APPLICATION OF DYNAMIC RHEOLOGY METHODS TO DESCRIBING VISCOELASTIC PROPERTIES OF WHEAT GLUTEN

Anna Pruska-Kędzior, Zenon Kędzior, Elżbieta Bera, Katarzyna Hryciuk, Justyna Golińska-Krysztofiak
Institute of Food Technology, The August Cieszkowski Agricultural University, Poznań, Poland

ABSTRACT

Viscoelastic properties of gluten obtained from two Polish cultivars of wheat of contrasted technological value (`Begra´ and `Wilga´) were studied by applying dynamic oscillatory rheometry. Mechanical spectra of gluten were registered in frequency window 0.001 - 200 rad/s and analysed using Cole-Cole functions to obtain quantitative characteristics of viscoelastic properties in terms of J_N⁰ (compliance of elastic plateau), the inverse of compliance equals to the network elasticity modulus G_N⁰, w₀ (central frequency of loss peak) and n (parameter related to broadness of loss peak). Differences in protein fractional composition appearing between individual studied samples are reflected in viscoelastic properties of wheat gluten and are related to weather conditions existing within respective wheat vegetation periods. Cole-Cole equations appeared suitable for obtaining concise quantitative rheological description of wheat gluten viscoelastic properties.

Key words: wheat, gluten, linear viscoelasticity, dynamic rheology measurements, Cole-Cole fit.

INTRODUCTION

Viscoelastic properties of gluten proteins matrix have fundamental impact on rheological properties of wheat bread dough. When dough is developed by mixing, the gluten proteins form continuous three-dimensional viscoelastic network throughout the dough embedding discrete rigid filler i.e. starch granules. This quasi-solid structure is infiltrated with aqueous phase composed of soluble polysaccharides, proteins and emulsified lipids, called dough liquor, and closes dispersed gas phase in form of air bubbles. Therefore, wheat gluten can be considered as a simplified model of viscoelastic skeleton of dough. Wheat gluten is composed essentially of polymerised glutenins and gliadins, which are recently referred to as wheat prolamins [15, 16]. Three-dimensional structure of gluten matrix is stabilised by covalent (disulfide) bonds, hydrogen and non-covalent ionic bonds and hydrophobic interactions [1, 12, 13, 15, 16]. Recognition and understanding of relationship between viscoelastic properties of gluten and both density and distribution of cross-links forming three-dimensional gluten structure remain still main objectives of cereal biochemistry and rheology [1, 2, 4, 8, 11, 12, 14, 15, 16].

Methods of small deformation dynamic rheology are often applied to studying viscoelastic properties of wheat gluten. The results of these measurements are presented as mechanical spectra, i.e. curves representing values of storage modulus G´ and loss modulus G" as a function of angular frequency, G´, G" = f (w). Usually, mechanical spectra are published and interpreted qualitatively in terms of G´ and G" versus frequency without any following mathematical analysis. For physical reasons, only limited frequency window 0.001 - 200 rad/s can be applied in dynamic rheological tests for studying wheat gluten in which most of mechanical spectra are not strongly differentiated. Therefore, it is useful to represent this spectra as a function of storage compliance J´ and loss compliance J" versus frequency what is much more informative as compared to G´ and G" moduli [4, 9, 13, 14].

Much of knowledge from polymer science studies can be usefully applied to increase understanding viscoelastic properties of gluten and their correlation to the protein composition. It was found for synthetic linear and branched polymers viscoelastic liquids that on their mechanical spectra represented as a function J´, J" = f(w) appear two characteristic zones: a loss peak on J" curve at high frequency region of the experimental window and an elastic (rubbery) plateau on J´ curve at the intermediate and low frequency region [5]. For analysis of these two zones Cole-Cole functions are applied [17]. Parameters of Cole-Cole functions describe quantitatively "mechanical strength" of a polymer liquid network (density of crosslinks forming the network) and permit to estimate the length of molecular chains between crosslinks.

However, synthetic polymers have well-defined repetitive monomer structure whereas biopolymer molecules have more complex structure composed of repetitive and non-repetitive monomer motifs, randomly varying depending on genetic and environmental conditions of proteins or polysaccharides synthesis. Also, biopolymer networking involves other mechanisms (covalent and hydrogen bonds, hydrophobic interactions etc.) than this of synthetic polymers (entanglements network). Biopolymers have also higher molecular weights than synthetic polymers. Macromolecules forming gluten structure represent molecular weights distribution ranging from 10⁴ - 2.10⁶ daltons. Therefore, rheological methods applied to studying synthetic polymers cannot be applied automatically to biopolymer systems.

The aim of present work was to evaluate the suitability of Cole-Cole method for describing gluten viscoelastic properties and to estimate the effects of environmental factors (cultivar and climatic conditions) on viscoelastic properties of gluten matrix. The results represent a fragment of multi-year studies of variability of wheat gluten rheological properties of two selected Polish wheat cultivars [11, 14].

MATERIALS AND METHODS

Experimental material

Grain of two wheat cultivars of contrasted technological value, cv. `Begra´ and cv `Wilga´, harvested in 2000 and 2001 (Plant Breeding Station DANKO in Choryń) was milled into laboratory flours using a Quadrumat Senior laboratory mill. Wheat gluten samples were washed-out manually, freeze-dried and stored at 4°C under oxygen-reduced atmosphere.

General technological and chemical characteristics of studied samples

Basic technological tests characterizing impact of protein system on baking quality of wheat flour were made (gluten yield and spreadability, sedimentation value according to Zeleny, farinographic characteristics). Dry matter content and protein content (N×5.7) were determined both for laboratory wheat flours and gluten samples. Protein fractional composition of laboratory wheat flours was determined by three-step extraction procedure according to Coates and Simmonds as modified by Jankiewicz [3, 6].

Rheological studies of wheat gluten

Gluten samples were prepared by rehydrating freeze-dried samples of gluten in 0.1 mole/dm³ aqueous solution of N-ethylmaleimide (NEMI) to prevent chemical changes of gluten structure induced by sulfhydryl - disulfide interactions [4, 11, 14].

Rheological measurements were carried out at 20°C using a Rheometric Scientific stress-controlled rheometer model SR 500. Cone-plate measurement geometry was used (cone diameter 25 mm, cone angle 0.1 radian). The samples were covered with 0.1 mole/dm³ NEMI to avoid drying. The range of linear viscoelasticity of studied gluten samples was established in preliminary tests. All samples showed linear viscoelastic response up to at least 10% strain value. Mechanical spectra of wheat gluten samples were measured for frequency range 0.001 - 200 rad/s. The amplitude of strain was kept close to 3%.

Interpretation of rheological measurements

Values of storage modulus G´ and loss modulus G" were transformed into storage compliance J´ and loss compliance J" (equation 1 i 2).

(1)

(2)

Mechanical spectra expressed in terms of J´ and J" can be fitted with Cole-Cole equations (equation 3, 4 and 5, respectively).

(3)

(4)

(5)

Equations (3-5) are applied to obtain parameter J_N⁰ (compliance of elastic plateau), w₀ (central frequency of loss peak) and n (parameter related to broadness of loss peak). At the zone of elastic plateau the inverse of compliance equals to the network elasticity modulus G_N⁰ (1/J_N⁰ = G_N⁰) [5, 17]. G_N⁰ can be taken as a measure of the network elasticity. Parameters w₀ and n characterize viscous dissipation of energy resulting from the movement of fragments of polymer chains.

These parameters describe viscoelastic properties and provide information on networking state of polymer structure in upper transient zone of mechanical spectrum, so called "softening region", which begins at the highest experimental frequencies and extends over experimentally accessible frequency window.

RESULTS AND DISCUSSION

General characteristics of studied samples

Samples of wheat grain used in the present study differed significantly in their baking quality. Wheat cv. `Begra´ represented material of good technological value and wheat cv. `Wilga´ represented material of poor technological quality (Table 1). Studied wheat cultivars showed also differences in their protein content (N×5.7) and protein fractional composition (Table 1 and Table 2). Wheat cv. `Begra´ contained below 19% of albumins and globulins (protein extract in 0.01 mole/dm<sup>3</sup> pyrophosphate buffer pH 7.0), while cv. `Wilga´ contained over 26% of these proteins. Compared wheat cultivars differed also in the content of gliadins and glutenins and quantitative ratios of these protein fractions (protein extracts respectively in 0.05 mole/dm³ acetic acid and in 0.1 mole/dm³ NaOH). Wheat cv. `Begra´ contained evidently more glutenins than wheat cv. `Wilga´, and its gliadin-to-glutenin ratio showed lower variability than those for wheat cv. `Wilga´. Gliadin-to-glutenin ratio for wheat cv. `Begra´ was close to 1.9:1 while for wheat cv. `Wilga´ varied in the range 2.2:1 - 2.5:1. Gluten samples isolated from two studied wheat cultivars differed in their technological quality and, due to differences in wheat flour protein fractional composition, also in gliadin and glutenin protein fractions content and their quantitative ratios (Table 1, Table 2).

Table 1. General characterization of wheat flours and vital gluten obtained from wheat cultivars `Begra´ and `Wilga´

Parameter	Wheat sample
	Begra 2000	Begra 2001	Wilga 2000	Wilga 2001
Wheat flour characteristics
Humidity [%]	13.3 ± 0.3	14.2 ± 0.3	13.5 ± 0.2	13.6 ± 0.1
Protein content (N×5.7) [%]	11.0 ± 0.1	13.1 ± 0.4	9.4 ± 0.1	11.7 ± 0.3
Sedimentation number(Zeleny test) [ml]	30.0 ± 0.5	22.0 ± 0.3	16.0 ± 0.3	14.0 ± 0.3
Farinographic characteristics
Water absorption (at 500 B.U.) [%]	54.4 ± 0.3	52.8 ± 0.3	52.4 ± 0.3	49.0 ± 0.3
Time of dough development [min]	2.0	1.7	1.5	2.0
Dough stability period [min]	5.0	2.5	2.0	1.3
Gluten characteristics
Wet gluten yield [%]	29.1 ± 0.9	28.0 ± 0.8	23.8 ± 0.6	29.0 ± 0.8
Gluten spreadability [mm]	2.0	4.0	6.0	7.0
Gluten value	54.0	50.0	38.0	45.0
Protein content [%]	83.9 ± 1.7	88.3 ± 1.2	78.6 ± 1.2	81.5 ± 1.3

Table 2. Comparison of protein fractional composition of wheat cultivars `Begra´ and `Wilga´

Wheat flour	Protein extracted from wheat flour [% of total protein content in flour]				Gliadin/glutenin ratio
	0.01 mole/dm3 Pyrophosphate buffer pH 7.0	0.05 mole/dm3 Acetic acid	0.1 mole/dm3 NaOH	total
Begra
2000	18.87 ± 0.58	47.72 ± 1.80	25.27 ± 0.88	91.86	1.89
2001	14.30 ± 0.68	45.44 ± 1.27	24.36 ± 0.99	84.10	1.87
Wilga
2000	34.88 ± 0.99	41.41 ± 1.39	19.21 ± 0.89	95.50	2.16
2001	26.06 ± 0.82	41.57 ± 0.87	16.45 ± 0.74	84.08	2.53

Rheological characterisation of wheat gluten samples

Mechanical spectra determined for studied wheat gluten samples are presented on Fig. 1 as G´,G" versus w. All gluten samples represent characteristic features of viscoelastic liquid [4, 5, 7, 8, 11, 13, 14, 17]. However, significant differences in G´ and G" patterns occurred depending on varietal and climatic conditions (effect of cultivar and year of harvest). Gluten able to forming stronger crosslinked networks (cv. `Begra´) showed weaker dependence of rheological properties on climatic conditions. Viscoelastic properties of gluten obtained from technologically weaker wheat cultivar (`Wilga´) varied in broad range. In case of cv. `Wilga´ 2001 gluten sample, besides the fact that the storage (G´) and loss (G") moduli are altogether shifted to considerably lower values as compared to `Begra´, the spectrum is qualitatively different because this time the upper crossover of moduli occurs at angular frequency ca 10 rad/s. On Fig. 2 the same mechanical spectra are shown as the curves of compliances J´ and J", as a start point to data analysis according to Cole-Cole method. Comparing to curve patterns of G´ and G" moduli, J´ and J" curves demonstrate more clearly qualitative differences in viscoelastic properties of studied gluten samples.

Fig. 1. Effect of varietal factor and climatic conditions on viscoelastic properties of wheat gluten (mechanical spectrum G´,G" = f(w)). Symbols: gluten sample from wheat harvested in 2000: - G´, - G"; gluten sample from wheat harvested in 2001: - G´, - G"

Fig. 2. Mechanical spectra of wheat gluten cv. `Begra´ and `Wilga´ presented as dependence: storage compliance J', loss compliance J" versus frequency w. Symbols: gluten sample from wheat harvested in 2000: - J´, - J"; gluten sample from wheat harvested in 2001: - J´, - J"

These results show clear positive dependency of gluten rheological properties on climatic conditions occurring within the wheat breeding period in 2000 and 2001. Vegetation period in 2000 was eight days longer and was also warmer and a little more humid than vegetation period in 2001. Cumulated temperature measured at 2 m height was equal within the vegetation period to 1783.1°C (daily average 16.0°C) and 1494.7°C (daily average 14.5°C) respectively in 2000 and 2001. Cumulated minimal temperature measured at the soil level was equal to 743.4°C (daily average 6.7°C) and 460.7°C (daily average 4.4°C) respectively within the vegetation period 2000 and 2001. Cumulated atmospheric precipitation within vegetation periods 2000 and 2001 was equal respectively to 174.2 and 125.6 mm (data provided by Meteorological Station of PAN in Turew, nearby Choryń).

Next, mechanical spectra of wheat gluten samples were submitted to mathematical analysis according to Cole-Cole method. Graphical solution of this method involves fitting to experimental data presented as J´ vs J" arc of a circle passing by the origin of the coordinates [4, 9, 13, 14, 17]. Examples of fitting of Cole-Cole arc to experimental data are shown on Fig. 3. As the result of Cole-Cole fitting procedure parameters J_N⁰ and G_N⁰, w₀ and n, have been found (Table 3).

Fig. 3. Cole-Cole fit. Comparison of wheat gluten of strong (cv. `Begra´) and weak (cv. `Wilga´) structure. Symbols:  - experimental points fitted to the arc,  - experimental points out of the arc, --------- - the Cole-Cole arc

Table 3. Parameters of Cole-Cole fit for gluten samples obtained from wheat cultivars `Begra´ and `Wilga´

Sample	JN0 [Pa-1]	GN0 [Pa]	w0 [rad/s]	n	r
Begra
2000	5.81.10-4	1721.0	2.3714	0.4498	0.9953
2001	1.20.10-3	837.2	0.4208	0.4209	0.9971
Wilga
2000	1.14.10-3	874.7	1.2793	0.4734	0.9990
2001	4.09.10-3	244.3	0.6801	0.5816	0.9999

Very high correlation coefficients were obtained for Cole-Cole fit for analysed mechanical spectra, ranging respectively from r = 0.995 (`Begra´ 2000) to r = 0,9999 (`Wilga´ 2001). Lowest compliance J_N⁰ showed gluten obtained from wheat cv. `Begra´ 2000, and the highest one - wheat gluten `Wilga´ 2001 (respectively 5.81.10^-4 and 4,09.10^-3 Pa^-1). Samples of gluten obtained from wheat harvested in 2001 had clearly weaker three-dimensional crosslinking of network than gluten from wheat harvested in 2000. It is proved by lower values of central frequency of loss peak w₀ and higher values of J_N⁰ found for gluten samples from wheat harvested in 2001 as compared to those harvested in 2000.

On the basis of fit parameters for each gluten sample curves of compliances J´ and J" were calculated from Cole-Cole equations. On Fig. 4 patterns of calculated J´ nad J" are compared to respective experimental curves for two (`Begra´ 2000 and `Wilga´ 2001) extremely different in their rheological properties gluten samples. On calculated curves location and broadness of loss peak (J" calculated) as well as location of initial zone of elastic plateau in the region of lowest applied frequencies (J´ calculated) are clearly seen.

Fig. 4.Comparison of loss peaks and zones of elastic plateau of strong (cv. `Begra´) and weak wheat gluten (cv. `Wilga´). Symbols: - J´ experimental, - J" experimental, --------- - J´ calculated from equation 3, - - - - - J" calculated from equation 4

As it was demonstrated for studied wheat gluten samples Cole-Cole approximation is satisfactory for quantitative describing and distinguishing viscoelastic properties of gluten network.

At present, there is no analytical method, which could enable determining in situ true density and distribution of disulfide and hydrogen bonds and hydrophobic interactions forming three-dimensional structure of gluten. Similarly, it remains still impossible determining true degree of in situ polymerisation of particular fractions of gliadins and glutenins, which constitute gluten structure. The only method of understanding the impact of gluten structure on its rheological properties consists in indirect inferring on the basis of parallel studying fractional composition of gluten protein system and correlating these results with quantitative rheological parameters [4, 9, 13]. Results being obtained recently in studies combining biochemical analysis of gluten protein system and its rheological properties permit to hope for achieving in future satisfying approximation of correlation between structure of individual types of proteins (mainly gliadins and glutenin subunits of high and low molecular weight), their quantitative content in given gluten preparation and rheological properties of three-dimensional structures formed by these proteins [4, 8, 9, 12, 13-16].

CONCLUSIONS

Studied samples of wheat cultivars `Begra´ and `Wilga´ show contrasted technological properties, respectively good and poor ones, and different susceptibility to environmental conditions what is proved by technological parameters and general protein fractional composition.

Differences in protein fractional composition appearing between individual studied samples are reflected in viscoelastic properties of gluten characterised with dynamic rheology method and described in form of mechanical spectra at the frequency window 0.001 - 200 rad/s. Positive relation is seen between technological and rheological properties of wheat gluten and climatic conditions of wheat breeding.

Mechanical spectra of gluten show a loss compliance peak and elastic plateau, indicating a network-type structure within the experimental frequency window. In the case of all gluten samples tested, experimental window frames only a section of the elastic plateau. Mechanical spectra show clearly that the elastic plateau of studied gluten samples extends beyond the lowest frequency of the experimental window. Full range of elastic plateau cannot be measured using dynamic method because the dynamic measurement cannot be performed at frequencies below 0.001 rad/s, especially with controlled stress rheometers. Hence, moduli or compliances at frequency below 0.001 rad/s have to be deduced from the data of retardation tests.

Cole-Cole equations describing viscoelastic properties of networked polymer systems can be applied to studying biopolymer systems like wheat gluten, enabling concise quantitative rheological characterisation of studied material. It has been evaluated for gluten obtained from wheat of good and weak baking quality.

REFERENCES

1. Belton P.S. (1999): On the elasticity of wheat gluten. J. Cereal Sci., 19, 103-107.

2. Bloksma A.H. (1990): Rheology of the breadmaking process. Cereal Foods World, 35, 228-236, 959-960.

3. Coates, J.H., Simmonds, D.H. (1962): Proteins of wheat and flour extraction, fractionation and chromatography of the buffer soluble proteins of flour. Cereal Chem., 38, 256-272.

4. Cornec M., Popineau Y., and Lefebvre J. (1994): Characterisation of gluten subfractions by SE-HPLC and dynamic rheological analysis in shear, J. Cereal Sci., 19, 131-139.

5. Ferry J. (1980): Viscoelastic Properties of Polymers, 3rd ed., (John Wiley and Sons, Inc., New York).

6. Jankiewicz, M., Pomeranz, Y. (1965): Isolation and characterisation of wheat flour proteins. I. Separation of salt- and acid dispersible proteins by gel filtration, polyacrylamide gel electrophoresis and sucrose gradient ultracentrifugation. J. Sci. Food Agric., 16, 644-652.

7. Janssen A.M., van Vliet T., Vereijken J.M. (1996): Rheological behaviour of wheat glutens at small and large deformations. Comparison of two glutens differing in bread making potential. J. Cereal Sci., 23, 19-31.

8. Janssen A.M., van Vliet T., Vereijken J.M. (1996): Rheological behaviour of wheat glutens at small and large deformations. Effect of gluten composition. J. Cereal Sci., 23, 33-42.

9. Hargreaves J., Popineau Y., Cornec M., Lefebvre J. (1996): Relation between aggregative, viscoelastic and molecular properties in gluten from genetic variants of bread wheat. Int. J. Biol. Macromol. 18, 69 - 75.

10. Lefebvre J., Popineau Y., Pruska-Kędzior A., Kędzior Z. (2000): A phenomenological analysis of fully hydrated gluten viscoelastic behaviour over a large frequency range. W: D.M. Binding, N.E. Hudson, J. Mewis, J.-M. Piau, C.J.S. Petrie, P. Townsend, M.H. Wagner, K. Walters (editors): Proceedings of the XIIIth International Congress on Rheology (Cambridge, UK, August 2000), vol. 4: 395-397.

11. Lefebvre J., Pruska-Kędzior A., Kędzior Z., Lavenant L. (2003): A phenomenological analysis of wheat gluten viscoelastic response in retardation and in dynamic experiments over a large time scale. J. Cereal Sci., 38, 257-267.

12. Lindsay M.P., and Skerrit J.H. (1999): The gluten macropolymer of wheat flour doughs: structure-function perspectives. Trends Food Sci. Technol., 10, 247-253.

13. Popineau Y., Cornec M., Lefebvre J., Marchylo B. (1994): Physical-chemical characterisation of glutens and gluten subfractions of near isogenic lines of wheat "Sicco": influence of HMW glutenin subunits on the glutenin polymerisation and on rheological properties of glutens. J. Cereal Sci., 19, 231-241.

14. Pruska-Kędzior A., Lefebvre J., Kędzior Z., Bojakowska E., Jankiewicz M. (2000): Study of shear viscoelasticity of wheat gluten and mixed wheat gluten - pea protein systems by dynamic assay and creep and recovery analysis. W: Fischer P., Marti I., Windhab E.J. (editors): Proceeding of 2^nd International Symposium on Food Rheology and Structure, ISFRS 2000, March 12-16. 2000, Zurich, Switzerland, 309-313.

15. Shewry P.R., Miles M.J., and Tatham A.S. (1994): The prolamin storage proteins of wheat and related cereals. Prog. Biophys. Mol. Biol., 61, 37-59.

16. Shewry P.R., Tatham A.S. (1997): Disulphide bonds in wheat gluten proteins. J. Cereal Sci., 25, 207-227.

17. Tschoegl N.W. (1989): The phenomenological theory of linear viscoelastic behavior. An introduction. Springer Verlag, Berlin, 433-435.

---

Anna Pruska-Kędzior
Institute of Food Technology,
The August Cieszkowski Agricultural University, Poznań, Poland
Wojska Polskiego 31, 60-624 Poznań, Poland
email: kedziorm@au.poznan.pl

Zenon Kędzior
Institute of Food Technology,
The August Cieszkowski Agricultural University, Poznań, Poland
Wojska Polskiego 31, 60-624 Poznań, Poland

Elżbieta Bera
Institute of Food Technology,
The August Cieszkowski Agricultural University, Poznań, Poland
Wojska Polskiego 31, 60-624 Poznań, Poland

Katarzyna Hryciuk
Institute of Food Technology,
The August Cieszkowski Agricultural University, Poznań, Poland
Wojska Polskiego 31, 60-624 Poznań, Poland

Justyna Golińska-Krysztofiak
Institute of Food Technology,
The August Cieszkowski Agricultural University, Poznań, Poland
Wojska Polskiego 31, 60-624 Poznań, Poland

---

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.

---

Main  -  Issues  -  How to Submit  -  From the Publisher  -  Search  -  Subscription