(314d) Evaluation of Segregation Intensity of Pharmaceutical Blends Using Near Infrared Spectroscopy

Segregation of components, especially of an active pharmaceutical ingredient (API), in a pharmaceutical blend during secondary manufacturing processes is considered to be one of the major bottlenecks in achieving tablet content uniformity, ultimately leading to batch rejections. The main reasons for segregation such as differences in particle size, density, shapes, solid-solid interactions and segregation induced due to external interferences such as gravitational, shear or vibrational forces are well established, but there are limited tools available to quantify the API segregation intensity at a manufacturing scale. Determining segregation intensity using spectroscopic technique can help in identifying segregation risks early on in product development. This research work focusses on quantifying segregation intensity of pharmaceutical blends using SPECtester (an instrument with an in built near-infrared (NIR) capability).

The instrument has two different modules capable to induce segregation to the pharmaceutical blends. The first module uses vibrational force and the second module uses compressed air to fluidize the particles in the system. In the first module, a simulating condition to that of large scale manufacturing process was created by dropping the blend from the highest point (high drop chute -16.69”)of the equipment into the NIR testing section of the instrument. Around 0.8 to 1.25 liters of the blend is required to fill the hopper. The material was dropped from the hopper through the feed chute to form pile using vibratory feeder. The speed of material flow in the chute was controlled by regulating the frequency of vibration. In the second method, the blends to be analyzed were placed in the Fluidization Analysis Module (FAM) test hopper. Appropriate fluidization settings (the flow required to fluidize the material) were selected and it was confirmed that materials reached fluidized bubbling regime. The pile height of material after fluidization were recorded and fed in the SPECTester software. The NIR analysis methodology in both these segregation modules were followed in the same way. The system allows the measurement of blend containing up to 6 components. Trays filled with individual components were placed adjacent to the testing hopper section. These component trays were used for collecting NIR spectra of pure samples of the components in the mixture. The instrument’s operational parameters such as pile height, the size of the view box (window) to scan, and number of measurements in a square box were optimized to suitable values to acquire segregation intensity data. Spectra of the blend was taken about 6 mm below the pile surface horizontal to the top. SPECTester primarily measures the concentration of components at these discrete number of square spaces in the poured pile and collectively computes the segregation intensity. The instrument superimposes the spectra of pure components on the spectra of components of the segregated blend and the concentration of components in the blend in each of these data points or measurements were then computed by tracking the changes in the absorption intensities in each of these spectrum.

Formulation blends using different types of APIs with different particle size ranges were explored. 200gm of blend weight was found to be adequate to ensure reliable data collection. Blends of these APIs with the excipients were prepared by mixing the constituents for 15min in a Turbula blender (This formulation composition was also used to make tablets in a manufacturing scale by direct compression process and to further study its content uniformity).The segregation induced due to the above mentioned modules were then evaluated for individual APIs and excipients at different API concentrations (w/w) ranging from 2 to 15%. The concentration profile of formulation components (along the pile length) were also determined. API segregation intensity (obtained using SPECTester) for corresponding formulation blends were compared with tablets content uniformity produced in the manufacturing scale.

The measured segregation potential corroborated with the content uniformity results. In other words, formulation blends with high segregation potentials generated tablets having higher deviation from the content specifications. Hence, this small scale prediction tool assessed the segregation potentials of formulation blends in an early formulation development phase and also proved its ability to predict the segregation issues experienced at large scale manufacturing.