- Email: [email protected]
Dough characteristics of Irish wheat varieties I. Rheological properties and prediction of baking volume
LWT - Food Science and Technology 44 (2011) 594e601
Contents lists available at ScienceDirect
LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt
Dough characteristics of Irish wheat varieties I. Rheological properties and prediction of baking volume A. Ktenioudaki a, b, *, F. Butler a, E. Gallagher b a b
UCD School of Agriculture, Food Science & Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland Teagasc, Ashtown Food Research Centre, Ashtown, Dublin 15, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 May 2009 Received in revised form 5 November 2010 Accepted 9 November 2010
Eight wheat varieties suitable for cultivation in Ireland were examined for the rheological properties and baking quality. Large deformation extensional rheology was used employing the Extensograph, the Kieffer extensibility rig, the Alveograph and biaxial extension by uniaxial compression. Similar discrimination between the wheat samples was achieved with both uniaxial extension methods used. Stress during uniaxial extension was higher at all strains measured than biaxial extension and the difference between uniaxial and biaxial stress increased with increasing strain. Also, differences existed in the strain hardening index between uniaxial and biaxial extension which depended on the sample tested. Significant correlations were established between loaf volume and the rheological properties determined in both uniaxial and biaxial extension. It was found that high loaf volumes can be achieved when biaxial and uniaxial extensibility is high and biaxial extensional viscosity is low. The importance of considering the standard error for validating the value of the correlation for prediction purposes was also demonstrated. When the standard error of the correlation was considered, the value of the correlation for prediction purposes was limited in practice. Nevertheless, the rheological tests provided useful information, and the results were valuable for screening flours for baking quality. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Dough rheology Biaxial extension Uniaxial extension Baking
1. Introduction Breadmaking is a complex process that involves many physicochemical and structural transformations which lead to the production of an aerated baked product from basic ingredients such as flour, water, yeast and salt. Generally, the breadmaking process is divided into three main steps: mixing, fermentation (proving), and baking. During all these stages, deformations of different magnitudes take place and rheology is involved in every step of the process. During mixing, extreme deformations take place above the point of rupture as a result of forces exerted on the dough from the mixer, whereas during proving and baking the deformations are much smaller and are due to the difference in pressure between the gas cells and the atmosphere (Bushuk, 1985). The importance of rheology in breadmaking and the impact of dough rheological properties on the quality of the baked product have led to the development of numerous rheological tests, employed both in research and in the cereal industry. Extensional
* Corresponding author. Teagasc, Ashtown Food Research Centre, Ashtown, Dublin 15, Ireland. Tel.: þ353 8 51469941; fax: þ353 1 8059550. E-mail address: [email protected] (A. Ktenioudaki). 0023-6438/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2010.11.014
flow measurements such as uniaxial and biaxial extension are considered more appropriate as extensional flow is present in many operations such as mixing, sheeting, moulding, and bubble expansion. According to Gras, Carpenter, and Anderssen (2000), the mixing action that occurs in a Mixograph can be interpreted as a series of extension tests. The extension to rupture occurring during an extension test corresponds to the elongate - rupture that occurs during mixing of the elongate-rupture-relax oscillations that form the bandwidth of the Mixograph test. Therefore the bandwidth recorded from the Mixograph can be used to better determine the evolving rheology of the dough during mixing. Ideally, every rheological test that aims at predicting the behaviour of the material should be taking place at relevant deformation conditions to what the material experiences during processing (Dobraszczyk, Smewing, Albertini, Maesmans, & Schofield, 2003). According to van Vliet, Janssen, Bloksma, and Walstra (1992), the dough around a growing cell is tangentially extended in two directions parallel to the surface of the gas cell and compressed radially perpendicular to the surface of the gas cell due to the radially acting compression force caused by excess pressure inside the gas cells. Biaxial extension at large strains and low strain rates is considered to be the relevant deformation that takes place around a growing cell and it has been used to study the rheological
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
properties of dough (Janssen, van Vliet and Vereijken, 1996; Kokelaar, van Vliet, and Prins 1996; Rouille, Valle, Lefebvre, Sliwinski, and van Vliet., 2005; Sliwinski, Kolster, and van Vliet, 2004a). These tests make possible the calculation of a rheological property known as strain hardening, which van Vliet et al. (1992), have proposed it provides the dough films of expanding gas cells with greater stability and protects them against premature rupture. Therefore, the extent of strain hardening will determine the breadmaking potential. Many researchers have studied the strain hardening phenomenon and have demonstrated that it relates to the baking potential of flour (Dobraszczyk & Salmanowicz, 2008; Sliwinski, Kolster, and van Vliet, 2004b). Numerous tests exist to measure the uniaxial extensional properties of dough. The test methods include the Simon extensometer, the Brabender Extensograph, and the Stable Micro Systems Kieffer dough and gluten extensibility rig. The Extensograph is an empirical test that has been used for years to perform uniaxial extension on dough and the measured properties have been correlated with baked volumes. However a few disadvantages exist such as: the force and the extension are not expressed in Newton and strain respectively, the amount of dough deformed increases with extension, and therefore the conversion of the Extensograph curves to stress and strain curves is restricted (Bloksma & Bushuk, 1988). An apparatus similar to the Extensograph was developed the Kieffer dough and gluten extensibility rig. With this apparatus only a small amount of dough is required (about 0.8 g), the force is measured in Newton and the speed of the test can be adjusted. This way more relevant strain rates can be applied and the results can be expressed in stress and strain data (Dunnewind, Sliwinski, Grolle, & van Vliet, 2004). Grausgruber, Schoggl, and Ruckenbauer (2002) compared the Kieffer extensibility rig with the Extensograph and found significant relationships between the parameters obtained from the two methods. Many researchers have used uniaxial extension tests to study the rheological properties of doughs (de Bruijne, de Looff, van Eulem, & Carter, 1990; Collar, Santos, & Rosell, 2007; Dobraszczyk & Salmanowicz, 2008; Sliwinski et al. 2004b; Suchy, Lukow, & Ingelin, 2000; Tronsmo et al., 2003) and good correlations with baking quality have been reported (Sliwinski et al., 2004b; Suchy et al., 2000; Tronsmo et al., 2003). Sliwinski et al. (2004b) studied the large deformation and fracture properties of flour in uniaxial extension using the Kieffer rig. They found that high loaf volumes were achieved from those varieties that exhibited intermediate stress levels at large strains. They also found that the relationship between loaf volume and stress and strain fracture is rate dependant. Tronsmo et al. (2003) studied the rheological properties at small and large deformations of four wheat cultivars. Two varieties had the High Molecular Weight glutenin subunits (HMWgs) pair 5 þ 10 and therefore resulted in good breadmaking quality and the other two had the HMWgs pair 2 þ 12 that is known to result in poor performance during baking. They found that maximum resistance to extension (Rmax) measured with the Kieffer rig and the biaxial strain hardening as measured with the Dobraszczyk/Roberts inflation rig could discriminate between these varieties. Anderssen, Bekes, Gras, Nikolov, and Wood (2004) studied the rheological properties of eight wheat varieties using a uniaxial micro-extension test. They indicated the need to relate rheological properties such as strength and extensibility to the glutenin composition of the flours in order to achieve better understanding of their relation to baking performance. The eight flours were divided into three categories (weak, intermediate, strong) with distinct differences in the HMW glutenin subunit allelic composition. Distinct morphological differences in the extensograms were found leading to the conclusion that the alleles significantly affect
595
the rheology of the flours. It was also shown that the measurement of maximum resistance to extension (Rmax) clearly separates the flours depending on their HMW glutenin subunit allelic composition. They also calculated the extensibility at the point of Rmax (ERmax)as well as the extensibility at the point of rupture(Erupture) and found that parameters (ERmax and Erupture ERmax) other than the ones traditionally used were important in interpreting rheologically the extensograms and could significantly differentiate between the varieties. The objective of this paper was to conduct a comprehensive study of the rheological properties of Irish wheat varieties and to identify the rheological properties that would successfully discriminate between varieties and predict their baking quality. Eight varieties suitable for growing in the Irish climate were chosen, as well as a Canadian and a German blend of varieties (for comparative purposes). Four large deformation extensional rheological tests were employed to provide empirical as well as fundamental measurements both in uniaxial and biaxial extension. 2. Materials & methods 2.1. Materials Eight wheat varieties suitable for cultivation in Ireland were selected. The varieties were: Gulliver, Solstice, NSLWW89, Einstein, Consort, Equation, Trappe, and Raffles. These are considered to be breadmaking varieties, with the exception of Consort and Equation, which are considered to be suitable for biscuits and feed respectively, and were included in the study to achieve a wider variation in wheat properties. Four varieties were supplied from a UK breeding company (Einstein, Gulliver, Solstice, and NSLWW89), and the rest were provided by Irish suppliers (Goldcrop, SeedTech, and Germinal seeds) from various geographical locations. The seeds were from the 2007 harvest and approximately 50 kg of grains were received for each variety. A blend of Canadian varieties and also a blend of German varieties were included in the study, obtained from an Irish milling company (Odlum Ltd., Dublin). The varieties were cleaned using a sample cleaner (SLN3, A/s Rationel Kornservice, Denmark) and conditioned to 16% moisture content prior to milling. The samples were milled using a Bühler mill (Bühler, Switzerland) to a milling degree of 70% following the AACC Bühler method (AACC, 1988). Analysis for moisture content (ICC standard Method No. 110/1, 1976), a-amylase activity (ICC standard method No. 107, ICC, 1968), protein content using a Leco protein analyzer (Leco FP-428 Nitrogen Analyzer, Leco Corporation, St. Joseph, MI, U.S.A), and for starch damage using an amperometric method (SDmatic, Chopin, France) (ICC standard method No 172, ICC, 2007) took place. Duplicate measurements were carried out for the chemical analysis and the results were averaged. The mixing properties were investigated using the Brabender Farinograph (Brabender OHG, Duisberg, Germany) following the Farinograph British Standard method No. 4317-20:1999 (British standards, 1999) (one test per wheat type). 2.2. Dough preparation Dough samples were prepared in a Farinograph mixing bowl using 300 g flour, 6 g salt (Pure Dried Vacuum salt, Ineos Enterprises, UK), and 3 g of emulsified bread fat (Irish Bakels Ltd., Dublin, Ireland). The amount of water added was as per the Farinograph water absorption value. The mixing times were based on the development time plus 1 min. A new batch of dough was mixed for each rheological test. All the experiments were replicated three times.
596
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
2.3. Uniaxial extension (Extensograph) A Brabender Extensograph (Brabender OHG, Duisberg, Germany) was used to perform uniaxial extension on the doughs. The method as described in Ktenioudaki, Butler, and Gallagher (2010a) was followed. Duplicate measurements were made for each replicate and the results were averaged. 2.4. Uniaxial extension (Kieffer dough and gluten extensibility rig)
et al. (2010a). Prepared dough was placed in the Alveograph mixer only to extrude the sample pieces. The results of each replicate were the average of five repeats of the same batch. The parameters recorded included the biaxial extensibility of dough (L), the maximum overpressure (P, height of the curve), the swelling index (G) (amount of air to inflate the bubble), the deformation energy (W) (area under the curve), and the configuration ratio (P/L). 2.6. Biaxial extension (compression method)
Uniaxial extension using the Kieffer dough and gluten extensibility rig (Stable Micro Systems, Surrey, UK) was carried out similar to the method described by Dunnewind et al. (2004). A texture analyzer (TA-XT2i, Stable Micro Systems, Surrey, UK) fitted with the Kieffer extensibility rig and equipped with a 5 kg load cell was used. The samples were prepared following the method described in the literature accompanying the Kieffer extensibility rig (Stable Microsystems, UK). A piece of dough (25 g) was placed to rest in a proofer (Koma SDCC-1P/W, Koma Koeltechnische Industrie B.V., The Netherlands) with controlled temperature (30 C) and relative humidity (85%) for 20 min. The sample was then rolled by hand into a cylindrical shape, placed in the lubricated Teflon mould and compressed with the lubricated top Teflon plate. The sample was allowed to rest for another 45 min in the proofer at 30 C and 80% relative humidity. Before the start of the test the sample was clamped between the plates of the Kieffer rig. The maximum force in tension and the displacement were recorded. Three different displacement speeds were used: 0.4, 2.0 and 3.3 mm/s. Three dough samples were used for each speed and the results were averaged. According to Dunnewind et al. (2004), clamping of the sample results in a certain degree of sagging of the dough and therefore the actual extension of the dough piece starts somewhat above the surface of the lower plate. Corrections were applied to account for this by measuring the distance the hook travelled from a predetermined starting point to the point where it touched the dough. The difference between this distance and the known distance from the starting point to the top of the bottom plate gave the height of sagging. Adding this to the point of the top surface of the bottom plate, the point of when the actual extension started was determined. The point when the hook touched the dough was determined as the point which the measured force deviated from zero. The fundamental rheological parameters Hency strain 3H, strain rate 3, stress s, and the apparent extensional viscosity hE were calculated as described by Dunnewind et al. (2004). Finally, similar steps were followed to calculate the strain hardening index as those for the biaxial extension described by Kokelaar et al. (1996). 2.5. Biaxial extension (Alveograph) Biaxial extension was performed using an Alveograph (Chopin S.A., Villeneuve la Garenne, France) as described in Ktenioudaki
The method as described in Ktenioudaki, Butler, and Gallagher (2010b) was followed using a texture analyzer (TA-XT2i, Stable Micro Systems, Surrey, UK). Three compression speeds were used (0.2, 1.0 and 2.0 mm/s) and the dough samples were compressed to 1.5 mm height. Nine samples (three for every speed) were prepared in total for each replicate from the same batch of dough and the results were averaged. 2.7. Baking The baking method used was as described in Ktenioudaki et al. (2010b). Briefly, pup loaves (65 g dough pieces in 100 55 35 mm tins) were produced following a straight dough baking procedure. The water added and the mixing time, were the same as used for the doughs for the rheological tests and the samples were prepared in the Farinograph mixing bowl. Six loaves were produced per batch. The loaves were allowed to cool for 2 h after baking and placed into plastic bags. After 24 h, the loaves were tested for loaf volume using a volume measurer (BVM-L370, TexVol Instruments, Sweden). 2.8. Statistical analysis One way analysis of variance was carried out using Minitab (Minitab version 15.1.1.0., Minitab Ltd., UK) to determine significant differences in the measured properties. Pearson’s correlation analysis between selected parameters was carried out using Minitab (Minitab version 15.1.1.0., Minitab Ltd., UK). Stepwise multiple regression was carried out using SAS software (SAS 9.1.3., SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Flour characteristics The data from the flour analysis are presented in Table 1. The protein content varied from 6.6 g/100g (Trappe) to 12.9 g/100g (Canadian blend). The typical range for Irish wheat varieties is between 8 and 10 g/100g. Consort and Trappe appeared to have medium to high a-amylase activity whereas the rest of the varieties
Table 1 Physical and chemical characteristics of flours tested. Variety
Protein g/100g (at 14g/100g mc)
Falling No (s)
Moisture g/100 g
Starch damage
Water absorption %
Development time (min)
Stability (min)
Degree of softening (BU)
Consort Equation Trappe Raffles Einstein Gulliver Solstice NSLWW89 Canadian blend German blend
7.4 9.4 6.6 9.3 9.2 10.0 9.4 9.3 12.9 12.5
253 327 266 454 433 471 447 425 428 456
13.6 13.6 14.0 13.9 13.8 13.4 13.4 12.8 13.4 13.3
9.31 27.51 30.03 39.10 36.60 40.43 41.10 46.45 36.97 37.52
50.0 58.5 58.0 58.0 58.0 62.5 62.7 64.7 65.5 63.0
1.5 2.0 2.0 2.5 2.0 3.0 2.0 2.0 6.0 2.5
1.0 2.0 2.0 2.5 3.5 4.0 2.0 1.5 9.0 5.0
180 140 110 100 90 90 100 150 50 70
The results of protein content, moisture and Falling No are the mean values of two repeats.
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
would benefit from the addition of a-amylase. The low water absorption of the Consort flour was expected as it is considered to be a biscuit variety whereas the absorptions of the remaining samples were in the normal range for bread varieties. The high development time for the Canadian blend is typical of strong Canadian varieties.
597
samples tested (58 mm). Significant differences were observed in terms of the maximum resistance to extension measured. The Canadian blend had the highest value (615 BU). The biscuit variety Consort together with Equation and Solstice gave the lowest values varying between 185 and 292 BU whereas the values for the rest of the varieties ranged between 305 and 453 BU (P < 0.001).
3.2. Extensograph 3.3. Kieffer dough and gluten extensibility rig Fig. 1a shows typical extensograms for all of the wheat varieties examined. The plot for the Canadian blend was clearly separated from the rest of the samples as the one with the highest energy (largest area under the curve). Consort flour gave a typical extensogram of a biscuit variety with high extensibility and low maximum resistance to extension. Significant differences (P < 0.001) were observed among the samples for both the extensibility and the maximum resistance to extension results. The Canadian blend, the German blend, Einstein, Gulliver and Solstice produced doughs with the highest extensibilities (between 183 and 200 mm). However, only the Canadian blend was accompanied by equally high resistance to extension. The rest of the varieties had similar extensibilities (varying between 153 and 173 mm) with the exception of Trappe which had the lowest extensibility of all the
Force-displacement curves similar to the ones from the Extensograph were obtained and are shown in Fig. 1b for the displacement speed of 3.3 mm/s. The Canadian blend stands out again as the strongest flour. Flour from Trappe variety had the lowest extensibility and the smallest area under the curve, in a similar way as presented in the extensogram in Fig. 1a. Both force and the displacement increased with increasing the displacement speed. Similar trends were observed by Sliwinski et al. (2004a) and Dunnewind et al. (2004). At the 3.3 mm/s displacement speed, Equation, Einstein, Gulliver, Solstice and the German blend exhibited the highest extensibility (from 102 to 113 mm) whereas the Canadian blend exhibited the highest maximum force recorded (0.27 N). The lowest extensibility was recorded for the Irish variety
Fig. 1. Typical extensograms of the wheat varieties during uniaxial extension with a) the Brabender Extensograph b) the Kieffer extensibility rig at 3.3 mm/s displacement speed.
598
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
Trappe (52 mm) whereas the lowest force for the biscuit variety Consort was 0.12 N. A similar pattern of results were obtained at the other two displacement speeds (results not shown). Stress and strain were calculated from the force and displacement data, accounting for the changing dimensions of the sample and assuming a constant volume. At all three speeds, stress increased with increasing strain and the stress was higher at higher displacement speeds. A constant strain rate was chosen (3 ¼ 0.01 s1) and stressestrain curves were produced. Differences in stress between varieties were higher at higher strains. At a Hencky strain of 1.20, the stress levels are similar for the varieties and only the Canadian blend exhibited significantly higher stress from the Solstice flour (P < 0.05). At a higher strain of 1.75, the Canadian blend exhibited significantly higher stress levels (19.8 kN/m2) than the rest of the samples (P < 0.001). Similarly to the calculations for the biaxial extension test, the natural logarithm of the stress was plotted against the Hencky strain (Fig. 2a) and the slope of these curves corresponds to the uniaxial strain hardening index. The values of strain hardening index (y) are presented in Table 2 together with the calculated stress and strain at the point of fracture. Dough from the Canadian blend had the highest strain hardening index (1.9) whereas the Irish variety Trappe had the lowest value (1.5). Stress and strain at the point of fracture increased with increasing deformation rate. Fracture stress was higher for the Canadian and the German blend whereas it was the lowest for Trappe for all three displacement speeds. Less variation was observed in the strains at
the point of fracture where Trappe had the lowest value (P < 0.001) at all three speeds. 3.4. Biaxial extension (Alveograph) Doughs made from the Trappe variety had the highest P value (P < 0.001), followed by Raffles (67 and 54 mm respectively), and the lowest extensibilities L (26 and 81 mm respectively). The biaxial extensibility L varied between 26 mm (Trappe) to 131 mm (Solstice). The P/L ratio (2.6) for Trappe was unacceptably high for baking flours, whereas there was no difference between the other flours (mean P/L value was 0.4). Significant differences were noted (P < 0.001), in terms of the W values, with the Canadian blend having the highest value (215 106 J) and Trappe having the lowest value (69 106 J). Finally, G values varied significantly between 11 ml (Trappe) to 26 ml (Gulliver) (Data not shown). 3.5. Biaxial extension The same analysis as proposed by Kokelaar et al. (1996) was carried out to calculate stress-strain curves from the forcedisplacement measurements. As stress increased with strain, the differences between the samples became more obvious. At low strains (0.1e0.5) Trappe exhibited the highest stress (2.23 kN/m2, P < 0.001) whereas Consort exhibited the lowest (915 N/m2). At higher strains (3b ¼ 1.3) Trappe still exhibited high stress levels
Fig. 2. Logarithmic plots of stress-strain curves of doughs made from all the wheat types, during a) uniaxial extension with the Kieffer extensibility rig and b) biaxial extension (x Consort, B Equation, > Trappe, C Raffles, O Einstein, * Gulliver, þ Solstice, , NSLWW89, - Canadian blend, : German blend).
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
599
Table 2 Uniaxial and biaxial strain hardening index (y), uniaxial stress (s) and strain (3) at the point of fracture, and biaxial extensional viscosity of all the wheat types examined. Sample
Uniaxial extension y
Consort Equation Trappe Raffles Einstein Gulliver Solstice NSLWW89 Canadian blend German blend
1.6a,b 1.6a,b 1.5a 1.7b 1.6a,b 1.6a,b 1.6a,b 1.6a,b 1.9c 1.7b
Biaxial extension
3H fracture
sfracture (kN/m2)
0.4 mm/s
3.3 mm/s
0.4 mm/s
3.3 mm/s
2.1b 2.3b,c 1.6a 2.3b,c 2.5c 2.3c 2.4c 2.2b,c 2.4c 2.4c
2.4b 2.6b,c 2.0a 2.4b 2.7c 2.6b,c 2.7c 2.4b,c 2.5b,c 2.7c
12a,b 17a,b,c 7a 23b,c,d 26c,d 22b,c 24b,c,d 16a,b,c 36d 30c,d
30a,b 41b 17a 47b,c 62c,d 49b,c 46b,c 35a,b 69d 74d
y
Biaxial extensional viscosity kN s/m2
1.8b,c 1.7b,c 1.4a 1.3a 1.5a,b 1.5a,b 1.6a,b 1.4a,b 2.0c 1.6a,b
90.8a 111.1a,b 219.1d 171.6c 137.5b,c 111.0a,b 116.6a,b 140.4b,c 126.2 b, 113.2a,b
Different superscripts indicate significant difference (P < 0.001). The values are the results of three replicates.
followed by the Canadian blend and significant differences were found with Consort, Equation, Gulliver, Solstice and the German blend (P < 0.001). The dependency of stress to strain (strain hardening index, y) was calculated from the slopes of the curves shown in Fig. 2b, where the natural logarithm of stress is plotted against the strain, and are shown in Table 2. The Canadian blend had the highest strain hardening index (2.0) whereas Trappe and Raffles had the lowest (1.4 and 1.3 respectively). The biaxial extensional viscosity was calculated for varying strains at a strain rate 3b ¼ 0.01 s1. The results for the biaxial extensional viscosity for 3b ¼ 0.5 and 3b ¼ 0.01 s1 are shown in Table 2. Trappe had the highest biaxial extensional viscosity (222 kPa s) from all the samples tested (P < 0.001). On the other hand Consort had the lowest (92 kPa s) although the difference between most of the samples was not significant (Table 2). 3.6. Baking Most of the flours tested resulted in breads of medium to high volume (180e190 ml) except Trappe (140 ml), Raffles (168 ml), NSLWW89 (172 ml), and Einstein (176 ml) which gave loaves of lower volume (P < 0.001). 4. Discussion 4.1. Uniaxial extension methods The two uniaxial extension methods undertaken yielded similar extension curves which led to a successful discrimination of the varieties (e.g. typical curves for a biscuit variety were obtained for the consort flour with both methods, i.e. low resistance and high extensibility). Correlation analysis between the extensional properties of the two methods showed that the relationship was biased by the results of the Trappe variety. Trappe performed poorly in all the tests giving significantly different values at each measured property from the rest of the samples. This was mainly due to the poor condition of the grains (high grain moisture-unacceptable milling quality). The correlation between displacement (Kieffer) and extensibility (Extensograph) deteriorated after excluding Trappe (r ¼ 0.7). However a high correlation was observed between force (at all displacement speeds measured with the Kieffer rig), and maximum resistance (measured with the Extensograph) (r ¼ 0.9). This is in agreement with Grausgruber et al. (2002), Mann et al. (2005), and Suchy et al. (2000), who reported that the strength parameters were best correlated between the two instruments whereas extensibility parameters were not. However, the poorer correlation in extensibility can be explained when
considering the differences in sample size and also in strain rates applied with the two instruments. In the extensograph method, the sample is 150 g as opposed to approximately 0.8 g (0.4 g extended) in the Kieffer method. The size will influence the time needed for a crack to progress and fracture to take place (in larger samples energy dissipation is more extensive leading to slow progression of a crack). Differences in strain rate could also have an input in this low correlation observed. To achieve the same strain rate in Kieffer as in the Extensograph the displacement speed should be 410 mm/ min (Dunnewind et al., 2004). Moreover, uniaxial strain hardening index, maximum force, and stress at fracture at all three displacement speeds of the Kieffer rig were correlated with the protein content (r ¼ 0.7, 0.8 and 0.8 respectively). Similarly, maximum resistance to extension and energy measured with the Extensograph were correlated with protein content (r ¼ 0.8 and 0.9 respectively). Both tests were able to discriminate among the varieties. The Kieffer extensibility rig has the advantage of the small sample size requirement and therefore it is useful for early generation wheat selection.
4.2. Comparing uniaxial and biaxial extension Stress for a strain rate of 0.01 s1 was calculated for both biaxial and uniaxial extension and various strains, and the results for dough made from the Gulliver flour are shown in Fig. 3. Stress during uniaxial extension was higher at all strains measured and the difference between uniaxial and biaxial stress increased with increasing strain. The same trend was observed for all the samples. This finding is in agreement with Sliwinski et al. (2004a) who found stress increased more than proportionally with strain and that stress was always higher during uniaxial extension. According to the authors, this indicates that dough becomes specifically stronger in the direction in which it is stretched and therefore orientation processes are very important in dough rheology. The same findings were reported by Meissner (1997) for synthetic polymers. The differences in strain hardening index between uniaxial and biaxial extension depended on the sample tested. Strain hardening index was the same for the Solstice dough; it was higher in biaxial extension for Canadian, Consort and Equation; and it was lower in biaxial extension for Raffles, NSLWW89 and Gulliver. The reason for this phenomenon is not clear but when Sliwinski, van der Hoef, Kolster and van Vliet (2004c) tested doughs made from gluten in uniaxial and biaxial extension, they found that strain hardening index was always higher in uniaxial extension, which indicates that the gluten fraction is mostly what large deformation properties depend on but the other components of flour such as starch and lipids obviously influence the extensional properties.
600
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
Fig. 3. Stress plotted against strain for uniaxial and biaxial extension for doughs made from the Irish variety Gulliver (strain rate ¼ 0.01 s1). The results show the average of three replications.(- Uniaxial extension, : Biaxial extension).
4.3. Correlations with loaf volume Trappe was excluded from the correlation analysis because it was skewing the correlations, resulting in significant relationships between loaf volume and flour properties where no real relationship existed. No correlation was established between loaf volume and protein content for the remaining flour varieties. In addition, there was no correlation between loaf volume and Farinograph data. The extensibility at rupture as measured in the Extensograph was correlated with loaf volume (r ¼ 0.8) but a poorer correlation between loaf volume and the extensibility as measured with the Kieffer rig was observed (r ¼ 0.6). This could be explained due to the fact that the Extensograph produced more reproducible results in terms of extensibility (Coefficient of variation varied from 2 to 7%) than the Kieffer rig (CV varied from 2 to 12%). The relationship though between extensibility and loaf volume improved when the test displacement speed increased (r was equal to 0.2, 0.3 and 0.6 for displacement speeds of 0.4, 2 and 3.3 mm/s respectively). Sliwinski et al. (2004b) found that the correlations between stress at fracture and strain at fracture with loaf volume were dependant
on the displacement speed and that with increasing displacement speed the relationship between strain fracture and loaf volume deteriorated whereas the relationship with fracture stress improved. In the current study, even though the variety with the lowest strain and stress at fracture gave the lowest loaf volume, no strong correlation was found. The strain hardening index in uniaxial extension was the highest for the strongest variety (Canadian blend) and equal to 1.9 but no correlation was established between uniaxial strain hardening and loaf volume. Poor correlation was established with biaxial strain hardening and loaf volume (r ¼ 0.5). This was possibly due to the small variation observed in baking volume of the varieties. However the strongest and the weakest varieties were discriminated when the strain hardening was considered (the strain hardening index was equal to 2 for the Canadian blend and equal to 1.4 and 1.3 for Trappe and Raffles respectively). Biaxial extensional viscosity had a negative correlation with loaf volume which improved at lower strains (r was equal to 0.4, 0.4, 0.6, 0.7 and 0.7 for strains of 1.3, 1.15, 0.95, 0.75 and 0.5 respectively). Correlations were achieved between loaf volume and biaxial extensibility L (r ¼ 0.7), the swelling index
Fig. 4. Frequency plot for the distribution of loaf volume data (27 samples). Arrowed line indicates the 95% prediction interval for loaf volume for a given value uniaxial of extensibility.
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
(r ¼ 0.7), and the P/L ratio (r ¼ 0.7), all measurements from the Chopin Alveograph. However there was no correlation between W and loaf volume (r ¼ 0.3) or between P and loaf volume (r ¼ 0.5). Stepwise multiple regression analysis was carried out to establish the best predictors for loaf volume. The input parameters were: protein content, water absorption, development time, Extensograph parameters (Erupture, Rmax), parameters from the Kieffer rig (Max. Force, DisplacementFracture, StrainFracture, StressFracture, Strain hardeningUniaxial), Alveograph parameters (P, L, and G), and also the parameters measured during biaxial extension with the compression method (Strain hardeningBiaxial, biaxial extensional viscosity for 3b ¼ 0.01 s1, 3b ¼ 0.5). Two parameters were included in the output of the analysis and according to the model the best predictor was the extensibility at rupture as measured with the Extensograph (r2 ¼ 0.54). Including biaxial extensional viscosity measured at a strain of 0.5 for 0.01 s1 strain rate increased the r2 of the model to 0.7. A number of correlation coefficients have been reported in this study between rheological properties and loaf volume, as well as in the published literature. However the standard error and how it can affect the precision of the correlation is rarely reported. For example the data regarding the extensibility measured in the extensograph and the loaf volume data were used to calculate the 95% upper and lower prediction interval for loaf volume for a given value of extensibility (Mendenhall & Sincich, 1995). Fig. 4 shows the frequency plot for the distribution of loaf volume. The 95% prediction interval, indicated by the arrowed line, was 9 ml (overall prediction interval was 18 ml) and represented 74% of the total variation. Uniaxial extensibility was the best predictor for loaf volume but however from the analysis above, the predictive value is still limited in practice. The same analysis was carried out using the specific volume of the breads and the prediction interval was reduced to 67%. This shows that high correlation coefficients do not necessarily translate into good prediction models when the standard error associated with the correlation is considered. 5. Conclusions In summary, similar discrimination between the wheat samples was achieved with both uniaxial extension methods used (Brabender Extensograph and Kieffer extensibility rig). This indicated that the Kieffer extensibility rig is suitable for discriminating wheat flours and has the advantage of requiring a small quantity of dough. However, the extensibility as measured with the Kieffer rig did not correlate well with the extensibility from the Extensograph or with the baking volume. Significant correlations were established between loaf volume and the rheological properties determined in both uniaxial and biaxial extension. It was found that high loaf volumes can be achieved when biaxial and uniaxial extensibility is high and biaxial extensional viscosity is low. However when the standard error was considered the value of the correlation for prediction purposes was limited in practice. Nevertheless, the rheological tests provided useful information, and the results were valuable for screening flours for baking quality. Acknowledgements This project was funded under Teagasc Walsh Fellowship Scheme and was carried out with the support of the Irish Department of Agriculture Fisheries and Food under the Food Institutional Research Measure. The authors would like to thank Odlum Group Ltd., Dublin), Nickerson Ltd. (Lincolnshire, UK), Goldcrop (Dunleer, Ireland), SeedTech Ltd. (Waterford, Ireland), Bolgers (Ferns, Ireland) and Germinal Seeds Ltd. (Tipperary, Ireland) for providing the wheat samples. Also, Mr M. Leo (Odlum Group) for the measurement of starch damage.
601
References AACC. (1988). Approved methods of the American association of cereal chemists (8th ed.). Method 26-20, approved April 1961, revised December 1988. The Association, St. Paul, Minnesota, USA. Anderssen, R. S., Bekes, F., Gras, P. W., Nikolov, A., & Wood, J. T. (2004). Wheat-flour dough extensibility as a discriminator for wheat varieties. Journal of Cereal Science, 39, 195e203. Bloksma, A., & Bushuk, W. (1988). Rheology and chemistry of dough. In Y. Pomeranz (Ed.), Wheat: Chemistry and technology (3rd ed.). (pp. 131e217) St. Paul, Minnesota: American Association of Cereal Chemists. British Standards. (1999). British standards institution method No 4317-20. 1999. London, UK: British Standards Institution. de Bruijne, D. W., de Looff, J., van Eulem, A., & Carter, R. E. (1990). The rheological properties of bread dough and their relation to baking. In R. E. Carter (Ed.), Rheology of food, pharmaceutical and biological materials with general rheology (pp. 269e283). London: Elsevier. Bushuk, W. (1985). Rheology: theory and application to wheat flour dough. In H. Faridi (Ed.), Rheology of wheat products. St. Paul, Minnesota: American Association of Cereal Chemists. Collar, C., Santos, E., & Rosell, C. M. (2007). Assessment of the rheological profile of fibre-enriched bread doughs by response surface methodology. Journal of Food Engineering, 78, 820e826. Dobraszczyk, B. J., & Salmanowicz, B. P. (2008). Comparison of predictions of baking volume using large deformation rheological properties. Journal of Cereal Science, 47, 292e301. Dobraszczyk, B. J., Smewing, J., Albertini, M., Maesmans, G., & Schofield, J. D. (2003). Extensional rheology and stability of gas cell walls in bread doughs at elevated temperatures in relation to breadmaking performance. Cereal Chemistry, 80, 218e224. Dunnewind, B., Sliwinski, E. L., Grolle, K., & van Vliet, T. (2004). The Kieffer dough and gluten extensibility rig - an experimental evaluation. Journal of Texture Studies, 34, 537e560. Gras, P. W., Carpenter, H. C., & Anderssen, R. S. (2000). Modelling the developmental rheology of wheat-flour dough using extension tests. Journal of Cereal Science, 31, 1e13. Grausgruber, H., Schoggl, G., & Ruckenbauer, P. (2002). Investigations on the validity of the micro-extensigraph method to measure rheological properties of wheat doughs. European Food Research and Technology, 214, 79e82. ICC. (1968). ICC standard method No 107. Vienna: International Association for Cereal Science and Technology. ICC. (1976). ICC standard method No 110/1. Vienna: International Association for Cereal Science and Technology. ICC. (2007). ICC standard method No 172. Vienna: International Association for Cereal Science and Technology. Janssen, A. M., Vliet, T.v., & Vereijken, J. M. (1996). Fundamental and empirical rheological behaviour of wheat flour doughs and comparison with bread making performance. Journal of Cereal Science, 23, 43e54. Kokelaar, J. J., Vliet, T.v., & Prins, A. (1996). Strain hardening properties and extensibility of flour and gluten doughs in relation to breadmaking performance. Journal of Cereal Science, 24, 199e214. Ktenioudaki, A., Butler, F., & Gallagher, E. (2010a). The effect of different mixing processes on dough extensional rheology and baked attributes. Journal of the Science of Food and Agriculture, 90(12), 2098e2104. Ktenioudaki, A., Butler, F., & Gallagher, E. (2010b). Rheological properties and baking quality of wheat varieties from various geographical regions. Journal of Cereal Science, 51(3), 402e408. Mann, G., Allen, H., Morell, M. K., Nath, Z., Martin, P., Oliver, J., et al. (2005). Comparison of small-scale and large-scale extensibility of dough produced from wheat flour. Australian Journal of Agricultural Research, 56, 1387e1394. Meissner, J. (1997). Rheology and rheometry of viscoelastic liquids. In J. Windhab, & B. Wolf (Eds.), Proceedings of 1st international symposium on food rheology and structure (pp. 27e36). Zurich: Swiss Federal Institute of Technology. Mendenhall, W., & Sincich, T. (1995). Statistics for engineering and the sciences (4th edition).. New Jersey: Englewood Cliffs. Rouille, J., Valle, G., Lefebvre, J., Sliwinski, E., & Vliet, T. V. (2005). Shear and extensional properties of bread doughs affected by their minor components. Journal of Cereal Science, 42, 45e57. Sliwinski, E. L., Kolster, P., & Vliet, T. V. (2004a). Large-deformation properties of wheat dough in uni- and biaxial extension. Part I. Flour dough. Rheologica Acta, 43, 306e320. Sliwinski, E. L., Kolster, P., & Vliet, T. V. (2004b). On the relationship between largedeformation properties of wheat flour dough and baking quality. Journal of Cereal Science, 39, 231e245. Sliwinski, E. L., van der Hoef, F., Kolster, P., & Vliet, T. V. (2004c). Large-deformation properties of wheat dough in uni- and biaxial extension. Part II. Gluten dough. Rheologica Acta, 43, 321e332. Suchy, J., Lukow, O. M., & Ingelin, M. E. (2000). Dough microextensibility method using a 2-g mixograph and a texture analyzer. Cereal Chemistry, 77, 39e43. Tronsmo, K. M., Magnus, E. M., Baardseth, P., Schofield, J. D., Aamodt, A., & Faergestad, E. M. (2003). Comparison of small and large deformation rheological properties of wheat dough and gluten. Cereal Chemistry, 80, 587e595. Vliet, T.v., Janssen, A. M., Bloksma, A. H., & Walstra, P. (1992). Strain hardening of dough as a requirement for gas retention. Journal of Texture Studies, 23, 439e460.
Contents lists available at ScienceDirect
LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt
Dough characteristics of Irish wheat varieties I. Rheological properties and prediction of baking volume A. Ktenioudaki a, b, *, F. Butler a, E. Gallagher b a b
UCD School of Agriculture, Food Science & Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland Teagasc, Ashtown Food Research Centre, Ashtown, Dublin 15, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 May 2009 Received in revised form 5 November 2010 Accepted 9 November 2010
Eight wheat varieties suitable for cultivation in Ireland were examined for the rheological properties and baking quality. Large deformation extensional rheology was used employing the Extensograph, the Kieffer extensibility rig, the Alveograph and biaxial extension by uniaxial compression. Similar discrimination between the wheat samples was achieved with both uniaxial extension methods used. Stress during uniaxial extension was higher at all strains measured than biaxial extension and the difference between uniaxial and biaxial stress increased with increasing strain. Also, differences existed in the strain hardening index between uniaxial and biaxial extension which depended on the sample tested. Significant correlations were established between loaf volume and the rheological properties determined in both uniaxial and biaxial extension. It was found that high loaf volumes can be achieved when biaxial and uniaxial extensibility is high and biaxial extensional viscosity is low. The importance of considering the standard error for validating the value of the correlation for prediction purposes was also demonstrated. When the standard error of the correlation was considered, the value of the correlation for prediction purposes was limited in practice. Nevertheless, the rheological tests provided useful information, and the results were valuable for screening flours for baking quality. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Dough rheology Biaxial extension Uniaxial extension Baking
1. Introduction Breadmaking is a complex process that involves many physicochemical and structural transformations which lead to the production of an aerated baked product from basic ingredients such as flour, water, yeast and salt. Generally, the breadmaking process is divided into three main steps: mixing, fermentation (proving), and baking. During all these stages, deformations of different magnitudes take place and rheology is involved in every step of the process. During mixing, extreme deformations take place above the point of rupture as a result of forces exerted on the dough from the mixer, whereas during proving and baking the deformations are much smaller and are due to the difference in pressure between the gas cells and the atmosphere (Bushuk, 1985). The importance of rheology in breadmaking and the impact of dough rheological properties on the quality of the baked product have led to the development of numerous rheological tests, employed both in research and in the cereal industry. Extensional
* Corresponding author. Teagasc, Ashtown Food Research Centre, Ashtown, Dublin 15, Ireland. Tel.: þ353 8 51469941; fax: þ353 1 8059550. E-mail address: [email protected] (A. Ktenioudaki). 0023-6438/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2010.11.014
flow measurements such as uniaxial and biaxial extension are considered more appropriate as extensional flow is present in many operations such as mixing, sheeting, moulding, and bubble expansion. According to Gras, Carpenter, and Anderssen (2000), the mixing action that occurs in a Mixograph can be interpreted as a series of extension tests. The extension to rupture occurring during an extension test corresponds to the elongate - rupture that occurs during mixing of the elongate-rupture-relax oscillations that form the bandwidth of the Mixograph test. Therefore the bandwidth recorded from the Mixograph can be used to better determine the evolving rheology of the dough during mixing. Ideally, every rheological test that aims at predicting the behaviour of the material should be taking place at relevant deformation conditions to what the material experiences during processing (Dobraszczyk, Smewing, Albertini, Maesmans, & Schofield, 2003). According to van Vliet, Janssen, Bloksma, and Walstra (1992), the dough around a growing cell is tangentially extended in two directions parallel to the surface of the gas cell and compressed radially perpendicular to the surface of the gas cell due to the radially acting compression force caused by excess pressure inside the gas cells. Biaxial extension at large strains and low strain rates is considered to be the relevant deformation that takes place around a growing cell and it has been used to study the rheological
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
properties of dough (Janssen, van Vliet and Vereijken, 1996; Kokelaar, van Vliet, and Prins 1996; Rouille, Valle, Lefebvre, Sliwinski, and van Vliet., 2005; Sliwinski, Kolster, and van Vliet, 2004a). These tests make possible the calculation of a rheological property known as strain hardening, which van Vliet et al. (1992), have proposed it provides the dough films of expanding gas cells with greater stability and protects them against premature rupture. Therefore, the extent of strain hardening will determine the breadmaking potential. Many researchers have studied the strain hardening phenomenon and have demonstrated that it relates to the baking potential of flour (Dobraszczyk & Salmanowicz, 2008; Sliwinski, Kolster, and van Vliet, 2004b). Numerous tests exist to measure the uniaxial extensional properties of dough. The test methods include the Simon extensometer, the Brabender Extensograph, and the Stable Micro Systems Kieffer dough and gluten extensibility rig. The Extensograph is an empirical test that has been used for years to perform uniaxial extension on dough and the measured properties have been correlated with baked volumes. However a few disadvantages exist such as: the force and the extension are not expressed in Newton and strain respectively, the amount of dough deformed increases with extension, and therefore the conversion of the Extensograph curves to stress and strain curves is restricted (Bloksma & Bushuk, 1988). An apparatus similar to the Extensograph was developed the Kieffer dough and gluten extensibility rig. With this apparatus only a small amount of dough is required (about 0.8 g), the force is measured in Newton and the speed of the test can be adjusted. This way more relevant strain rates can be applied and the results can be expressed in stress and strain data (Dunnewind, Sliwinski, Grolle, & van Vliet, 2004). Grausgruber, Schoggl, and Ruckenbauer (2002) compared the Kieffer extensibility rig with the Extensograph and found significant relationships between the parameters obtained from the two methods. Many researchers have used uniaxial extension tests to study the rheological properties of doughs (de Bruijne, de Looff, van Eulem, & Carter, 1990; Collar, Santos, & Rosell, 2007; Dobraszczyk & Salmanowicz, 2008; Sliwinski et al. 2004b; Suchy, Lukow, & Ingelin, 2000; Tronsmo et al., 2003) and good correlations with baking quality have been reported (Sliwinski et al., 2004b; Suchy et al., 2000; Tronsmo et al., 2003). Sliwinski et al. (2004b) studied the large deformation and fracture properties of flour in uniaxial extension using the Kieffer rig. They found that high loaf volumes were achieved from those varieties that exhibited intermediate stress levels at large strains. They also found that the relationship between loaf volume and stress and strain fracture is rate dependant. Tronsmo et al. (2003) studied the rheological properties at small and large deformations of four wheat cultivars. Two varieties had the High Molecular Weight glutenin subunits (HMWgs) pair 5 þ 10 and therefore resulted in good breadmaking quality and the other two had the HMWgs pair 2 þ 12 that is known to result in poor performance during baking. They found that maximum resistance to extension (Rmax) measured with the Kieffer rig and the biaxial strain hardening as measured with the Dobraszczyk/Roberts inflation rig could discriminate between these varieties. Anderssen, Bekes, Gras, Nikolov, and Wood (2004) studied the rheological properties of eight wheat varieties using a uniaxial micro-extension test. They indicated the need to relate rheological properties such as strength and extensibility to the glutenin composition of the flours in order to achieve better understanding of their relation to baking performance. The eight flours were divided into three categories (weak, intermediate, strong) with distinct differences in the HMW glutenin subunit allelic composition. Distinct morphological differences in the extensograms were found leading to the conclusion that the alleles significantly affect
595
the rheology of the flours. It was also shown that the measurement of maximum resistance to extension (Rmax) clearly separates the flours depending on their HMW glutenin subunit allelic composition. They also calculated the extensibility at the point of Rmax (ERmax)as well as the extensibility at the point of rupture(Erupture) and found that parameters (ERmax and Erupture ERmax) other than the ones traditionally used were important in interpreting rheologically the extensograms and could significantly differentiate between the varieties. The objective of this paper was to conduct a comprehensive study of the rheological properties of Irish wheat varieties and to identify the rheological properties that would successfully discriminate between varieties and predict their baking quality. Eight varieties suitable for growing in the Irish climate were chosen, as well as a Canadian and a German blend of varieties (for comparative purposes). Four large deformation extensional rheological tests were employed to provide empirical as well as fundamental measurements both in uniaxial and biaxial extension. 2. Materials & methods 2.1. Materials Eight wheat varieties suitable for cultivation in Ireland were selected. The varieties were: Gulliver, Solstice, NSLWW89, Einstein, Consort, Equation, Trappe, and Raffles. These are considered to be breadmaking varieties, with the exception of Consort and Equation, which are considered to be suitable for biscuits and feed respectively, and were included in the study to achieve a wider variation in wheat properties. Four varieties were supplied from a UK breeding company (Einstein, Gulliver, Solstice, and NSLWW89), and the rest were provided by Irish suppliers (Goldcrop, SeedTech, and Germinal seeds) from various geographical locations. The seeds were from the 2007 harvest and approximately 50 kg of grains were received for each variety. A blend of Canadian varieties and also a blend of German varieties were included in the study, obtained from an Irish milling company (Odlum Ltd., Dublin). The varieties were cleaned using a sample cleaner (SLN3, A/s Rationel Kornservice, Denmark) and conditioned to 16% moisture content prior to milling. The samples were milled using a Bühler mill (Bühler, Switzerland) to a milling degree of 70% following the AACC Bühler method (AACC, 1988). Analysis for moisture content (ICC standard Method No. 110/1, 1976), a-amylase activity (ICC standard method No. 107, ICC, 1968), protein content using a Leco protein analyzer (Leco FP-428 Nitrogen Analyzer, Leco Corporation, St. Joseph, MI, U.S.A), and for starch damage using an amperometric method (SDmatic, Chopin, France) (ICC standard method No 172, ICC, 2007) took place. Duplicate measurements were carried out for the chemical analysis and the results were averaged. The mixing properties were investigated using the Brabender Farinograph (Brabender OHG, Duisberg, Germany) following the Farinograph British Standard method No. 4317-20:1999 (British standards, 1999) (one test per wheat type). 2.2. Dough preparation Dough samples were prepared in a Farinograph mixing bowl using 300 g flour, 6 g salt (Pure Dried Vacuum salt, Ineos Enterprises, UK), and 3 g of emulsified bread fat (Irish Bakels Ltd., Dublin, Ireland). The amount of water added was as per the Farinograph water absorption value. The mixing times were based on the development time plus 1 min. A new batch of dough was mixed for each rheological test. All the experiments were replicated three times.
596
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
2.3. Uniaxial extension (Extensograph) A Brabender Extensograph (Brabender OHG, Duisberg, Germany) was used to perform uniaxial extension on the doughs. The method as described in Ktenioudaki, Butler, and Gallagher (2010a) was followed. Duplicate measurements were made for each replicate and the results were averaged. 2.4. Uniaxial extension (Kieffer dough and gluten extensibility rig)
et al. (2010a). Prepared dough was placed in the Alveograph mixer only to extrude the sample pieces. The results of each replicate were the average of five repeats of the same batch. The parameters recorded included the biaxial extensibility of dough (L), the maximum overpressure (P, height of the curve), the swelling index (G) (amount of air to inflate the bubble), the deformation energy (W) (area under the curve), and the configuration ratio (P/L). 2.6. Biaxial extension (compression method)
Uniaxial extension using the Kieffer dough and gluten extensibility rig (Stable Micro Systems, Surrey, UK) was carried out similar to the method described by Dunnewind et al. (2004). A texture analyzer (TA-XT2i, Stable Micro Systems, Surrey, UK) fitted with the Kieffer extensibility rig and equipped with a 5 kg load cell was used. The samples were prepared following the method described in the literature accompanying the Kieffer extensibility rig (Stable Microsystems, UK). A piece of dough (25 g) was placed to rest in a proofer (Koma SDCC-1P/W, Koma Koeltechnische Industrie B.V., The Netherlands) with controlled temperature (30 C) and relative humidity (85%) for 20 min. The sample was then rolled by hand into a cylindrical shape, placed in the lubricated Teflon mould and compressed with the lubricated top Teflon plate. The sample was allowed to rest for another 45 min in the proofer at 30 C and 80% relative humidity. Before the start of the test the sample was clamped between the plates of the Kieffer rig. The maximum force in tension and the displacement were recorded. Three different displacement speeds were used: 0.4, 2.0 and 3.3 mm/s. Three dough samples were used for each speed and the results were averaged. According to Dunnewind et al. (2004), clamping of the sample results in a certain degree of sagging of the dough and therefore the actual extension of the dough piece starts somewhat above the surface of the lower plate. Corrections were applied to account for this by measuring the distance the hook travelled from a predetermined starting point to the point where it touched the dough. The difference between this distance and the known distance from the starting point to the top of the bottom plate gave the height of sagging. Adding this to the point of the top surface of the bottom plate, the point of when the actual extension started was determined. The point when the hook touched the dough was determined as the point which the measured force deviated from zero. The fundamental rheological parameters Hency strain 3H, strain rate 3, stress s, and the apparent extensional viscosity hE were calculated as described by Dunnewind et al. (2004). Finally, similar steps were followed to calculate the strain hardening index as those for the biaxial extension described by Kokelaar et al. (1996). 2.5. Biaxial extension (Alveograph) Biaxial extension was performed using an Alveograph (Chopin S.A., Villeneuve la Garenne, France) as described in Ktenioudaki
The method as described in Ktenioudaki, Butler, and Gallagher (2010b) was followed using a texture analyzer (TA-XT2i, Stable Micro Systems, Surrey, UK). Three compression speeds were used (0.2, 1.0 and 2.0 mm/s) and the dough samples were compressed to 1.5 mm height. Nine samples (three for every speed) were prepared in total for each replicate from the same batch of dough and the results were averaged. 2.7. Baking The baking method used was as described in Ktenioudaki et al. (2010b). Briefly, pup loaves (65 g dough pieces in 100 55 35 mm tins) were produced following a straight dough baking procedure. The water added and the mixing time, were the same as used for the doughs for the rheological tests and the samples were prepared in the Farinograph mixing bowl. Six loaves were produced per batch. The loaves were allowed to cool for 2 h after baking and placed into plastic bags. After 24 h, the loaves were tested for loaf volume using a volume measurer (BVM-L370, TexVol Instruments, Sweden). 2.8. Statistical analysis One way analysis of variance was carried out using Minitab (Minitab version 15.1.1.0., Minitab Ltd., UK) to determine significant differences in the measured properties. Pearson’s correlation analysis between selected parameters was carried out using Minitab (Minitab version 15.1.1.0., Minitab Ltd., UK). Stepwise multiple regression was carried out using SAS software (SAS 9.1.3., SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Flour characteristics The data from the flour analysis are presented in Table 1. The protein content varied from 6.6 g/100g (Trappe) to 12.9 g/100g (Canadian blend). The typical range for Irish wheat varieties is between 8 and 10 g/100g. Consort and Trappe appeared to have medium to high a-amylase activity whereas the rest of the varieties
Table 1 Physical and chemical characteristics of flours tested. Variety
Protein g/100g (at 14g/100g mc)
Falling No (s)
Moisture g/100 g
Starch damage
Water absorption %
Development time (min)
Stability (min)
Degree of softening (BU)
Consort Equation Trappe Raffles Einstein Gulliver Solstice NSLWW89 Canadian blend German blend
7.4 9.4 6.6 9.3 9.2 10.0 9.4 9.3 12.9 12.5
253 327 266 454 433 471 447 425 428 456
13.6 13.6 14.0 13.9 13.8 13.4 13.4 12.8 13.4 13.3
9.31 27.51 30.03 39.10 36.60 40.43 41.10 46.45 36.97 37.52
50.0 58.5 58.0 58.0 58.0 62.5 62.7 64.7 65.5 63.0
1.5 2.0 2.0 2.5 2.0 3.0 2.0 2.0 6.0 2.5
1.0 2.0 2.0 2.5 3.5 4.0 2.0 1.5 9.0 5.0
180 140 110 100 90 90 100 150 50 70
The results of protein content, moisture and Falling No are the mean values of two repeats.
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
would benefit from the addition of a-amylase. The low water absorption of the Consort flour was expected as it is considered to be a biscuit variety whereas the absorptions of the remaining samples were in the normal range for bread varieties. The high development time for the Canadian blend is typical of strong Canadian varieties.
597
samples tested (58 mm). Significant differences were observed in terms of the maximum resistance to extension measured. The Canadian blend had the highest value (615 BU). The biscuit variety Consort together with Equation and Solstice gave the lowest values varying between 185 and 292 BU whereas the values for the rest of the varieties ranged between 305 and 453 BU (P < 0.001).
3.2. Extensograph 3.3. Kieffer dough and gluten extensibility rig Fig. 1a shows typical extensograms for all of the wheat varieties examined. The plot for the Canadian blend was clearly separated from the rest of the samples as the one with the highest energy (largest area under the curve). Consort flour gave a typical extensogram of a biscuit variety with high extensibility and low maximum resistance to extension. Significant differences (P < 0.001) were observed among the samples for both the extensibility and the maximum resistance to extension results. The Canadian blend, the German blend, Einstein, Gulliver and Solstice produced doughs with the highest extensibilities (between 183 and 200 mm). However, only the Canadian blend was accompanied by equally high resistance to extension. The rest of the varieties had similar extensibilities (varying between 153 and 173 mm) with the exception of Trappe which had the lowest extensibility of all the
Force-displacement curves similar to the ones from the Extensograph were obtained and are shown in Fig. 1b for the displacement speed of 3.3 mm/s. The Canadian blend stands out again as the strongest flour. Flour from Trappe variety had the lowest extensibility and the smallest area under the curve, in a similar way as presented in the extensogram in Fig. 1a. Both force and the displacement increased with increasing the displacement speed. Similar trends were observed by Sliwinski et al. (2004a) and Dunnewind et al. (2004). At the 3.3 mm/s displacement speed, Equation, Einstein, Gulliver, Solstice and the German blend exhibited the highest extensibility (from 102 to 113 mm) whereas the Canadian blend exhibited the highest maximum force recorded (0.27 N). The lowest extensibility was recorded for the Irish variety
Fig. 1. Typical extensograms of the wheat varieties during uniaxial extension with a) the Brabender Extensograph b) the Kieffer extensibility rig at 3.3 mm/s displacement speed.
598
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
Trappe (52 mm) whereas the lowest force for the biscuit variety Consort was 0.12 N. A similar pattern of results were obtained at the other two displacement speeds (results not shown). Stress and strain were calculated from the force and displacement data, accounting for the changing dimensions of the sample and assuming a constant volume. At all three speeds, stress increased with increasing strain and the stress was higher at higher displacement speeds. A constant strain rate was chosen (3 ¼ 0.01 s1) and stressestrain curves were produced. Differences in stress between varieties were higher at higher strains. At a Hencky strain of 1.20, the stress levels are similar for the varieties and only the Canadian blend exhibited significantly higher stress from the Solstice flour (P < 0.05). At a higher strain of 1.75, the Canadian blend exhibited significantly higher stress levels (19.8 kN/m2) than the rest of the samples (P < 0.001). Similarly to the calculations for the biaxial extension test, the natural logarithm of the stress was plotted against the Hencky strain (Fig. 2a) and the slope of these curves corresponds to the uniaxial strain hardening index. The values of strain hardening index (y) are presented in Table 2 together with the calculated stress and strain at the point of fracture. Dough from the Canadian blend had the highest strain hardening index (1.9) whereas the Irish variety Trappe had the lowest value (1.5). Stress and strain at the point of fracture increased with increasing deformation rate. Fracture stress was higher for the Canadian and the German blend whereas it was the lowest for Trappe for all three displacement speeds. Less variation was observed in the strains at
the point of fracture where Trappe had the lowest value (P < 0.001) at all three speeds. 3.4. Biaxial extension (Alveograph) Doughs made from the Trappe variety had the highest P value (P < 0.001), followed by Raffles (67 and 54 mm respectively), and the lowest extensibilities L (26 and 81 mm respectively). The biaxial extensibility L varied between 26 mm (Trappe) to 131 mm (Solstice). The P/L ratio (2.6) for Trappe was unacceptably high for baking flours, whereas there was no difference between the other flours (mean P/L value was 0.4). Significant differences were noted (P < 0.001), in terms of the W values, with the Canadian blend having the highest value (215 106 J) and Trappe having the lowest value (69 106 J). Finally, G values varied significantly between 11 ml (Trappe) to 26 ml (Gulliver) (Data not shown). 3.5. Biaxial extension The same analysis as proposed by Kokelaar et al. (1996) was carried out to calculate stress-strain curves from the forcedisplacement measurements. As stress increased with strain, the differences between the samples became more obvious. At low strains (0.1e0.5) Trappe exhibited the highest stress (2.23 kN/m2, P < 0.001) whereas Consort exhibited the lowest (915 N/m2). At higher strains (3b ¼ 1.3) Trappe still exhibited high stress levels
Fig. 2. Logarithmic plots of stress-strain curves of doughs made from all the wheat types, during a) uniaxial extension with the Kieffer extensibility rig and b) biaxial extension (x Consort, B Equation, > Trappe, C Raffles, O Einstein, * Gulliver, þ Solstice, , NSLWW89, - Canadian blend, : German blend).
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
599
Table 2 Uniaxial and biaxial strain hardening index (y), uniaxial stress (s) and strain (3) at the point of fracture, and biaxial extensional viscosity of all the wheat types examined. Sample
Uniaxial extension y
Consort Equation Trappe Raffles Einstein Gulliver Solstice NSLWW89 Canadian blend German blend
1.6a,b 1.6a,b 1.5a 1.7b 1.6a,b 1.6a,b 1.6a,b 1.6a,b 1.9c 1.7b
Biaxial extension
3H fracture
sfracture (kN/m2)
0.4 mm/s
3.3 mm/s
0.4 mm/s
3.3 mm/s
2.1b 2.3b,c 1.6a 2.3b,c 2.5c 2.3c 2.4c 2.2b,c 2.4c 2.4c
2.4b 2.6b,c 2.0a 2.4b 2.7c 2.6b,c 2.7c 2.4b,c 2.5b,c 2.7c
12a,b 17a,b,c 7a 23b,c,d 26c,d 22b,c 24b,c,d 16a,b,c 36d 30c,d
30a,b 41b 17a 47b,c 62c,d 49b,c 46b,c 35a,b 69d 74d
y
Biaxial extensional viscosity kN s/m2
1.8b,c 1.7b,c 1.4a 1.3a 1.5a,b 1.5a,b 1.6a,b 1.4a,b 2.0c 1.6a,b
90.8a 111.1a,b 219.1d 171.6c 137.5b,c 111.0a,b 116.6a,b 140.4b,c 126.2 b, 113.2a,b
Different superscripts indicate significant difference (P < 0.001). The values are the results of three replicates.
followed by the Canadian blend and significant differences were found with Consort, Equation, Gulliver, Solstice and the German blend (P < 0.001). The dependency of stress to strain (strain hardening index, y) was calculated from the slopes of the curves shown in Fig. 2b, where the natural logarithm of stress is plotted against the strain, and are shown in Table 2. The Canadian blend had the highest strain hardening index (2.0) whereas Trappe and Raffles had the lowest (1.4 and 1.3 respectively). The biaxial extensional viscosity was calculated for varying strains at a strain rate 3b ¼ 0.01 s1. The results for the biaxial extensional viscosity for 3b ¼ 0.5 and 3b ¼ 0.01 s1 are shown in Table 2. Trappe had the highest biaxial extensional viscosity (222 kPa s) from all the samples tested (P < 0.001). On the other hand Consort had the lowest (92 kPa s) although the difference between most of the samples was not significant (Table 2). 3.6. Baking Most of the flours tested resulted in breads of medium to high volume (180e190 ml) except Trappe (140 ml), Raffles (168 ml), NSLWW89 (172 ml), and Einstein (176 ml) which gave loaves of lower volume (P < 0.001). 4. Discussion 4.1. Uniaxial extension methods The two uniaxial extension methods undertaken yielded similar extension curves which led to a successful discrimination of the varieties (e.g. typical curves for a biscuit variety were obtained for the consort flour with both methods, i.e. low resistance and high extensibility). Correlation analysis between the extensional properties of the two methods showed that the relationship was biased by the results of the Trappe variety. Trappe performed poorly in all the tests giving significantly different values at each measured property from the rest of the samples. This was mainly due to the poor condition of the grains (high grain moisture-unacceptable milling quality). The correlation between displacement (Kieffer) and extensibility (Extensograph) deteriorated after excluding Trappe (r ¼ 0.7). However a high correlation was observed between force (at all displacement speeds measured with the Kieffer rig), and maximum resistance (measured with the Extensograph) (r ¼ 0.9). This is in agreement with Grausgruber et al. (2002), Mann et al. (2005), and Suchy et al. (2000), who reported that the strength parameters were best correlated between the two instruments whereas extensibility parameters were not. However, the poorer correlation in extensibility can be explained when
considering the differences in sample size and also in strain rates applied with the two instruments. In the extensograph method, the sample is 150 g as opposed to approximately 0.8 g (0.4 g extended) in the Kieffer method. The size will influence the time needed for a crack to progress and fracture to take place (in larger samples energy dissipation is more extensive leading to slow progression of a crack). Differences in strain rate could also have an input in this low correlation observed. To achieve the same strain rate in Kieffer as in the Extensograph the displacement speed should be 410 mm/ min (Dunnewind et al., 2004). Moreover, uniaxial strain hardening index, maximum force, and stress at fracture at all three displacement speeds of the Kieffer rig were correlated with the protein content (r ¼ 0.7, 0.8 and 0.8 respectively). Similarly, maximum resistance to extension and energy measured with the Extensograph were correlated with protein content (r ¼ 0.8 and 0.9 respectively). Both tests were able to discriminate among the varieties. The Kieffer extensibility rig has the advantage of the small sample size requirement and therefore it is useful for early generation wheat selection.
4.2. Comparing uniaxial and biaxial extension Stress for a strain rate of 0.01 s1 was calculated for both biaxial and uniaxial extension and various strains, and the results for dough made from the Gulliver flour are shown in Fig. 3. Stress during uniaxial extension was higher at all strains measured and the difference between uniaxial and biaxial stress increased with increasing strain. The same trend was observed for all the samples. This finding is in agreement with Sliwinski et al. (2004a) who found stress increased more than proportionally with strain and that stress was always higher during uniaxial extension. According to the authors, this indicates that dough becomes specifically stronger in the direction in which it is stretched and therefore orientation processes are very important in dough rheology. The same findings were reported by Meissner (1997) for synthetic polymers. The differences in strain hardening index between uniaxial and biaxial extension depended on the sample tested. Strain hardening index was the same for the Solstice dough; it was higher in biaxial extension for Canadian, Consort and Equation; and it was lower in biaxial extension for Raffles, NSLWW89 and Gulliver. The reason for this phenomenon is not clear but when Sliwinski, van der Hoef, Kolster and van Vliet (2004c) tested doughs made from gluten in uniaxial and biaxial extension, they found that strain hardening index was always higher in uniaxial extension, which indicates that the gluten fraction is mostly what large deformation properties depend on but the other components of flour such as starch and lipids obviously influence the extensional properties.
600
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
Fig. 3. Stress plotted against strain for uniaxial and biaxial extension for doughs made from the Irish variety Gulliver (strain rate ¼ 0.01 s1). The results show the average of three replications.(- Uniaxial extension, : Biaxial extension).
4.3. Correlations with loaf volume Trappe was excluded from the correlation analysis because it was skewing the correlations, resulting in significant relationships between loaf volume and flour properties where no real relationship existed. No correlation was established between loaf volume and protein content for the remaining flour varieties. In addition, there was no correlation between loaf volume and Farinograph data. The extensibility at rupture as measured in the Extensograph was correlated with loaf volume (r ¼ 0.8) but a poorer correlation between loaf volume and the extensibility as measured with the Kieffer rig was observed (r ¼ 0.6). This could be explained due to the fact that the Extensograph produced more reproducible results in terms of extensibility (Coefficient of variation varied from 2 to 7%) than the Kieffer rig (CV varied from 2 to 12%). The relationship though between extensibility and loaf volume improved when the test displacement speed increased (r was equal to 0.2, 0.3 and 0.6 for displacement speeds of 0.4, 2 and 3.3 mm/s respectively). Sliwinski et al. (2004b) found that the correlations between stress at fracture and strain at fracture with loaf volume were dependant
on the displacement speed and that with increasing displacement speed the relationship between strain fracture and loaf volume deteriorated whereas the relationship with fracture stress improved. In the current study, even though the variety with the lowest strain and stress at fracture gave the lowest loaf volume, no strong correlation was found. The strain hardening index in uniaxial extension was the highest for the strongest variety (Canadian blend) and equal to 1.9 but no correlation was established between uniaxial strain hardening and loaf volume. Poor correlation was established with biaxial strain hardening and loaf volume (r ¼ 0.5). This was possibly due to the small variation observed in baking volume of the varieties. However the strongest and the weakest varieties were discriminated when the strain hardening was considered (the strain hardening index was equal to 2 for the Canadian blend and equal to 1.4 and 1.3 for Trappe and Raffles respectively). Biaxial extensional viscosity had a negative correlation with loaf volume which improved at lower strains (r was equal to 0.4, 0.4, 0.6, 0.7 and 0.7 for strains of 1.3, 1.15, 0.95, 0.75 and 0.5 respectively). Correlations were achieved between loaf volume and biaxial extensibility L (r ¼ 0.7), the swelling index
Fig. 4. Frequency plot for the distribution of loaf volume data (27 samples). Arrowed line indicates the 95% prediction interval for loaf volume for a given value uniaxial of extensibility.
A. Ktenioudaki et al. / LWT - Food Science and Technology 44 (2011) 594e601
(r ¼ 0.7), and the P/L ratio (r ¼ 0.7), all measurements from the Chopin Alveograph. However there was no correlation between W and loaf volume (r ¼ 0.3) or between P and loaf volume (r ¼ 0.5). Stepwise multiple regression analysis was carried out to establish the best predictors for loaf volume. The input parameters were: protein content, water absorption, development time, Extensograph parameters (Erupture, Rmax), parameters from the Kieffer rig (Max. Force, DisplacementFracture, StrainFracture, StressFracture, Strain hardeningUniaxial), Alveograph parameters (P, L, and G), and also the parameters measured during biaxial extension with the compression method (Strain hardeningBiaxial, biaxial extensional viscosity for 3b ¼ 0.01 s1, 3b ¼ 0.5). Two parameters were included in the output of the analysis and according to the model the best predictor was the extensibility at rupture as measured with the Extensograph (r2 ¼ 0.54). Including biaxial extensional viscosity measured at a strain of 0.5 for 0.01 s1 strain rate increased the r2 of the model to 0.7. A number of correlation coefficients have been reported in this study between rheological properties and loaf volume, as well as in the published literature. However the standard error and how it can affect the precision of the correlation is rarely reported. For example the data regarding the extensibility measured in the extensograph and the loaf volume data were used to calculate the 95% upper and lower prediction interval for loaf volume for a given value of extensibility (Mendenhall & Sincich, 1995). Fig. 4 shows the frequency plot for the distribution of loaf volume. The 95% prediction interval, indicated by the arrowed line, was 9 ml (overall prediction interval was 18 ml) and represented 74% of the total variation. Uniaxial extensibility was the best predictor for loaf volume but however from the analysis above, the predictive value is still limited in practice. The same analysis was carried out using the specific volume of the breads and the prediction interval was reduced to 67%. This shows that high correlation coefficients do not necessarily translate into good prediction models when the standard error associated with the correlation is considered. 5. Conclusions In summary, similar discrimination between the wheat samples was achieved with both uniaxial extension methods used (Brabender Extensograph and Kieffer extensibility rig). This indicated that the Kieffer extensibility rig is suitable for discriminating wheat flours and has the advantage of requiring a small quantity of dough. However, the extensibility as measured with the Kieffer rig did not correlate well with the extensibility from the Extensograph or with the baking volume. Significant correlations were established between loaf volume and the rheological properties determined in both uniaxial and biaxial extension. It was found that high loaf volumes can be achieved when biaxial and uniaxial extensibility is high and biaxial extensional viscosity is low. However when the standard error was considered the value of the correlation for prediction purposes was limited in practice. Nevertheless, the rheological tests provided useful information, and the results were valuable for screening flours for baking quality. Acknowledgements This project was funded under Teagasc Walsh Fellowship Scheme and was carried out with the support of the Irish Department of Agriculture Fisheries and Food under the Food Institutional Research Measure. The authors would like to thank Odlum Group Ltd., Dublin), Nickerson Ltd. (Lincolnshire, UK), Goldcrop (Dunleer, Ireland), SeedTech Ltd. (Waterford, Ireland), Bolgers (Ferns, Ireland) and Germinal Seeds Ltd. (Tipperary, Ireland) for providing the wheat samples. Also, Mr M. Leo (Odlum Group) for the measurement of starch damage.
601
References AACC. (1988). Approved methods of the American association of cereal chemists (8th ed.). Method 26-20, approved April 1961, revised December 1988. The Association, St. Paul, Minnesota, USA. Anderssen, R. S., Bekes, F., Gras, P. W., Nikolov, A., & Wood, J. T. (2004). Wheat-flour dough extensibility as a discriminator for wheat varieties. Journal of Cereal Science, 39, 195e203. Bloksma, A., & Bushuk, W. (1988). Rheology and chemistry of dough. In Y. Pomeranz (Ed.), Wheat: Chemistry and technology (3rd ed.). (pp. 131e217) St. Paul, Minnesota: American Association of Cereal Chemists. British Standards. (1999). British standards institution method No 4317-20. 1999. London, UK: British Standards Institution. de Bruijne, D. W., de Looff, J., van Eulem, A., & Carter, R. E. (1990). The rheological properties of bread dough and their relation to baking. In R. E. Carter (Ed.), Rheology of food, pharmaceutical and biological materials with general rheology (pp. 269e283). London: Elsevier. Bushuk, W. (1985). Rheology: theory and application to wheat flour dough. In H. Faridi (Ed.), Rheology of wheat products. St. Paul, Minnesota: American Association of Cereal Chemists. Collar, C., Santos, E., & Rosell, C. M. (2007). Assessment of the rheological profile of fibre-enriched bread doughs by response surface methodology. Journal of Food Engineering, 78, 820e826. Dobraszczyk, B. J., & Salmanowicz, B. P. (2008). Comparison of predictions of baking volume using large deformation rheological properties. Journal of Cereal Science, 47, 292e301. Dobraszczyk, B. J., Smewing, J., Albertini, M., Maesmans, G., & Schofield, J. D. (2003). Extensional rheology and stability of gas cell walls in bread doughs at elevated temperatures in relation to breadmaking performance. Cereal Chemistry, 80, 218e224. Dunnewind, B., Sliwinski, E. L., Grolle, K., & van Vliet, T. (2004). The Kieffer dough and gluten extensibility rig - an experimental evaluation. Journal of Texture Studies, 34, 537e560. Gras, P. W., Carpenter, H. C., & Anderssen, R. S. (2000). Modelling the developmental rheology of wheat-flour dough using extension tests. Journal of Cereal Science, 31, 1e13. Grausgruber, H., Schoggl, G., & Ruckenbauer, P. (2002). Investigations on the validity of the micro-extensigraph method to measure rheological properties of wheat doughs. European Food Research and Technology, 214, 79e82. ICC. (1968). ICC standard method No 107. Vienna: International Association for Cereal Science and Technology. ICC. (1976). ICC standard method No 110/1. Vienna: International Association for Cereal Science and Technology. ICC. (2007). ICC standard method No 172. Vienna: International Association for Cereal Science and Technology. Janssen, A. M., Vliet, T.v., & Vereijken, J. M. (1996). Fundamental and empirical rheological behaviour of wheat flour doughs and comparison with bread making performance. Journal of Cereal Science, 23, 43e54. Kokelaar, J. J., Vliet, T.v., & Prins, A. (1996). Strain hardening properties and extensibility of flour and gluten doughs in relation to breadmaking performance. Journal of Cereal Science, 24, 199e214. Ktenioudaki, A., Butler, F., & Gallagher, E. (2010a). The effect of different mixing processes on dough extensional rheology and baked attributes. Journal of the Science of Food and Agriculture, 90(12), 2098e2104. Ktenioudaki, A., Butler, F., & Gallagher, E. (2010b). Rheological properties and baking quality of wheat varieties from various geographical regions. Journal of Cereal Science, 51(3), 402e408. Mann, G., Allen, H., Morell, M. K., Nath, Z., Martin, P., Oliver, J., et al. (2005). Comparison of small-scale and large-scale extensibility of dough produced from wheat flour. Australian Journal of Agricultural Research, 56, 1387e1394. Meissner, J. (1997). Rheology and rheometry of viscoelastic liquids. In J. Windhab, & B. Wolf (Eds.), Proceedings of 1st international symposium on food rheology and structure (pp. 27e36). Zurich: Swiss Federal Institute of Technology. Mendenhall, W., & Sincich, T. (1995). Statistics for engineering and the sciences (4th edition).. New Jersey: Englewood Cliffs. Rouille, J., Valle, G., Lefebvre, J., Sliwinski, E., & Vliet, T. V. (2005). Shear and extensional properties of bread doughs affected by their minor components. Journal of Cereal Science, 42, 45e57. Sliwinski, E. L., Kolster, P., & Vliet, T. V. (2004a). Large-deformation properties of wheat dough in uni- and biaxial extension. Part I. Flour dough. Rheologica Acta, 43, 306e320. Sliwinski, E. L., Kolster, P., & Vliet, T. V. (2004b). On the relationship between largedeformation properties of wheat flour dough and baking quality. Journal of Cereal Science, 39, 231e245. Sliwinski, E. L., van der Hoef, F., Kolster, P., & Vliet, T. V. (2004c). Large-deformation properties of wheat dough in uni- and biaxial extension. Part II. Gluten dough. Rheologica Acta, 43, 321e332. Suchy, J., Lukow, O. M., & Ingelin, M. E. (2000). Dough microextensibility method using a 2-g mixograph and a texture analyzer. Cereal Chemistry, 77, 39e43. Tronsmo, K. M., Magnus, E. M., Baardseth, P., Schofield, J. D., Aamodt, A., & Faergestad, E. M. (2003). Comparison of small and large deformation rheological properties of wheat dough and gluten. Cereal Chemistry, 80, 587e595. Vliet, T.v., Janssen, A. M., Bloksma, A. H., & Walstra, P. (1992). Strain hardening of dough as a requirement for gas retention. Journal of Texture Studies, 23, 439e460.