PAT SUPPLEMENT


the real manufacturing environment in order to avoid misinterpretations. The following examples will justify the


argument that process understanding plays a significant role in designing robust PAT analysers. Raman spectroscopy is preferred for measurements in aqueous systems compared to NIR spectroscopy as water is a weak Raman scatterer. The homogenisation process in an aqueous pharmaceutical suspension was monitored in-line using Raman Spectroscopy10


.


Raman spectroscopy was also utilised to quantify the amount of API in the suspension10


.


Various complex analytical set ups have been studied and found useful in monitoring crystallisation processes11


. However, owing to


improvements in process understanding even simple techniques can be utilised with excellent efficacy. For example, the Dow Chemical Company has used in-situ fibre optic turbidity probe technology for monitoring and control of unseeded crystallisation in a commercial batch crystalliser12


. Using these probes, they were able


to monitor the stages in the unseeded crystallisation process, namely nucleation, digestion and crystal growth. The process of crystallisation was controlled by adjusting process variables to follow a predetermined turbidity signal profile over the process time.


analysers and implementation of the PAT principles in


“ The development of PAT


pharmaceutical manufacturing is a herculean task ”


mechanism for correcting the process variables responsible for causing the values to deviate from the set limits. Most of the literature studies have been performed in isolation, considering only one or few of the above mentioned points. List and Steffens13


have demonstrated the use of


NIR spectroscopy for the in-line solvent content determination during granulation. Further, Rantanen et al.14


determination of moisture content during fluid showed that in-line NIR


Data acquisition, analysis and feedback mechanism The data sets generated during in-line or on-line monitoring can be extensive, hence data acquisition can be a tedious job. Utilising this data to make real time decisions is a further challenge. This demands a robust interface to continuously convert the generated data into meaningful values, definition of the acceptable limits for these values and the feedback


bed granulation can be achieved through the simultaneous detection of the NIR signal at only four wavelengths. An entire NIR spectrum is not necessary for determination of water enabling reliable and high-speed detection of moisture. The data generated is intrinsically multivariate in nature and the successful conversion of the collected data depends on the appropriate selection of numerical strategies which can reveal the chemical information or the process signature and depends on the level of process understanding. Chemometric techniques are very commonly utilised for this purpose. A further challenge is to ensure that all software is suitable for the specific method of analysis and can be validated. The development of analysers suitable for


use in industrial processes pose other practical problems to ensure that the devices are robust, meet safety requirements (for example for use in the presence of potentially flammable atmospheres) and the avoidance of interference by deposits on optics. In conclusion, the use of PAT can yield


dramatic quality improvements, time savings, and cost reductions. However, the development of PAT analysers and implementation of the PAT principles in pharmaceutical manufacturing is a herculean task requiring expertise in various


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17