(M1430-06-32) Transferring Laser Diffraction Particle Size Determination Methods in a Regulated Environment

Purpose: The particle size distribution of particulate systems in the sub-micron and micron range can be measured using the principle of laser diffraction. Samples are tested as either wet or dry dispersions. The exact methodology used to test a sample is specific to both the sample type and instrument setup. Phase appropriate method development must be performed to ensure subsequently generated data is accurate and reliable, and this method development should follow a systematic approach. One such approach being adopted in GMP (Good Manufacturing Practice) environments is ‘Quality by Design (QBD) for Analytical Methods’ as described in 2018’s ICH Q14: ‘’Analytical Procedure Development’’. The QBD for Analytical Methods approach encourages the pre-definition of objectives for method development in the form of an ‘Analytical Target Profile’ (i.e. the performance requirements of the method which must be met) and associated ‘Critical Method Attributes (CMAs)’ (i.e. those analytical parameters and conditions which significantly impact method performance).  This study highlights how QBD for Analytical Methods is not only limited to method development, as it can also be used to support method transfer between different generations of particle size analyser.

Methods: A method transfer of a particle size determination method for a test sample of milk powder from a Mastersizer 2000 with Scirocco dispersion unit to a Mastersizer 3000 with an Aero S dispersion unit was targeted. The method attributes related to a dry dispersion particle size measurement with these laser diffraction systems were summarized in an Ishikawa diagram. The method attributes expected to differ between the different generations of particle size analyser were selected on a risk assessment basis. The dispersion characteristics (dispersion pressure and geometry) were selected as the ‘Critical Method Attributes’. Particle size measurements were performed with the Mastersizer 2000 with Scirocco, Mastersizer 3000 with Aero S/standard venturi and Mastersizer 3000 with Aero S/high energy venturi at the following dispersion pressures: 0.1, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0 bar. The Dv50 particle size results were compared against a result from a reference wet measurement.

Results: At low pressures, a reduction in the Dv50 particle size with increasing pressure is observed for all three instrument configurations. This is consistent with dispersion and de-agglomeration of the milk powder test sample. The Dv50 results for the Aero S/standard configuration are consistently higher than the corresponding result for either the Scirocco or Aero S/high energy configurations. This is due to the Scirocco and Aero S/high energy configurations both comprising a 90° bend to promote more energetic ‘particle-wall’ collisions. This feature is missing from the Aero S/standard configuration and so collisions are mostly of the ‘particle-particle’ type. At 1 bar of air pressure, the Dv50 results from the Scirocco and Aero S/high energy configurations are consistent with the wet dispersion reference measurement. However, as the air pressure exceeds 1 bar the Dv50 results fall below the wet dispersion reference result. This indicates that particle attrition is now occurring and air pressures exceeding 1 bar on these systems are too aggressive. The Dv50 results for the Aero S/standard configuration only drop to the level of the wet dispersion reference result at approximately 3 bar. A further increase in dispersion pressure to 4 bar shows a very slight reduction in particle size.

Conclusion: A Quality by Design for Analytical Methods approach was successful in identifying the dispersion pressure and dispersion geometry as Critical Method Attributes for experimental investigation. Experimentation to investigate the impact of dispersion pressure and dispersion geometry highlighted that accurate particle size results could be achieved on the Mastersizer 3000 either by using the standard energy venturi between 3 to 4 bar or using the high energy venturi at 1 bar. Use of the standard energy venturi at 3.5 bar would be the most robust selection for progressing with the method transfer exercise. Future work to verify the new method and implement suitable control strategies would be the next steps.

An Ishikawa or 'fish bone' diagram to visualise the various analytical conditions and parameters impacting on the particle size results generated from a dry dispersion laser diffraction measurement. The conditions and parameters are sub-divided by type: analyst, instrument, measurement, environment, material and method. Dispersion pressure/geometry is highlighted in red, owing to its nature as a 'Critical Method Attribute'.



A schematic of the two venturi types investigated. The top schematic illustrates the design of the standard venturi used by the Aero S, where material passes along a linear path. The bottom schematic illustrates the design of the Scirocco or high energy venturi used by the Aero S, where material passes through a 90 degree bend. This bend encourages 'particle-wall' collisions, thus achieving a more energetic dispersion.



A plot of the 50th undersize percentile size (Dv50) versus dispersion air pressure for the different dry dispersion instrument configurations. A line to indicate the wet dispersion result is shown across the pressure range to signify the expected particle size and identify target air pressures required to achieve the correct dispersion.