The Problem

Over the past century, the materials used in the medical industry have changed drastically. Rubber tubing, glass bottles, and metal instruments have largely been replaced by equipment made of polymers. While polymers have a wide variety of advantageous properties, one downside is that generally, their behaviour is inherently more dependent on strain rate, temperature, and pressure. 

Of these variables, the one most relevant to medical devices is the strain rate. 

This was the motivation for creating material models dependent on these variables. The models would allow medical device designers to use FEA to simulate their design's behaviour at the strain rates they might experience during use and transport. This reduces the amount of physical testing that is required, decreasing the cost of developing medical devices and increasing the speed that they can get to market.

The Materials

The four polymers being modelled are POM, PET, PP, and HDPE. These materials are commonly used in the medical industry and were specifically recommended by the project's industrial partner, Crux Product Design.

POM (also known as Derlin) is a high-performance polymer. It is stiff, hard, and has low friction and high chemical resistance. This makes it ideal for joint reconstruction, traumatology, and joint replacement sizing trials. PP (specifically homopolymer PP) is mechanically rugged with high chemical and heat resistance. It is often used in syringes and sterilization trays. PET is most often used for packaging and containers. HDPE is most often used for protective shells around medical equipment and hardware. 

The Model

Material models have two components: the Constitutive Model and the Material Parameters. The material parameters are specific to the material and come from the practical tests. The constitutive model is the base model that the material data is used to calibrate. Since non-linear viscoelastic, time-dependant behaviour needs to be captured, the possible constitutive models are more limited. The models that performed the best across all four polymers were the PolyUMod Three Network Viscoplasticity (TNV) model and a linear viscoelastic (LVE) model. 

For this project, MCalibration® was used to map the material parameters onto the constitutive model. 

The Practical Tests

Three practical tests were conducted to gather the data to calibrate the material models: a drop tower test, a unixial tensile test, and dynamic mechanical analysis (DMA). These tests covered a range of strain rates.

Dynamic mechanical analysis also provides data for a wide range of temperatures, and using Time-temperature superposition, this can be used to approximate data for a much wider range of strain rates. More details on this are given on the practical testing page.