The pharmaceutical industry has long struggled with the problem of controlling the characteristics of a drying mixture, a critical step in the production of drugs and chemical compounds. Currently, two noninvasive characterization methods are commonly used: a sample is imaged and individual particles are counted, or researchers use scattered light to estimate the particle size distribution (PSD). The former is time-consuming and creates more waste, making the latter a more attractive option.
In recent years, MIT engineers and researchers developed a physics- and machine-learning-based scattered-light method that has been shown to improve pharmaceutical pill and powder manufacturing processes, increasing efficiency and accuracy and resulting in fewer failed product batches. A new open-access paper, “Noninvasive estimation of dust size distribution from a single speckle image”, available in the magazine Light: science and applicationsextends this work by introducing an even faster approach.
“Understanding the behavior of scattered light is one of the most important topics in optics,” said Qihang Zhang, PhD ’23, a research associate at Tsinghua University. “By advancing the analysis of scattered light, we also invented a useful tool for the pharmaceutical industry. Finding the problem and solving it by investigating the fundamental rule is the most exciting thing for the research team.”
The paper proposes a new PSD estimation method, based on pupil engineering, that reduces the number of frames required for analysis. “Our learning-based model can estimate the dust size distribution from a single speckle image, thereby reducing the reconstruction time from 15 seconds to just 0.25 seconds,” the researchers explain.
“Our main contribution in this work is to speed up a particle size detection method by 60 times, with collective optimization of both the algorithm and hardware,” says Zhang. “This high-speed probe is capable of detecting size evolution in fast dynamic systems, providing a platform to study process models in the pharmaceutical industry, including drying, mixing, and blending.”
The technique offers a low-cost, noninvasive particle size probe that collects backscattered light from powder surfaces. The compact, portable prototype is compatible with most drying systems on the market, provided there is an observation window. This in-line measurement approach can help control manufacturing processes, improving efficiency and product quality. Furthermore, the previous lack of in-line monitoring prevented the systematic study of dynamic models in manufacturing processes. This probe could provide a new platform for conducting serial investigations and modeling for particle size evolution.
This work, a successful collaboration between physicists and engineers, was born out of the MIT-Takeda program. The collaborators are affiliated with three MIT departments: Mechanical Engineering, Chemical Engineering, and Electrical Engineering and Computer Science. George Barbastathis, professor of mechanical engineering at MIT, is the senior author of the paper.
