The use of plasma and neural modelling to optimise the application of a repellent coating to disposable surgical garments

The modification of the surface properties of and the application of coatings to textile materials by means of exposure to plasma has attracted much attention in recent years. The advantages of using excited gases include low process cost and duration, and the avoidance of effluents such as solvents or chlorine. Low-pressure plasma treatment with hexafluoroethane has been shown to create a hydrophobic surface on cotton, which would normally be hydrophilic. Cotton is a popular material for surgical garments and drapes because of its comfort and low cost. The acquisition of hydrophobic behaviour will provide haemo-repellancy and the prevention of bacterial attachment. This paper describes a series of designed experiments to vary three parameters for the plasma process, gas concentration, power and duration, and to measure the resulting degrees of hydrophobic behaviour at the cotton surface by means of observing water droplets. Neural networks can provide rapid development of simulation models of processes by adaptation to observed conditions as inputs and the results as outputs. The data from the plasma trials has been used to develop a neural model to predict surface hydrophobic behaviour. The model is itself optimised for interpolative ability, and allows a search to be made through the data space to find the best possible combination of the process parameters to encourage optimal surface treatment, and thus make the cotton most hydrophobic. The work indicates how the environmentally-friendly approach of plasma treatment can be used to provide garments and drapes for surgery which are comfortable but also protective. 2002 © AUTEX.