Process analytical technology (PAT) provides real-time monitoring for pharmaceutical processes and products, offers valuable process information, and helps maintain high product quality. Utilizing inline Raman spectroscopy measurement and particle imaging, critical in-process parameters of a salt break process were identified and ranges explored. To prepare intermediate-grade API, a salt break was performed on the sodium salt of the API using HCl under low temperature. Raman trends showed gradual change during acid addition, and a rapid, sharp change during the subsequent heat-up to room temperature. The sharp change in the Raman trends was accompanied by a similarly quick change in habit, observed via inline particle imaging. The gradual reaction progression during HCl addition indicates the salt break reaction was kinetically limited at low temperature. And during the heat-up, the kinetic limit was removed, leading to quick neutralization of the sodium salt. The fast reaction/crystallization may not be ideal during scale-up since it poses difficulties in controlling the crystallization process. A laboratory-scale attempt was made to perform acid addition under room temperature for possible energy conservation, and to prevent sudden transformation to the API during the heat-up. However, XRPD results of the resulting product indicated the presence of unconverted sodium salt. Both Raman and particle imaging results showed only gradual changes in spectra trends and habit over the course of the acid addition. These PAT measurements indicated the room-temperature run is under a HCl-limiting condition, leaving only a fraction of salt particles in reaction environment. As the acid crystal grows, some fine salt particles become trapped in the aggressively growing acid crystal. Absence of sodium salt contamination when adding HCl at low temperatures could be explained by the kinetic limitation on the reaction during acid addition, making the reaction slurry well mixed until the start of reaction take-off and reducing the chances of particle entrapment. As a result of these findings, the process was recommended to proceed with low-temperature HCl addition towards scale-up, and plant-scale inline Raman was recommended to be used in the pilot plant.
At the pilot plant scale, the temperature range during HCl addition was initially relaxed for ease of operation and control. However, one pilot plant batch of this process was again found to contain a small amount of unconverted sodium salt in the product. The batch with the sodium salt impurity showed Raman trend resembling that of the room temperature laboratory-scale run. Meanwhile, batches that showed full conversion by XRPD had Raman trends similar to the laboratory-scale runs with the low-temperature hold. The Raman results indicate the impurity in the plant-scale run could be from operating at or above the upper limit of the temperature range during the HCl addition. As a result, the temperature range during acid addition was again restricted, and no sodium salt impurity was found under XRPD in any of the subsequent batches.
