Pick of the bunch — How multivariate powder characterisation assists with comprehensive datasets

Doug Millington-Smith, principal applications specialist at powder characterisation company Freeman Technology, considers the value of multivariate powder characterisation as a solution to the demand for comprehensive data.

Well-established techniques for material characterization may not provide the differentiation required to quantify variations in batches of excipients that contribute to variation in downstream process behaviour. A fundamental part of quality by design (QbD) is the need to establish a design space of acceptable raw material properties which requires understanding of a diverse range of excipient properties.

The importance of excipients to patients and producers

Patients clearly want to be treated with effective drugs, but this is not their only concern. Any therapeutic agent is more acceptable to those who rely on it if the side effects are minimized and the delivery system is as palatable as possible. If drugs taste better, and are easy to take, patients are more likely to continue using them. For instance, large tablets are often coated to make them easier to swallow, flavours and sweeteners can be used to mask unpleasant tasting active ingredients, and colours may be added to improve the aesthetics and aid identification. Optimising these properties through the management of excipients is important in helping patients adhere to their treatment schedule.

Excipients are equally important in pharmaceutical manufacturing, where their properties can be as vital to therapeutic performance as those of the active ingredient. Excipient properties are key to aspects such as product stabilisation and shelf-life, for example. The manufacturer can also choose to boost the effectiveness of the active ingredient by manipulating excipient properties to enhance solubility, reduce viscosity and increase absorption, among other things. Powder and solid doses, in particular, use a wide range of excipients to solve manufacturing problems. Anti-adherents, binders and lubricants are used to provide mechanical strength, prevent the ingredients from agglomerating and to protect tablets from sticking to the punch and die faces during processing.

Many excipients have been used for decades and their general characteristics closely defined. However, inter-batch variability can lead directly to variability in processing resulting in reworking or even scrappage. With recent QbD initiatives making it necessary to optimise production processes to ensure the consistency and reliability of final products, a robust method of quantifying the variations in batches of excipients that contribute to differences in downstream process behaviour is an absolute necessity if a design space of acceptable raw material properties is to be established. However, even well-established techniques for material characterisation (such as particle size analysis) do not always provide the required differentiation and typically only evaluate one physical property of the particles.

Reproducible properties measurement

The FT4 Powder Rheometer (Freeman Technology, UK), uses powder characterization techniques and fully automated test protocols to deliver a comprehensive and reproducible database of process-relevant powder characteristics.

As well as characterising the rheology, or flow properties, of powders, the inclusion of a shear cell allows the powder’s shear strength to be determined and quantification of how a powder shears with respect to the surfaces of process equipment can be undertaken using a wall friction kit. Bulk powder properties, such as density, compressibility and permeability, can also be measured.

The benefits of such an approach are easily illustrated. Consider a pharmaceutical manufacturer sourcing alternative supplies of an excipient based on a specification that covers only particle size and bulk density. A new material matching this specification is sourced but, when fed into the plant, problems arise and efficiency plummets: blockages become commonplace and final product quality is inconsistent.

In a case such as this, the excipient specification is clearly inadequate, incorrect, or insufficiently detailed, and does not reflect and reference all the powder variables that impact in-process behaviour. Analysing the conditions throughout the manufacturing process and identifying the key properties that correlate with these conditions is necessary.

An example of multivariate analysis of bulk excipient batches

Microcrystalline cellulose (MCC) has been used as a pharmaceutical excipient for many years, due to its abundance, ease of production and resistance to degradation. As with any excipient, however, variability exists between batches of supplied material that can, in turn, lead to similar variability in processing and the finished product.

Particle size analysis was undertaken to test the three grades MCC (sample A, B and C) used as an excipient in the production of pharmaceutical tablets by direct compression. Data generated illustrated almost identical D₅₀ values (100 µm) for each grade, and it was determined that all three samples were indistinguishable from each other in this context.

All three samples were further analysed using an FT4 Powder Rheometer and the results are presented below.

Dynamic testing: Aeration

Sample A and sample B exhibited similar responses, whereas Sample C generated a significantly higher Aeration Ratio (AR). This demonstrates that Sample C’s packing structure changes to a greater extent when air is introduced into the sample and typically indicates a lower degree of cohesive strength between particles.

Bulk testing: Permeability

Again, sample A and sample B were similar whereas sample C generated a considerably higher pressure drop across the powder bed, indicating that it is the least permeable of the three. Low permeability means that any air that becomes entrained in the bulk is less able to escape. Considering the example of a tabletting process, greater air content within the dose would typically lead to weight variation as well as capping and lamination in the final product.

Shear cell testing

Sample A exhibited markedly different behaviour when compared with the other two samples. It generated the highest shear stress values at low normal stress levels, but the lowest values when subjected to higher stress levels. This suggests that the way the powders perform in a given process would be heavily influenced by the stress levels they are subjected to, and demonstrates the need to characterize powders using process relevant techniques.

It is also interesting to consider the results for sample B. The yield locus, generated by a best-fit line through the data points, is so steep that it intercepts the y-axis below the origin. This leads to a negative value for cohesion, and generates no result for flow function (as the minor Mohr’s circle cannot be constructed to produce a value for unconfined yield strength). This illustrates the risks of relying on parameters derived from Mohr’s circle analysis alone, as the mathematical model used to obtain these parameters may not always provide data for comparison.

In summary, sample A exhibited the highest permeability and the lowest sensitivity to Aeration, and was less sensitive to changes in applied stress during shear cell tests. This suggests that it is the most efficiently packed of the three samples and is likely to exhibit more free-flowing behaviour across a range of conditions. Sample C generated the highest shear stress values, was most sensitive to aeration, and exhibited the lowest permeability, suggesting a greater sensitivity to the process conditions than the other samples and illustrating the need to understand the relationship between the process environment and the material properties.

The vital importance of powder flowability

In quantifying clear and repeatable differences between three grades of MCC , the advanced powder characterization capabilities of the FT4 identified likely differences in their response to various process conditions.

Although production conditions can be rigorously standardised and monitored, variations in the finished product can still occur. As a consequence of batch-to-batch variations, it is not uncommon for batches of the ‘same’ powder to perform differently in a process.

For pharmaceutical manufacturers, these results demonstrate the importance of understanding the physical characteristics of a powder which correlate with good performance within their particular process, and perhaps just as importantly, those that are associated with problematic performance or poor product quality.

Successful processing demands that powder and process are well-matched. Multivariate testing delivers relevant data that can be matched with process ranking to produce a design space of parameters that correspond to acceptable process behaviour. Excipient suppliers and users now have the opportunity to intelligently and reliably select excipient batches that exhibit the characteristics necessary for a specific manufacturing process.