Producing ceramics

Software predicts how a material will change during the sintering process

Manufacturers could produce ceramics with less time, cost and material waste using a new technique developed by researchers at Leicester University.

The new method, which is based on computer modelling, removes the conventional trial-and-error approach to manufacturing ceramics.

The Leicester University research team, led by Jingzhe Pan, focused its attention on the sintering stage of the ceramic manufacturing process.

In this stage, ceramic powders, which have been compacted into a solid, are fired and heated up to a temperature that causes them to adhere to one another. The materials repack more closely so that their overall volume shrinks while their density increases.

Traditionally, it has been difficult for manufacturers to predict exactly how these materials change in dimension during the sintering process, and therefore the final product is not always desirable.

The brittleness of ceramics makes them difficult to alter post-production so the erroneous product is usually scrapped.

In order to save manufacturers from such waste, the Leicester researchers have developed a computer software program that can predict how ceramics will change in dimension during the sintering process.

By providing basic data about the ceramic material, Pan said that their software will tell manufacturers how to set their sintering process so that it yields a product with their desired shape and dimensions.

The program uses data based on the density measurements of different ceramics during sintering. These measurements are gained by firing several equally sized samples of a ceramic to 1,600°C in a kiln. Every few hours one of these samples is taken out of the kiln and its dimension are measured.

Pan added that the dimension of the sample can be related to its density. By studying the dimensions of each sample, he said, the density change over time can be calculated and inputted into the software.

‘Our computer software can predict changes in dimensions even before production begins,’ he added.

Pan, who has been working on this process for 10 years, said that previous methods for predicting sintering deformation required complex constitutive equations. These equations are more commonly used in Finite Element Method (FEM) for stress and strain analysis of structures such as bridges and aeroplane airframes.

Pan added that many in the ceramics industry believed constitutive equations were ideal for the prediction of sintering deformation. Even though the process for gathering the necessary input data for the equations was extremely challenging.

‘The analysis could only be performed by studying the effects of forces on sintering ceramic powders,’ he said.

‘Applying a force on a fragile powder is very difficult,’ he added. ‘The force itself changes and alters the material properties.’

This makes it difficult to achieve accurate information. ‘It is so expensive to get this data and it takes so much effort, many manufacturers just prefer trial and error,’ said Pan.

While trying to simplify the constitutive method, Pan and his research team realised that the data gained from applying a force on the ceramic was superfluous and not needed to predict how a material would shrink or deform during the sintering process.

‘We discovered you don’t need a full constitute equation,’ he said. ‘If you get the densification data right, nothing else is needed.’

Pan added that the software has potential to improve the manufacture of products ranging from alumina-based hip implants to PCBs and has received enquiries from two undisclosed ceramic manufacturers in the US interested in his team’s assistance.

Siobhan Wagner