Python program for analysing planar X-ray waveguides.

I kludged this together as a part of a 3rd year physics research project at Monash University (they are really awesome at getting undergrads doing real research!) with Daniele Pelliccia (http://monash.edu/science/about/schools/physics/people/research/pelliccia.html) and David Paganin (http://monash.edu/science/about/schools/physics/people/academic/paganin.html).

If you've got any questions or comments about this (or if you're trying to use it and it's completley broken!), please do email me at kjtsa1@student.monash.edu! It wasn't really intended to be something generally useful when I wrote it, but I figured I may as well throw it on Github anyway in case I was wrong. For a discussion on the physics behind this, look at characterisation_of_an_x-ray_waveguide.pdf in the repo.

All the heavy lifting is done by the most excelent NumPy, SciPy and matplotlib libraries (http://www.numpy.org/, http://www.scipy.org/ and http://matplotlib.org/)

PyGuide is a command-line program that accepts inputs predominantly from the command line and outputs plots as image files and tables as comma-separated text files.

All functions of PyGuide can accept the following parameters-

-v, --verbose Display calculation information as the program is running

-wg [wg-name], --waveguide [wg-name] -wgf [wg-file], --waveguidefile [wg-file] These options determine the characteristics of the waveguide that is being analysed. If the --waveguidefile flag is specified, then the waveguide characteristics are read from the specified file. Otherwise, if the --waveguide flag is specified, the waveguide is read from a file named [wg-name].xml from the waveguides folder. The details of this waveguide format are provided later in this document.

-o [filename | dirname], --output [filename | dirname] File in which to store the output of this program. If multiple files are generated, this name will be taken as a directory name to create and store the output files in.

PyGuide can calculate the guided modes of your waveguide by invoking it as python PyGuide [general-options] modesolver [output-type] [modesolver-options]. The [output-type] flag is mandatory and specifies what kind of output will be generated.

-kx, --wavevectors Produces a CSV table containing the wavevector value and grazing angle of each of the guided modes in the waveguide

-ip, --intensityplot Produces plots of the square modulus of each guided mode wavefunction as individual image files. The longitudinal distance at which to calculate this is specified by the --distances flag

-wfp, --wavefunctionplot Produces plots of the entrance surface wavefunction of each guided mode as individual image files.

-pp, --poyntingplot Produces plots of the Poynting vector at the entrance surface as a function of transverse distance of each guided mode as individual image files.

-ap, --argandplot Produces plots of the entrance surface wavefunction of each guided mode on an argand diagram as individual image files. It looks cool, but I never understood what this meant physically.

The following options specify aspects of the behaviour of this mode calculation

-m [int list], --modes [int list] A list of numbers representing which guided modes should have plots generated for them (ignored for the --wavevectors table). Don't specify this if you want to see plots of all modes; the syntax of it looks like --modes 3 4 5.

-d [distance], --distance [distance] The longitudinal distance (in metres) at which to generate the relevant plots; this is only used for the --poyntingplot plots. Exponential notation works (e.g. --distance 5e-3 is 5mm).

-l [wavelength], --wavelength [wavelength]. The wavelength of light for which to calculate the guided modes (applies to all output types). This is specified in metres and again exponential notation is OK.

The waveguide file format is pretty self-documenting; look in the waveguides folder for some examples. Some things to note:

• the type parameter of the opening <waveguide> tag was supposed to describe the geometry of the waveguide; I only implemented planar waveguides, so this value actually is ignored
• The <length> tag is the length of the waveguide in metres
• The <slabgap> tag is the width of the core material in metres
• The <coreindex> tag is the complex refractive index of the core material
• The <claddingindex> tag contains a space seperated table consisting of three columns: wavelength (in nm), real part of refractive index, and complex part of refractive index. This table describes the refractive index of the cladding material as a function of wavelength. This table is interpolated to yield the refractive index at an arbitrary wavelength.
• Note that I just fit a straight line to the log of the refractive index and use it for the interpolation. This is no good if you have an absorption edge; you could do something different by changing PlanarWaveguideInterp.__init__() and PlanarWaveguideInterp.cladding_index() in waveguides.py.