Data supporting "Understanding the Plasmonics of Nanostructured Atomic Force Microscopy Tips"

This archive contains the data presented in the above paper.  It is present in two formats:
firstly, tab-separated-values for the various 1D and 2D graphs, and secondly an HDF5 file
containing more extensive data, for example the hyperspectral images from which slices are 
plotted.  HDF5 is an open format, extensively supported by scientific analysis packages.
A free viewer is available at https://support.hdfgroup.org/products/hdf5_tools/
Plugins are also available for Matlab, Python, Igor Pro, and many more.

The datasets are listed below by figure:

Tab-separated value data:
Figure 3:
(a-c) are not include in TSV format, as they belong to a 3- or 4-dimensional dataset.
The traces of scattering spectra vs wavelength are tabulated in fig_3d*.tsv and fig_3e*.tsv
Figure 3f shows three spectra vs wavelength, which are presented in three TSV files as the
wavelength points are not the same between all three traces.

Figure 4:
(a) the two spectra are tabulated against wavelength in fig_4a_spectra.tsv
(b-c) the images at the wavelengths shown are extracted in tab-separated-value format.
(d) the X- and Y-axis values are given in separate 1D TSV files, and the 2D scattering
vs neck size and wavelength data is given in a third file.  The fitted mode positions and 
intensities are tabulated against neck size.  The columns in the file are neck size, 
then the position of each of three modes, then the intensity of each mode.

Data in the HDF5 file is organised by dataset rather than by figure, though they correspond
closely.  HDF5 files are heirarchically organised into "groups" analogous to folders in a
filesystem:

scattering_spectra/<tip>/:
There is a folder for each of the four tips presented in figure 3.  In each of these folders,
there are two hyperspectral images - three or four dimensional datasets showing scattering
intensity against X, Y, (optionally) Z, and wavelength.  The X, Y, Z and wavelength values
are given in the correspondingly named datasets.  NB the nanotools tip was scanned in 3D 
while the others were scanned only in 2D.  The z_slice value (5) indicates that the fifth
slice was taken out of the array, i.e. the third index of the array is set to 5.

The spectra shown in figure 3(d-e) are extracted from the hyperspectral images above.  The
region over which the spectra were averaged is indicated by the "tip_roi" dataset, which has
the left, right, top, bottom values of the box used to capture the tip.  Note that these are
in the same units (metres) as the X, Y datasets, but that they are relative to the centre of
the image rather than being in the same coordinate system as the X, Y datasets.  The extracted
spectra are also included in the group for each tip.

For each tip, two spectrometers were used to acquire spectra in the axial and transverse
directions; axial polarisation is saved in spectra and the corresponding wavelength values are 
in wavelength.  Transverse polarisation is in spectra2 and wavelength2 respectively.

The images and spectra in the paper have been smoothed to remove noise and scanning artefacts.
Data included in this archive is saved before this smoothing step.

sers_background/:
The six datasets in this folder correspond simply to the X and Y values for each of the three
traces in Figure 3(f).

simulation/:
The spectra plotted in fig. 4(a) are included here; NB their wavelength values are the same.
The images in fig. 4(b-c) are included in their own data groups in the "image_as_shown"
dataset.  This is extracted from a larger hyperspectral image, in the dataset "spectra".  The
wavelength values for the third index of this array are in "wavelengths".  The wavelength of
the extracted image is included for completeness.
The neck width simulation for figure 4(d) has its own group, with the spectra simulated for 
each neck size in "neck_spectra", and the neck size and wavelength values included in separate
datasets.  The intensity and position of each of three modes is stored in "mode_intensities"
and "mode_positions", corresponding to the nine neck sizes in "neck_sizes".