Developing a process analytical technology for monitoring the particle size distribution in twin screw granulation
Bibliographic record
Abstract
Twin screw wet granulation (TSG) has been studied as a continuous manufacturing alternative to batch granulation for nearly twenty years. One of the main differences between batch granulation and TSG lies in the exiting granules being presented as a bimodal particle size distribution (PSD) in the latter case. Current process analytical technologies (PAT) can monitor a monomodal distribution well but there have been no techniques disclosed in the public domain so far that can accurately monitor this unusually shaped PSD. Acoustic emissions (AE) has been identified as a PAT of interest due to its ease of use (lack of calibration), low cost, and non-invasive design relative to other PATs used for monitoring PSDs. Hence the goal of this thesis was to develop AE as a process analytical technology (PAT) capable of estimating the full distribution of produced granules by TSG in real time. The first research study of this thesis focused on the development of the new technology. The AE PAT consisted of an acoustic sensor, an impact plate, and software to convert the time-domain signal of particle collisions into a time-averaged frequency-domain spectrum to be subsequently used to estimate a weight-averaged particle size distribution. A novel and much required addition to the PAT was inclusion of a digital filter based on particle mechanics parameters to overcome auditory masking which hindered accurately converting the cumulative sounds of impact into a PSD. The PAT was tested in this study with granulated lactose monohydrate and with the new digital filter, obtaining a maximum error of 1 wt% across all particle sizes tested. In the second research study, as more formulations commonly used in the industry were tested, the filter proved unable by itself to account for the differences in impact mechanics and therefore needed to be modified to incorporate the more inelastic behaviour now being seen. Two micromechanical models were explored, and the Walton-and-Braun model was found to be the most suitable for the AE PAT – reducing its error from 8 wt% down to 2.75 wt% across four formulations producing coefficients of restitution from 0.79 to 0.24. In the last research study in this thesis, the now-functional inline PAT was used to reveal mechanistic details related to the transition state in granulation as a TSG starts up, to improve the field’s understanding of the granulation mechanism. The technique was able to estimate the PSD over much shorter periods of material collection compared to sieving, allowing the evolution of the PSD as a function of time to be examined for varying degrees of fill (DF) and liquid-to-solids ratios. It was determined that the time to steady state, at both DF tested, occurred at approximately 5 times the mean residence time of the process by both PAT and sieving analyses. Particle sizes between 102-2230 μm were then tracked as a function of time below 120 s and variations of granule growth were seen for each degree of fill which added to the understanding of the granulation mechanism. This PAT shows great promise as a monitoring tool to implement quality by design principles for TSG in pharmaceutical manufacturing.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".