Optimization of Acoustic-Resonance Spectrometry for Analysis of Intact Tablets and Prediction of Dissolution Rate

When a new spectrometric method is introduced in the literature, the method is investigated initially with instruments with a number of different designs. Each new application of the spectrometric method requires a set of trial experiments to optimize the instrumental parameters. Eventually the different applications and instrumental designs assist in establishing the fundamental advantages and limitations of the spectrometric method. The requirement for trial experiments and optimization extends to existing instruments that are used in a manner that deviates significantly from the literature. In cases where the spectrometric method and physics are not well defined, such as ultrasonic-resonance spectroscopy,’ it may be difficult to predict the effects of instrumental modifications in an application. For example, a change that increases the sensitivity of an instrument to ultrasonic absorbance in a sample may decrease sensitivity to variations in ultrasonic velocity in the same sample. Trade-offs must be made, often between parameters with different and apparently incompatible units (such as absorbance and velocity). The Universal Numeric Optimization (UNO) introduced in this report allows simultaneous optimization of different parameters (even parameters with different units) by means of a nonparametric multivariate algorithm. The first reported ultrasonic-resonance spectrometer consisted of a V-shaped quartz rod with two identical ceramic piezoelectric transducers (one transducer for