Documentation

Features

Handles Static and MAS quadrupolar line shapes.
Handles CSA line shapes — uses Herzfeld-Berger Convention, i.e span = δ₃₃ − δ₁₁ and skew = 3×(δ₂₂ - δ_iso)/ span
Handles distributions (Gaussian, Czjzek and Flat) of quadrupolar and CSA interaction parameters.
Broadens lines using gaussian or lorentzian distributions.
When you hit return it is equivalent to clicking "calculate"
Supports shifting spectra vertically. Helps if you have a DC offset.
Handles static CSA and Quadrupolar lines with distribution of interaction parameters
Iteration of single or multiple lines with as many parameters you would like to iterate. Be careful with how many parameters you step and how many steps as it could take a VERY long time or just run out of memory when allocating memory space.

Key Shortcuts

Up / Down – change between lines (sites)
For the rest, no modifier is up by a large amount, shift changes to a small amount and control changes to down.

H – Height
S – Shift
B – Broadening
U – C_Q centre or if there isn't a distribution then the coupling constant of the line.
I – C_Q width
O – Asymmetry Centre
P – Asymmetry Width

And when using combined CSA and quadrupolar:
E – Span Centre
R – Span Width
T – Asymmetry (CSA) Centre
Y – Asymmetry (CSA) Width
J – Alpha
K – Beta
L – Gamma

Useful Comments

When saving a file for loading into QuadFit it must be in one of two formats.
1:-Two columns separated by 2 spaces, the first is a column of shifts in ppm (in decending order) and the second is a column of data. There must also be an even number of points. This is the format that the "save_ascii" module in spinsight uses, as written by Dr Kevin Pike.
2:-The output format used by Topspin when saving as text file. The file must contain only real data.
If when opening a spectrum no data is shown, the data in the file might be in reverse order.
When running iteration, if you have too many parameters to iterate with too many steps then the program will do nothing. To get around this set the number of iterations to a higher number and reduce the number of steps. What this will do is to redo the iteration the "number of iterations" amount of times, thus increasing the accuracy, this is quicker than increasing the number of steps.
The chemical shift convention is the Herzfeld-Berger convention. This is explained and can be converted here: Chemical Shift Tensor Conventions
If you find the program is running out of memory, you can increase the maximum allowed by using the modifier -Xmx. i.e. "java -jar -Xmx512m QuadFit.jar" will give you a maximum of 512 megs allocated to the program.

The widths of the distributions of interaction parameters is the full width at half height.

Gaussian and Lorentzian broadening are both full width at half height.

Limitations

Can only have 100 lines in total

Still to be implemented:
Delete button needs to be implemented, for now either re-use the line or change its intensity to 0
Loading old fit and thus saving parameters

Bugs

When changing a line height from 1 to another value and back to one, the start and finish lines are not the same height.
This is caused because when you change the height value the line is broadened and then multiplied by the height, if no broadening has been applied the line shape is unchaged but due to not having some of the correction factors in the broadening routean the line changes height, this will get fixed at some point but is not a major issue at the moment.

When the program opens in windows you must click on it as if to change the size of the box.
When using the key shortcuts when in a focused text box the shortcut is appended to that box. Click on a button (delete is probably best as it does nothing) before using shortcuts.

References

II.D.1 P.109 Experimental Pulse NMR, A Nuts and Bolts Approach, E.Fukushima S.B.W.Roeder, Addison-Wesley Publishing Company 1981
Single-Crystal Cobalt 59 NMR Study of Tris(2,4-pentanedionato-O.O')cobalt(III) Klau Eichele et. al. J. Phys. Chem. A 1997, 101, 5423-5430

Department of Physics