(203f) Assessment of Powder Tribo-Charging during Twin-Screw Feeding Operations and Its Sensitivity to Environmental Conditions

The tribo-charging effects during pharmaceutical manufacturing can significantly impact process performance and final product quality. During twin-screw feeding operations, powder tribo-charging can cause particle adhesion to the feeder hopper and at the screw outlet, thus affecting feeding accuracy and consistency^1,2. Pharmaceutical powders can get charged during feeding processes by the numerous particle collisions and separations within the powder bed and with the feeder surface. The severity of the tribo-electrification depends on several factors, including environmental conditions and intrinsic material attributes of the contacting surfaces³. Relative humidity (RH) has been reported to be the environmental factor with a prominent impact on tribo-charging. However, the underlying mechanisms of its effect are not yet fully clear. For diverse material types, it has been reported that there is a specific RH threshold (per material type) after which a reduction in charge magnitude occurs⁴. This study investigates the effect of material-water interactions (induced by changes in RH) and their impact on the powder tribo-charging tendency. The investigated powders were additionally evaluated for their processability during twin-screw feeding with a particular focus on the tribo-charging effects occurring during feeding.

MATERIALS AND METHODS

Materials

Four pharmaceutical powders were investigated in this study: D-mannitol (MAN, Parteck^®, Merck), microcrystalline cellulose (MCC, Avicel^®, DuPont), micronized paracetamol (PMIC, Mallinkrodt) and magnesium stearate (MgSt, Parteck^® LUB, Merck). For the excipients (i.e., MAN and MCC), two different grades were used (i.e., M100 and M200 for MAN; PH101 and PH200 for MCC), whereas for the API (i.e., PMIC) and the lubricant (i.e., MgSt) only one grade was considered.

Methods

All the powders were homogenized and sieved to break large agglomerates prior to the storage at ambient temperature (i.e., 22 ± 2 °C) in desiccators at three levels of RH (i.e., 23, 53 and 75% RH) for 48-72 h. The amount of water present in the samples was determined via loss-on-drying (LOD) and Karl Fischer (KF) titration. Particle size distribution (PSD) was analysed via laser diffraction for fine powders (i.e., MgSt and PMIC) and via dynamic image analysis for the coarse ones (i.e., MAN and MCC grades). Likewise, particle shape was determined via static and dynamic image analysis. The specific surface area (SSA) of the powders was measured via nitrogen physi-sorption and the powder bulk properties were characterized with the FT4 powder rheometer, equipped with different available accessories. For the evaluation of the powder charging tendency at the different RH levels, the GranuChargeâ„¢ apparatus (GranuTools, Belgium) was used. The electrostatic charge of the powders was measured after letting flow through the stainless-steel tubes. With regard to the twin-screw feeding experiments, a small-scale feeder (MT-S Hyg, Brabender, Germany) was used to alternately feed the material on a scale (i.e., for mass flow determination) and in a Faraday cup (i.e., for charge measurement). Feeding runs were performed at controlled environmental conditions (i.e., 22 ± 2 °C and 53 ± 4% RH) and the charge acquired after 5 and 12 min from the process start was measured. Statistical analysis was performed to identify relevant correlations between the material charging tendency and both environmental conditions and material attributes.

RESULTS

Powder characterization

Powders were characterized by distinct particle and bulk properties as well as different levels of hygroscopicity. Particularly, MCC and MAN powders consisted of considerably large particles characterized by high aspect ratios and narrow particle size distributions. In contrast, smaller particles with lower aspect ratios and broader distributions were observed for MgSt and PMIC. After storage of the powders at different RH levels, a significant increase in the water content was observed for the hygroscopic MCC grades. For the other non-hygroscopic materials, slight increases were noticed only at high RH (i.e., 75% RH). The increase in the moisture content of the materials at progressively higher RH induced changes in the material bulk properties. For instance, a consistent increase of wall friction angle (WFA) was observed for MAN powders from FT4 measurement, which was found to correlate with a progressively higher charge density measured.

Impact of RH on powder tribo-charging behavior

The measured charge was normalized by the specific surface area of the samples to allow an appropriate comparison among the materials. All the powders were found to charge negatively after the contact with stainless steel, however distinct charge magnitudes were obtained for the different materials at three levels of RH (Figure 1a). Overall, PMIC exhibited the highest propensity to charge, followed by MCC and MAN grades, whereas a minimal charging tendency was observed for MgSt. RH was found to impact the tribo-electrification of the materials to different extents and distinct tribo-charging trends per material type were observed, namely: i) MCC powders showed an increase in charge density from 23 to 53% RH, followed by a pronounced decrease at higher RH (i.e., 75% RH); ii) MAN grades showed a progressive increase in charge density with the increase in RH; iii) PMIC exhibited a decrease in tribo-charging propensity only at high RH; iv) MgSt had a slight decrease in charge density with increasing RH.

Powder tribo-charging during twin-screw feeding operations

All the powders, except MgSt, showed a negative charge polarity also during twin-screw feeding operation (Figure 1b). Moreover, for these materials charge values remained relatively constant between the 2 measurement times (i.e., 5 and 12 min from the process start). For MgSt, an initial negative charge was obtained, which decreased over time and turned positive during the second measurement. This observation was attributable to the particular lubrication mechanism of this material, which allows the material to form a lubrication layer on stainless-steel surfaces and consequently causes an alteration to particle contacts. When comparing the charge values measured after feeding with the one obtained from the GranuChargeâ„¢ device at the same RH (i.e., 53% RH), distinct trends were observed. Particularly, the charge density was found to be lower for the excipients (i.e., MAN and MCC grades), whereas PMIC showed to charge to a higher extent. This is in agreement with qualitative observations for this powder with a higher tendency to adhere to surfaces, such as feeder hopper, outlet and screws.

CONCLUSION

This study demonstrates that, depending on the material type, RH impacts powder tribo-charging behavior distinctly. The amount of water sorbed and its distribution depends on intrinsic material properties and thus, affects material bulk properties such as powder tribo-electrification to a different extent. Additionally, the type of unit operation considered can also impact the magnitude and polarity of the acquired charge by changes in the surface contact characteristics and contact mode. Both effects should therefore be considered during the formulation and process development to avoid possible risks in production lines.

REFERENCES

(1) Muehlenfeld, C.; Thommes, M. Small-Scale Twin-Screw Extrusion - Evaluation of Continuous Split Feeding. J. Pharm. Pharmacol. 2014, 66.

(2) Stauffer, F.; Vanhoorne, V.; Pilcer, G.; Chavez, P. F.; Schubert, M. A.; Vervaet, C.; De Beer, T. Managing Active Pharmaceutical Ingredient Raw Material Variability during Twin-Screw Blend Feeding. Eur. J. Pharm. Biopharm. 2019, 135, 49–60.

(3) Naik, S.; Mukherjee, R.; Chaudhuri, B. Triboelectrification: A Review of Experimental and Mechanistic Modeling Approaches with a Special Focus on Pharmaceutical Powders. Int. J. Pharm. 2016, 510, 375–385.

(4) Schella, A.; Herminghaus, S.; Schröter, M. Influence of Humidity on Tribo-Electric Charging and Segregation in Shaken Granular Media. Soft Matter 2017, 13, 394–401.