Semester

Fall

Date of Graduation

2024

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Chemical and Biomedical Engineering

Committee Chair

Dr. Rakesh Gupta

Committee Co-Chair

Dr. Mahesh Padmanabhan

Committee Member

Dr. Sushant Agarwal

Committee Member

Dr. Ruifeng Liang

Committee Member

Dr. Charter Stinespring

Abstract

Dough is the starting material for a large number of baked products, including cookies, crackers and bread. In commercial food processing operations, it is desired to obtain the same product and with the same dimensions in a reproducible manner in order to minimize waste. In the case of dough, the flour used, and the nature and amount of ingredients employed depend on the specific product and the market served. This research was aimed at relating cracker dough rheology to sheeting behavior between two counter-rotating rollers. The sheet that is formed is used to punch out the crackers that are then baked in a subsequent step. The focus here, though, is on relating the material, geometrical and processing variables characterizing sheeting to the experimentally measured sheet thickness and its variation in the flow and transverse directions. Flour is a complex mixture of soluble and insoluble proteins, carbohydrates, lipids, enzymes and ash. From a rheological point of view, however, properly mixed dough can be considered to be a suspension of starch granules inside a protein network. As a consequence, observed flow behavior can be considered to be a combination of the behavior representative of a solid-in liquid suspension and a rubbery network, and this behavior can be represented mathematically with the help of equations and models familiar from polymer rheology. As a practical matter, one observes a yield stress, shear thinning viscosity and fluid elasticity.

Although polymer melt rheology is determined essentially by polymer molecular weight, molecular weight distribution, chain branching, temperature and deformation rate, such is not the case with dough rheology. Since wheat is an agricultural product, there are regional and seasonal variations in composition, and these influence flow behavior. In addition, the mixing protocol, water content, time of mixing and subsequent resting affect dough rheology which, in turn, affects sheeting behavior. These changes in flow properties should be reflected in changes in material constants and material functions appearing in any chosen rheological constitutive equation. However, others have found that it is not easy to properly characterize dough rheology because a steady state is not attained either in shearing at a constant shear rate or in uniaxial stretching even at low stretch rates. Given that sheeting involves flow through a set of rollers which results in a compressive flow of short duration and large strains, it was decided to employ constant area squeeze flow tests for rheological characterization.(

Here, dough was prepared in an internal mixer, and dough variables included water content (17 wt% and 20 wt%) and sodium metabisulfite (SMS) content (0 wt% and 0.05 wt%). Squeezing flow experiments were conducted at 42 ℃ on freshly prepared dough as well as on dough that had been allowed to rest for four hours at 42 ℃. An Instron™ machine fitted with roughened, parallel plate fixtures was used for these no-slip, constant-area squeeze flow tests; squeezing speeds were as high as 200 mm/min and as low as 2 mm/min, and strains as high as 90% were imposed. Force versus time plots were obtained during squeezing and upon subsequent cessation of squeezing. As expected, force levels increased with decreased water content, reduced SMS content, increased squeezing speed and increased strain. The influence of conditioning time was much less pronounced. Upon cessation of squeezing at high speeds, the built-up stresses relaxed over a time scale of less than 0.5 s, but typically 0.1 to 0.2 s. This showed that cracker dough exhibits some elasticity. For squeezing at low speeds, the stresses did not relax completely, and there was a residual stress indicating the presence of a yield stress. However, the residual stress kept decreasing over a period exceeding one hour.

In terms of quantitatively representing the squeeze-film data, it was found that, over the time scale of the experiments, these followed the Herschel-Bulkley equation. The yield stress predicted by this equation, and computed from the measured residual force upon cessation of low-speed squeezing, was relatively small. As a result, all the data obtained at squeezing speeds of 120 mm/min and 200 mm/min could be represented by the Scott equation, demonstrating that dough behaved like a power-law fluid under these conditions; the power-law index ranged from 0.3 to 0.5. The only exception was behavior at the highest speed, largest strain and lower water content where large stresses were observed, possibly due to jamming.

Sheeting experiments were conducted on a custom-built, instrumented laboratory sheeting device. It consisted of two 9-inch long and 3-inch diameter rollers whose surfaces were carefully machined and chrome plated. Each roller is driven independently and could be rotated at variable speeds of up to 80 rpm; the nip gap between the rollers could be varied in steps from 1 to 10 mm. The rollers are hollow inside and special shafts and feeding mechanism were installed to allow for the circulation of tempered water to control the temperature of the roller surface at 42 °C. Quantities that were varied in the sheeting experiments were roller speed and the nip gap, and these were such that the deformation rates were comparable to those experienced during squeezing flow. Thick sheets of dough of the same initial thickness were fed by hand to the rollers, and the exiting sheet thickness was measured at ten different locations using a laser sensor. Over the range of variables examined, the ratio of the final sheet thickness to the nip gap ranged between slightly more than unity to 1.7, with minor exceptions; for a fluid having a power-law index of about 0.3, the theoretical value is about 1.25 for a very thick incoming sheet. The variable influencing the sheet thickness the most was found to be the gap size; at a given roller speed, the sheet thickness increased progressively as the nip gap was reduced. The sheet thickness increased some more when the roller speed increased. Differences in formulation, and these affected the power-law parameters, seemed to have only a slight influence on the measured sheet thickness. The practical outcome of this comprehensive investigation is that it may not be essential to control dough formulation very tightly when producing thin sheets from cracker dough in industrial operations by the process of calendering.

Download

Share

COinS