ANALYSIS OF MASS TRANSFER FOR THE DRYING OF EXTRUDED DURUM SEMOLINA (PASTA)
Abstract
The purpose of this study was to analyze and measure mass transfer for the drying of extruded durum semolina. A laboratory apparatus monitored the sample weight as conditioned air was circulated around a pasta noodle. The desorption isotherms and the effective diffusivities were determined for seven temperatures ranging from 45(DEGREES) to 125(DEGREES)C and for a range of moisture contents from 0 to 25% (db). Effective diffusivities were from 3.0 x 10('-13) to 1.5 x 10('-10) (m('2)/s). The isotherm results were regressed to fit a modified form of the Henderson isotherm equation. The effective diffusivities were expressed with an Arrhenius temperature dependence and a moisture dependence. The isotherm and diffusivity expressions were used in a finite difference solution of the diffusion equation to predict drying curves and moisture profiles. By slicing frozen samples, the transient moisture profiles were measured for slabs of pasta during drying. Samples were removed from the test dryer, immersed in liquid nitrogen, sliced on a microtomecryostat, and tested for moisture by an air-oven test. The spatial resolution was 0.1 mm between data points. These tests were conducted for drying temperatures from 40(DEGREES)C to 110(DEGREES)C. Drying rates were also measured for gas flow at a partial vacuum pressure of 0.45 atm at three temperatures, 40(DEGREES), 60(DEGREES), and 80(DEGREES)C. The diffusion equation successfully predicted the drying curves and profiles for low temperature, atmospheric pressure conditions. For higher temperatures and at the partial vacuum pressures, the model underpredicted the drying rate. It is expected that a pressure gradient occurred in the product at these conditions, and the pressure gradient induced higher rates of mass flux and flatter moisture profiles than those predicted by the model. Another factor that may have contributed to these phenomena was the possible presence of a surface or skin resistance that was much greater than the resistance of the rest of the material. A drying model that is based on non-equilibrium thermodynamic and mechanistic principles was also evaluated. Unique coefficients for the liquid and vapor conductivities in the model could not be determined. Nonunique solutions could be found by regression of experimental results, but the coefficients varied widely with the boundaries imposed in the regression search technique. The difficulty in measuring and calculating unique values for the model coefficients raises questions about the utility of this model.
Degree
Ph.D.
Subject Area
Agricultural engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.