 |
|
Introduction
Thrust B: Design, Scale up and Optimization of Manufacturing Processes#Top
Turning designer particles into designer delivery forms
Thrust Leader: James D. Litster (Purdue)
Project Leaders: Fernando Muzzio (Rutgers), Boris Khusid (NJIT),
Jim Litster (Purdue), Alberto Cuitiño (Rutgers)
A. Thrust Objective and Goals
The vision for this thrust is to bring the manufacturing processes that produce controlled structure delivery forms to the point of clear, quantitative engineering design through
· development of new design models that predict delivery form properties as a function of process variables and
formulation properties, and
· development of new or improved process designs for continuous manufacturing.
B. Technical scientific barriers/grand challenges for Thrust B
There has been considerable progress in our quantitative understanding of powder processing in the last 10 years. We have reasonable macroscopic constitutive equations for many systems, good understanding of macroscopic mechanisms and dramatically improved characterization tools from the nanometer to millimeter scale. However, the current knowledge base still falls well short of good design models that allow us to predict the product attribute distributions at a manufacturing scale in a quantitative manner with minimal pilot scale experimentation. In addition, current processes have poor controllability due to (1) multiple rate process occurring together the unit process, and (2) very broad velocity and stress fields. Therefore, the design space in terms of achievable combinations of product attributes is severely limited.
There are a number of barriers to progress:
- Quantitative understanding of how microstructure is developed during processing of complex multiphase delivery forms is lacking.
- There is an absence of predictive and quantitative scaling rules between macroscopic properties and single particle properties.
- Process design models that track multidimensional distributions of properties are still relatively rare.
- Carefully designed and executed experiments at laboratory scale and in test bed pilot plants to validate the design models are generally absent.
- Robust on-line techniques for measuring microstructure and distributions of important properties are lacking.
- Current equipment designs are based on heuristics, rather than in depth process understanding and batch process equipment is not easily converted to continuous processing.
- In some industries (eg. pharmaceuticals), are resistant to changing existing equipment and practices is strong.
C. Strategic Plan and Thrust Connectivity for Thrust B
The framework for a predictive design model is illustrated in Figure 2P-9. The model must incorporate the important physics of the processes so that the impact of changes to feed formulation, process parameters and process scale on the product attributes can be predicted. The application of the physics to the process requires well characterized flow and stress fields within the processing equipment through some combination of discrete and continuum modeling. The integrated design model will then incorporate a series of rate equations for the key transformations within a macroscopic framework, such as a population balance, to predict distributions of product attribute distributions and/or microstructure. Figure 2P-9 also illustrates how the development of such models is inherently linked to the objectives of thrusts A, C, and D.
Experimental validation of the models is at two levels. The sub-models are validated with elegant and simple laboratory scale experiments eg. compaction experiments under very controlled stress and strain conditions; flow visualization experiments for validating DEM or CFD models. On the other hand, the test bed pilot plants will be used to validate the integrated models and establish the design space for the processes.
The model development described above will highlight limitations of current processes eg. inability to independently control two different product properties; inability to achieve an attribute in the desired range; inability to maintain key attributes on changing production scale. Overcoming these limitations may be possible through process modifications driven by a clear understanding of both the required product outcomes and the physics of the processes involved. For example, modification of the feeding system in a roll compactor might completely eliminate cross ribbon density variation and eliminate completely dust generation in the dry granulate. Such changes are designed to radically move the design space for the processes. The sub-models developed for process design of existing equipment can be used to test new design ideas in silico before prototypes are designed and built within the test bed pilot plants.
D. Key Scientific Deliverables for Thrust B
- Improved quantitative understanding of microstructure development during manufacture for multi-component, multiphase delivery forms based on careful characterization of single particle property distributions and excellent understanding of velocity and stress profiles in the manufacturing processes.
- New process design models to reduce the need for extensive pilot scale experiments and reduce development time for new products. These models will predict product properties as a function of formulation particle properties and process parameters.
- New techniques for on-line measurement of microstructure and other property distributions.
- Carefully designed and executed experiments at laboratory scale and in test bed pilot plants to validate the design models.
- Reduced order models suitable for use in real time process management.
- Improvements to existing process designs for better product quality, improved design space and better operability.
- Completely new processes to produce product structures and property distributions that are currently difficult, or even impossible, to achieve.
- Individual projects’ scientific focus is designed to help establish the knowledge base of the thrust, hence, this thrust makes strong contributions to the bottom and middle planes of the 3-plane chart as shown in Figure 2P-10. However, the project deliverables also serve to support specific test bed (shown in the top plane in Figure 2P-10) needs in addition to the thrust level goals.
Figure 2P-11 shows 15 major milestones for Thrust B addressing scientific and technological deliverables listed above. The current list of projects support processes associated with test bed 1 and test bed 2. Deliverables and milestones in the next two to three years are associated with these projects. Starting in approximately 2011-12, we expect several of the existing projects to be replaced by projects supporting personalized medicine through drop on demand (test bed 3), and extensions to test bed 1 eg. continuous wet granulation. The center intends to develop a new test bed from 2012, and projects in Thrust B will support this development from 2013 to 2016.
Project B-1: Continuous powder feeding#Top
Faculty: Fernando Muzzio (Lead) (Rutgers), Marianthi Ierapetritou (Rutgers), Carlos Velázquez (UPRM)
Postdoctoral Fellows: Zhenya Jia, Rafael Méndez, Kalyana Pingali
Mentors: James Cartwright (GSK), George Sienkiewicz (Pfizer), Cecile Forness (Boehringer Ingelheim), Richard Hsia (Sepracor), Eli Crossman, Steve Glassmeyer (P&G)
Post Doc's: Luis G. Obregon (UPRM), Rafael Méndez (Rutgers), Kalyana Pingali (Rutgers)
Graduate Students: William Engisch (Rutgers)
Goals
The long-term goal of this research is to develop both a fundamental and a practical understanding of the impact of powder material properties, device design, and operation conditions, on the variability in powder feed rate and on the effect on powder material properties for two types of continuous powder feeders: Gravimetric feeders, and feed frames used in tablet presses.
Deliverables
· Scientific Deliverables
o Understand effect of raw material properties (composition, cohesion, size distribution, moisture content, electrostatic behavior) on flow rate variability of output stream
o Understand effect of process parameters (feeder system design, screw/blade speed, flow rate) on blend properties and flow rate variability of output stream.
o Develop predictive models that allows design and optimization of feeder equipment using readily obtained powder characteristics (shared with D-5)
· Test Bed Deliverables - Gravimetric feeders
o Select, install, and qualify feeder equipment.
o Characterize equipment performance for parametric space (excipient and API particle size, flow rate, feeder size, screw speed, screw design, screen, agitator design)
o Develop set of recommended operating conditions.
o Develop statistical (response surface) model of feed rate variability and dynamical (transfer function) model of feeders as a function of powder properties and processing parameters suitable for process design and control (shared with D-5)
· Test Bed Deliverables - Feed Frames
o Select, install, and qualify feed frame equipment.
o Characterize equipment performance (effect on powder properties, hydrophobicity, PSD, RTD) for parametric space (excipient and API particle size, flow rate, feed frame design, feed frame rotation rate, tablet size and tableting speed)
o Examine effect of feed frame design (size, paddle design, number of chambers)
o Develop set of recommended operating conditions
o Develop statistical (response surface) model of feed rate variability and dynamical (transfer function) model of feed frame as a function of powder properties and processing parameters suitable for process design and control (shared with D-5)
Project B-2: Continuous Powder Mixing #Top
Faculty: Fernando Muzzio (Lead) (Rutgers), Carlos Velázquez (UPRM), Rodolfo Romañach (UPRM), Marianthi Ierapetritou (Rutgers), Carl Wassgren (Purdue),
Mentors: Mayur Lodaya (GSK), Bill Ketterhagen (Pfizer), Pius Tse (Schering Plough), Cecile Forness (Boheringer Ingelheim), Michael Abaskhroun (Pepsi), Brett Alexander (Aztra-Zeneca), Vinit Murthy, Steve Glassmeyer (P&G)
Postdoctoral Fellows: Luis G. Obregon (UPRM), Michael L. Ramírez (UPRM), Atul Dubey (Rutgers)
Graduate Students: Aditya Vanarese (Rutgers), Yijie Gao (Rutgers)
Goals
The principal goal of this research is to develop a mechanistic understanding of the performance of continuous powder mixers as a function of powder material properties, mixer design, and mixer operational parameters, suitable for the optimal design and control of the continuous mixing process.
Deliverables
· Scientific Deliverables
o Characterize fundamental processes (convection, dispersion, shear, segregation) that govern the mixing process
o Understand effect of material properties (composition, cohesion, size distribution, moisture content, electrostatic behavior) on blend homogeneity
o Understand effect of process parameters (mixer design, fill level, agitator speed, blade pattern, residence time) on blend homogeneity and blend properties.
o Develop a predictive (response surface) model as a function of material and process parameters (and their interactions) that allows design and optimization of the continuous mixing process (shared with D-5)
· Test Bed deliverables
o Select, install, and qualify equipment for API blending
o Characterize equipment performance for parametric space (raw material properties, speed, angle of inclination, weir position, blade pattern, fill level, residence time) for API blending.
o Develop set of recommended operating conditions for API blending
o Select, install, and qualify equipment for lubricant blending
o Characterize equipment performance for parametric space (raw material properties, speed, angle of inclination, weir position. blade pattern, fill level, residence time, point of insertion of lubricant into blender) for lubricant blending.
o Develop set of recommended operating conditions for lubricant blending
o Understand the effect of feeder flow rate variability on blend homogeneity
o Develop effective sampling protocol for continuous blending,
o Develop online sensing method for monitoring the concentration of the blend
o Develop a DEM model that allows effective characterization of mixing mechanisms as a function of design parameters
o Develop reduced order (response surface) model suitable for process control
Project B-3: Film Manufacture #Top
Faculty: Boris Khusid (Lead) (NJIT), Rajesh Dave (NJIT), Somnath Mitra (NJIT)
Consultants: Paul Takhistov (Rutgers), Rodolfo Pinal (Purdue), Lynne Taylor (Purdue)
Mentors: Todd Darrington (Boehringer Ingelheim)
Graduate Students: Maneesh Merwade (NJIT)
Goals:
· Develop scalable film-forming methods effective at the manufacturing scale to produce pharmaceutical films loaded with micrometer or smaller poorly water-soluble drug particles that meet regulatory requirements for the film uniformity.
· Develop a basic understanding of a subtle interplay between the composition and properties of a film and film-forming parameters to provide guidelines for producing film strips with accurate dosing and the desirable dissolution rate and stability of APIs.
Deliverables:
- Compile and categorize market information about bench-top and pilot scale machinery commercially available for the production of pharmaceutical films, depending on the film composition, rheological and thermodynamic properties of the film precursor, and requirements for the pharmaceutical product.
- Develop understanding of the effect of the composition of multi-component film precursors produced by Thrust A’s projects (water, organic solvent, micrometer or smaller poorly water-soluble drugs, surfactants, polymers, etc.) and processing conditions (freeze-drying, vacuum-drying, etc.) on properties of the processed film precursors (porosity, texture, thermodynamic and rheological properties, spatial drug distribution, etc).
- Develop conceptual process-line designs and lab-scale experimental setups for forming pharmaceutical films from film precursors provided by Thrust A’s projects.
- Develop a basic understanding of a subtle interplay between properties of the processed film precursors, such as texture and rheological properties, parameters associated with film forming regimes, and characteristics of pharmaceutical films (mechanical properties, spatial drug distribution, drug dissolution rate, etc.).
- Coordinate with Catalent (or other manufacturers) the adjustment of their technology of manufacturing pharmaceutical products for the development of a simple table-top process train to form pharmaceutical films from appropriate film precursors provided by Thrust A’s projects.
- Integrate a knowledge base about film formulations and film forming technologies at a technology partner site, followed by processing pharmaceutical films in a lab-scale system from film precursors provided by Thrust A’s projects.
Project B-4: Optimization and design modeling for continuous roll compaction granulation #Top
Faculty: James Litster (Lead) (Purdue), Carl Wassgren (Purdue), Rodolfo Pinal (Purdue)
Mentors: Renuka Nair (GSK), Dan Blackwood (Pfizer), Ashleth Sheth (Schering Plough), Carlos Ortiz (Lilly), Dongmei Qiang (Boehringer Ingelheim), Steve Glassmeyer, Eli Crossman (P&G)
Graduate Students: Ryan John McCann (Purdue), Tuhin Sinha (Purdue)
Goals
The overarching goal of this research is to develop a mechanistic approach to formulation and device design and operation for continuous roller compaction.
Deliverables
- Integrated process design model to predict ribbon density, density distribution and milled granule size and density distribution given process parameters and material properties. Sub models will include:
o Model to predict compact microstructure and macroscopic properties given single particle properties, applied stress history and appropriate mixing rules based structure-function models developed in Thrust C
o FEM simulation for roll press that is flexible enough to cover many geometries
- Validation of the design model and establishment of the full design space for roll compaction using the test bed 1 pilot plant.
- New feeder designs to dramatically improve ribbon uniformity and milled granulate properties Development and/or validation of on-line measurement techniques for ribbon density distribution and dry granule size distribution
Project B-5: Design of tabletting operations for optimal performance #Top
Faculty: Alberto Cuitino (Lead) (Rutgers)
Mentors: Kanji Meghpara (GSK), Bruno Hancock (Pfizer), Ashleth Sheth (Schering Plough), Ian Leavesley (Lilly), Luying Wang (Boehringer Ingelheim), Steve Glassmeyer, Eli Crossman (P&G)
Postdoctoral Fellows: TBD
Graduate Students: TBD
Goals:
· to quantify the impact of the processing parameters and material properties on tablet properties
· to develop a modeling platform to predict the mechanical behavior of tablets
· to develop strategies for selecting tabletting operating conditions to improve tablet performance
Deliverables:
Development of a Finite Element Method (FEM) platform to simulate the compression process including the effect of process parameters (precompression & main compression force, precompression & compression roll geometry, press speed, tooling shape, dwelling time) and material properties (elasticity, plasticity, fracture).
· Development of multidimensional response surfaces
· Development of compact models for process control
· Model validation and determination of material parameter for constitutive models
|
|
|
|