CHANGES IN FUNCTIONALITY OF SOY-BASED EXTRUDATES DURING SINGLE-SCREW EXTRUSION PROCESSING

The effects of defatted soy flour content, feed moisture, screw speed and temperature on the functionality of soy-based extrudates were studied. Defatted soy flour and corn flour were mixed with ratios from 10% up to 30% of soy flour. The moisture contents of the blends (wet basis) have three levels: 15%, 18% and 21%. The barrel temperatures and screw speeds of the single screw extruder were controlled from 155 to 185°C and 180 to 220 rpm range, respectively. Shear force, color of the extrudates, expansion ratio, bulk density and water absorption were measured. Higher soybean flour content significantly increased shear force, bulk density and Hunter a* values and significantly decreased expansion ratio and Hunter L* values. High feed moisture significantly increased bulk density and Hunter L* values while significantly decreasing shear force and Hunter a* values. High process temperature significantly increased shear force and Hunter a* values and decreased bulk density. Increasing screw speed resulted in significantly increased bulk density. Understanding the relationship between extrusion parameters and physicochemical properties of the extrudates provides a basis for effective product development.

INTRODUCTION

Snacks have long been a part of the American diet and are rapidly increasing in popularity.[1] In developing countries where protein energy is deficient, snacks are also widely consumed. [2] A prime target of consumer advocates and nutritionists is to effectively improve the quality and nutrition of snack foods.[3] Extrusion cooking is currently utilized to produce expanded snacks, textured soy protein, ready-to-eat cereals, soup and drink bases, etc.4-5] This process is flexible, inexpensive and rapid. It exposes the raw materials to high temperatures, pressure and shear force mix and cause chemical changes, which constitutes cooking and gives extruded snacks a variety of shapes and flavors. Extruded snacks are frequently low in nutrition density and high in fat and starch. Soy protein fortification has a high prospect to improve the nutritional profile of starch-based extruded snacks. Populations with high levels of soybean intake have lower rates of coronary heart disease, breast cancer and osteoporosis than populati ons with low levels of soybean intake.[6] Soy-corn flour blends are a promising proteinaceous feed material because of their availabil ity and low cost. The isoflavones in soybeans have been shown to contain anti-proliferative, anti-fungal, antioxidative and estrogenic/antiestroge nic effects.7-9

There are few studies reported in the literature on the physical and rheological properties of soy-containing snacks.[2], 10-12 The soy-corn flour blend has been successfully extruded to produce low-cost nutritious blended foods.[13] Park studied the characteristics of extrudates containing defatted soy flour, corn starch and beef and got convex curves of expansion ratio, bulk density and shear force with moisture. Omueti and Morton[2] mixed maize, soybean and condiments and tested the shear force, moisture content and product diameter. They reported that higher feed moisture led to lower hardness of the product. Diaz-Mercado[14] studied on the stability of isoflavones during extrusion processing by a twin-screw extruder from soy protein concentrate and corn meal mixture. However, there is no comprehen sive report on the functional characteristics of extrudates composed of corn flour and soy flour and the importance of isoflavones in the soy-based extrudates.

The purpose of this study was to investigate the effects of extrusion parameters such as soy flour content, feed moisture content, screw speed and temperature on the bulk density, expansion ratio, shear force, Hunter L*, a*, and b* color values and water absorption properties of extrudates produced by a single-screw extruder. This research will have importance for soybean processing as well as snack production and development.

MATERIALS AND METHODS

Samples

Corn flour and defatted soy flour were purchased from Archer Daniels Midland Co. (Decatur, IL). The moisture contents of the flours were tested before mixing. The flours were mixed using a Hobart mixer (Model A-200, Hobart Instruments, Inc., Columbus, Ohio) before extrusio n at ratios of 10, 20, and 30% defatted soy flour. Appropriate amounts of water were added to the mixture to reconstitute the mixture to the required moisture for extrusion (15, 18, and 21%). After moisture was added, the samples were sealed in polyethylene bags and held for 24 h at 4–5°C before extrusion.

Extrusion

Extrusion was carried out using a single-screw extruder (C. W. Brabender 3/4′′, model PL 2000, L/D ratio 20:1) with the following conditions: screw compression ratio 3:1, die diameter 3 mm. The temperatures at the second barrel section studied were 155, 165, 175 and 185°C. Screw speeds were controlled to be 180, 200, and 220 rpm at the four different barrel temperatures.

Functional Properties

Shear Force

Each sample was cut to obtain three 6 cm-long strands, which were placed in a Kramer shear cell. The force required to shear the sample was recorded using a Sintech 2/D universal test machine (Sintech Inc., Raleigh, NC). The cross-head speed was 20 cm/min and full scale load was 1000 kg. The results were reported as peak force, measured in kg, divided by the mass of the samples. Triplicate measurements were made for each sample.

Hunter Color

Colors of the extrudates were measured with a Minolta CM-2000 spectroph otometer (Minolta, Ramsey, NJ) and reported as Hunter L*, a*, and b* color values. Three measurements on each sample were taken.

Expansion Ratio

Expansion ratio is the ratio of the diameter of the extrudates to the diameter of the die. Twenty randomly selected segments of each sample were measured using a digital caliper (Mitutoyo Inc., Tokyo, Japan).

Bulk Density

The samples were chopped with a handy chopper (Black & Decker Canada Inc., Brockville, Ontario). Chopped extrudates that passed through a US no. 4 sieve but were retained on a US no. 6 sieve were collected for bulk density analysis. Chopped extrudates were transferred into a 25 mL-graduated cylinder. The bottom of the graduated cylinder was tapped gently on a table until no further reduction of sample volume was possible. Triplicate measurements were done and the bulk density was determined as the weight of sample/unit volume (g/L).

Water Absorption

Saturated salt solutions with different water activities (a_w), shown in Table 1,[15] were used to provide constant a_w conditions. Saturated salt solutions were prepared in plastic jars with distilled water to obtain a_w 0.6–0.93 at room temperature. The solutions were kept in four sealed jars at room temperature (25°C) overnight to ensure saturation. Five 5 cm-long strands were put into each jar and equilibrium was continued until the mass of samples didn't change by more than 0.01 g. Water absorption was calculated as WA (%) = (weight gain after equilibrium/d ry weight)×100. Each sample was determined in duplicates.

Table 1. Water Activities (a_w) of Saturated Salt Solutions at Room Temperature (25°C)

Salt Solutions	aw
NH4NO3	0.60
NaCl	0.75
KCl	0.86
KNO3	0.93

Statistical Analysis

Data were analyzed using SAS statistical package (SAS Institute, Cary, NC). Proc Anova was used to separate out significant effects. Differenc es between the sample means were analyzed by Fisher's least significant different (LSD) test.

RESULTS AND DISCUSSION

Shear Force (SF)

The analysis of variance across all feed moistures, soy flour contents, screw speeds and process temperatures showed that the feed moisture, soy flour content, screw speed and temperature were all a significant (p<0.05) source of variation individually. A significant interaction between independent variables was observed. The effects of each factor on SF are presented in Table 2. The SF decreased with the increase of feed moisture. The SF increased with increased defatted soy flour level and barrel temperature. The SF also increased with increasing screw speed up to 200 rpm and then decreased with the further increase of screw speed. Significant (p<0.05) interactions between feed moisture and soy flour, feed moisture and extrusion temperature, and soy flour and temperature were observed. Low feed moisture and high soy flour content resulted in high SF, as did low moisture content and high temperature. High soy flour content and high temperature resulted in high SF.

Table 2. Effects of Each Factor on Shear Force¹

Moisture		Soy		Screw Speed		Temperature
(%)	SF^2(kg/g)	(%)	SF (kg/g)	rpm	SF (kg/g)	°C	SF (kg/g)
15	40.97^a	10	34.06^a	180	35.17^b	155	29.48^a
18	34.44^b	20	36.13^b	200	36.88^a	165	31.74^b
21	33.09^c	30	36.99^c	220	35.13^b	175	38.11^c
						185	41.50^d

¹Data are means of three replicates.

²Means within the same column with the different superscript letters are significantly different (p<0.05).

Hunter Color (HC)

The analysis of variance across all feed moistures, defatted soy flour contents, screw speeds and process temperatures showed that the feed moisture, soy flour content and screw speed were significant (p<0.05) sources of variation individually on Hunter color values of L*, a*, and b*. A significant interaction between independent variables was observed. The effect of extrusion temperature was significant on a* values but not on L* and b* values.

The effects of each factor on Hunter L* value are shown in Table 3. The L* value decreased with increasing defatted soy flour content and decreasing feed moisture. The L* value also increased with increasing screw speed up to 200 rpm and then decreased with further screw speed increases. The L* value didn't show significant changes at the four different temperature levels. Significant (p<0.05) interactions between feed moisture and soy flour level and between soy flour level and temperature were observed. High feed moisture and low soy flour content resulted in high L* value, as did low soy flour content and low temperatures.

Table 3. Effects of Each Factor on Hunter L* Color Value¹

Moisture		Soy		Screw Speed		Temperature(%)
(%)	L*^2	(%)	L*	rpm	L*	°C	L*
15	56.72^a	10	60.55^a	180	55.41^a	155	58.47^a
18	57.81^b	20	58.58^b	200	63.99^b	165	58.87^a
21	60.85^c	30	56.73^c	220	56.46^c	175	58.74^a
						185	58.34^a

¹Data are means of three replicates.

²Means within the same column with the different superscript letters are significantly different (p<0.05).

The effects of each factor on Hunter a* value are shown in Table 4. The a* value increased significantly with increasing soy flour content and process temperature. The a* value decreased with the increase of moisture content. Screw speed didn't show significant influence on a* value. A significant (p<0.05) interaction between feed moisture and soy flour level was observed. Low feed moisture and high soy flour content resulted in high a* value. Significant interaction between soy flour level and temperature was also observed. High soy flour content and high temperature resulted in high a*value.

Table 4. Effects of Each Factor on Hunter a* Color Value¹

Moisture		Soy		Screw Speed		Temperature
(%)	a*^2	(%)	a*	rpm	a*	°C	a*
15	4.25^a	10	1.24^a	180	2.03^b	155	0.64^a
18	2.43^b	20	2.41^b	200	2.26^a	165	1.79^b
21	0.19^c	30	2.68^c	220	2.04^b	175	2.49^c
						185	3.03^d

¹Data are means of three replicates.

²Means within the same column with the different superscript letters are significantly different (p<0.05).

The effects of each factor on Hunter b* value are given in Table 5. The b* value increased for certain ranges of feed moisture, defatted soy flour content, and screw speed, then decreased with further increases of these factors. Temperature didn't show significant influence on b* value.

Table 5. Effects of Each Factor on Hunter b* Color Value¹

Moisture		Soy		Screw Speed		Temperature
(%)	b*^2	(%)	b*	rpm	b*	°C	b*
15	18.65^b	10	18.55^b	180	16.40^b	155	18.80^a
18	19.02^a	20	19.09^a	200	23.36^a	165	18.83^a
21	18.53^b	30	18.58^b	220	16.46^b	175	18.72^a
						185	18.63^a

¹Data are means of three replicates.

²Means within the same column with the different superscript letters are significantly different (p<0.05).

Expansion Ratio (ER)

The analysis of variance across all feed moistures, defatted soy flour contents, screw speeds and process temperatures showed that the feed moisture, soy flour content, screw speed and process temperature were all significant (p<0.05) sources of variation individually. The effects of each factor on ER are shown in Table 6. The ER decreased (p<0.05) significantly with increased defatted soy flour content and increased process temperature. The ER increased with increasing feed moisture and screw speed individually in a certain range and then decreased with further increases in these factors. No significant interactions were noted for ER.

Table 6. Effects of Each Factor on Expansion Ratio¹

Moisture		Soy		Screw Speed		Temperature
(%)	ER^2	(%)	ER	rpm	ER	°C	ER
15	2.89^b	10	3.20^a	180	2.88^b	155	2.96^a
18	2.93^a	20	2.90^b	200	2.92^a	165	2.94^a
21	2.87^b	30	2.59^c	220	2.89^b	175	2.89^b
						185	2.82^c

¹Data are means of twenty replicates.

²Means within the same column with the different superscript letters are significantly different (p< 0.05).

Bulk Density (BD)

The analysis of variance across all feed moistures, soy flour contents, screw speeds and process temperatures showed that feed moisture, soy flour content, screw speed and process temperature were all significant (p<0.05) sources of variation individually. Significant interactions between independent variables were observed. Table 7 shows the influence of each factor on BD using least significant difference analysis. The BD decreased significantly (p<0.05) with increasing process temperature. The BD increased significantly (p<0.05) with increasi ng feed moisture, soy flour content and screw speed individually. Significant (p<0.05) interactions between feed moisture and soy flour level, feed moisture and screw speed, feed moisture and process temperature, soy flour content and temperature, and screw speed and temperature were observed. BD increased with high feed moisture and high soy flour content, high soy flour content and high temperature, high feed moisture and low temperatur e, high soy flour content and low temperature, and high speed and low temperature.

Table 7. Effects of Each Factor on Bulk Density¹

Moisture		Soy		Screw Speed		Temperature
(%)	BD^2(g/l)	(%)	BD(g/l)	rpm	BD(g/l)	°C	BD(g/l)
15	81.30^a	10	86.40^a	180	95.95^a	155	126.33^a
18	98.09^b	20	95.34^b	200	97.83^b	165	98.14^b
21	108.90^c	30	110.58^c	220	98.54^c	175	88.30^c
						185	86.63^d

¹Data are means of three replicates.

²Means within the same column with the different superscript letters are significantly different (p< 0.05).

Water Absorption (WA)

The experimental data for WA is listed in Table 8. Soy flour content, process temperature, and feed moisture were significant (p<0.05) sources of variation. Increased soy flour content and increased feed moisture resulted in low WA. Increasing process temperature resulted in increasing WA.

Table 8. Effects of Each Factor on Water Absorption¹

Moisture		Soy		Screw Speed		Temperature
(%)	WA^2(%)	(%)	WA (%)	rpm	WA (%)	°C	WA (%)
15	6.19^a	10	6.05^a	180	5.73^a	165	5.08^a
18	5.42^b	20	5.52^b	220	5.49^b	185	5.72^b
21	4.95^c	30	4.99^c

¹Data are means of two replicates for NH₄NO₃.

²Means within the same column with the different superscript letters are significantly different (p< 0.05).

Effects of Soy, Moisture, Temperature, and Screw Speed on Functional Properties

Soy content has a significant influence on the physical properties of the extrudates. Park et al.[12] reported that ER increased with increasing corn starch level. That might be due to the easy-to-expand properties of corn starch. Low corn flour increased SF, BD, Hunter a* values but decreased ER, WA and Hunter L* values. High feed moisture increased BD and Hunter L* values and decreased SF, WA and Hunter a* values. High process temperature increased SF, Hunter a* values and WA and decreased BD.

Water inside the feed vaporizes and the extrudates expand under lower pressure when exiting the die nozzle. Water in the extruder can reduce shear strength and make it easy for the feed to move along the barrel.[16] Hunter L* value reflect the lightness of samples. Hunter L* value (0=black, 100=white) increased 8% when feed moisture increased from 15% to 21%. This is most likely due to lower feed moisture resulting in more browning and the extrudat es exhibiting darker color. Hunter a* value with a positive value indicates redness and negative values greenness. Hunter b* value with a positive value indicates yellowness and negative value blueness. With the increase of feed moisture, Hunter a* value decreased significantly from 4.25 to 0.19. Feed moisture didn't significantly influence Hunter b* value. With decreasing feed moisture, SF increased significantly. With the increase of feed moisture from 15% to 21%, BD increase d significantly. The increase of feed moisture caused a decrease in WA. This might be due to the higher water content in the extrudates, which absorb less water.

Process temperature also influences the rheological properties of extrudates. Temperature influences the vaporization of feed moisture and the chemical reactions in the raw feed. As reported, product moisture was inversely related to barrel temperature and directly related to feed mois ture.[17] Because of the increase of temperature, more water in the feed is vaporized, which resulted in significantly increase d SF and significantly decreased BD. Temperature changes didn't show signific ant influence on Hunter L* and b* values. Increasing process temperat ure from 155°C to 185°C increased Hunter a* value significantly. WA increased when process temperature increased. That is due to extrudates at high process temperatures containing less water and having more ability to absorb water.

Screw speed showed the retention time of the feed in the extruder. A screw speed 200 rpm showed highest b* color value compared to the other two screw speeds. This can be interpreted as a screw speed of 200 rpm gives the extrudates the most yellow appearance. BD increased significantly accompanied by the increase of screw speed from 180 rpm to 220 rpm.

CONCLUSIONS

Based on the results of this study, it may be concluded that: Higher soy flour significantly increased shear force, bulk density and Hunter a* values and but significantly decreased expansion ratio and Hunter L* values. High feed moisture significantly increased bulk density and Hunter L* values and significantly decreased shear force and Hunter a* values. High process temperature significantly increased shear force and Hunter a* values and decreased bulk density. Increasing screw speed resulted in a significant increase in bulk density.

ACKNOWLEDGMENT

We thank the Ethel Austin Martin Program for their financial support to carry out this study. Published with the approval of the Director of the South Dakota Agricultural Experiment Station as Publication Number 3312 of the Journal Series.