Turbidity and Endoscopy Assisted Monitoring of Solvent Mediated Polymorphic Transformation Processes

Polymorphism is a feature which characterizes most of the active pharmaceutical ingredients and influences the solubility, shape and color of crystals. The process of solvent mediated polymorphic transformation is a series of events such as crystal dissolution, nucleation and growth and several on-line PAT sensors have been evaluated to characterize the transformation process by monitoring the solid or liquid phase properties. 

The focus of this work is to compare low-cost on-line PAT sensors with more complex instruments such as the FBRM, FT-Raman, ATR-FTIR and FT-NIR spectroscopy. The target of the investigation is to evaluate whether these probes are able to detect changes in solid concentration due to dissolution-nucleation-growth events and it is not expected that these give information about the ratio of the polymorphic forms. In the family of low-cost sensors are included the turbidity sensor and the bulk video imaging method which can be implemented using an endoscope. Since the turbidity sensor is already found in many industrial crystallizers and it is used to detect nucleation and growth its evaluation for polymorphic transformation monitoring is obvious. Moreover, note that no additional investment and installation costs occur. Recently, GSK presented a case-study where turbidity and FT-Raman spectroscopy was successfully used to follow the polymorphic transformation of an API [1].

 The endoscopy based process monitoring is a low-cost PAT tool in evaluation phase [2]. It is based on imaging and recently it was proposed for nucleation detection applications. Depending on the image analysis method used it can detect the first formed crystals as objects; alternatively, when the mean gray intensity of the images is calculated its operation is more similar to that of the turbidity probes. Note that one image can provide the equivalent information of several thousands of spatially distributed turbidity probes. In this work the endoscopy images are analyzed using texture analysis based methods. Simple image texture descriptors have been previously evaluated in the form of mean, minimum and maximum intensities [3,4] for the purpose of nucleation detection. Since the turbidity probe operates at 880 nm which is in the NIR range FT-NIR spectroscopy was used to gain further information about the operation of this probe. This work presents a quantitative evaluation of the detection sensitivity of the dissolution and nucleation/steps of the sensors mentioned above. Furthermore, parameters and sensor design issues which influence the monitoring sensitivity are discussed in detail.

References:

[1] Barnes, S., Gillian, J., Diederich, A., Burton, D., Ertl, D., In Process Monitoring of Polymorphic Form Conversion by Raman Spectroscopy and Turbidity Measurements, American Pharmaceutical Review, http://americanpharmaceuticalreview.com/ViewArticle.aspx?ContentID=3386.

[2] Simon, L. L., Nagy, Z. K., Hungerbuhler, K., 2009b, Endoscopy-Based in Situ Bulk Video Imaging of Batch Crystallization Processes, Organic Process Research & Development, Special Issue on Polymorphism and Crystallization, 13: 1254-1261.

[3] Simon, L. L., Abbou Oucherif, K., Nagy, Z. K., Hungerbuhler, K., 2010, Bulk Video Imaging Based Multivariate Image Analysis, Process Control Chart and Acoustic Signal Assisted Nucleation Detection, Chem Eng Sci, 65: 4983-4995.

[4] Simon, L. L., Nagy, Z. K., Hungerbuhler, K., 2009a, Comparison of External Bulk Video Imaging with Focused Beam Reflectance and Ultra Violet-Visible Spectroscopy for Crystallization Nucleation Detection and Metastable Zone Identification, Chem Eng Sci, 64: 3344-3351.