Loomis Wood style plots can be produced with PGOPHER; they work by plotting the same spectrum several times, with successive spectra horizontally displaced by an amount given by a function in terms of the plot number, m. With the correct choice of function (typically a polynomial), successive transitions in a branch will appear lined up.
This is done by adding a Loomis Wood object in
the Overlays window (brought up
with "View, Overlays"); they can be created at the top
level (in which case multiple windows are created) or under a
specific plot, in which case the plot is duplicated in a single
window. Right click on the top level object or the specific plot
as required, and select "Add New...", "Loomis Wood".
The required function is given in the comment in terms of the
running plot number, m. The default polynomial is:
b(m-mLowest) + d(m-mLowest)2but any sensible function of m can be given. (See expressions for more on what is possible.) For a simple diatomic molecule b = 2B will line up all the lines in a P or R branch.
To determine the parameters of the function a built in fitting
process can be used, similar to that used for calibrating spectra. A Loomis Wood
window can be brought up by right clicking on the Loomis-Wood
object, and selecting "View...". When this is visible,
and selecting an active Loomis-Wood object, a series of peaks
making up a branch can be measured. Right clicking and dragging
over a peak will add the peak to the series for that object,
with the m value set according to the spectrum the peak
is measured in. Pressing "Fit" will do a least squares fit on
the parameters in the function using the given peaks, and then
update the display to reflect the new parameters. To avoid
undesirable correlation in the parameters of the function, the
values fitted to are offsets from the position of the peak with
the lowest value of m. The mLowest parameter
(and the corresponding variable in the expression) is set to
this value of m at the start of the fit. Other
operations in the Loomis-Wood window include:
|True to produce a Loomis-Wood plot; there can be multiple
Loomis-Wood objects, but only one should be active at a
|If non zero, write the values of all the variables and the value of the function on each evaluation to the log window. See debugging for options on the output.
|The number of plots to produce
|The m values used in the plots are displaced by
this. The plots have m values of mOffset
|If false, the simulation occupies half the
plot. If false, the simulation shrinks to 1/(nPlots+1)
of the plot.
|For future functionality
|The vertical offset between plots, expressed as a fraction
of the vertical height of one plot. Possible values are in
the range -1.1 ... +1.1.
|This parameter is set to the lowest m
value for the peaks included in the fit at the start of each
|Parameters in the polynomial given in the
comment are shown here.