MATLAB scripts for ICPOES data reduction from Qtegra raw data.

Ángel Rodés, SUERC (2021)
angelrodes.com

These files must be in the same folder:

• ICP data in .csv
• ICPOESdatared_vX.m
• linear_regression_chisq_fn.m
• normrnd_BoxMuller.m

1. After running the Thermo ICP-OES, export data including Raw.Average, Raw.STD and ExtCal.StandardConcentration in a .csv file with semicolon (;) as separator:

Raw.Average;(...);Raw.STD;(...);ExtCal.StandardConcentration

See example file: AR20200205Aliquots.csv

You can export other data too. They will be ignored.

2. Modify data in ExtCal.StandardConcentration columns if needed. E.g. if you want to use some standard data not stated in Qtegra.

3. Run ICPOESdatared_vX.m in MATLAB/Octave and select your file. A lot of graphical output will appear, and a text file (.txt) containing a few tables will be created (concentrations, calibrations, LOD/LOQ; by analyte, by element, etc.).

v2

• Some code was added to transform Qtegra format to standard ICP format (comma separated values with one analyte per line)

v2.4 (2021.09.08)

• If Raw.STD is not included, Raw.RSD is used instead (Raw.STD=Raw.Average*Raw.RSD/100)
• Samples named 'blank" or 'blk' are considered blanks
• Standard's nominal values ara copied to all locations if they have the same name.

v3 (2022.01.21)

• Lots of comments added.
• Two new functions defined (normrnd_BoxMuller and normpdf_local) to replace normrnd and normpdf just in case the statistic package is not available in Octave.
• Some redundant code was deleted.
• New dialog allows the user selecting standards and blanks.
• Maximum chi-squared formula was simplified to min-chi-squared+DOF in linear_regression_chisq_fn.m to avoid using functions form the statistic package. (original code is commented).