Rheological Behavior of Hot-Air-Puffed Amaranth Seeds

Abstract

Force-deformation and force-relaxation experiments were performed on amaranth seeds puffed at 290, 330 and 370°C. Less force and energy was required to cause a given deformation in seeds processed at 290°C than in those puffed at 330 and 370°C. It was also observed that the forces and energies required to produce a given deformation did not differ significantly (p ≤ 0.05) for seeds puffed at 330 and 370°C. The three-element generalized Maxwell model and Peleg model were applied for modeling force relaxation of puffed amaranth seeds. It was found that the generalized Maxwell model predicted the experimental data better than the Peleg model. The elastic parameters and asymptotic residual force of the generalized Maxwell model were significantly affected by puffing temperature, showing an increase with its rise. Relaxation times were not significantly affected by the puffing temperature. It was concluded that a higher puffing temperature resulted in a more rigid material and less viscous behavior.

Keywords:

• Amaranth seeds
• Hot-air-puffing
• Rheological behavior

Introduction

Amaranth seeds are gaining worldwide recognition as a source of food and contain many valuable nutritional components which can enrich human diet.[1] There have been numerous research reports published on the composition of amaranth seeds,[2, 3] their color,[4] dietary fiber[5] and starch[3] characteristics, as well as functional properties of amaranth seed flour[6] and use of puffed amaranth seeds as food ingredients.7–9 The rheological behavior of high-temperature processed seeds changes during processing, and depends on the final moisture content and temperature applied.[10] The properties deducted from experimental stress-strain curves, such as maximum stress, maximum strain, elastic modulus, and total deformation energy of foods provide useful information about the texture of the material. If food particles are irregular in shape, force-displacement curves can be used instead of stress-strain curves. Quasi-static compression tests may be performed to determine the rheological characteristics of processed seeds. The viscoelastic behavior of high-temperature processed seeds also changes during processing and depends on the temperature applied.[11] Murthy and Bhattaharia[12] studied the moisture-dependent physical and compression characteristics of black pepper seeds. Singh and Goswami[13] studied the mechanical properties of cumin seeds under compressive loading. Engleson and Fulcher[14] investigated the mechanical behavior of oat groats. Despite a lot of research correlated on the viscoelastic properties of foods,15–17 the rheological behavior of puffed amaranth seeds has not been investigated in detail yet, and literature data on this topic are scant. The objective of this article was to determine the effects of air temperature on the rheological properties of hot-air-puffed amaranth seeds. The data obtained in this study may be used in the bakery and breakfast cereal meals industry to optimize the puffing of cereals and thus achieve a better texture of the final product.

MATERIALS AND METHODS

Raw Material Preparation

Amaranth seeds, var. MT-3, were provided by a local professional producer. The seeds contained 14% protein, 56% starch, 6.5% lipids, and 16% water.[18] The seeds were cleaned from mechanical impurities (particles of inflorescence) and divided into classes using sieves with orifices 0.63, 0.80, and 1.00 mm in diameter. Only seeds 0.80 and 1.00 mm in diameter were used in the experiments. In order to ensure uniform distribution of moisture content in all samples, the seeds were put into a hermetic glass container and kept refrigerated at 5°C for two days. Before the puffing experiments, the samples, 500 ± 5 g each, were taken out of the refrigerator and left at an ambient temperature (20°C) for 24 hours. The moisture content of seeds prepared for puffing was 0.163 ± 0.005 kg water/kg dry matter. It was determined as mass loss of a sample dried at 130°C for 2 hours. After 2 hours of drying, changes in mass were not observed.

Amaranth seeds were puffed during pneumatic transportation in hot air along a vertical pipe, 35 mm in inner diameter and 2 m in length. Seed feed rate was 0.400 ± 0.008 g/s for raw material and 3.90 ± 0.04 g/s for hot air. Air temperature was 290, 330, and 370°C. Fresh air supplying the experimental stand was sucked from the surroundings by a blower, and pressed through a rotameter to electric heaters. Air feed rate was maintained with control valves, and temperature was controlled with an autotransformer. A vibrating feeder enabled to control seed feed rate. The product was separated from the air in a cyclone and fell into a container underneath. The time of seed conveyance through the tube varied from 2 to 3 seconds, and was sufficient for full puffing of most of the seeds. After puffing the material was put into a plastic bag and placed in the dark at 5°C for 2 months. All puffing experiments were carried out using a laboratory-scale experimental stand located at the Department of Machines and Applications for Food Industry, Technical University in Bialystok, Poland.[19]

Compression and Force Relaxation Measurements

Two types of experiments were carried out. In the first step, quasi-static compression tests of amaranth seeds were performed. In the second step, compressive force relaxation tests were conducted on puffed amaranth seeds. All test were carried out using an Instron Universal Testing Machine (High Wycombe, Bucks, UK), model 4301, operating alternatively in the compression or compression-force-relaxation mode, fitted with a parallel plate fixture for uniaxial compression and a 100 N load cell. Cross-head speed was 2.5 mm/min during compression experiments, and 25 mm/min during the loading phase of force relaxation tests. Amaranth seeds were randomly selected from three batches of raw material puffed at three different temperatures, sixty from each batch. Before each experiment a single seed was placed on the bottom parallel plate in random spatial orientation. The results of preliminary studies performed at our laboratory showed that the rupture point, which may be correlated with macroscopic damage of the material was not observed during compression of puffed amaranth seeds. Based of that observation the maximal displacement of the upper plate was limited to 1.3 mm during compression tests. The averaged force and energy required to cause 0.4, 0.8, and 1.2 mm deformation were determined on the basis of force-deformation curves. During the force-relaxation test the displacement rate of the upper plate was 25 mm/min, and maximum strain was 15% of the averaged width of grains, relative to the initial position defined by the distal end of a seed. The residual force required to maintain constant strain was measured as a function of time for 120 seconds. Relaxation curves (force versus time), obtained as a function of puffing temperature, provided useful information on the viscoelastic behavior of puffed amaranth seeds. All compression and compression-force-relaxation tests were carried out in 30 replications.

Mathematical Modeling

For small deformations solid foods can be assumed to behave as linear viscoelastic materials. Typical experimental methods for studying the effect of time on the mechanical properties of foods are creep and stress (or force) relaxation tests. In the present study time-dependent changes in force during force-relaxation tests were examined, and force-relaxation curves for individual amaranth seeds were described using the generalized Maxwell model, described by Eq. (1), consisting of two Maxwell elements in parallel with a residual spring.[20]

Parameters F ₁ and F ₂ in the generalized Maxwell model Eq. (1) were assumed to be proportional to the apparent elastic modulus of the material. The last term, F_AM , in Eq. (1) describes a hypothetical value of asymptotic force determined according to the Maxwell model. The relaxation times, τ₁ and τ₂, of each parallel Maxwell element are defined as:[20]

Peleg[21] suggested that stress relaxation data can be calculated as a normalized stress (a normalized force term is also acceptable) and fitted into the following linear equation:

where reciprocal k₁ in Eq. (3) describes the initial decay rate in the relaxation force, and slope k₂ describes a hypothetical value of the asymptotic normalized force determined according to the Peleg equation. Asymptotic force determined according to the Peleg model may be calculated based on equation:

Eqs. (3) and (4) may be applied for studying the viscoelastic properties of foodstuffs which exhibit non-linear behaviors caused by large deformation, usually over 10% in strain.[10, 21] Therefore, in the present study the Peleg equation was also used to describe the viscoelastic behavior of amaranth seeds as an alternative to the Maxwell model.

Statistical Analysis

A one-factor analysis of variance (ANOVA) was carried out to test whether puffing temperature had a significant effect on compression resistance, energy absorbed and parameters of force relaxation models. An interaction is referred to as significant only when p ≤ 0.05. The best adjustment of the models to force relaxation data was determined calculating the determination coefficient (R ²) and relative root mean square errors (RMS) according to Eq. (5).

The results were analyzed statistically using STATISTICA 6.0 (StatSoft Inc.) software.

RESULTS AND DISCUSSION

At each deformation measured during compression tests and at each sampling during force-relaxation tests, the corresponding values of forces were taken for 30 samples of amaranth seeds. Results for all 30 samples were averaged and are presented in this section. The standard error calculated for three levels of puffing temperature was found to be less than 10% during compression tests, and less than 7% during force-relaxation tests.

Force-Deformation Characteristics

Table 1 contains the averaged values of the force and energy required to cause 0.4, 0.8, and 1.2 mm deformation of amaranth seeds puffed at 290, 330, and 370°C. It was found that less force and energy was required to cause a given deformation in seeds processed at 290°C than in those puffed at 330 and 370°C. It was also observed that the forces and energies required to produce a given deformation did not differ significantly (p £ 0.05) for seeds puffed at 330 and 370°C. The mechanical properties of the natural outer layer of a single kernel, as well as the properties of starch, are the main factors influencing the mechanical behavior of thermally processed amaranth seeds. Ciesielski and Tomasik[22] showed that 250°C is an approximate temperature of the beginning of starch decomposition, and that a further rise in starch temperature results in its degradation. This means that starch contained in amaranth seeds may be partly transposed during puffing, which may affect the mechanical properties of seeds. The higher the air temperature applied during puffing, the more starch can be transposed to a state characterized by altered mechanical properties. This can partly explain temperature-dependent compression resistance of puffed amaranth seeds.

Table 1 Averaged values of force and energy required to cause 0.4, 0.8, and 1.2 mm deformation of amaranth seeds puffed at 290, 330, and 370°C.

T (°C)	F0.4 (N)	F0.8 (N)	F1.2 (N)	E0.4 × 10^3 (J)	E0.8 × 10^3 (J)	E1.2 × 10^3 (J)
290	0.90^A, *	1.78^A	3.11^A	0.18^A	0.72^A	1.65^A
330	1.49^B	2.76^B	5.17^B	0.30^B	1.13^B	2.62^B
370	1.47^B	3.16^B	6.27^B	0.28^B	1.21^B	2.92^B

Fig. 1 presents typical force-deformation characteristics of hot-air-puffed amaranth seeds for different puffing temperatures. If overall changes in the loading force were taken in consideration, i.e., a general tendency of force changes rather than local changes in force, it would become evident that there was no clear single peak corresponding to the force required for seed rupture for every force-deformation curve, but there were more than one intermediate peaks in force-deformation curves for several seeds. Usually the first peak corresponds to the bio-yield point at which seed damage is initiated. However, this is not the case with puffed amaranth seeds because continuous collapse of the material structure starts immediately after loading. Therefore, the values of the peaks were not considered in the analysis. Each force-deformation curve consisted of many quasi-discrete changes in force. This step-like nature of changes in force may be due to step collapse of the material microstructure, progressing with deformation. The duration of the hot-air-puffing process is very short, measured in single seconds, and changes in seed volume are considerable. Konopko[19] showed that starch contained in amaranth seeds is melting and swelling rapidly during the hot-air-puffing process, which makes the structure of a puffed amaranth seed resemble the structure of a honeycomb-like conglomerate of viscoelastic starch bubbles. Continuous destruction of the bubble-like skeleton microstructure of a seed during compression, which started immediately after loading, is probably responsible for the step-like nature of force changes.

Figure 1 Typical force-deformation characteristics of hot-air-puffed amaranth seeds.

Force-Relaxation Characteristics

Force relaxation curves obtained for amaranth seeds puffed at 290, 330 and 370°C are presented in Fig. 2. These curves are typical force relaxation curves of viscoelastic solid materials where stress/force decreases exponentially with time. The bars show the standard error of the experiments. Since each experimental point represents a mean of thirty measurements, sample heterogeneity caused a considerable dispersion at some experimental points. The solid line in the curves represents the generalized Maxwell model (Eq. 1). It shows that both the maximum and residual forces measured during relaxation were higher for higher puffing temperatures. Tables 2 and 3 present the viscoelastic properties of puffed amaranth seeds as a function of temperature for both the generalized Maxwell model and the Peleg model, and adjusted parameters RMS and R ².

Figure 2 Force relaxation curves for amaranth seeds puffed at 290, 330, and 370°C. (•)—average experimental data at 290°C; (n)—average experimental data at 330°C; (▾)—average experimental data at 370°C; the bars show the standard error of 30 measurements.

Table 2 Averaged values of the parameters of the generalized Maxwell model (Eq. 1) for amaranth seeds puffed at 290, 330, and 370°C.

T (°C)	τ1 (s)	τ2 (s)	F0 (N)	F1 (N)	F2 (N)	FAM (N)	R^2	RMS
290	34.5^A, *	1.99^A	0.834^A	0.178^A	0.206^A	0.450^A	0.996	0.008
330	35.3^A	1.98^A	1.063^B	0.214^B	0.263^B	0.585^B	0.996	0.007
370	34.4^A	1.78^A	1.261^C	0.259^C	0.318^C	0.684^C	0.996	0.007

Table 3 Averaged values of the parameters of the generalized Peleg model (Eq. 2) for individual amaranth seeds puffed at 290, 330, and 370°C.

T (°C)	k1 (s)	k2	F0 (N)	FAP (N)	R^2	RMS
290	10.1^A, *	2.14^A	0.852^A	0.454^A	0.952	0.026
330	9.80^A	2.20^A	1.086^B	0.592^B	0.951	0.026
370	9.17 ^A	2.16^A	1.228^C	0.693^C	0.944	0.028

The determination coefficient, R ², of the generalized Maxwell model was 0.996 for all three puffing temperatures, and root mean squares, RMS, varied from 0.007 to 0.008. In the Peleg model R ² varied from 0.952 at 290°C to 0.944 at 370°C, and RMS ranged from 0.026% to 0.028%. It can be deducted from Tables 2 and 3 that both the Maxwell and Peleg models predicted the same (p ≤ 0.05) amount of force that remained unrelaxed. A comparison between the determination coefficient R ² and root mean squares RMS for the force relaxation curves received on the basis of the generalized Maxwell model, Eq. (1) and the Peleg model, Eq. (3) indicates that the generalized Maxwell model predicted experimental data slightly better than the Peleg model. However, the Peleg model has only two parameters whereas the Maxwell model has five parameters, and the Peleg model has more physical meaning. Thus, taking into account the fact that both models enabled good prediction of experimental results, it may be concluded that both of them (the three-element Maxwell model and the Peleg model) can be considered suitable for characterizing the viscoelastic properties of hot-air puffed amaranth seeds.

It can be seen in Fig. 2 that hot-air-puffed amaranth seeds display viscoelastic behavior. Table 2 shows that both elastic parameters, F ₁ and F ₂, and asymptotic residual force, F_AM , of the generalized Maxwell model were significantly (p ≤ 0.05) affected by puffing temperature, and increased with its rise. This Table also shows that the relaxation times τ₁ and τ₂ were not significantly (p ≤ 0.05) affected by puffing temperature and remained on the same level for all puffing temperatures, which means that all samples relaxed at the same rate. At a higher puffing temperature this resulted in a more rigid material and, on the basis of Eq. (2), in less viscous behavior. Similar results were also observed by Krokida et al.[23] They determined the viscoelastic characteristics of dehydrated carrots and potatoes applying stress relaxation tests, and observed that at low moisture content the material regains its elasticity, showing collapsed structure. In the structural sense, the honeycomb-like microstructure of puffed starch contained in amaranth seeds can be perceived as a semi-rigid sponge filled with air. The solid matrix of puffed amaranth starch is made up of consolidated, elastic open and closed cells. A temperature-dependent process of starch degradation probably also influences the elastic properties of puffed amaranth seeds.[22] Amaranth seeds contain 3.1 to 11.5% lipids Eq. (2), contributing a viscous component to the overall viscoelastic behavior of these seeds. Force relaxation experiments differentiated the viscoelastic nature of puffed amaranth seeds due to puffing temperature. Force relaxation tests can be used for systematic studies on the functionality of various puffed cereals.

CONCLUSION

The results obtained indicate that the rheological behavior of hot-air-puffed amaranth seeds is influenced by puffing temperature. Less force was required to cause a given displacement rise in seeds puffed at 290°C than in those puffed at 330 and 370°C. The rupture point, which may be correlated with macroscopic damage of the material, was not observed during compression of puffed amaranth seeds. Each force-deformation curve consists of many step-like changes in force, which can be explained by step collapse of the material microstructure, progressing with increasing deformation. It was observed that hot-air-puffed amaranth seeds displayed viscoelastic behavior. The viscoelastic characteristics of hot-air-puffed amaranth seeds were significantly influenced by puffing temperature. This resulted in an increase in the elastic parameters and a decrease in the viscous parameters with a rise in puffing temperature. Samples puffed at a higher temperature exhibited higher elasticity and the same relaxation times, which is a characteristic of more rigid materials with less viscous behavior. It was also found that the generalized Maxwell model predicted the experimental data slightly better than the Peleg model, but both the three-element Maxwell model and Peleg model can be considered suitable for characterizing hot-air puffed amaranth seeds.

NOMENCLATURE

τ	=	Relaxation time, (s)
η	=	Viscosity, (Pa s)
E	=	Modulus of elasticity, (Pa)
d	=	Deformation, (mm)
F	=	Force, (N)
k	=	Coefficient in Peleg's equation
R 2	=	Determination coefficient
RMS	=	Roots mean squares
t	=	Time, (s)

SUBSCRIPTS

0	=	Initial value
0.4,0.8,1.2	=	Deformation in millimeters
1,2	=	Indexes in equations
AM, AP	=	Asymptotic value determined according to the Maxwell and Peleg models
exp	=	Experimental
mod	=	Based on mathematical model

Notes

*Values followed by the same letters in a column are not significantly different (P ≤ 0.05).

*Values followed by the same letters in a column are not significantly different (P = 0.05).

*Values followed by the same letters in a column are not significantly different (P ≤ 0.05).

18. Konopko H. Wymiana ciepla i masy w procesie ekspandowania nasion. [Heat and mass transfer during process of puffing of seeds]. Project No. 3 P06T 003 22 – The Final Report, State Committee for Scientific Research, Warszaw 2004 1–80