J. L. Kardos

Polymeric flow through fibrous media

Proper description of the resin flow through fibrous media is a very important input in the modeling of composite manufacturing processes. In this study, the complex phenomena associated with the flow of viscoelastic fluids through fibrous media are examined. In particular, the effects of fluid and fiber bed properties on the pressure drop observed during such flows were investigated, in conjunction with the specific conditions of typical composite manufacturing processes. To gain some insight into the problem, a theoretical and experimental investigation of Newtonian and non‐Newtonian flows through different geometric cylinder arrangements designed to simulate the actual fiber configurations was carried out. Particular care was taken to select the appropriate dimensionless flow parameters to demonstrate that the onset of the ‘‘excess’’ pressure drop due to fluid elasticity is independent of the flow geometry. Employment of this dimensionless group, which is related to the total strain on the macromolecul...

Modeling and Control of an Injection Pultrusion Process

Injection Pultrusion (IP) is a novel process that combines the best features of traditional Resin Transfer Molding (RTM) and wet bath pultrusion for manufacturing polymer composites. Few researchers have dealt with the quality control aspects of this process which is extremely crucial to the plant economics. The main contribution of this paper is the study of the IP process from the point of view of product quality control while maximizing production rates. A mathematical model that includes resin flow and cure, and heat transfer in the part and the die is used to determine sensitivity of the quality variables to important processing variables and parameters. Similarities and differences with traditional pultrusion process are pointed out. Control-relevant features of the process are identified and the requirements necessary for a control system are presented. A cascaded control strategy which uses an on-line process model is proposed.

Structure property relations in short-fiber reinforced plastics

Abstract As the number of uses for variously reinforced plastics has grown, the design engineers need for dependable approaches to property prediction for these systems has grown tremendously. It is quite clear now from experiment that one has to know not only the volume fraction of reinforcement, but also how to account for the effects of reinforcement shape and distribution in space. Most importantly, having accounted for the significant variables, any workable prediction method must be straightforward enough to use in designing without consuming huge amounts of expensive computing time.

Shrinking horizon model predictive control applied to autoclave curing of composite laminate materials

Motivation for batch chemical reactor optimization and control is provided. Autoclave curing of composite laminate materials is defined as an important batch reactor example. Shrinking horizon model predictive control (SHMPC) is presented as a means of online batch reactor control. The autoclave curing process is illustrated for a specific miniature autoclave. Results from curing 112 composite laminates from Hercules 8551-7A fiberglass/epoxy prepreg, with variable manipulated inputs, are summarized. Quantification of off-line control and SHMPC of the autoclave curing process is then proposed as a concluding effort to this work.

Electropolymerization of pyrrole on PAN-based carbon fibers: experimental observations and a multiscale modeling approach

Aqueous electropolymerization of pyrrole on PAN-based carbon fibers has been investigated. Using experimental data obtained from gravimetric analysis and scanning electron microscopy (SEM), a continuum-level electrokinetic-diffusion model has been developed. For short time, the coating process is reaction-limited. Consequently the weight gain increases linearly with time. However, as the reaction time is increased, the weight gain becomes proportional to the square root of time suggesting that diffusion of the monomer onto the reactive sites of the growing aggregates becomes the controlling mechanism. These observations motivated a multiscale approach for the simulation of mesoscopic coating morphology in a model process where the monomers diffuse to a heterogeneous surface that consists of growing surface-bound polymeric chains with reacting ends. A diffusion-limited, aggregation (DLA)-based approach is used to derive transition probabilities consistent with continuum-level conservation principles and used in lattice Monte Carlo simulations. This approach is illustrated for two-dimensional lattices. The scaling laws obtained for this process such as the thickness of the coating as a function of the number of particles are compared with those for classical DLA. The influence of effective diffusion coefficient and reaction rate constant on the surface coverage, maximum and bulk values of coating density, as well as the boundary layer thickness, is examined in detail.

A flexible approach to modeling and simulation of polymeric composite materials processing using object oriented techniques

Abstract Process modeling and simulation is a key requirement for improved design and quality control of the evolving composite manufacturing technology. However in order to effectively use the available simulation software in this field a thorough knowledge of mathematical modeling techniques and material science is required. This work explores the use and ramifications of the object-oriented modeling and design (OOD) methodology in developing flexible process simulators for composites processing. Specifically, we have developed an object-oriented based simulator that possesses the required flexibility to model a large class of composite materials processing operations. The flexibility of the simulation software is demonstrated by simulating two widely used composite manufacturing processes (i.e. injected pultrusion and autoclave curing). In addition, through these examples we have clearly demonstrated that the object-oriented framework greatly facilitates development of hi-fidelity process models with significantly less time, efrort and expertise than the conventional procedural coding approaches.

Ecologically and economically conscious design of the injected pultrusion process via multi-objective optimization

Injected pultrusion (IP) is an environmentally benign continuous process for low-cost manufacture of prismatic polymer composites. IP has been of recent regulatory interest as an option to achieve significant vapour emissions reduction. This work describes the design of the IP process with multiple design objectives. In our previous work (Srinivasagupta D et al 2003 J. Compos. Mater. at press), an algorithm for economic design using a validated three-dimensional physical model of the IP process was developed, subject to controllability considerations. In this work, this algorithm was used in a multi-objective optimization approach to simultaneously meet economic, quality related, and environmental objectives. The retrofit design of a bench-scale set-up was considered, and the concept of exergy loss in the process, as well as in vapour emission, was introduced.The multi-objective approach was able to determine the optimal values of the processing parameters such as heating zone temperatures and resin injection pressure, as well as the equipment specifications (die dimensions, heater, puller and pump ratings) that satisfy the various objectives in a weighted sense, and result in enhanced throughput rates. The economic objective did not coincide with the environmental objective, and a compromise became necessary. It was seen that most of the exergy loss is in the conversion of electric power into process heating. Vapour exergy loss was observed to be negligible for the most part.

Autoclave processing of long fiber thermoplastic composites

Abstract Models for management of volatiles and resin flow during autoclave processing of long fiber thermoplastics composites are developed and described. Comparison of model predictions with data is favorable.