[Ewald] ...

Former-commit-id: e9f98d06ca7ccfad3d9181d083919d1a146f8860
This commit is contained in:
Marek Nečada 2017-09-11 08:07:23 +00:00
parent 9f9da628cc
commit 814ea36415
1 changed files with 122 additions and 4 deletions

View File

@ -1489,6 +1489,8 @@ Let's polish it a bit more
\end_inset
\size footnotesize
\begin_inset Formula
\begin{multline}
\pht n{s_{q,k_{0}}^{\textup{L}1,c}}\left(k\right)=\frac{k^{n}Γ\left(2-q+n\right)}{k_{0}^{q}\left(c-ik_{0}\right)^{2-q+n}}\left(\frac{-k^{2}}{\left(c-ik_{0}\right)^{2}}\right)^{-\frac{n}{2}}\left(1+\frac{k^{2}}{\left(c-ik_{0}\right)^{2}}\right)^{\frac{q}{2}-1}P_{q}^{-n}\left(\frac{1}{\sqrt{1+\frac{k^{2}}{\left(c-ik_{0}\right)^{2}}}}\right),\\
@ -1497,6 +1499,8 @@ k>0\wedge k_{0}>0\wedge c\ge0\wedge\lnot\left(c=0\wedge k_{0}=k\right)\label{eq:
\end_inset
\size default
with principal branches of the hypergeometric functions, associated Legendre
functions, and fractional powers.
The conditions from
@ -1525,6 +1529,10 @@ reference "eq:2D Hankel transform of regularized outgoing wave, decomposition"
\end_layout
\begin_layout Standard
\begin_inset Note Note
status collapsed
\begin_layout Plain Layout
Let's do it.
\begin_inset Formula
\begin{eqnarray*}
@ -1534,6 +1542,11 @@ Let's do it.
\end_inset
\end_layout
\end_inset
One problematic element here is the gamma function
\begin_inset Formula $\text{Γ}\left(2-q+n\right)$
\end_inset
@ -1561,6 +1574,76 @@ polynomial
\end_inset
.
In other cases, different expressions can be obtained from
\begin_inset CommandInset ref
LatexCommand ref
reference "eq:2D Hankel transform of exponentially suppressed outgoing wave as 2F1"
\end_inset
using various transformation formulae from either DLMF or
\begin_inset ERT
status open
\begin_layout Plain Layout
\backslash
begin{russian}
\end_layout
\end_inset
Прудников
\begin_inset ERT
status open
\begin_layout Plain Layout
\backslash
end{russian}
\end_layout
\end_inset
.
\end_layout
\begin_layout Standard
In fact, Mathematica is usually able to calculate the transforms for specific
values of
\begin_inset Formula $\kappa,q,n$
\end_inset
, but it did not find any general formula for me.
The resulting expressions are finite sums of algebraic functions, Table
\begin_inset CommandInset ref
LatexCommand ref
reference "tab:Asymptotical-behaviour-Mathematica"
\end_inset
shows how fast they decay with growing
\begin_inset Formula $k$
\end_inset
for some parameters.
The only case where Mathematica did not help at all is
\begin_inset Formula $q=2,n=0$
\end_inset
, which is unfortunately important.
But if I have not made some mistake, the expression
\begin_inset CommandInset ref
LatexCommand eqref
reference "eq:2D Hankel transform of exponentially suppressed outgoing wave expanded"
\end_inset
is applicable for this case.
\end_layout
\begin_layout Standard
@ -2712,6 +2795,41 @@ reference "eq:prudnikov2 eq 2.12.9.14"
\end_layout
\begin_layout Section
Major TODOs and open questions
\end_layout
\begin_layout Itemize
Check if
\begin_inset CommandInset ref
LatexCommand eqref
reference "eq:2D Hankel transform of exponentially suppressed outgoing wave expanded"
\end_inset
gives a satisfactory result for the case
\begin_inset Formula $q=2,n=0$
\end_inset
.
\end_layout
\begin_layout Itemize
Analyse the behaviour
\begin_inset Formula $k\to k_{0}$
\end_inset
.
\end_layout
\begin_layout Itemize
Find a general algorithm for generating the expressions of the Hankel transforms.
\end_layout
\begin_layout Itemize
Three-dimensional case.
\end_layout
\begin_layout Section
(Appendix) Fourier vs.
Hankel transform
@ -2761,7 +2879,7 @@ where the spherical Hankel transform
2)
\begin_inset Formula
\[
\bsht lg(k)\equiv\int_{0}^{\infty}\ud r\,r^{2}g(r)j_{l}\left(kr\right).
\bsht lg(k)\equiv\int_{0}^{\infty}\ud r\, r^{2}g(r)j_{l}\left(kr\right).
\]
\end_inset
@ -2771,7 +2889,7 @@ Using this convention, the inverse spherical Hankel transform is given by
3)
\begin_inset Formula
\[
g(r)=\frac{2}{\pi}\int_{0}^{\infty}\ud k\,k^{2}\bsht lg(k)j_{l}(k),
g(r)=\frac{2}{\pi}\int_{0}^{\infty}\ud k\, k^{2}\bsht lg(k)j_{l}(k),
\]
\end_inset
@ -2784,7 +2902,7 @@ so it is not unitary.
An unitary convention would look like this:
\begin_inset Formula
\begin{equation}
\usht lg(k)\equiv\sqrt{\frac{2}{\pi}}\int_{0}^{\infty}\ud r\,r^{2}g(r)j_{l}\left(kr\right).\label{eq:unitary 3d Hankel tf definition}
\usht lg(k)\equiv\sqrt{\frac{2}{\pi}}\int_{0}^{\infty}\ud r\, r^{2}g(r)j_{l}\left(kr\right).\label{eq:unitary 3d Hankel tf definition}
\end{equation}
\end_inset
@ -2838,7 +2956,7 @@ where the Hankel transform of order
is defined as
\begin_inset Formula
\begin{equation}
\pht mg\left(k\right)=\int_{0}^{\infty}\ud r\,g(r)J_{m}(kr)r\label{eq:unitary 2d Hankel tf definition}
\pht mg\left(k\right)=\int_{0}^{\infty}\ud r\, g(r)J_{m}(kr)r\label{eq:unitary 2d Hankel tf definition}
\end{equation}
\end_inset