Work described in this thesis has been undertaken in a collaborative project involving SWECAST, CAST CRC and several industrial foundries in Sweden. SWECAST is a semigovernment research institute in Jonkoping Sweden that provides foundry industry technical support in areas like environment and product quality and collaborates with international research institutes. The aim of this study was to examine the relationship between the moulding sand composition and the type of reaction products that result when these sands are used in typical industrial environments. This research will help in the design and selection of sand systems which will minimise the environmental and health impacts while maintaining or improving product quality.
A selection of industrially relevant moulding sands, predominantly recycled greensands were chosen for this study after consulting with project stakeholders. Samples of virgin greensand mixed with virgin additives and phenolic-urethane sand were included in this study for comparison purposes. Greensands used in this study differ from each other in not only the base type (the fundamental-sand composition), but also in the type of additives. After collecting samples, they were analysed for gas and condensate evolution under controlled conditions in a purpose-built experimental rig, the COGAS analyser. This equipment consists of a small induction furnace which holds molten cast iron and replicates the casting environment by immersing sand-samples into molten cast iron and capturing the evolved gases. In addition to these experiments, conventional tests such as loss on ignition (LOI), total carbon content and the methylene blue clay test were used for comparative purposes. Surprisingly, there was not a strong relationship between the type of base-sand or binder used in a particular greensand and the amount or composition of both gasses and condensates. It was however shown, that greensand additives which claim to have higher organic carbon contents have a higher volume and rate of gas evolution and also result in condensates that are fairly high in carcinogenic material. It is concluded that, using the current generation of industrial moulding-lines, there are likely environmental and health benefits to be obtained from minimising the amount of organic carbon within the sand mix. It is further concluded that traditional methods of sand characterisation such as LOI and the iii total carbon content, do not correlate well with the gas evolution information obtained from newer generation equipment such as the COGAS analyser.