Analytical thinking: Real-time analysis of granulation

Pharma Manufacturing News +

Tim Freeman, managing director, Freeman Technology on how to optimise granulation using real-time analysis. 

High shear wet granulation (HSWG) transforms fine powder blends into free-flowing granules with improved compression properties during the tabletting process. Granulation also enhances the uniformity of a blend, increases density and reduces dust levels, an important health and safety benefit. The pharmaceutical industry is one of several which rely heavily on HSWG processes, most especially in the manufacture of oral solid dosage forms.

A Quality-by-Design (QbD) approach in drug development and manufacture relies on data collection at relevant points in the process to control the critical quality attributes (CQAs) of the product. Granules are typically an intermediate, rather than the end product, making it more difficult to identify the critical process parameters (CPPs) that impact the CQAs of the finished product. Research has shown that HSWG processes can be successfully optimised using at-line techniques such as dynamic powder testing, but continuous, real-time analysis has the ability to provide further information by measuring the wet granulated mass, in-process, without interrupting the granulation operation.

The Lenterra Flow Sensor (LFS) System (Lenterra Inc. USA) is a Process Analytical Technology (PAT) that provides high frequency, high resolution in-line flow force measurement. In this article we look at how it works and examine its potential application in HSWG monitoring and control, via an experimental study. Correlation of in-line data with results generated in the lab using an FT4 Powder Rheometer (Freeman Technology, UK) suggests that both can be used at different stages of development and manufacturing to optimise HSWG processes.

The challenge of HSWG

During HSWG a blend of active ingredients and excipients are energetically combined with liquid, often water, to form relatively large, homogeneous granules. These undergo further processing to produce an optimal feed for downstream tablet manufacture. A typical objective for tableting would be to produce homogeneous granules that enable high throughput on the press and result in tablets with target CQAs.

The properties of granules are controlled through manipulation of a number of processing parameters, including:

  • Quantity of water added
  • Water addition rate
  • Impeller/hopper speed
  • Granulation time

Altering these variables enables the optimisation of granule properties and is usually a lengthy empirical process that relies on the implementation of statistical design-of-experiment (DoE) studies to correlate CPPs with the critical quality attributes of the manufactured granules. The resulting correlations tend to be scale-dependent, and process scale-up is a widely recognised challenge.

Producing granules with properties that are well-matched to the requirements of subsequent processes relies on being able to measure relevant properties during the course of the granulation. Dynamic powder testing, an at-line technique that measures bulk powder/granule flowability, has been shown to be valuable [1]. Its application enables characterisation of the developing granules, wet or dry, in terms of parameters that directly correlate with, for example, the CQAs of finished tablets [2]. The development of equally successful techniques for in-line measurement has the potential to deliver more responsive HSWG control and process scale-up.

In-line HSWG monitoring

A number of the PAT options proposed for granulation monitoring are based on the measurement of specific properties of the developing granules, for example, particle size. However, rather than focusing on individual particle parameters, which may or may not be relevant, measuring bulk properties, such as flowability, can have potential benefits.

The LFS system incorporates a drag force flow (DFF) sensor for in-line measurement of flow and an optical sensor interrogator for data processing and analysis. The sensor is a thin hollow cylindrical needle, in the order of 1-4 mm diameter, that can be mounted inside processing equipment such as a mixer, granulator or feeder, to provide real-time local measurement of the flow forces within the in-process material. Material flow causes a deflection of the pin, the magnitude of which is measured using two optical strain gauges, fixed to the inner surfaces of the sensor. The resulting measurements correlate directly with fundamental parameters of the material, such as density and shear viscosity, and can therefore be used to track the progression of a process, such as HSWG, that changes these attributes. Complementary temperature measurements enable the automatic correction of any temperature-related drift in measurement baseline.

Key advantages of this type of sensor are minimal intrusion into the flow, and relative insensitivity to the adherence of material on the sensor surface; an important attribute for robust continuous operation within a granulator. Furthermore, the probe has no moving parts, enhancing its inherent reliability, and measures at a sufficiently high frequency, up to 500 samples per second, to provide a high resolution data-stream that is able to successfully detect the evolving properties of the in-process materials. In a mixer or granulator it is typically installed through a standard port in the lid of the vessel, above the impeller and, like all in-line technologies, offers the opportunity to monitor the process without stopping and sampling. This is an important gain for process optimisation studies, and process control.

A DFF sensor reports data in the form of the Force Pulse Magnitude (FPM) which characterises the flow force associated with the passing of in-process material, and associated deflection of the pin. Changes in FPM therefore correlate directly with bulk properties of the process material - wet mass consistency and/or densification in the case of an HSWG process. FPM is a differential measurement and therefore not subject to baseline drift - an important additional benefit for process monitoring.

The following case study illustrates how this technology works and the relevance of in-line FPM measurements.

Case study: Monitoring HSWG with in-line and at-line technology

HSWG trials were carried out to investigate correlations between in-line flow force data gathered using an LFS system (Measurement range +/- 3N, Lenterra Inc, USA) and dynamic powder properties measured using an FT4 Powder Rheometer (Freeman Technology, UK).

Two kilogram batches of dry powder were made up according to the compositions shown in table 1 and granulated with 800g of water in a 10 L high shear wet granulator (Pharma-Connect®, GEA). Each granulation run consisted of 3 min of dry mixing followed by 3 min of water addition and up to 5 min of wet massing. The impeller and chopper of the granulator were switched on from dry mixing through to the end of the wet massing time with impeller tip speed kept constant at 4.8 m/s for all tests, and chopper speed maintained at 1000 rpm. These conditions were set with reference to previous optimisation studies [2].

The in-line sensor was installed via the granulator lid and was positioned 2.5 cm above the

granulator blade, 8.2 cm off the blade rotation axis (see figure 1). For each formulation, the granulation was stopped at a predetermined time after the start of water addition, to enable the gathering of three representative samples of the wet mass, for immediate at line measurement with the FT4 Powder Rheometer. All at-line dynamic testing was completed within 1 hour of removing the sample from the process. A new granulation was started with the same formulation to produce samples at later time points, giving six granulation batches in total (at 1, 2, 3, 4, 6, and 8 min) for each of the three formulations. In-line data were gathered for each of the batches (profiles B to G, fig 2). Data were also measured for the dry powder blend (profile A fig 2). All dynamic testing was carried out on the FT4 using standard methodologies.

Figure 2 shows how FPM varies as a function of time for each formulation. These profiles are generated using a rolling average over 300 blades to smooth the raw data; the passing of 300 blades corresponds to a time period of 20 s. The superimposition of the data for each batch demonstrates high reproducibility and clear differences are observed in the behaviour of the three different formulations. Flow energy data from the FT4 for the three formulations (figure 3) similarly highlights differences in their properties.

Both the in-line and at-line techniques successfully detect differences in granule wet mass consistency over time showing a rising profile during water addition, to a maximum that generally occurs soon after the end of water addition, with subsequent decay of the signal afterward. There are close similarities in the profiles generated, especially for the datasets produced at an HPC concentration of 5%, and both techniques show a similar trend with respect to the effect of HPC concentration. The magnitude of the rise associated with water addition is relatively small in terms of flow energy relative to FPM, particularly for the lower HPC concentrations. For these formulations BFE values appear to peak during water addition, while the FPM profile peaks shortly after water addition is complete however, the data demonstrate how both techniques can be used to quantify properties of the wet mass in order to track the granulation process.

Looking ahead

Wet granulation is a much valued process in the pharmaceutical industry, but is notoriously difficult to control and scale-up. There is considerable value in developing analytical techniques that can reliably monitor HSWG processes. It is increasingly recognised that the quality of granules is not easily defined in terms of a single granule property but rather is a function of a number of parameters including density, porosity, surface adhesion, and size. Techniques that focus on bulk properties, as a function of water content, formulation or process changes, therefore have considerable potential.

These studies provide further evidence of the value of dynamic powder testing and in-line flow force measurement for monitoring HSWG processes. While dynamic powder testing is an at-line technique, flow force measurement has been successfully implemented as an in-line sensor, via the LFS system.  This PAT tool for real-time continuous measurement offers robust technology for routine HSWG monitoring and control during scale-up and into manufacture.