Off-campus South Dakota State University users: To download campus access theses, please use the following link to log into our proxy server with your South Dakota State University ID and password.

Non-South Dakota State University users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Document Type

Thesis - University Access Only

Award Date

2006

Degree Name

Master of Science (MS)

Department

Agricultural and Biosystems Engineering

First Advisor

Kasiviswanathan Muthukumarappan

Abstract

The effect of incorporation of Distillers Dried Grains with Solubles (DDGS) as an alternative source of protein in the production of aquaculture feed through extrusion processing was studied. Ingredient blends containing different levels of DDGS along with soy flour, com flour, fish meal, vitamin mix, mineral mix and cheese whey with net protein adjusted to 28% protein were formulated. The ingredient blends were extruded both in single screw and twin screw extruders and the properties of extrudates were studied. The experiments conducted in the Brahender single screw extruder includes: effect of changing the level of DDGS (20, 30 and 40%) along with other ingredients without cheese whey, moisture content (15, 20 and 25%) and screw speed (100,130 and 160 rpm) on properties of extrudates such as pellet durability, unit density, bulk density, porosity, color; effect of changing levels of DDGS (20, 30 and 40%) along with other ingredients with cheese whey, moisture content (15,20 and 25%), and screw speed (130 and 160 rpm) on properties of extrudates such as pellet durability, unit density, moisture content, and water activity; development of a viscosity model incorporating a correction factor for DDGS along with correction factor for moisture content and temperature; effect of die dimensions on the properties such as unit density, bulk density, pellet durability, water absorption index, water solubility index, sinking velocity of the extrudates containing 40% DDGS and extrusion processing parameters such as mass flow rate, torque, specific mechanical energy, apparent viscosity, temperature of dough at barrel and die. Experiments were conducted in a Wenger TX52 twin screw extruder to study the effect of changing the levels of DDGS (20, 40 and 60%) along with other ingredients with net protein adjusted to 28% on expansion ratio, pellet durability, unit density, bulk density, true density, moisture content, water solubility index, water absorption index and color of extrudates. Regression models and Neural Network (NN) models were developed to predict the extrudate properties and extrusion processing parameters from diameter, length, length to diameter ratio of die insert, temperature at the barrel, moisture content of the ingredient blend, screw speed and data collected from viscosity modeling and die effects experiments.
Increasing the DDGS content from 20% to 40% without cheese whey as a binder resulted in 37.1, 3.1, 8.4% decrease in pellet durability, specific gravity, porosity respectively and 7.5% increase in bulk density of the extrudates. Increasing the screw speed from 100 rpm to 130 rpm resulted in 20.3, 8.8% increase in pellet durability, porosity respectively, and 12.9% decrease in bulk density of the extrudates. Increasing the moisture content of the feed ingredient mix from 15% (wb) to 25% (wb) resulted in 28.2% increase in pellet durability and 8.3, 8.5% decrease in specific gravity and porosity of the extrudates, respectively. Increasing the DDGS content from 20% to 40% resulted in 52.9, 51.4% increase in fiber and fat content, respectively and 7.2 % decrease in nitrogen free extracts in the extrudates.
In the experiment with different levels of DDGS along with cheese whey as a binder it was found that the pellet durability and unit density of the extrudates were increased significantly by addition of cheese whey at 5% on weight basis. Increasing the DDGS content from 20 to 40% resulted in 6.5, 8.0, 5.2% decrease in moisture content, water activity, brightness respectively, 8.5,4.2% increase in redness and yellowness of the extrudates respectively. Increasing the moisture content of the ingredient blends from 15 to 25% resulted in 15.4, 4.1,19.9, 22.6, 8.4% decrease in moisture content. Water activity, brightness, redness and yellowness of the extrudates respectively. Changing the screw speed, from 100 to 160 rpm had no significant effect on most of the properties of the extrudates studied in the experiments except pellet durability, unit density and moisture content of the dried extrudates. During nutrient analysis, an increase of 31.8, 66.0% in fiber content, fat content and a decrease of 7.3% in nitrogen free extracts in the extrudates was observed when the DDGS content was increased from 20 to 40%. At higher temperatures in the barrel and die, the viscosity of the dough was lower, leading to lower torque and specific mechanical energy requirements. Viscosity modeling resulted in a positive coefficient for moisture content, while the measured apparent viscosity did not show an increasing trend as the moisture content of the ingredient mix was increased from 15% (wb) to 25% (wb). The coefficient for DDGS in the viscosity model was found to be positive and the measured apparent viscosity increased as the DDGS content was increased. Increasing the DDGS content resulted in a higher mass flow rate and decreased absolute pressure inside the die. The coefficient for the DDGS in the viscosity model was much higher compared to the coefficient for moisture content, indicating that the viscosity can be controlled readily by adjusting the level of DDGS. Modeling of the viscosity of dough in the barrel resulted in flow behavior index 'n' value <1 for reference viscosity values between 0 and 3.
In the experiment to determine the effect of die dimensions, it was found that increasing the L/D ratio of the die nozzle resulted in an increase in bulk density, L*, a*, and torque, hut a decrease in unit density, pellet durability, water absorption index, sinking velocity, b*, mass flow rate, dough temperature and apparent viscosity. Increasing the moisture content of the ingredient mix from 15 to 25% (wb) resulted in a 2.0, 16.0, 16.3, 22.9,18.5, 32.5, and 63.7% decrease, respectively, in bulk density, water solubility index, sinking velocity, L*, b*, mass flow rate, and absolute pressure, as well as 11.6,16.2, and 7% increase, respectively, in pellet durability, water absorption index, and a*. Increasing the barrel temperature from 100 to 140°C resulted in a 17.0, 5.9, 35.4, 50.6, 28.8, 33.9, and 33.9% decrease, respectively, in unit density, pellet durability, sinking velocity, absolute pressure, specific mechanical energy, torque and apparent viscosity, but a 49.1 and 16.9% increase, respectively, in dough temperature and water absorption index.
In the twin screw extruder, increasing the DDGS content from 20% to 60%, resulted in 36.7% decrease in the radial expansion leading to 159, and 61.4% increase in the unit density and bulk density of the extrudates respectively. Increasing the DDGS content resulted in significant increase in the water absorption index (WAX) and significant decrease in the water solubility index (WSI) of the extrudates. Changing the screw speed from 350 to 420 rpm and moisture content from 15 to 19% had no effect on the radial expansion ratio, but resulted in significant difference in the bulk density of the extrudates and might be due to longitudinal expansion. Even though changing the moisture content and screw speed had no effect on the WSI of the extrudates, significant difference in the WAI of the extrudates was observed. Color changes in the extrudates was mostly due to color changes in the ingredient components and moisture content of the extrudates and screw speed had no effect on the color of the extrudates.
In general both regression and NN models predicted the extrusion processing parameters with better accuracy than the extrudate properties. The NN models developed with 3 input variables (L/D ratio of die, moisture content and temperature gradient) predicted the extrusion processing parameters and extrudate properties with better accuracy than the regression models developed with the same 3 input variables. In NN modeling increasing the number of input variables from 3 to 5 (D, L, L/D ratio of die, moisture content and temperature gradient) resulted in better accuracy of prediction for both extrudate properties and extrusion processing parameters. The highest accuracy of prediction (R^ values from 0.901 to 0.991) was observed for the NN models developed to predict the extrusion processing parameters with 6 input variables (D, L, L/D ratio of die, moisture content, temperature gradient and screw speed) and combining the results of both the experiments. Even though increasing the number of input variables resulted in better R values in regression modeling, there was no significant decrease in the coefficient of variation between the measured and predicted values.
The results of the experiments shown that floating aquaculture feed can be produced with ingredient mix containing 60% DDGS with net protein content adjusted to 28% using twin screw extrusion technology.

Library of Congress Subject Headings

fishes
feeding and feeds
distillers feeds
extrusion process
distilling industries
by-products

Description

Includes bibliographical references (233-251)

Format

application/pdf

Number of Pages

283

Publisher

South Dakota State University

Rights

Copyright © 2006 Nehru Chevanan. All rights reserved

Share

COinS