Research paper
Scale-Up of pharmaceutical Hot-Melt-Extrusion: Process optimization and transfer

https://doi.org/10.1016/j.ejpb.2019.07.009Get rights and content

Abstract

Hot-Melt-Extrusion on Twin-Screw-Extruders has been established as a standard processing technique for pharmaceutical products. A major challenge is the transfer from a lab to a production level, since the combination of several unit operations within one apparatus leads to complex conditions for such a continuous manufacturing process. Here the residence time distribution is a crucial measure, which reflects the different mechanisms, e.g. dissolution, mixing or degradation, during processing. In the first part of a Scale-Up study, a methodology for the optimization of an extrusion process with respect to the load and throughput is presented. The developed concept was applied for different extruder scales in order to compare the identified processing windows. A deviation of the dominant material heating mechanisms was observed for the different scales, while the constraints for the transfer of a process to a different scale by the developed optimization concept is demonstrated. Finally, a sufficient operating point on a reference extruder is identified and in the second part of this study, different concepts from literature are applied for the transfer of this Hot-Melt-Extrusion process to two larger scales. The focus of the investigations was on the impact of the different approaches on the residence time distribution and the comparison. The determined results revealed a change of the most sufficient approach for the two different extruder sizes. The impact on the location in the time domain and form of the distribution are discussed and additionally evaluated by the fit to a RTD-model. In conclusion, the ratio of the applied energy for transport to mixing is identified as valuable addition in this context.

Introduction

Within the last decade Twin-Screw-Extrusion (TSE) has been a focus of research interest within the field of pharmaceutical technology [1], [2]. A key aspect in this context is a shift of batch production to continuous manufacturing [3], [4], which is driven by the overall aim of process optimization and an enhanced cost efficiency [5]. One approach to this is the specific design of production processes based on the correlation of desired quality attributes and process parameters. The corresponding concept Quality-by-Design is based on the work of Juran from 1992 [6].

In this context, TSE offers a vast potential. Due to the modular set-up [7] and the implementation of different screw element types [8], [9], [10] several unit operations can be combined within one apparatus. This reduces the machine footprint and invest costs. At the same time, this continuous technology is robust, which reduces set-up times and enhances the overall process stability. Therefore, Hot-Melt-Extrusion (HME) with co-rotating Twin-Screw-Extruders has been a standard process for the production of solid dispersions [11]. This specific dosage form is an approach for one major challenge of the pharmaceutical industry nowadays: the bioavailability enhancement of poorly water soluble drugs [12], [13].

Solid dispersions consist of a drug dispersed within a matrix, typically a polymer [14]. During HME the dissolution of the active pharmaceutical ingredient (API) within the carrier melt is crucial. An indicator for this mechanism is the residence time distribution (RTD) [15], since the on-set is a surrogate for the minimum contact time of drug and polymer melt. Additionally, the width of the distribution indicates the backmixing of material [16]. This is related to the distributive mixing performance, which is essential for product homogeneity. Finally, the off-set of the RTD symbolizes the maximum duration of thermal and mechanical stress applied to the processed material [17]. Overall, this is linked to degradation processes [18], [19]. In consequence, the RTD reflects critical mechanism for extrusion.

Another major challenge for pharmaceutical technology and TSE is the Scale-Up [20], [21] of such a production method from a lab to an industrial level. From an economic point of view, the process should be optimized regarding throughput during early stage development level in a first step in order to ensure a maximum degree of utilization for the machine. During Scale-Up the maximizing of the output should prevent unnecessary investment cost and during later production. In this context, the fill level [22] is crucial, since it is a measure for the utilization of an apparatus.

At the same time, the throughput and fill level have a direct impact on the RTD [23], which is influenced for a constant formulation by machine as well as process parameters, e.g. the screw configuration, the free volume inside the extruder, the screw speed, mass flow or temperature profile. This complexity contributes to the general challenge of a process transfer from a development stage to an industrial production level. Several Scale-Up concepts for TSE are presented within literature and can be classified in approaches related to geometric or energetic aspects. However, these are not focusing on the RTD and the variation of this parameter by the scale transfer.

The aim of this study is on the one hand to present a systematic methodology for enhancing the workload of a TSE process and finally identify a sufficient process parameter set with respect to various limitations. These are either linked to material properties, e.g. the degradation temperature, or the performance of the extruder itself, e.g. the drive power. The systematic approach is used for three machine sizes of extruders in order to highlight the universality of it, while at the same time general constraints for the application are identified. For reasons of comparability, the screw configuration was kept geometrically constant as well as the temperature profile for all experiments. Finally, the impact of maximizing the throughput on the residence time is demonstrated and compared for different extruder sizes. Therefore, the RTD is determined inline via UV–VIS spectroscopy.

The second focus of this study is to characterize the effect for different Scale-Up concepts from literature on the RTD and the individual potential for the preservation of this critical process parameter. The approaches from literature are applied for a reference point on the smallest extruder to larger machine sizes at two levels. The screw configuration was kept geometrically constant in order to emphasize the same unit operations and maintain the general process mechanisms. Correlations between direct process parameters (mass flow, screw speed) and the impact on the RTD as an indirect process parameter during Scale-Up are identified and additionally characterized by the fit to a model and the corresponding model parameters.

Section snippets

Hot-Melt-Extrusion on co-rotating Twin-Screw-Extruders

The experiments were carried out on three different intermeshing, co-rotating Twin-Screw-Extruders (Leistritz, Nuremberg, Germany) from the ZSE-series with a nominal external screw diameter Da of 17.8 mm (ZSE 18), 27 mm (ZSE 27) and 39.7 mm (ZSE 40). For all applied extruders the ratio of Da to the inner screw diameter Di was in the range of 1.5 and the relative length L of the processing section was 36 Da (see Table 1). Screw configuration and temperature profile are based on a common set-up

Scale independent optimization strategy

A generalized approach for the optimization of a TSE process was developed based on the concept of Kolter et al. [29], which has been adjusted with respect to temperature management and process stability.

The main focus of this Scale Independent Optimization Strategy (SIOS) is the throughput expressed by the mass flow ṁ and the workload. These are linked by the specific feed load (Eq. (3)), which is a dimensionless surrogate for the fill level of the extruder and is defined as the ratio of mass

Conclusion

Within this study a general strategy was presented for optimizing a Twin-Screw-Extrusion process with focus on maximizing the feed load and the throughput. This concept was applied for three different extruder scales. Therefore, the screw configuration as well as the barrel temperature profile was kept constant. The obtained results revealed similar operating windows with comparable limitations related to the maximum feed load, which was restricted by the intake mechanisms. While the

Declaration of Competing Interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

References (34)

  • K.B. Bischoff et al.

    Fluid dispersion—generalization and comparison of mathematical models—II comparison of models

    Chem. Eng. Sci.

    (1962)
  • M. Wilson et al.

    Hot-melt extrusion technology and pharmaceutical application

    Therapeut. Deliv.

    (2012)
  • I. Major et al.

    Hot melt extruded and injection moulded dosage forms: Recent research and patents

    Recent Pat. Drug Deliv. Formul.

    (2015)
  • M. Teżyk et al.

    Recent progress in continuous and semi-continuous processing of solid oral dosage forms: A review

    Drug Dev. Ind. Pharm.

    (2015)
  • S.L. Lee et al.

    Modernizing pharmaceutical manufacturing: from batch to continuous production

    J. Pharmaceut. Innovat.

    (2015)
  • L.X. Yu et al.

    Understanding pharmaceutical quality by design

    AAPS J.

    (2014)
  • J.M. Juran

    Juran on Quality by Design: The New Steps for Planning Quality into Goods and Services

    (1992)
  • Cited by (28)

    • Determining local residence time distributions in twin-screw extruder elements via smoothed particle hydrodynamics

      2022, Chemical Engineering Science
      Citation Excerpt :

      For example, in pharmaceutical Hot Melt Extrusion (HME), the process RTD was used to identify the affected product in the event of an production anomaly or disturbance (Gao et al., 2012; Engisch and Muzzio, 2016). Moreover, studies related to the design, scale-up and optimization of pharmaceutical HME identified the process RTD as a critical process parameter that strongly affects the product quality (Kolter et al., 2012; Liu et al., 2010; Thiry et al., 2015; Matić et al., 2019; Wesholowski et al., 2019). Often, the product quality depends on several kinetic processes that differ from application to application.

    View all citing articles on Scopus
    View full text