{|m|}}$, with $(x)_n$ the Pochhammer symbol
  • $₁F₁(α,β;z)$ is the confluent hypergeometric function of the first kind
  • $p$ is the radial index of the mode, $p ∈ ℤ, p ≥ 0$
  • $m$ is the azimuthal index of the mode, $m ∈ ℤ$
  • $ξ_x$ and $ξ_y$ give the polarization (choosing $ξ_x=1$ and $ξ_y=0$ for example, would give a linearly polarized field along the $x$ axis while taking them $1/\sqrt{2}$ and $±\mathrm{i}/\sqrt{2}$ would give right and left-handed circular polarization)
  • $w(z)$ is the beam radius at $z$
  • $w_0$ is the beam waist
  • $k$ is the wavenumber
  • $R$ is the radius of curvature
  • $\arctg{\frac{z}{z_R}}$ is the Gouy phase
  • $ϕ_0$ is the initial phase

The magnetic field field is given by

\[\newcommand{\ii}{\mathrm{i}} % imaginary unit \begin{aligned} B_x(r,z) &= -\frac{1}{c}E_y(r,z)\\ B_y(r,z) &= \frac{1}{c}E_x(r,z)\\ B_z (r,z) &= -\frac{\ii}{ck} \left\{\left[-2\frac{1+\ii\frac{z}{z_0}}{w^2(z)} + \frac{4p}{(|m|+1)w(z)^2}\ ₁F₁^{-1}\left(-p, |m|+1; 2\left(\frac{r}{w(z)}\right)^2\right)\right](y E_x + x E_y) - \frac{|m|}{x+\ii y} (E_y \mp \ii E_x)\right\} \end{aligned}\]

Choice of normalization

The normalization was chosen such that the normalization factor $N_{pm}$ is adimensional and the result is consistent with the choice for the Gaussian mode. This way it is easy to obtain the electric field from the mode, as the dimensions are given through $E_0$.

The Laguerre-Gauss modes are expressed via the Gauss ones in order to reuse code and together with the choice for $N_{pm}$ ensures consistent normalization preferences.

Note

In the expression of the Laguerre-Gauss modes above, the confluent hypergeometric function of the first kind was used instead of the standard associated Laguerre polynomials. This helps reduce the number of factorials required for the normalization constant. The relation between the two is given by

\[L_n^α(z) = \frac{(α+1)_n}{n!} ₁F₁(-n,α+1;z),\]

where $(x)_n$ is the Pochhammer symbol