Estrategia para la optimización de un compuesto de caucho con aceite de soja utilizando diseños de experimentos

ESTEBAN NICOCIA; SALVATORI, PABLO E.; CARLOS E. BOSCHETTI; GASTÓN SÁNCHEZ; ALDO LOMBARDI

REVISTA MATéRIA

UNIV FED RIO DE JANEIRO

Año: 2018 vol. 23

1517-7076

When a rubber component is made, is necessary to start from a ?rubber compound? which may be composed of one or more polymers and various additives or fillers. The development of an elastomer mixture to produce a compound with a specific end-use is complicated in the field of rubber technology. Therefore, the successful development of a compound requires a systematic work, being crucial to use a statistical design of experiments (DoE).By implementing the DoE, the variables for the system under investigation should be selected; i.e., the observed or measured variables (responses, compound properties to be controlled) and the independent variables describing the system (the ingredients and amounts in the formulation of the compound). The purpose of the DoE is to specify the value of the independent variables, as well as to select the appropriate response.The DoE strategy was already tested using two grades of E-SBR rubber (emulsion styrene-butadiene), E-SBR 1712, E-SBR 1721 and a BR rubber (polybutadiene) high-cis, in order to find a mixture of the three with properties comparable to those of a high performance tire. In this paper we will show the use of the DoE strategy working with the extender oil added to the compound. A mixture of three types of oils, two petrochemical oils (Pq1 and Pq2) and soybean oil are used, seeking to maximize the amount of the latter.The tested properties are those related to features required for the tread of a tire: the glass transition temperature (Tg); the ratio between the viscous modulus and the elastic modulus (tanδ@60°C); the Mooney Viscosity and the tensile-elongation properties. The values obtained for each measured property were fit to statistically significant models, yielding response surfaces for the properties. The surfaces are combined in the desirability function, where values are set for each property and under the condition that the amount of soybean oil is maximum in the mixture. Applying this desirability function, it was possible to predict the optimal composition of oils (3.17 pHq of oil Pq1, 5.90 pHR of oil Pq2 and 7.86 pHR of soybean oil) to be added to the compound to achieve the desired properties.The successful experimental validation of the proposed models shows the ability of DoE as a useful strategy for the optimization of properties in a rubber compound with complex formulation.