Alex Mit / Shutterstock
In the history of engineering ceramics applications there have been numerous examples where imitated designs based on the performance of metals have led to great failures of ceramic components.
In ceramic engineering design it is necessary to consider the operating environment, material properties and the fabrication process used. Dimensional tolerances, cost, risk of premature failure, reliability and component geometry are the limiting factors in ceramic designs.
A precondition of any ceramic engineering design is a clear and prioritized specification of the application requirements. What function is the component supposed to carry out and what are the conditions it will undergo? For a majority of the engineering applications, the following considerations will be involved:
- The temperature of operation including transients
- The load the component will experience and more particularly the stress distribution that develops
- The potential for abrasion and impact
- The chemistry related to the operating environment
- Other external environmental factors like high electrical or magnetic fields and ionizing radiation
The application conditions are compared with the appropriate properties of the material to complete the first stage of ceramic material selection. This seemingly easy task is usually made difficult by the dearth in property data, specifically data measured under the proper conditions.
The typical sources of property data for ceramics are reference works that are usually restricted in scope and indicate classes of materials, or suppliers’ data sheets that offer standard values or ranges of properties attained by particular materials. With the lack of relevant data, designers must be ready to face the need to measure the properties of the materials on their own.
Prediction of Performance
In the next step of the material selection process the performance of the ceramic in the application is predicted. This will take into account the dependence of various properties on component-specific features like size and shape, as well as the interfaces with adjacent components. Finite element and other mathematical modeling approaches are very important in predicting the behavior of ceramic components, mainly in intricate stress conditions or where transient thermal and mechanical conditions exist.
In several cases, there are well-documented performance predictions that are based on material properties. In the remaining cases, the easiest and most cost-effective approach is to create the component and assess its performance directly in the application.
Fabrication and Manufacturing Issues
The fabrication method selected for the component is dictated by its size and shape, as well as the economics of manufacture, which itself is largely governed by the manufacturing rate.
Furthermore, the fabrication method determines the type and population of process-associated defects, which restrict the properties and reliability that can be realized. For example, identical parts that are produced by slip casting and pressing and firing will, despite having the same density, show dissimilar shape and size microscopic defects (for example, bubbles in the slip cast item and small cracks in the pressed item). These will result in variations in toughness and strength which could make the component manufactured by one method inappropriate for the application.
The reliability of the measured properties is one of the most vital features affected by the fabrication method. The statistical scatter of test results produced by a batch of ceramic test specimens can be applied to forecast the likelihood of failure of a component under specified operating conditions. As a result, components can be developed (over-engineered) to achieve a tolerable failure rate. What makes up a tolerable failure rate will be dependent on the importance of the application, with applications where life is at stake being the most challenging.
After selecting the fabrication method, the designer should evaluate the property data and determine that it is still applicable and that any mathematical modeling that has been performed remains valid.