Powders are widely acknowledged as being particularly challenging and unpredictable. For example, they may not flow consistently through the different stages of a manufacturing process and certain blends may be more prone to attrition or segregation.
High Shear Wet Granulation (HSWG) is a critical step in the pharmaceutical industry, widely used to transform the multiple constituents of fine powder blends into homogeneous free-flowing granules ideal for downstream processing and, especially, oral solid dosage form production. Since granulation enhances blend uniformity, increases density, reduces the opportunity for segregation and, importantly, benefits health and safety through reduced dust levels, it is unsurprising that the pharmaceutical industry relies extensively on this process. Furthermore, granulation also improves flow and compression properties that help optimize processes such as tableting.
Importantly, granules are typically not the end product in drug development and manufacture. Rather, they are an important but intermediate step in the production process. This can make it more difficult to identify the critical process parameters (CPPs) that impact the critical quality attributes (CQAs) of the finished product, and more challenging for a Quality-by-Design (QbD) approach that relies on the collection of data at all relevant points in the process.
Furthermore, the manufacture of pharmaceutical solid dosage forms is increasingly focused on becoming a continuous process, or at least a series of continuous operations, rather than the traditional batch process. In this regard, integration of HSWG and associated drying techniques, such as an agitated fluidized bed, can be seen as part of a switch to continuous manufacture.
While at-line techniques, such as dynamic powder testing, are already successfully employed for optimizing HSWG processes, this article explores the application of a continuous, in-line analytical technique to evaluate the wet granulated mass in real-time, a key benefit for both continuous and batch manufacture. It considers the importance of HSWG, introduces the concept of Drag Force Flow (DFF) measurement, and examines its potential for monitoring and controlling the HSWG process. An experimental study is also presented that investigates in-line DFF data with lab-based data produced using an FT4 Powder Rheometer® and suggests that the two techniques are highly complementary in optimizing development and production, while DFF potentially offers enhanced process control opportunities.
The Importance of HSWG
HSWG energetically combines a formulated blend of active ingredients and excipients with liquid, usually water, to form homogeneous granules with properties suited to further downstream processing. For instance, in tableting the objective may include producing homogeneous granules with properties that enable high throughput and which yield tablets with desired CQAs.
The HSWG process parameters that impact granule properties include the amount of water added to the powder blend, the rate at which the water is added, the impeller and/or chopper speed, and the granulation time. Altering one or more of these variables changes the properties of the granulate produced. Understanding the role played by each of the variables usually involves time-consuming empirical studies designed to correlate the CPPs with the CQAs of the granule. Dynamic testing of the flow properties of the powder/granulate during the granulation process has been shown to produce valuable data that correlate granule properties with the finished tablet CQAs.1 However, scaling lab-based or pilot scale studies to process plant is a widely recognized challenge and additional in-line, real-time characterization is likely to be extremely beneficial both in monitoring/controlling HSWG and implementing successful scale-up.
Drag Force Flow for Real-Time Monitoring of HSWG
Figure 1. Drag Force Flow SensorA Drag Force Flow (DFF) sensor is a thin, hollow, cylindrical needle, approximately 1-4 mm diameter, which provides real-time measurement of the local flow forces within the in-process material when mounted inside mixers, granulators, feeders, or other processing equipment (Figure 1).
As powders or granules flow against the needle it is deflected, and the magnitude of this deflection is captured using two optical strain gauges fixed to the inside walls of the needle. The gauges are composed of Fiber Bragg Gratings (FBGs) that shift in relative spectra in proportion to the amount they are compressed or stretched by the needle’s deflection, with the signal transferred by optical fibers to an interrogator that assesses the response. As the degree of deflection is measured in-line and in real-time, it allows operators to immediately determine the attributes of the material being processed and whether adjustments are required.
DFF measurements correlate directly with fundamental parameters of the material, such as density and shear viscosity. In the high shear wet granulator, DFF will track the progressive change in these attributes as the wet mass densifies into granules and the force of the deflection increases. DFF signals are reported as a Force Pulse Magnitude (FPM), a differential measurement which is not subject to baseline drift, and complementary temperature measurements enable the automatic correction of any temperature-related variation.
Figure 2. In-line sensor installation within a granulatorSensor sensitivity is defined by the length, diameter, and material of the needle, with tip deflection as low as one micrometer being detectable. The sensors employ no moving parts, have no material traps, and are relatively insensitive to material build-up on the sensor surface. Also, the small diameter of the sensor offers minimal intrusion to the flow of material and high frequency measurement rates of up to 500 samples per second are achievable. Crucially, this high frequency, high resolution, in-line flow sensor system provides a data stream without having to stop the HSWG process to sample for off-line measurement (Figure 2).
Case Study: Monitoring HSWG with In-Line And At-Line Technology
Figure 3. FT4 Powder Rheometer®Trials were undertaken to determine whether in-line drag force flow (DFF) measurements can be used to track the granulation process in the same way that off-line dynamic powder properties are currently used for assessing granule development. The HSWG study investigated the relationships between DFF data gathered using an LFS system (measurement range +/- 3N and measurements of basic flowability energy (BFE) using an FT4 Powder Rheometer® – Figure 3).
Methods
Tests were carried out on two kilogram batches of three placebo pharmaceutical formulations produced with different levels of hydroxypropyl cellulose (1% w/w HPC, 3% w/w HPC, and 5% w/w HPC). Each batch was granulated with 800g of water in a 10 L high shear wet granulator (Pharma-Connect®). The in-line data was gathered using a DFF sensor mounted in the granulator lid, positioned 8.2 cm off the blade rotation axis and 2.5 cm above the granulator blade (see Figure 2).
Each granulation run consisted of 3 minutes of dry mixing, followed by 3 minutes of water addition and up to 5 minutes of wet massing. The impeller tip speed and chopper speed of the granulator were maintained for all the batches, and from dry mixing through to the end of the wet massing phase for each individual batch, at 4.8 m/s and 1,000 rpm respectively. These conditions were set with reference to previous optimization studies.2
At-line dynamic measurements of BFE using the FT4 Powder Rheometer were conducted for each of the three formulations by stopping granulation at a predetermined time after the start of water addition (at 1, 2, 3, 4, 6, and 8 minutes ) to enable the gathering of three representative samples of the wet mass. All dynamic testing was completed within 60 minutes using standard methodologies. Data for later time points was gathered using samples from new granulation batches of the same formulation.
In-line DFF measurements were gathered during each of these granulations runs (Profiles B to G, Figure 4) for each of the HPC concentrations plus the dry powder blend (Profile A, Figure 4).
Figure 4. Real-time DFF data clearly show the change in FPM measurements as the consistency of the granulating mass changes with the addition of water (a) 1% w/w HPC b) 3% w/w HPC c) 5% w/w HPC).Results
The derived FPM values varied with respect to time reflecting the change in consistency of the granulating mass during the process (Figure 4). Repeat tests also exhibited high levels of repeatability. The FPM profiles shown in Figure 4 are generated using a rolling average over 20 s to smooth the raw data; with each 20 s corresponding to 300 passes of the blade. Superimposing the data for each batch demonstrates high reproducibility and the three formulations produced comparable profiles. The initial work (shear) to mix the dry powders did not result in significant changes in FPM, however, FPM rises rapidly as water is added and larger, denser, less compressible, and more adhesive granules are generated. A peak in FPM is observed shortly after the end of water addition, after which FPM declines as the continued mixing generates smaller granules. It was also observed that FPM increases with respect to HPC percentage, suggesting that higher binder concentration results in stronger, denser, and larger granules.
Comparing the DFF profiles with the BFE profiles generated from the FT4 tests (Figure 5), it can be observed that the BFE profiles support the FPM data. Again, a rising profile is seen during water addition with a subsequent decay as the addition of water ends. Also, as the content of the HPC binder increases, BFE measurements also increase. Data gathered from the highest concentration of HPC binder (5%) showed close similarities between the two techniques The sensitivity of DFF measurement is illustrated by the magnitude of the increase in signal associated with water addition compared to the corresponding increase in BFE. For the formulations with the two lower concentrations of HPC binder, BFE values appear to peak during water addition. In comparison, the BFE values for the highest concentration of HPC and all three FPM profiles peak shortly after water addition is complete. In general, the data demonstrate how both techniques can be used to track granule development in HSWG processes.
Figure 5. Comparison of in-line FPM measurements from the DFF sensor with at-line BFE measurements from the FT4 Powder Rheometer.Combining In-Line and At-Line Measurements
The case study presented demonstrates the value of at-line dynamic powder characterization and in-line DFF measurements for monitoring and controlling a HSWG process.
The data build on previous studies that have shown how at-line powder rheology can provide valuable information for optimizing HSWG processes. Granule quality is a function of several properties, including density, porosity, surface adhesion, and size, rather than a single granule parameter. Techniques that evaluate properties of the bulk with respect to process variables such as moisture content, formulation changes, or process settings therefore have considerable potential.
In-line FPM measurements have been demonstrated to correlate with dynamic powder properties measured off-line, illustrating how variations in granule properties can be monitored in real-time during an HSWG process in order to generate granules suitable for downstream processing.
HSWG is a valued process across many manufacturing sectors, including the pharmaceutical industry, but is recognized as challenging to control and scale-up. The robust, real-time, continuous measurement capability offered by DFF sensors for routine process monitoring and control is potentially invaluable.
References
- Freeman, T. (2014) In Pursuit of Wet Granulation Optimization. Pharmaceutical Manufacturing.
- Narang, AS. (2016) Process Analytical Technology for High Shear Wet Granulation: Wet Mass Consistency Reported by In-Line Drag Flow Force Sensor Is Consistent with Powder Rheology Measured by At-Line FT4 Powder Rheometer®. Journal of Pharmaceutical Sciences. 105:185-187