Investigation of a drying process of needle-shaped particles using particle size analysis techniques and non-invasive Raman spectrometry

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2011
Thesis identifier
  • T13019
Qualification Level
Qualification Name
Department, School or Faculty
  • The attrition of needle-shaped cellobiose octaacetate (COA) particles during drying was studied using particle size analysis and in situ spectrometry techniques. Firstly, three particle size analysis techniques were evaluated for their efficacy relating to needle-shaped particles. Laser scattering techniques provided qualitative information on average particle size, however, it was shown that the Feret Max dimension obtained from dynamic image analysis was the best indicator of needle length, and allowed quantitative observations to be reported. A bespoke lab-scale drier was used to determine the effect of varying the process conditions on drying time and the extent of attrition using two approaches, with noninvasive wide illumination Raman measurements recorded in situ. A design of experiments was carried out to determine the effect of three drying parameters on the total drying time and extent of attrition. All the parameters had an effect, however, attrition was mainly affected by the agitation strategy. Variation of the pressure filtration time prior to vacuum agitated drying produced a range of particle wetness that allowed the relationship between particle wetness, agitation and attrition rates to be examined. In situ Raman spectrometry allowed the drying curves from both the solvent and the particles to be monitored in real-time. The magnitude of the Raman signal of COA during drying was affected by changes in the bulk density of the particle bed. By interpretation of the COA drying curve, it was possible to detect the onset of aggregation and then the break-up of the aggregates to powder. It was not possible to estimate the particle size from the Raman spectra during drying as changes in the particle bed's bulk density dominated the variation in the Raman signal. In the final part, the learning from the laboratory experiments was applied successfully to an industrial case study at GSK.
Resource Type
Date Created
  • 2011
Former identifier
  • 947886