Oil spill dispersants are used to enhance the rate of natural dispersion of an oil spill at sea. Dispersants remove the oil slick from the sea surface and dilute the oil as small droplets in the water column. The large increase in the oil-water interface due to oil droplet formation increases the biodegradation of the oil by natural occurring micro-organisms. Mixture de-Ž . w sign simplex-centroid and response surface methods have in an earlier simulation study P.J. Brandvik, Statistical simulation as an effective tool to evaluate and illustrate the advantage of experimental designs and response surface methods, Ž .
x Chemometrics and Intelligent Laboratory Systems submitted for publication . proved to be an effective tool to enhance optimisation of oil spill dispersant and to reduce the number of experiments needed for development of new products. This proposed multivariate method is representing a new approach within the development of oil spill dispersants. The main objective for the work presented in this paper was to verify the performance of this new approach within this area on real laboratory data. This combined technique using mixture design and response surface methods has been verified to be a powerful and cost reducing approach in dispersant optimisation. New dispersant formulations for both crude oils and bunker fuels have been formulated and verified by measurements to have high effectiveness.