To represent molecules and states making up the simulation, PGOPHER makes use of a
hierarchy of objects. This works rather like a nested directory
structure containing files except that each level can only
contain certain types of object. These are no particular limits
to the number of objects at each level, so that arbitrary
combinations of spectra can be constructed, and multiple
interacting (perturbing) states can be simulated. To view or
edit any of these objects, use the tree view in the Constants window (View,
Constants). The topmost object is a
which contains settings controlling details of the calculation,
such as the units used. The name of the mixture is normally the
filename, as it corresponds to a single .pgo
mixture contains one or more:
- Species which contain one or more
- Molecules These specify the
settings such symmetry and statistical weight which are
common to all the states within a molecule. There are four
different types available, depending on the type of
molecule: Linear, Symmetric Top, Asymmetric Top and Vibrating Molecules (for
vibrational structure only). Molecules contain three types
- Manifolds provide a way of
grouping states together,
allowing perturbations to be simulated. Manifolds contain
two types of object:
- States which contain the band
origin and rotational constants of a particular state.
Again, these are specific to the type of molecule, with
Linear, Symmetric Top, Asymmetric Top and Vibrating Molecule
versions, depending on the Hamiltonian used. If nuclear
hyperfine structure is to be simulated, then each state
- Nucleus which contains the
hyperfine parameters for each nucleus for the parent
- Perturbation which
describes an interaction or perturbation between two
states. There is one of these for each perturbation
- Transition Moments
which describe the spectroscopic transitions between or
within manifolds. One of these objects is required for
each pair of manifolds that have transitions between them.
They contain one or more
Partition Functions. This (optional) object can be
used if the partition function is to be input externally,
rather than automatically calculated.
- Simulation objects, which
describe the experimental conditions. One of these will be
generated automatically and relate to the main window, and
additional simulation windows and the corresponding objects
are created if additional plot windows are created from View, Another Plot.
- Variables objects, which contain
global parameters typically used to constrain fits.
- Table objects which contain tables
of values for use in expressions.
- An optional Correlations
object, which contains the correlation between parameters in
the most recent least squares fit. It is automatically created
and updated on every fit.
Duplicate names for objects
under any given object should be avoided; they can be used when setting
up the calculation in the first place, but some operations may fail if
duplicates are allowed to persist.
Overlays and Experimental Data
Experimental data used to overlay simulated spectra are also
described by a similar, if simpler hierarchy of objects. These are
not saved in the .pgo file, but can be saved separately in a .ovr
file if required. These can be individually inspected using the
tree view in the Overlays window (Overlays,
Properties). The topmost object is:
To manipulate these see the File Menu,
the Overlays Menu or the Overlays window. Some operations can
also be carried out by right clicking on the overlay on the plot
window, depending on the format of the plot.
(corresponding to a .ovr
file if saved) which can
contain three types of plots
- Experimental Plots,
which contain the (frequency, intensity) points of
an experimental spectrum
- Bitmap Plots, which contain
an image for use where only a picture of the spectrum is
available, but not the raw data.
- Calibration plots, containing built in line lists - see Calibrating Spectra.