Linear Fitting of Time-Varying Signals in Static Noble Gas Mass Spectrometry Should Be Avoided.

This Perspective highlights a malpractice during data processing in static noble gas mass spectrometry, viz. retrodicting the noble gas signals at "time zero" based on the linear fitting results of the time-varying analytical responses. Linear fitting is the most commonly used by noble gas analysts mainly on the consideration of high coefficient of determination (R), while it poses risks of inaccurate estimation of required information. Here, we appeal to re-examining the signal variation mechanisms before deciding the way of fitting instead of merely choosing one with good R. The primary processes during the static mass spectrometry measurement are considered, and a mechanism-based exponential fitting is recommended for the relevant data processing. For the same given data set, exponential fitting is a more robust processing strategy than linear fitting because not only does the former show perfect goodness of fitting but also it contributes to better consistence of retrodicted signals. For the sake of maintaining a high level of data quality in analytical science, we propose to apply the mechanism-based exponential fitting to retrodict the "time zero" signal in static noble gas mass spectrometry, and the commonly used linear fitting should be avoided.