Using cure modelling to design optimal processes for composite production

Composites are the materials of choice for applications in which low weight is important. As automobiles and other vehicles electrify, composites will play an increasingly large role, because lower vehicle weight translates directly to greater range, and indirectly to lower costs (lighter vehicles require less battery capacity for a given range).

However, there is a trade-off: compared to metals and other materials, the design and processing of composites is complex, and optimizing manufacturing processes for mass production can be costly in terms of both money and time.

Process modelling and computer simulation are key tools that can help to reduce the number of real-world trials required to design optimal manufacturing processes, and thus to bring new products into series production much faster and more cheaply.

Cure modelling

One common application of process modelling is cure kinetics, a fairly mature technology that has been extensively studied for many years. Cure simulations are commonly applied to the area of liquid composite molding, specifically to manufacturing processes such as vacuum infusion and resin transfer molding. Cure kinetic modelling can provide important insights into how the behavior of a resin system impacts the quality of the parts produced.

Cure-induced shrinkage is generally something to be avoided in composite manufacturing processes, as it can result in localized stresses and/or deformation of parts. Numerous studies have provided a body of knowledge about the cure kinetics of various resin systems, and tools have been developed to facilitate analysis and control of the curing process.

In order to understand the impact of the curing process on the mechanical properties of composite parts, special sensors such as fiber Bragg gratings (FBGs) can be embedded into the composite materials. These sensors can yield information on the degree of cure and the development of any undesired stress points inside the material.

Components for automotive applications can be very complex, and process designers must often balance conflicting goals—for example, choosing an optimal trade-off between process time and part quality.

For example, in wet compression molding (WCM) or dynamic fluid compression molding (DFCM), an advanced variation of the WCM process, resins with cure times as low as 30 seconds can be used. However, in order to ensure consistent part quality, it is critical to fully understand the cure kinetics of the resin system used. Process simulation enables designers to optimize both cure time and part quality quickly and efficiently.

Process modelling for production of Type 4 pressure vessels

Type 4 pressure vessels feature all-composite wrapping and a polymer liner, as shown in Figure 1. Composite materials used for this application tend to be thick, and when cure times are short, the reaction enthalpy creates high resin temperatures, which can cause the thermoplastic liner to melt or become deformed. Typical materials used for liners are polyethylene, which can withstand a processing temperature of no more than 110° C, and polyamide, which can be used at somewhat higher temperatures. Process modeling is a critical tool to establish the shortest cure times that will not result in excessive operating temperatures.

Representation of processing by proper setup of boundary and initial conditions

Figure 2 represents a 2D axisymmetric model of temperature distribution on the end section of a pressure vessel, which incorporates a metal boss used for filling. Figure 3 depicts the cylindrical section, based on a very simple 1D model, which enables extremely fast simulations. Using simpler simulations allows engineers more time to test different options.

Both on the physical production line and in the virtual world, avoiding unnecessary complexity saves time and money. Running highly accurate simulations requires long calculation times, and in many cases, approximate simulations are quite adequate to provide the information needed to determine optimal process parameters.

Practical validation of simulations

Once an array of simulations has been performed, the chosen parameters must be validated by real-world testing. Figure 4 shows the results of a simulation in which an epoxy resin cured with a medium-reactivity amine is used to manufacture a 40 mm-thick pressure vessel using the wet filament winding process. The resin was selected for a long winding time of up to 2 hours, and the processing temperature was limited to 120° C—the maximum safe temperature for the thermoplastic liner. Figure 5 shows a real-world validation of the simulation, and confirms that the maximum peak temperature was accurately predicted.

Impact of cure-temperature domain – low- and high-temperature cure

Evaluating methods of reducing cure times

Process engineers often look to reducing cure times as a way to speed up manufacturing processes and reduce costs. However, for thick laminates, such as those commonly used to produce pressure vessels, a faster resin reaction may cause a higher temperature overshoot. Simulations can be used to predict the results of various acceleration scenarios in order to demonstrate whether it is possible to speed up the reaction without undesirable side effects.

Figure 6 illustrates this behavior in the context of a 40 mm-thick laminate. Adding an accelerating factor to the resin reaction causes peak temperature at the liner to increase along with reactivity. However, while the effect of acceleration on peak temperature is substantial at an acceleration factor of 2, smaller increases in temperature are seen beyond this point (higher factors).

Figure 7 illustrates a different scenario, in which an anhydride cured epoxy resin is used for a 25 mm GFRP laminate. The simulation demonstrates that higher acceleration actually reduces peak temperature in this case.

Cure simulation can be a powerful tool to find ways to reduce cure time by using faster resin systems while avoiding undesired temperature overshoots. However, material characteristics, including thermal stability, vitrification and reactivity, interact in highly complex ways, so devising effective simulations requires deep knowledge of the materials and manufacturing processes to be used. Manufacturers and material suppliers need to work closely together at every stage of the process.