October AMP_Digital

A D V A N C E D M A T E R I A L S & P R O C E S S E S | O C T O B E R 2 0 1 8 2 1 NONLINEAR MODELS SPEED DEVELOPMENT OF CROSS- LINKED POLYETHYLENE FOAM Relating composition, process, and dimensional variables with materials properties using nonlinear modeling provides a valuable tool for efficient materials and product development. Abhay Bulsari Nonlinear Solutions Oy, Turku, Finland Markku Talja NMC Termonova Oy, Inkoo, Finland C ross-linked low density polyeth- ylene (LDPE) foams are used for several purposes including pack- aging and thermal insulation. A variety of cross-linked LPDE foams with a wide range of densities are produced using the Furukawa process (Furukawa Electric Co. Ltd., Japan). In this continuous pro- cess, raw LDPE polymer blended with a cross-linking agent, blowing agent, and coloring agents is fed into a twin-screw extruder to produce a thin, wide sheet called amatrix (Fig. 1). The matrix is fed into an oven where it is first cross-linked at a tem- perature between 130° and 180°C de- pending on the cross-linking agent used. The temperature in the second half of the oven is set above 200°C, where the foaming agent releases a gas (typically nitrogen), resulting in the cross-linked foam. The foam is cooled with water as it exits the oven and is wound into rolls. Material properties depend on the composition and amount of various raw materials added, process variables, and dimensional variables (Fig. 2). Materi- al properties include density, strength, thermal conductivity, refractive index, and electrical, magnetic, and chemi- cal properties. Because one material property often comes at the expense of another, and could affect produc- tion economics, it is desirable to con- sider the wider picture, which includes several consequences of operating the process. When it is known how com- position and process and dimensional variables affect the results, good and/or optimal values of the variables can be determined to achieve a desired combi- nation of material properties and pro- duction rate and cost. MATHEMATICAL MODELING Relationships between variables are usually not simple or linear and polymeric materials are typically more complicated than other materials. Cre- ating equations to describe these re- lationships is accomplished by mathe- matical modeling. Fig. 1 — Test samples of rawmatrix (upper squares) and cross-linked polyethylene foam (lower squares) from experiments 12 and 28; matrix thickness in experiment 28 is twice that of experi- ment 12, resulting in similar densities but different elastic moduli.