AbstractDegermed, dehulled cornmeal grits were extruded through a self‐heating single screw extruder with inner restrictions. Constant degree of screw fill was tested at 32.5, 35, and 37.5% w.b. moisture content. The screw speeds tested were 100, 200, and 300 rpm. A novel two hole die was used and showed the melt viscosity to decrease with increased screw speed. The inline data was compared to offline capillary rheometer measurements and showed a decrease in melt viscosity. For example, at 35% moisture content and roughly 110°C the melt viscosity decreased by 32% from offline to 100 rpm for a shear rate of 100 s−1. The melt viscosity change suggested that each process transforms the material differently into two different materials which was supported by RVA pasting profiles. The extruder die temperature increased with increased screw speed and decreased with increased moisture content. The melt viscosity behaved nonlinearly with moisture content due to high temperature rise and starch degradation. The flow behavior index increased with increased screw speed. Overall, the novel two hole die showed general trends of the melt viscosity behavior with operating conditions and made comparisons to offline data.Practical ApplicationsThe extrusion process is an efficient process that can produce a wide range of products at various throughputs depending upon operating conditions and extruder design. Originally developed for the plastics industry, extrusion has been applied to food and pharmaceutical applications. This application comes with increased challenges due to heterogeneous materials that consist of biological macromolecules that are Non‐Newtonian in rheological behavior as well as being susceptible to temperature and shear degradation. The product rheological properties are dependent on extruder design and operating conditions. Unfortunately, measuring the viscosity while extruding can be complicated, time consuming, and expensive. Thus, to facilitate the design of effective future extruders, the viscosity behavior must be able to be predicted from an offline method which is more cost and time effective.
Read full abstract