Simultaneous calibration in flow-injection analysis using multiple-injection signals evaluated by partial least squares

In this contribution, a new calibration technique for flow-injection analysis is presented. The technique is based on a multiple-injection system complemented with multivariate evaluation. A standard, the sample and another standard solution are injected in a fast sequence. A partial mixing of the solutions occurs due to dispersion. However, the overlapping measurement signals can be deconvoluted reliably employing partial least-squares regression. The simultaneous calibration technique enables a fast adaptation to changes in the reaction system, while the time lost by rapid threefold injection and the signal evaluation is minimal. Applying simultaneous calibration the change of the sensor sensitivity is considered inherently. To test the simultaneous calibration technique, it was applied to measurements of a flow-injection system for the determination of glucose while the temperature of the reaction coil was changed.