Abstract

Particle density and size distribution are critical process parameters in many of today's industrial processes that involve dispersed phase process streams. For example, granulation, crystallization, emulsion polymerization, and liquid-liquid extraction lack the appropriate monitoring techniques to facilitate process control and optimization. Optical sensors involving dynamic light scattering, turbidity, and backscatter have been proposed in the past. Yet these optical sensors are restricted to process streams that are either relatively non-scattering by industry standards, or require calibration to the process streams which they are intended to monitor. The latter is especially detrimental for successful implementation of optical sensors in the chemical based industries. When process disturbances occur, the optical sensor is intended to provide important and accurate information to the controller for corrective action. If the calibration of the sensor is invalidated due to changing process stream composition (as might happen during process disturbances), then the accuracy of the feedback signal may cause catastrophic results if actuation is based upon it. Thus, a robust, self-calibrating optical sensor is required for optimization of many processes in the chemical-based industries.

© 1996 Optical Society of America