Thermoluminescent Glow Curve Deconvolution

On this page a few scripts can be found which can be used for a quick and coarse glow curve deconvolution. The scripts were written using the octave scripting language, which is a free open-source version of MatLab. (The .m files presented here may work with MatLab as well.)

Data Preparation

Currently this package consists of multiple files. First, the raw GC curve data - assuming a format like [bin] [counts] - are loaded with
data = load("my_gc_curve.dat");

afterwards the data are formatted to a array of 701 elements, where the element number is equivalent to the temperature in Kelvin. The value of the element is the number of counts. This is done with the script [gc_dataprep.m]. i.e. data = gc_dataprep(data);

The data can be plotted with
axis([350 700]);
plot(data);

this image shows a glow curve acquired from a TLD-700 tablet.

Deconvolution

The actual deconvolution is done with the script [gc_deconv.m] which also needs the kernel function [gc_kernel.m] based on first order kinematics, see reference [1].

yy = gc_deconv(data);
This files also prints the characteristic temperature and the integral of the peaks found.

With plot((1:701),data,(1:701),yy,'b'); one can see how well the deconvolution worked out.

Additional Deconvolution Scripts

Since I had hundreds of glow curves I had to deconvolute, I automated the process with two additional scripts:
[deconv_script.m] and [deconv_script.sh].

You have to adjust the path in the shell script to the place where your .m files are located. Then execute the main script devonv_script.sh "*.dat" in order to deconvolve all files ending with ".dat".

References

[1] G. Kitis, J.M. Gomez-Ros and J.W.N. Tuyn, "Thermoluminescence glow-curve deconvolution functions for first, second and general order kinetics", J. Phys. D: Appl. Phys. 31, (1998) 2636-2641.

21.12.2004 - Niels Bassler - http://www.phys.au.dk/~bassler