In roller compaction, particle distribution is recognised as one of the most critical parameters affecting downstream process performance and product quality.
A roller compaction process is designed to yield consistent downstream tablet compression resulting in uniform dissolution and content uniformity. A successful process produces a granule with consistent particle size distribution, density and porosity control. However, inconsistencies occur during granulation scale-up due to an in change raw materials or process dynamics.
Collaboration with Patheon demonstrates FBRM at-line to map design space and optimize a series of roller compaction runs while varying vertical/horizontal feed speed, roller compaction force and mill speed. Characterising particle distribution allows users to link process control parameters directly to product quality. By designing a robust process, consistent processing from dry granulation to tablet compression is achieved.
experiment design
A 19 batch DoE was performed in order to understand processing parameters affecting downstream product quality. FBRM technology was used to measure and control changes in particle count and dimension. FBRM is typically inserted in-line in a collection funnel downstream of the Comil and powder flows over the probe tip providing a representative measurement due to measuring in-line or at-line within concentrated particle systems, increasing sample size and providing high sensitivity to fine particles.
In this case, an at-line method was used. Ten grams of powder was sampled downstream and dispersed in 100g of mineral oil. Due to concentrated sample size, a representative measurement was achieved. Sample repeatability was <1% on the median (50th percentile) statistic.
results
The pre-blend distribution had fewer coarse particles compared with the distributions after roller compaction and milling. Trial 10, 12, 13 and19 had the highest number of fine particles, high porosity and density. They also corresponded to 4000lb/inch roller force and 1000 rpm mill speed. The fines population is an early indication of downstream flow properties and possible dissolution inconsistency. Trials 6 and 11 had the highest number of coarse particles, low porosity and density. They also corresponded to 8000lb/inch roller force and 2000 rpm mill speed.
statistical results
Particle distribution mean, number of fine particles counted per second (0-50μm) and number of coarse particles (200-2000μm) are early predictors of high sensitivity to compact porosity and changes in upstream roller compaction force and mill speed parameters. The particle mean and number of fine and coarse particles counted per second may also be early indicators of downstream flow and dissolution rate or disintegration time. In general, roller compaction force significantly affected ribbon density, porosity compacts and milled compact particle size.
mean vs porosity correlation
By measuring granule dimensions in real time, roller compaction process conditions can be controlled to target specific mean dimensions. Since mean dimensions correlate to granule porosity, there is an opportunity for real time control ensuring consistency.
conclusion
Roller compaction is a complex process with competing mechanisms of breakage and agglomeration. Using Mettler Toledo’s FBRM, it is possible to quantify the effect of critical process parameters and correlate this to ribbon parameters. By characterising these effects, tools (FBRM) can be used to reduce scale-up time, minimise upsets and troubleshoot issues. In this study, high roller compaction force and mill speed resulted in coarser particles with lower porosity and density, whereas low compaction force and mill speed corresponded to a higher level of fine particles, porosity, and density. Inline particle characterisation is also used to identify screen breaks and hardware malfunctions reducing manufacturing costs.
