(668c) Modeling of Residence Time Distribution in Continuous Solid Oral Dose Pharmaceutical Manufacturing Processes
- Conference: AIChE Annual Meeting
- Year: 2014
- Proceeding: 2014 AIChE Annual Meeting
- Group: Pharmaceutical Discovery, Development and Manufacturing Forum
- Time: Thursday, November 20, 2014 - 1:20pm-1:45pm
Modeling of Residence Time Distribution in Continuous Solid Oral Dose Pharmaceutical Manufacturing Processes
M. Sebastian Escotet-Espinoza, Amanda Rogers, Fernando J. Muzzio and Marianthi Ierapetritou
Department of Chemical and Biochemical Engineering, Rutgers University, Piscataway, NJ
The pharmaceutical regulatory environment has recognized the need for a systematic approach to process development through the implementation of engineering modeling and optimization techniques in order to manufacture safe and effective products with their quality built into the process . This philosophy of Quality-by-Design (QbD) has been applied in other industries to reliably create products with the desired quality attributes. Quality is assured through the development of reliable and robust processes, designed based on the knowledge of process principles. The transition from batch to continuous pharmaceutical operations can facilitate the development of processes within the QbD framework. Continuous processing allows real-time control, mitigates the need for scale-up studies and can help to reduce product variability . In continuous manufacturing processes the major challenges are the coordination of equipment input-output properties, and the formulation of predictive process control strategies. To achieve equipment coordination and predictive capabilities, the relationships between critical quality attributes (CQAs), material properties (CMPs) and process parameters (CPPs) need to be correlated sequentially between the multiple unit operations in product manufacturing. Mathematical models characterizing individual unit operation performance as a function of CPPs (e.g., screw speed, residence time) and CMPs (e.g., particle size distribution) have been developed. In addition, multi-unit processes can be characterized with respect to the temporal-space location of particles in the system .
The residence or space-time is a critical equipment and process parameter that characterizes the period of time a particle stays in one or more unit operations in a continuous system. The residence time distribution (RTD) is defined as the probability of particles exiting the equipment at a given time period based on the flow patterns inside of the unit . Mathematical models for the RTD have previously been developed for chemical engineering applications in order to characterize the influence of operating conditions, material properties and unit geometry on the degree of non-ideal behavior (i.e., back-mixing). Pharmaceutical unit operations for which the RTD has been experimentally studied include feeding, blending, and granulation processes . In previous work, the RTD of convective mixers was extensively studied and correlated to lumped and discrete parameters such as the Peclet number and effective particle diffusivity .
In this work, we focus on developing the mathematical modeling tools for RTDs in continuous solids-handling unit operations as a predictive tool for process outcomes and characterization of unit performance. Our research focuses on modeling the residence time distribution in feeding (i.e., feeders) and blending (i.e., convective mixers) operations as a function of operating conditions. The overall goal of the research is to model the RTD using historic and recently developed RTD models. Data from both pulse and step change experiments for the convective blender have been reduced to fundamental unit operation parameters (i.e., Peclet number) using the Fokker-Planck Equation (i.e., PFR with axial mixing). The effective diffusivity of the system has been studied with respect to CMPs of acetaminophen-excipient blends in order to establish an initial method for correlating material properties to unit operation. Critical unit operation parameters (e.g., mass hold-up, blade speed) and blend properties (e.g., composition, particle size distribution) can then correlated to in order to develop new RTD methods. The relative standard deviation (RSD) of the convective mixer with respect to the pulses experiments has been studied in conjunction to the RTD in order to establish a predictive control strategy for the unit operation. RTD can be used to characterize the propagation of transient disturbances across sequential unit operations in flowsheet simulations. The RTD can provide traceability along the continuous line and delineate continuous products based on the feed strategy. Manufactured products created using different batches of starting material can be traced in order to comply with process validation. This will facilitate the prediction and rejection of out-of-spec (OOS) products by tracing disturbances along the process. Overall, RTD models can add significant value to the transition from batch to continuous operations by providing reliable predictive tool for behavior of unit operations based on their residence time.
1. USFDA, Pharmaceutical cGMPs for the 21st century: a risk-based approach, U.S.D.o.H.a.H. Services, Editor. 2002: FDA, Rockville, MD.
2. Boukouvala, F., et al., Computer-Aided Flowsheet Simulation of a Pharmaceutical Tablet Manufacturing Process Incorporating Wet Granulation. Journal of Pharmaceutical Innovation, 2013. 8(1): p. 11-27.
3. Gernaey, K.V. and R. Gani, A model-based systems approach to pharmaceutical product-process design and analysis. Chemical Engineering Science, 2010. 65(21): p. 5757-5769.
4. Danckwerts, P.V., Continuous flow systems: Distribution of residence times. Chemical Engineering Science, 1953. 2(1): p. 1-13.
5. Gao, Y., F.J. Muzzio, and M.G. Ierapetritou, A review of the Residence Time Distribution (RTD) applications in solid unit operations. Powder Technology, 2012. 228(0): p. 416-423.
6. Gao, Y., et al., Characterizing continuous powder mixing using residence time distribution. Chemical Engineering Science, 2011. 66(3): p. 417-425.